(introduction...)

June 1998

In response to requests from people living with HIV/AIDS, health professionals and governments, WHO held an informal consultation on the implications of antiretroviral treatments in April 1997, in collaboration with UNAIDS. Participants included clinicians, researchers, PLHA, national AIDS programme managers, representatives of ministries of health, the pharmaceutical industry, NGOs, donors, UNAIDS, other UN agencies, and various WHO programmes. A report of the consultation* was produced in September 1997. The report summarised the essential points raised during the consultation and derived from discussions, presentations and background papers.

The modules are addressed mainly to "middle level" decision makers at academic and training institutions, major hospitals and district hospitals. However, in all modules there are issues which may be of concern to technical decision makers in government ministries such as health, planning or finance.

The modules form a complete set which together address the major issues relating to the use and provision of ARVs but each is designed to "stand alone". Depending on further developments in the treatments and experience with their use in different settings, there may be requests for guidance in other areas. Other modules may then be added to the set.

WHO/ASD/98.1
UNAIDS/98.7
Distr: General

Module 1. Introduction to Antiretroviral Treatments

Acknowledgements

The World Health Organization and UNAIDS would like to thank Dr Rachel Baggaley, Dr Susan Fernyak, Ms Alison Martin Katz and Dr Eric van Praag, Office of HIV/AIDS and Sexually Transmitted Diseases, WHO, and Dr Joseph Perri, Policy, Strategy and Research, UNAIDS, for preparing the text of this module.

©World Health Organization and Joint United Nations Programme on HIV/AIDS, 1998

This document is not a formal publication of the World Health Organization (WHO) and Joint United Nations Programme on HIV/AIDS (UNAIDS), and all rights are reserved by them. The document may, however, be freely reviewed, abstracted, reproduced, or translated in part, but not for sale or for use in conjunction with commercial purposes. It may also be reproduced in full by non-commercial entities for information or for educational purposes with prior permission from WHO and UNAIDS.

For authorization to translate or reproduce the work in full, and for any use by commercial entities, applications and enquiries should be addressed to the Office of HIV/AIDS and Sexually Transmitted Diseases, World Health Organization, 1211 Geneva 27, Switzerland, which will be glad to provide the latest information on any changes made to the text, plans for new editions, and the reprints and translations that are already available.

The mention of specific companies or of certain manufacturers’ products does not imply that they are endorsed or recommended by either WHO or UNAIDS in preference to others of a similar nature that are not mentioned. Errors and omissions excepted, the names of proprietary products are distinguished by initial capital letters.

The views expressed in these documents by named authors are solely the responsibility of those authors.

HIV/AIDS and ARVs in developing countries: the reality

John found out he was HIV positive in 1989 when he was offered a place at university in the USA. A requirement before taking up his place was a medical examination which included an HIV test. He was devastated when the positive result came back as he not only lost the opportunity of studying abroad but he also had to face the knowledge that he had a terminal illness with very little hope of any treatment.

He attended counselling and after some months felt able to use his skills to help others learn about HIV. Over the following five years he visited numerous schools, workplaces and clinics to talk about HIV and use his personal experience to educate others and destigmatise HIV. He remained well until 1994 when he got TB. This was successfully treated and he was able to go back to work. His health deteriorated again in late 1995 and he had a bout of pneumonia and recurrent abscesses. He continued to be active in the PLHA network and was always very well informed about the latest development in HIV treatments, particularly ARVs. He even arranged a meeting for PLHA groups where ARVs were extensively discussed as “hope for people with HIV”. Not a single person at the meeting was on ARVs or had any likelihood of ever being so. John’s health has continued to deteriorate and he is unable to work. Despite having given more than seven years of his life to educating people about HIV he is now unable to buy food let alone any medication.

Stephen, a medical doctor, developed HIV related symptoms in January 1992. By March 1992 he was getting weaker, requiring frequent hospital admissions and by July he had to give up work. In February 1996 he was admitted in critical condition and remained unconscious for eight days. He was eventually discharged after three weeks reasonably well. It was then that he expressed the desire to try some of the antiretroviral drugs. He was counselled about the expense and possible side effects and decided to discuss it with the family. Initially his family discouraged him from spending such a large amount of money when had had a big family to look after. His decision was much swayed by the media stories about the protease inhibitors which were run during and after the Vancouver conference.

In August 1996 he started on indinavir and ZDV only because he could not afford triple therapy. Indinavir had to be obtained from the USA through friends, costing him about $600 for a month’s supply. He managed to acquire ZDV locally at a cost of $200 for a month’s supply. By the time he started the combination therapy he was a very sick man. No basic tests such as CD4 cell counts or viral measurements had been done. After 2 months treatment he had a bad spell where he was taking the medicines irregularly either because he was vomiting or because he was taking too many other drugs. In January he went two weeks without indinavir because his supply did not arrive in time. He was getting it through people who were travelling to Uganda to minimise on courier expenses. When the medicine did arrive he resumed taking it but now he was getting weaker. He developed septicaemia and died on February 9th 1997 with the January bottle of indinavir hardly used.

Maureen is a 28 year old single mother with two daughters, aged 3 and 5. Her husband, a police officer, died last year and his family took virtually all their possessions after the burial. “Property grabbing” is common. Maureen is a teacher and until recently was teaching the final year of primary school. With her take-home pay she was just about able to pay for the rent of her 2 room compound home and feed herself, the children and her younger sister who had come to live with her to look after the children whilst she was at work. Her house is not electrified and water is collected from a communal pump more than 100 yards from the house. She had an HIV test following her husband’s death as she suspected that he had died from AIDS. He had widespread KS and eventually died of meningitis. Maureen’s main concern is the future of her children. Whilst she was working, life was tolerable but now that she is too weak to work she cannot support her family. Her sister has been forced to sell sex merely to get enough money to feed them all. Maureen is aware of the “wonder drugs” that are available for people with HIV. She has read about them in the newspaper. Her concern is how can she feed her children and she is anxious about who will look after them when she dies.

Anthony, a relatively successful banker came to the company clinic after a loan to build a house for his wife and new son had been refused on medical grounds. After much heartache, he decided to be tested for HIV and found that he was seropositive. His wife was also tested and was seronegative. He wanted the doctor to get hold of antiretrovirals whatever the cost. As he earned approximately $1000 a month he could put his entire salary into ARVs or build a house for his wife and new son. He decided to build the house and start ARVs only when he became immunocompromised. He remained well for 3 years and built his house.

By late 1995 his CD4 count had fallen to 160 and he started on ZDV and ddI. His CD4 remained stable until March 1997 when he had a severe attack of herpes zoster. He was prescribed high dose acyclovir and he made a quick and complete recovery. But by June 1997 his CD4 had fallen further. He badgered his doctor to change his regimen to include a protease inhibitor. Through an American mail order company the doctor managed to obtain indinavir at 50% the local cost. He started on triple therapy and by October his CD4 count had risen to 380 and he was remarkably well.

His “success” must in part be due to his wife. She kept him to his regimen religiously. She dealt with her own stresses in resigning herself to only one child in the face of extreme pressure from relatives (who do not know her husband is HIV-positive) to have more children. How long his good health will continue is anyone’s guess.

Foreword

1996 and 1997 were marked by fresh optimism about finding a solution to the devastating problem that HIV/AIDS has come to represent - in terms of serious social and economic disruption and appalling human suffering. This new hope came from the development of a growing number of new approaches that promise a longer life, of higher quality, supported medically and socially, for people living with HIV infection. The development of new antiretrovirals (ARVs), particularly the combination therapies of three ARVs including one protease inhibitor, has been greeted with enormous enthusiasm.

WHO, in collaboration with UNAIDS, responded to these important developments with an Informal Consultation on the Implications of Antiretroviral Treatments, held in April 1997 in Geneva. As an immediate follow up to the Consultation, a set of guidance modules has been prepared for decision makers at health planning units of government ministries, major hospitals, and academic and training institutions, on major issues relating to antiretroviral treatments.

The majority of people living with HIV/AIDS in developing countries have no access to these costly treatments - they will perhaps obtain a couple of courses of antibiotics to treat opportunistic infections when these arise and if they are lucky, pain killers in the last stages of disease. Many people living with HIV/AIDS learn about the new treatments through the media, are filled with hope and prepared to mobilise whatever financial resources are needed to obtain the drugs as soon as possible. Medical requirements to ensure that the treatment can be prescribed safely and effectively may be scarcely considered. How to pay for the drugs on a long term basis and the impact on the family of devoting all resources to buying drugs are likewise secondary considerations, perhaps to be dealt with “tomorrow”.

UNAIDS and WHO are committed to increasing access to new technologies which have been shown to be effective in preventing and treating HIV/AIDS and improving length and quality of life, to all those in need. We share the general enthusiasm about the remarkable advances in ARV treatments. But this enthusiasm is tempered by two major concerns: one, is their long term benefit - and only clinical experience over several years at least will provide evidence that a reliable, long term treatment has been found. The other is the accessibility of these expensive and difficult treatments to the vast majority of the world’s HIV infected people, namely those living in the developing countries where resources for health care are often very limited.

New compounds are being synthesised and developed at a tremendous rate. It is our hope that cheaper and easier treatments will become available in the near future. Meanwhile, UNAIDS and WHO will regularly assess progress and disseminate information on the state-of-the-art of the new treatments, and promote and facilitate access to the drugs to all in need, through advocacy with national governments and the pharmaceutical industry, and through their international networks of researchers, health professionals, and NGOs. The guidance provided in these modules will be regularly updated to keep up with the latest advances.

Brief history

Since the human immunodeficiency virus (HIV) was first identified as the cause of AIDS, enormous research efforts have concentrated on identifying and developing compounds to suppress its replication.

In 1987, zidovudine (ZDV, formerly known as AZT) was approved by the US Food and Drug Administration. In the years that followed, four other drugs of the same family were introduced. The principal problems with these drugs, including ZDV, is their limited potency, their toxicity and their time-limited benefit - largely due to the development of resistance.

Large multicentre clinical trials then showed that double therapy with two of these drugs was superior to monotherapy in terms of disease progression and survival. Greater and more sustained decreases in plasma levels of HIV-1 RNA were achieved with double combinations.

Significantly larger reductions in viral load were achieved by adding a new class of agents, the protease inhibitors, which became available in early 1996. Over the past two years, the number of antiretroviral agents available has expanded substantially and trials of the new antiretroviral treatments, particularly the triple therapies using combinations of drugs from different classes have shown impressive short term results decreasing both morbidity and mortality and offering real hope for people living with HIV/AIDS of a longer and a better life. Consequently, combination therapy has rapidly become the gold standard of care in antiretroviral therapy.

With regimens of three drugs, over three quarters of HIV-infected patients had levels of virus in plasma suppressed to below the level at which it could be detected and this persisted throughout follow up for as long as one year in ARV naive patients with good adherence.

It is now recognised that the sustainability of the therapeutic response to ARVs depends on the level of suppression of viral replication. Regimens which only partially suppress replication will ultimately promote the emergence of resistance. The aim of therapy therefore is the full suppression of replication to below detectable levels by the most sensitive assays available. In practical terms, this implies today the use of triple therapy usually including a powerful protease inhibitor.

The latest research shows that even with sustained suppression of HIV replication, there are still reservoirs of HIV in patients under treatment. Furthermore, the virus recovered may not be latent but still replicating at a slow rate. Viable virus has been retrieved from patients who have been well suppressed on tritherapy for up to two years. Patients who are well controlled with undetectable viral loads show massive rebound viremia when for some reason (such as an accident) they suddenly stop taking ARVs.

Recent studies are reporting high levels of resistance in relatively short periods of time, even with good adherence to the regimen. Resistance develops more readily in patients who have already received ARVs and in those with low CD4 counts. The most favourable results have been obtained from studies in patients with optimal adherence. It is likely that treatment failure in normal clinical practice will be higher.

In recognition of these limitations, researchers are already seeking alternatives to triple therapy - quadruple therapy and new classes of drugs. Approaches to HIV treatment are in rapid evolution as work continues on disease pathogenesis and viral dynamics.

Critical issues in ARV treatments for HIV infection still require clarification: when to start and change treatment, and with which combinations, what drug sequence strategy to adopt. With the rapid progress in drug development, treatment guidelines are likely to have to be frequently updated.(For current recommendations on the use of antiretroviral treatments see Module 4)

The use of zidovudine to prevent mother to child transmission of HIV was shown as long ago as 1994 to be effective (reducing transmission by 50-70%). Monotherapy for the treatment of HIV infection is now regarded as obsolete because of serious problems of resistance.

ARVs are also used in post-exposure prophylaxis for health care workers (See Module 7). Guidelines have been developed in industrialised countries all of which recommend ARV therapy following accidental exposure to blood which may be contaminated by HIV. It should be noted that data on efficacy is limited and that the average risk of seroconversion following exposure, even without treatment, is very low. Minimising risks of exposure through strict observance of universal precautions remains the priority.

Treatment for the few: but a concern for all

The major challenges in antiretroviral treatments are access, correct and supervised use, adherence and the development of resistance. The new treatments are not available to the vast majority of the world’s HIV infected people, most of whom live in the developing countries. Triple therapy is very expensive and the related services required to ensure their safe and effective use are complex. Adherence requires a strict individual protocol and reliable psychosocial and material support.

Despite this, ARVs are present in most poor countries of the world, but accessible to a very small, and in the main, wealthy, minority. Efforts to obtain the drugs, often in an irregular and haphazard fashion, may absorb the entire resources of the family. In these situations, problems of incorrect and unsafe use, unreliable supply, and a black market in both good quality and counterfeit drugs are likely to appear. This has serious public health implications because of the almost certain development and spread of resistant strains. It is therefore urgent for policy makers, ministries of health, health professionals and NGOs to address all aspects of ARV treatments including access, irrespective of the wealth and circumstances of their country.

Over and above the public health concern, UNAIDS, WHO and its member states are committed to the principle of universal access to effective treatments for all in need. All efforts therefore will be made to assess the currently available treatments, the desirability and feasibility of introducing them into national health systems in different settings, and likely future research developments; to increase access to these treatments, to press for further research into cheaper and easier regimes, to negotiate with industry to devise mechanisms to reduce prices; and to maintain and reinforce prevention and care activities for all those affected by HIV/AIDS.

WHO Informal Consultation on the Implications of Antiretroviral Treatments

In response to requests from people living with HIV/AIDS, health professionals and governments, WHO held an informal consultation on the implications of antiretroviral treatments in April 1997, in collaboration with UNAIDS. Participants included clinicians, researchers, PLHA, national AIDS programme managers, representatives of ministries of health, the pharmaceutical industry, NGOs, donors, UNAIDS, other UN agencies, and various WHO programmes. A report of the consultation* was produced in September 1997 and French and Spanish versions will be available in early 1998. The report summarised the essential points raised during the consultation and derived from discussions, presentations and background papers. The essential points raised during the consultation are summarised below; the full report is available from Distribution and Sales, WHO, 20 Avenue Appia, 1211 Geneva 27 (fax: 41 22 791 4857; email: publications@who.ch) or from UNAIDS Information Centre, 20 Avenue Appia, 1211 Geneva 27 (fax: 41 22 791 4187; email unaids@unaids.org).

Technical guidance modules

Audience

The modules are addressed mainly to “middle level” decision makers at academic and training institutions, major hospitals and district hospitals. However, in all modules there are issues which may be of concern to technical decision makers in government ministries such as health, planning or finance.

Topics

Key content areas for the modules were identified from the issues raised and discussed at the consultation. The nine guidance modules which make up the complete set are:

1. Introduction to antiretroviral treatments
2. Introducing antiretroviral treatments into health systems: economic considerations
3. Antiretroviral treatments: planning and integration into health services
4. Safe and effective use of antiretrovirals
5. Laboratory requirements for the safe and effective use of antiretrovirals
6. The use of antiretroviral drugs to reduce mother to child transmission of HIV
7. Treatments following exposure to the human immunodeficiency virus
8. Antiretrovirals: regulation, distribution and control
9. Ethical and societal issues relating to antiretroviral treatments

The modules form a complete set which together address the major issues relating to the use and provision of ARVs but each is designed to “stand alone”. Depending on further developments in the treatments and experience with their use in different settings, there may be requests for guidance in other areas. Other modules may then be added to the set.

* The Implications of Antiretroviral Treatments: Informal Consultation, Geneva, April 1997. Edited by Eric van Praag, Susan Fernyak and Alison Martin Katz

Summary of the Report of the WHO Informal Consultation on the Implications of Antiretroviral Treatments

Aims:

· To review the findings on ARVs - their efficacy, side effects, problems of resistance , adherence and cost benefits;

· To discuss minimum requirements for their safe and effective use;

· To examine the clinical, social, financial and ethical implications of providing ARVs, especially in low and middle income countries;

· To make preliminary recommendations.

Currently available antiretroviral drugs

Current antiretroviral agents for HIV/AIDS can be divided into two major classes of drugs: reverse transcriptase inhibitors (RTIs) and protease inhibitors (PIs). RTIs are further divided into nucleoside (NRTI) and non-nucleoside (NNRTI) subclasses. A third class, integrase inhibitors, is under development, as are drugs with other mechanisms of action (e.g. hydroxyurea, chemokines and interleukin 2).

Eleven antiretroviral agents are currently available in different parts of the world. Registration of drugs differs from country to country. Ten of these agents have been shown to be useful in the treatment of HIV/AIDS by slowing the progression of the disease and prolonging survival. Table 1 lists the currently available agents, by brand name, class and estimated market cost, based on US pricing in June 1997.

Table 1

*brand names may vary between countries
** ranges reflect dosing variations

Best treatment option available

Evidence has shown that a combination of ARVs including at least one PI, is the most effective treatment available to suppress the replication of HIV. This is reflected clinically, in decreased incidence of opportunistic infections, decreased hospitalisations and ability to return to normal functions of daily living. Serologically, it is reflected in decreased viral loads (often to undetectable levels) and increases in the number of CD4 cells.

Cautious optimism

As yet it is not known how long these benefits will last. Long term clinical outcomes (spanning several years of observation) have not been demonstrated; resistance may occur even with triple therapy, and “sanctuary” sites may exist where drugs cannot suppress viral replication even in individuals under treatment. In addition, when treatment is interrupted, viral rebound may occur leading to rapid clinical deterioration. Finally, in only a proportion of patients on triple therapy, with available techniques, is viral load reduced to undetectable levels.

The combination therapies are very expensive (US$ 1000-1500 per month in May 1998) and require a rigorous and difficult daily regimen in order to avoid the emergence of drug resistance. Fifteen to twenty tablets are to be taken over the day. Unpleasant side effects are common, drug interactions have to be considered, and clinical and laboratory monitoring is required to detect adverse reactions.

Individual needs and public health considerations

During the consultation there was a shared understanding that ARVs can improve the life prospects of PLHA and that combination therapy represents the best treatment option available currently. Participants recognised the primacy of the need to respond to the individual suffering of PLHA while acknowledging concerns about long term sustainability and safety of ARVs, and responsibilities for the health needs of the general population.

Guiding principles

· Universal access to care and treatment, a principle which WHO promotes as a human right, is the ultimate aim. The fact that access might not be achieved immediately and universally does not preclude progressive introduction of ARVs, recognising that they are already present in countries. The stark contrast between industrialised countries where patients are usually receiving these treatments through public and private insurance schemes and developing countries where the majority of those requiring treatment cannot access these drugs, is ethically unacceptable.

· As an overarching ethical principle, it must be recognised that combination ARV therapy with at least 3 ARV currently represents the most advanced therapy of choice for people living with HIV/AIDS. There is therefore an obligation for governments to make every effort to increase access to these treatments within the context of national health programmes which address competing health care needs. Governments must ensure non-discrimination and equity in access among PLHA and among those living with other life-threatening conditions requiring comparably costly and complex therapies.

· The provision of ARVs must not divert resources from HIV prevention activities (e.g., control of STDs, promotion of safer sex including condoms, and ensuring a safe blood supply), the treatment of opportunistic infections, and research to develop preventive tools, particularly vaccines and microbicides. Neither should it divert resources from other essential public health programmes.

· ARV treatments must be supported by a functioning health and social system which ensures adequate diagnosis and treatment of opportunistic infections, appropriate pain management, and correct use of and adherence to ARVs to avoid the emergence of drug resistance and transmission of resistant strains.

· Adherence is a required component of effective treatment, not a precondition for access to treatment. Patients should not be excluded a priori from ARV treatment because they belong to groups regarded as unlikely to adhere (e.g. injecting drug users, the homeless and the poor). Health providers’ roles and responsibilities in ensuring adherence must also be recognised.

· The supply, distribution and provision of ARVs require national quality assurance systems to be in place. Governments have a regulatory and supervisory role in these areas

· Cost is a major issue. Collaboration between and within countries to enable those who are most in need to obtain treatment, should be promoted and mechanisms/strategies put in place to effect such solidarity. It is ever more urgent for governments to reconsider the level of the health budget in relation to the budget for other sectors.

· Commitment to continue basic science research into drugs and clinical outcomes is essential as none of the currently available ARVs are ideal and a number of promising drugs are under development.

· The pharmaceutical industry is a key partner in health care, and has social responsibilities as do all civil organisational and institutional bodies. Their collaboration is essential.

· Prevention and care are two sides of the same coin. All prevention efforts reduce or limit the final number of infected people needing care. Whenever care is provided, there is a golden opportunity to provide or reinforce prevention education.

Minimum requirements

· Availability of reliable, inexpensive tests to diagnose HIV infection and access to sites providing voluntary and confidential counselling and testing.

· Adequate management of opportunistic infections including availability of affordable drugs for their treatment and prophylaxis.

· Laboratory facilities to monitor adverse reactions including liver function test, complete blood count, amylase, and electrolytes.

· Appropriate training for clinicians and nurses in the correct use of ARVs.

Support of a social network to help patients adhere to the regimen.

· Strengthening of health and social services in a continuum of care, from the home, through community health centres, to district/city hospitals, with flexible and timely referral and co-ordination procedures; provision of material and psychosocial support for those taking ARVs at various levels of the continuum,

· Reliable, long-term and regular supply of drugs at the health centre level.

· Joint decision making between physician and patient on all aspects of ARV treatment, including the decision to begin ARVs.

Recommendations

· Long term affordability based on scenarios which assume full coverage and total adherence is not achievable at present in most countries. A useful approach might be to ensure access at a limited number of centres of excellence where clinical, laboratory and social support can be guaranteed. Allocating different therapies to subsets of people with HIV-related conditions (e.g. ARVs for prevention of MTCT and PEP), taking into account the realities of different settings, may represent a rational strategy offering affordable starting points.

· Because of the high daily turnover of virus production, resistant strains develop very easily. Strict adherence to the drug regimen is therefore essential. Emergence of resistance should be carefully monitored through a network of collaborating laboratories. WHO/UNAIDS could establish such a co-ordination mechanism.

· In order to minimise unregulated supply and distribution leading to irregular provision, unsupervised use and the emergence of resistance, there need to be effective national drug regulatory authorities.

· The diagnosis and treatment of opportunistic infections remains an essential component of the clinical management of HIV/AIDS patients. As in the recent revision of the WHO essential drug list in December 1997, governments should also consider adding drugs for opportunistic infections, other HIV-related conditions and pain management to their Essential Drugs Lists.

· International technical agencies and programmes should collect and widely disseminate information on all aspects of ARV treatments including the drugs themselves, rational approaches to problems of adherence and resistance, prescription guidelines, health system requirements, pricing policies and financing options.

· Clinical studies in all countries should follow a basic set of ethical guidelines. All clinical trials should aim to provide meaningful estimates of the gain in quality-adjusted life years, reflecting not only changes in survival but also improvements in the quality of life.

· The currently available nucleic acid amplification assays (DNA and RNA) used to measure viral load are not equally sensitive and/or equally efficient in amplifying all known HIV subtypes. The tests need to be improved to allow accurate detection of the HIV variants most prevalent in Africa, Latin America and Asia. The standardisation of these assays is also an urgent priority.

· Viral load measurement is highly desirable to monitor efficacy of ARV treatments in asymptomatic patients.

· Operational research on how best to implement ARV treatments in settings where resources are limited and adherence beset with difficulties, is needed in low and middle income countries.

· Training of health professionals in the correct use of ARVs is essential if the decision to introduce them is made. Clinical guidelines should be developed for professional use. Training should be undertaken to reach all care providers as a team.

Tremendous optimism has been generated by the recent development of new antiretrovirals, particularly the triple combination therapies including one protease inhibitor, which promise a longer and better life for people living with HIV/AIDS (PLHA). In response to requests for the treatments from PLHA, and for policy and technical guidance from health professionals and governments, WHO, in collaboration with UNAIDS, held an Informal Consultation on the Implications of Antiretroviral Treatments with particular reference to low and middle income countries, in April 1997.

Participants at the consultation, ministries of health, health professionals, PLHA, donors and NGOs working in HIV/AIDS have all called for technical and policy guidance for health planners and policy makers, and decision makers in training institutions, central and district hospitals on antiretroviral treatments, as an immediate follow up to the consultation.

In collaboration with UNAIDS, WHO has produced a set of nine guidance modules on the following aspects of antiretroviral treatments:

1. Introduction to antiretroviral treatments
2. Introducing antiretroviral treatments into national health systems: economic considerations
3. ARV treatments: planning and integration into health services
4. Safe and effective use of antiretrovirals
5. Laboratory requirements for the safe and effective use of antiretrovirals
6. The use of antiretroviral drugs to reduce mother to child transmission of HIV
7. Treatments following exposure to HIV
8. Antiretrovirals: regulation, distribution and control
9. Ethical and societal issues relating to antiretroviral treatments

Module 2. Introducing Antiretroviral Treatments into Health Systems: Economic Considerations

Acknowledgements

The World Health Organization and UNAIDS would like to thank Dr Katherine Floyd and Dr Charles Gilks, Liverpool School of Tropical Medicine, UK, for preparing the text of this module;

Dr Thierry Mertens, Office of HIV/AIDS and Sexually Transmitted Diseases, WHO, Dr German Velasquez, Action Programme on Essential Drugs, WHO, and Dr Joe Kutzin, Division of Analysis, Research and Assesment, WHO, for their most thoughtful reviews of the draft text; and

Dr Rachel Baggaley, Dr Susan Fernyak, Ms Alison Martin Katz, Ms Marguerite Nguyen, and Dr Eric van Praag, Office of HIV/AIDS and Sexually Transmitted Diseases, WHO, and Dr Joseph Perri, Policy, Strategy and Research, UNAIDS, for coordinating the project.

©World Health Organization and Joint United Nations Programme on HIV/AIDS, 1998

This document is not a formal publication of the World Health Organization (WHO) and Joint United Nations Programme on HIV/AIDS (UNAIDS), and all rights are reserved by them. The document may, however, be freely reviewed, abstracted, reproduced, or translated in part, but not for sale or for use in conjunction with commercial purposes. It may also be reproduced in full by non-commercial entities for information or for educational purposes with prior permission from WHO and UNAIDS.

For authorization to translate or reproduce the work in full, and for any use by commercial entities, applications and enquiries should be addressed to the Office of HIV/AIDS and Sexually Transmitted Diseases, World Health Organization, 1211 Geneva 27, Switzerland, which will be glad to provide the latest information on any changes made to the text, plans for new editions, and the reprints and translations that are already available.

The mention of specific companies or of certain manufacturers’ products does not imply that they are endorsed or recommended by either WHO or UNAIDS in preference to others of a similar nature that are not mentioned. Errors and omissions excepted, the names of proprietary products are distinguished by initial capital letters.

The views expressed in these documents by named authors are solely the responsibility of those authors.

Introduction

This module is structured in seven main sections, which are:

Affordability (1), in which cost, GNP, health expenditure and HIV prevalence data from various parts of the world are presented to illustrate how it is possible to assess whether ARVs are affordable at national, regional, district or institutional level, and in which suggestions for how to make similar assessments for individuals/ households are made. Ways in which ARVs may be made more affordable are also discussed;

Cost-effectiveness (2), in which existing evidence that can be used to assess whether or not ARVs represent a good investment of public resources is provided, problems with the application of cost-effectiveness analysis to ARV treatments are described, and reasons why governments and social insurance schemes may decide that financing ARVs is worthwhile even if they do not appear to pass the test of “cost-effectiveness” are explained. Ways in which the difficulties with cost-effectiveness analysis can be reduced are also suggested;

Availability of necessary human/physical resources (3), in which the importance of assessing whether the inputs needed for provision of ARV treatment are actually available or mobilisable is discussed;

Equity (4), which addresses the issue of who is most likely to benefit from ARV treatment, using some preliminary evidence from Cd’Ivoire and Mexico. This emphasises that the likely distribution of resources is an important issue to consider with ARV treatments, and has particular implications for financing policy;

Financing (5), in which the main options for funding ARV treatments are assessed, including public funding, social and private insurance, user fees/other out-of pocket payments, donor agencies and community financing. This section draws on the analysis of issues of affordability, cost-effectiveness and equity in (1), (2) and (4);

Provision (6), in which the potential role of different health care providers in delivery of ARV treatments is analysed; and

ARVs, economics and health sector reform (HSR) (7), in which the interactions among various aspects of HSR, economic issues, and the introduction of ARVs are considered. Although some of these are touched upon in earlier sections, the importance of HSR initiatives merits a separate discussion.

Figure 1 summarises how these issues relate to decision-making on ARV treatments. However, sometimes, the answers to questions 1,2 and 3 are not simply yes or no. ARVs are likely to be affordable to some, but not to all. The bullet points illustrate the most important considerations in each case.

When using this module, it is important to bear in mind that there are large differences between countries, regions, communities and individuals in terms of the AIDS epidemic and in terms of many socio-economic factors. The module provides a range of examples from Latin America, sub-Saharan Africa, the Caribbean and Asia which highlight these differences. Evidence is limited at present and at this stage it may only be possible to propose general principles rather than specific guidance.

1. Affordability

Affordability is a continuum, not an absolute yes or no. It can be assessed from a number of perspectives, including that of:

· a country
· a region
· a district
· an individual/household
· an institution, such as a national insurance scheme or a social security organisation.

This section presents both cost data and a methodology for using these cost data, which in combination can be applied to assess the affordability of ARV treatments at each of these levels. It also discusses whether ARV treatments could result in some cost-savings, and suggests how ARV treatments may be made more affordable in future.

1.1 The cost of ARV therapy

The costs of providing ARV therapy fall into two major categories:

(a) drug costs; and
(b) costs associated with safe and effective use.

Table 1: Anti-retroviral drug costs (US$) at market prices in 1997

¹World wide web page of AIDSRx, 1997

The costs of the drugs are shown in Table 1. These are largely derived from USA sources, so it is important to note that under market conditions the costs of drugs may be significantly higher in developing countries. Double combination therapies range in costs from US$ 4836 to US$ 9276 per year, while the triple combination therapies range from US$ 7944 to US$ 11916, increasing to up to US$ 17,580 if ritonavir is added. Country-specific data on these costs needs to be assembled wherever possible.

Data concerning the other important costs associated with providing therapy are shown in Table 2. These include: HIV tests to establish whether someone is HIV-infected and hence eligible for therapy; pre- and post-test counselling; regular out-patient visits to monitor patients for side-effects and to issue supplies of drugs; laboratory tests such as CD4 counts, complete blood counts, viral loads, and chemistry panels to monitor patient health status; and out-patient visits/hospitalisations associated with adverse drug effects. Again, country-specific data on these costs should ideally be assembled since they will allow more accurate assessments of affordability to be made.

Table 2: Non-drug costs associated with antiretroviral therapy (US$)

²Source recent literature and worldwide web on AIDSRx

It is worth emphasising that in addition to the costs shown in Table 2, some developing countries’ laboratory services may have to be strengthened considerably (including substantial investment in quality assurance) if they are to be capable of providing the diagnostic and monitoring support necessary for provision of ARV therapy (see Module 5). Developing such capacity may have large costs attached to it. It is also possible that there will be costs associated with either basic or specialised in-service staff training.

1.2 A methodology for assessing whether or not ARVs are affordable

There is no “right” way to assess affordability. Two approaches are described here. The first applies to affordability at a level beyond the individual or household, and is thus relevant to assessing affordability at national, regional, district or insurance institution level. The second applies to affordability at the level of the individual/household, and is thus relevant to both financing policy on ARVs and considerations of equity (see also sections 4 and 5). If ARVs are not affordable at an individual/household level, public funding or insurance coverage will be necessary if ARVs are to be made widely available, rather than accessible to a small, relatively wealthy minority of the population only (see also section 5).

Affordability at national, regional, district or insurance institution level

Three key pieces of information are needed to assess affordability at either national, regional, district or insurance institution level. These are:

· how many people in the country/area/institution are eligible for ARV treatment? The numbers should be defined over a specified time period, usually a year. Numbers can be assessed (a) overall, for all indications or (b) separately for the three major indications - MTCT, PEP, all HIV+ individuals/people with AIDS;

· what is the likely cost per person for ARV treatment? Again, this needs to be specified for a particular time period or episode. A year would be appropriate for treatment of all HIV+ people/people with AIDS; one episode of treatment would be appropriate for MTCT and PEP;

· what are the total resources available per year in the country/district/region/ institution? This can be measured in terms of (a) total income (GNP for a country as a whole, with a similar measure of total wealth also relevant to regions/districts; total value of annual contributions for an insurance scheme) (b) existing health expenditure (c) total annual drug expenditure and (d) total spending on HIV/AIDS prevention and care. For a country, region or district, it may also be relevant to consider total public sector health spending. This is because if ARVs are found unlikely to be affordable at individual/household level, the affordability of public sector funding will need to be considered in its own right.

These pieces of information can be used to assess how the likely total costs of ARV treatment compare with available resources, in turn allowing conclusions concerning affordability to be drawn. Such an assessment, for ARV treatment for all HIV+ people, is illustrated at national level for a number of countries in the table below. However, the same analytical principles apply equally to the other levels too, and to other indications for ARV treatment.

Assumptions are likely to be needed in such analyses, and the analysis illustrated below is no exception. It is important to be explicit about these assumptions, because this enables the implications of changed assumptions (e.g. the numbers eligible for treatment, the cost of drugs) to be explored as a key part of an assessment (see also section 2 on the role of sensitivity analyses in cost-effectiveness studies). In the illustrative analysis below, the assumptions are:

· the annual cost of therapy is approximately US$10 000, which includes both drug costs and the cost of other inputs identified in Table 2 (more accurate assessments should be possible when performing such an analysis for a specific country);

· and all HIV+ people are eligible for treatment.

It is also worth highlighting the fact that the data in Table 3 are out-of-date, and this may be a problem common to many such analyses. It is almost inevitable when a common source is used (in this case “AIDS in the World II³”), because it is time-consuming to compile data for all countries. However, if the principles illustrated here are applied in 1998 or at a later date within a particular country/region/district/ institution, it should be feasible to incorporate more up-to-date figures.

³AIDS in the World II, 1996 eds. Mann J and Tarantola D.

The countries have been chosen to show how the affordability of ARV treatments is likely to depend critically on two factors:

· the magnitude of the HIV/AIDS epidemic (the numbers of people with HIV);

· and income levels.

It is clear, for example, that ARV treatment is much more affordable in Latin America and Asia than it is in sub-Saharan Africa. This reflects the fact that in Latin America and Asia the HIV seroprevalence is currently relatively small, and that Latin American and at least some Asian countries (e.g. Thailand) have much higher overall levels of income. In turn, this helps to explain why several Latin American governments or social security schemes in Brazil, Mexico, Costa Rica and Uruguay have been able to decide to finance ARVs, why Thailand is publicly funding ARVs to some extent, and why there is virtually no public funding of ARV treatment in sub-Saharan Africa. It also shows that countries at similar average income levels (e.g. Bangladesh and Uganda) can have very different capacities to afford ARVs because of substantial differences in the magnitude of the HIV/AIDS epidemics.

Table 3: Affordability assessment for a range of countries

The example above is for the most expensive indication for ARVs, and it is based on the assumption that all those HIV+ are eligible. In practice it is likely that far less than 100% of those HIV+ will come forward for treatment even if it is available, so that this could be seen as a worse-case scenario. On the other hand, unless the epidemic has already stabilised, the numbers of people infected are likely to increase for some years to come.

It is also worth emphasising that ARVs for MTCT and PEP are relevant to much smaller numbers of people, and that they are less expensive. It is therefore conceivable that while ARV treatment for all those HIV+ may not be affordable in a given area/institution, these two indications may well prove affordable. It is thus critical to assess all three indications, and to assess them separately. For example, in Thailand it has been estimated that an MTCT intervention would consume only 20% of the National AIDS programme budget, making it appear relatively affordable (Prescott, 1997).

A second example of an affordability assessment is provided in Table 4 below, which shows the proportion of the national drug budget which would be accounted for by different ARV treatment indications in a variety of countries.

Table 4: Estimated annual ARV drug costs as a percentage of total annual public drug budget

Affordability at individual/household level

At the level of the individual or household, a useful way to assess affordability is to consider:

· what level of individual/household income is required to make purchase of ARVs affordable?

· how many people fall into those income bands which enable purchase of ARVs?

It is important to remember that the totality of an individual/household’s income will not be available for health care. Some basic needs e.g. food will always need to be paid for. It is the individual/household’s discretionary expenditure - the expenditure available once basic needs have been met - that is of most relevance. If this level can be estimated, and if income distribution data are available, it should be possible to assess what proportion of the population could realistically pay for ARV treatment themselves. In turn, this will be useful for informing financing policy (see section 5). For example, in a recent assessment for Thailand, it was estimated that 50% of the population could afford ARV treatment for AIDS at the subsidised price of US$500 per year.

1.3 Could ARV treatments be cost-saving?

Providing treatment to people with HIV-related health problems e.g. tuberculosis may be associated with high costs. If ARV treatments prevent some of these problems, they will result in some cost-savings. In developed countries such as the USA, there is some evidence that the costs associated with AIDS patients may be so high that ARV treatments result in net cost-savings, particularly when indirect costs (i.e. the costs associated with lost income due to ill-health or premature death, as opposed to direct costs which are those associated with provision of health care) are considered. For example, the average wage of a US citizen is much higher than the annual cost of ARV treatment. It is easy to see how the costs of ARV treatment could be outweighed by the monetary benefits resulting from such treatments, if ARV treatment allows someone to remain healthy enough to work. The high costs of paediatric AIDS care also appear to make MTCT cost-saving in the USA.

In developing countries, cost-savings are much less likely to result from ARV treatments. It has been recently estimated that AIDS care costs approximately 1.5 times GNP/capita (Sheperd et al, 1997), and this is in line with earlier summaries of cost estimates (Martin, 1996; see Table 5 below). Pre-AIDS care is likely to cost less. This means that even if ARV treatments prevented HIV-related illness altogether, they would certainly not be cost-saving for the health system in any country with an average income of below US$ 7300 (given that ARV treatment costs approximately US$ 10,000 per year).

Table 5: Costs of medical care for AIDS patients in different countries

Even if indirect costs are considered, a country would need to have an average annual per capita income of at least US$4 000 to make ARV treatments cost-saving (savings resulting from avoided illness would then be US$4 000 in wages and US$6 000 in health care costs per year). Moreover, over the long-run the figure may be higher, given that saved treatment costs only apply over the expected lifetime of HIV infection in the absence of ARV treatment, whereas it is likely that ARV treatments have to be undertaken over a more extended time period.

Overall, this means that ARV treatments are not likely to be cost-saving, on average, in developing countries. However, this should not conceal the fact that for some individuals ARV treatments are likely to be cost-saving. For the highest income earners in any developing country, annual income will exceed the costs of ARV treatments. This has implications for the role of government in ARV treatment, particularly in the areas of financing and regulatory policies (see sections 5 and 7).

1.4 How may ARVs be made more affordable?

As drugs constitute approximately 90% of total per patient annual costs for triple combination therapy, making ARVs more affordable is dependent on achieving reductions in drug costs.

How might such reductions be achieved? One option is for governments to purchase drugs in bulk. Table 6 below shows the drug costs that have been negotiated in Uruguay on the basis of bulk-purchase.

Table 6: Drug costs (US$) in Uruguay when drugs are purchased in bulk

This shows that annual treatment costs approximately two-thirds of those quoted above have been achieved for two of the drugs, and the cost of ZDV has been reduced to 28% of the 1997 US market price.

In addition to bulk-purchase arrangements, international partnerships among national governments, international organisations and drug companies may facilitate reductions in prices which enable the drug companies to earn reasonable profits so that research and development costs are recouped. For example, drug companies have recently announced a reduction in prices of ARVs for MTCT in developing countries.

Finally, it is important to note that over time companies may substantially reduce their prices - ZDV was reported to cost more than US$10 000 per year in the late 1980s but now costs much less (Table 1). Furthermore, as patents expire generic products will be produced at lower prices. Affordability should therefore be periodically re-assessed.

2. Cost-effectiveness

If ARV treatments are affordable, a second key economic consideration is their cost-effectiveness. Cost-effectiveness analysis is useful for deciding whether ARV therapies are likely to represent a good use of resources managed on behalf of a group⁴ - for example, a country’s population or the members of an insurance scheme. It is therefore particularly relevant to government decision-making, since governments allocate resources on behalf of large population groups. If the likely cost-effectiveness of ARVs compares favourably with that of standard, accepted health interventions - particularly public health interventions - there are strong grounds for providing them. If the cost-effectiveness of ARV therapies compares unfavourably, then it is harder (but not impossible) to justify their provision.

⁴It is important to distinguish this from decisions made by an individual, where use of available income is simply based on willingness and ability to pay.

2.1 Evidence concerning cost-effectiveness of ARV treatments

Table 7 summarises current evidence on the cost-effectiveness of different ARV treatments. The Disability Adjusted Life Year (DALY) is used as the outcome measure as this allows comparisons between interventions for different kinds of health conditions.

Table 7: Costs and cost-effectiveness of ARV treatments

The figures in Table 7 show that of the ARV interventions available, prevention of MTCT is the most cost-effective.

2.2 Cost-effectiveness of other health interventions which are widely implemented

To enable comparison with other health interventions, examples of the cost/DALY for a wide range of health interventions are shown in Table 8 below. These are taken from Jamison et al, 1993 Disease Control Priorities in Developing Countries⁵ and a recent publication on the cost-effectiveness of improved STI treatment in Mwanza, Tanzania.

⁵The cost/DALY figures quoted in the World Development Report 1993 are also taken from the studies presented in this book.

Table 8: Cost-effectiveness of a range of health interventions

⁶Data from the Mwanza study of improved treatment services for STIs in preventing HIV infection based on Tanzanian life expectancy

The most obvious benchmark for comparison is the cost-effectiveness data from the Mwanza study of improved treatment services for STIs in preventing HIV infection. This showed that the cost per HIV infection averted was $217.62 and $10.33 per DALY saved (based on Tanzanian life expectancy) or $9.45 per DALY saved if the assumptions used in the 1993 World Development Report were applied.⁷ Measured against this standard, ARV treatments do not appear to be very cost-effective.

⁷These results are in line with a theoretical estimate made by Over and Piot of between $0.56 and $50 per DALY saved, depending on the various assumptions made.

Making comparisons with other HIV/AIDS interventions is constrained by the limited evidence on their cost-effectiveness. However, it has recently been suggested that if the cost/DALY of a health intervention is less than per capita GNP, the intervention is likely to be cost-effective (Sheperd et al, 1997). Although this is not, conceptually, the way that cost-effectiveness analysis is to be used (since it should be used to reflect the relative worth of interventions relative to each other rather than to any absolute standard), it seems to be a useful ‘rule of thumb’ . Using this as a general guide, Table 8 suggests that:

· ARV prophylaxis for HIV infected pregnant women may be cost-effective in countries with a GNP/capita above US$285;

· ARV triple-combination therapy may be cost-effective in countries with per capita incomes of over US$8 000;

· prophylaxis of people exposed to HIV infection may be cost-effective in countries with a per capita GNP of over US$10 000.

Caution is needed when using cost-effectiveness data. There are a number of difficulties with applying cost-effectiveness analysis to HIV/AIDS interventions, including ARV treatments. These need to be borne in mind when conducting cost-effectiveness studies or when using results derived from them. The major issues are discussed below.

Calculations very sensitive to drug prices

Cost-effectiveness calculations are highly sensitive to changes in ARV drug prices, which will significantly alter any cost/DALY calculation. Sensitivity analyses are therefore very important in cost-effectiveness analyses related to ARV treatments, as are re-assessments of cost-effectiveness following any reductions in drug prices.

Quantifying secondary benefits is difficult

ARV interventions may have important secondary benefits, beyond their direct impact on the individual treated or on MTCT. For example, in the case of interventions to reduce MTCT, decreased HIV transmission to infants may not be the only beneficial outcome. MTCT will lead to larger numbers of pregnant women receiving testing and counselling. Whether the result is positive or negative, this may encourage more responsible behaviour, and may also help promotion of interventions such as Vitamin A supplementation. In addition, if more people are aware of their HIV status as a result of MTCT, normalisation of HIV is more likely to occur. This should enhance prevention, which is currently hampered by denial and stigma. Knowledge of HIV status also allows PLHA early access to psychological and medical care including TB preventive therapy and cotrimoxazole prophylaxis. In the case of PEP, health workers are likely to have increased awareness of HIV infection, which may encourage safer practices in the work setting, in sexual behaviour, and in general infection control. Provision of PEP may also have important beneficial effects on staff morale (see Module 7 for further details). It is difficult to quantify any of these secondary benefits, but it is important to bear them in mind when using cost-effectiveness studies. It may be judged that these benefits justify providing some ARV treatments even when cost-effectiveness comparisons based only on primary effects are not favourable to ARV interventions.

Extrapolating from cost/DALY figures is not straightforward

It is not always easy to extrapolate cost/DALY figures to many country settings, because the studies on which they are based were generally conducted in one or only a few countries.

Effectiveness of ARV treatments is uncertain

It is difficult to estimate the effectiveness of ARV treatments for people with HIV/AIDS because the long-term impact is unknown. For some treatments, clinical trials have not been set up, are still in progress, or have yet to be designed and funded.

Furthermore, effectiveness data concerning ARVs is currently limited to recent trials. This means that at this stage comparisons can only be between results of pilot research interventions and established programmes. There are often considerable differences between what can be achieved in well-resourced and managed pilot research projects, and what can be achieved in more typical operational settings.

Few cost-effectiveness studies have been undertaken and the costs included vary between studies

Few cost-effectiveness studies of ARVs have been undertaken in developing countries. Those that have been done have focused on MTCT. There is far less information about PEP and HIV/AIDS treatment.⁸ In addition, cost-effectiveness calculations will vary according to what costs are included. At present there are no agreed upon guidelines concerning what costs should be included. These are needed, but in the meantime it is important to read cost-effectiveness studies carefully to identify what costs have been considered. One particular issue is whether or not the start-up costs associated with VCT services, which are needed for interventions such as MTCT, are included in cost-effectiveness analyses.

⁸one reason for this is that comprehensive strategies have not yet been formulated.

Providing coverage for “catastrophic financial risks” may be considered more important than cost-effectiveness

It can be argued that it is appropriate for governments/insurance agencies to provide coverage for “catastrophic financial risks”, which the individual/household cannot afford to pay for, even if the care entailed is not cost-effective. Interventions that are not for “catastrophic risks” e.g. many treatments provided at primary care level, it could be argued, should be funded by individuals/households - even if they are far more cost-effective than interventions for “catastrophic risks”. Such arguments have implications for financing policy (see section 5).

Political imperatives

Pressure on governments to “do something” may make it almost impossible to strictly apply the results of cost-effectiveness analysis in practice.

Cost-effectiveness analyses only allow comparisons within the health sector

Cost-effectiveness analysis in the health sector uses health-specific outcome measures. This means that it can be used to inform decisions about resource allocation within the health sector, but not the decisions about whether more resources should be allocated to the health sector as a whole. This issue appears to be of particular concern in Latin America, where AIDS activists have reportedly argued that governments are already funding non-cost-effective policies in other sectors, so that ARV treatments should not be competing for funds with other health programmes, but with other public budgets. This is an important argument to bear in mind when using cost-effectiveness to inform decisions on the introduction of ARVs, or any other kind of public policy (including health) intervention.

2.3 How can the difficulties with cost-effectiveness analysis be reduced?

It may be especially helpful to supplement cost-effectiveness analyses with other types of economic assessment. A particularly useful approach is to consider the “opportunity costs” of investing in ARV treatments. This means that the question is not “Are ARV treatments cost-effective?” but “What is given up by investing in ARV treatments?”. If the cost of an ARV intervention can be estimated and if the costs of other types of HIV/AIDS care and prevention activities can also be estimated, it is possible to illustrate what potential benefits would be lost by investing resources in ARV treatments. For example, what prevention activities could be funded for the cost of ARV interventions? What level of counselling and testing, and treatment of opportunistic infections, could be funded for the cost of ARV treatments? The answers to such questions should both (a) inform decision-making and (b) help to provide justifications for the choices made. If such data are not already available, assessments should be undertaken.

This type of analysis could also be extended beyond HIV/AIDS prevention and care: for example, how much malaria control could be provided, how many tuberculosis cases could be treated, how many children could be immunised, with the funds required for ARV treatments? Similar questions could be asked for non-health sector activities too. How many children could be educated at primary level, what basic infrastructure could be installed or repaired, for the cost of ARV treatments? These kinds of questions that are of importance in any particular case depend on the relevant unit of decision-making(e.g. national, provincial, district health system, particular programme etc.)

3. Are the necessary human/physical resources available?

It is important to emphasise that cost is not the only economic constraint to the provision of ARV treatments. Even if ARVs are affordable, and whether or not decisions concerning their provision are influenced by cost-effectiveness studies, successful provision in practice depends on whether the necessary human and physical resources are available and/or capable of being mobilised. Under some circumstances, inadequate capacity - for example too few staff or insufficient laboratory services - may to be more of an obstacle to ARV treatment provision than is the securing of funding.⁹

⁹for example, in the World Bank’s recent publication “Confronting AIDS”, it is observed that “effective anti-retroviral therapy requires a highly trained, specialised physician working in a well-equipped clinic with experience performing a wide range of sophisticated tests and procedures, all of which are in critically short supply in most developing countries”.

The sort of capacity which is likely to be critical is shown in Box 2 below.

A recent study in rural South Africa has estimated that there needs to be a 10-15% increase in the number of midwives, a 50% increase in laboratory staff and a 100% increase in counselling staff to implement an MTCT initiative. This is at a time when HIV-related staff sickness and premature death is on the increase and is already reducing the pool of skilled health workers. Furthermore, if the intervention is implemented simultaneously in rural and urban areas, there will not be enough trained mid-wives. Difficulties in strengthening basic infrastructure and recruiting additional staff may seriously impede the process of introducing and implementing ARV interventions.

Problems of capacity are likely to be less acute in parts of the world where health staff are in greater supply and where educational levels are relatively high - for example Latin America, the Caribbean, and parts of Asia. However, even in areas where capacity is relatively strong, the problem is still daunting given the complexities of the interventions.

4. Equity

Equity is not an easy term to define, but is related to ideas of “fairness” and “justice”. In the case of ARVs, there are two particularly important questions to ask:

· who is likely to have access to ARV treatments?
· is such access fair?

In economic terms, the issue is therefore one of the distribution of resources for ARV treatments. What is this distribution likely to be, and is it fair? At present, there is limited evidence concerning who is likely to have access to ARV treatments, while the question of fairness is a value judgement. Nonetheless, it is possible to suggest the following:

· access to ARV treatments will be closely connected to financing policy (see section 5). If treatments are not provided through public funding, only the highest income earners in a country, and those who are members of social security schemes which have agreed to pay for ARVs (see Box 3), will be able to obtain treatment;

· while it could be argued that access based on ability to pay is unfair, most goods and services are consumed on this basis. Furthermore, even though much health care is funded publicly, access to public services is not equal: it would need to be carefully justified why ARV treatments should be treated as a special case. If there is already unequal access to services, it is also likely that it would be hard to achieve more equal access with ARV treatments, particularly since they are likely to be relatively difficult to deliver;

· particular care is required if ARV treatments are heavily subsidised or completely funded by government. ARV treatments may then be provided to people who would have been prepared to pay for them in the private sector. If this happens, public resources will be unnecessarily diverted away from other services, which might have benefited people less able to access treatment in the private sector, including those without HIV infection. There is some evidence from CD’Ivoire that this situation does occur in practice (see Box 3);

· since ARV treatments require specialised staff and equipment, urban populations are more likely to have access to them. Access will probably always pose problems for rural populations and some marginalised and vulnerable groups (see Module 3).

5. Financing

Decisions regarding financing policy are closely connected to issues of affordability, cost-effectiveness, and equity; and even if financing is available, it will only be successfully translated into treatment provision if the necessary human and physical resources are available. Therefore while decision-makers may regard financing as the most important issue on which they require guidance, it cannot be divorced from these other issues.

This section takes each possible financing mechanism in turn. The most important factors to consider in the context of ARVs are suggested, and the advantages and disadvantages of each option are discussed. The options considered are public funding, national insurance/social security schemes, private insurance, voluntary sector/charitable funding, community financing, out-of-pocket payment (private funding), external assistance from donor agencies, and development loans.

To set the discussion in context, Box 4 and Table 9 highlight key features of health financing in developing countries. They suggest that:

· public expenditure is relatively important in Latin America and Africa;

· private spending is most extensive in Asia, particularly India, but is significant everywhere;

· national insurance and social security schemes are mainly important in Latin America;

· public expenditure as a percentage of total drug expenditure is relatively low;

· most drug expenditure in developing countries is out-of-pocket expenditure;

· per capita incomes in all countries suggest that without public funding, ARVs will be difficult to afford for many people with or exposed to HIV infection; and

· external assistance is an important source of funding in only a few countries.

Such assessments will be useful in any country considering the introduction of ARV treatments, since their financing is likely to be linked to, and/or constrained by, existing financing strategies.

Table 9: Patterns of health financing in African, Latin American, Caribbean and Asian countries in 1990

5.1 Public financing

The main issues to consider in the case of public financing of ARVs are:

· what are the justifications for public, as opposed to private, funding of health care, and how do these relate to ARVs?

· would public funding of ARVs be consistent with overall health sector policies?

· can the public sector afford to provide ARVs, and if so for all or just some indications; would this be a cost-effective use of scarce public resources?

· do budget constraints and taxation policies make generation of additional public revenues for ARVs feasible?

· are the consequences of not providing any public funding acceptable?

· will public funding result in subsidised care being provided to people who would otherwise have accessed such care in the private sector?

Each of these is discussed below.

Justifications for public funding of health care in general

Public funding of health care can be justified on the following grounds:

· externalities: for some health conditions, there are important “externalities” - that is, the costs of that condition, or the benefits of providing treatment for it, extend beyond the individual directly affected by it. For example, immunisation benefits people other than the child immunised; tuberculosis treatment benefits people other than the person with the disease, because treatment prevents further transmission. In such instances, where private costs/benefits are exceeded by societal costs/benefits, there will be inefficiently low investment in such interventions if decisions are left up to individuals. This is because individuals are expected to make decisions based on private costs and benefits, not social ones. Therefore, public funding may be required in cases where there are important externalities, in order to ensure that an efficient level of provision occurs;

· catastrophic risks and insurance market failure: some health conditions are very costly to treat, and will be beyond the capacity of most individuals to finance. This means that there is a case for insurance, so that such potentially “catastrophic financial risks” can be shared (see also section 2). Furthermore, since markets in private health insurance may fail - some people may be unable to obtain insurance because they are “bad risks” or already have chronic conditions, and cost containment is difficult - a case can be made for public funding;

· health care is a special case: it may be judged that, unlike most goods and services, access to health care should not be based on ability and willingness to pay. Health care is a “basic right” to which everyone should have access, and the consequences of basing access to health care on ability to pay are unacceptable.

To what extent do justifications for public sector funding apply to ARVs?

The above arguments could be used to justify provision of ARVs, at least in some instances. Externalities exist in the case of MTCT and PEP, because prevention of transmission to children and infection of people exposed to HIV infection will have benefits beyond the individuals treated. HIV/AIDS is also a catastrophic financial risk - most of those affected cannot afford to purchase treatment themselves. And like other health interventions, it may be considered that access to ARVs should not be a function of ability to pay. Private financing is likely to restrict access to a high-income minority (see also section 1), and this may not be considered acceptable.

Consistency

Nonetheless, even if these arguments are accepted, it is important to be consistent. Are such arguments used for other aspects of health care provision? As Box 4 and Table 9 show, much care is not financed by the public sector: is HIV/AIDS different enough to be seen as a special case? Public provision of ARVs may also not be consistent with certain health sector reform initiatives: for example, increased emphasis on user fees, cost containment, essential health packages where cost-effective services have first call on public resources, and private sector provision (see also section 7). As Box 4 shows, public funding of drugs is low: yet with ARVs, public funding would mean that most, if not all, of the drug costs would have to be covered.

Affordability and cost-effectiveness

If ARV treatments are likely to consume a large share of health sector resources, public funding may not be practical, and if they do not appear to be very cost effective, it may be hard to justify their provision. Affordability will also be influenced by wider government priorities, such as taxation levels and public sector borrowing targets.

The consequences of not providing public funding

The consequence of not providing public funding is that access to ARVs will be limited to those who can pay for them on an individual basis i.e. those in the highest income brackets.

Access issues may arise even with public funding

Public funding does not guarantee that access will not vary according to socio-economic status or geographical area. For example, because specialised skills and sophisticated equipment are required, it may only be feasible to supply ARVs in a few locations - often capital cities and other large urban centres. And because compliance to drugs is crucial, entry criteria to treatment programmes may have to favour those with higher socio-economic status (see Box 3 for examples from Mexico and Cd’Ivoire).

5.2 Social health insurance

Social health insurance involves insurance schemes where membership is compulsory. These can be national insurance schemes, to which everyone (or at least those in formal employment) must contribute; or social security schemes, which can be national in scope but are frequently linked to employment in a particular company or sector. Social insurance is an important source of health sector funding in Latin America, but is of minor importance at this stage in Africa and Asia (see Box 4). Therefore, social insurance has most potential to fund ARV treatments in Latin America.

One advantage of social insurance, as opposed to private health insurance, is that people cannot be excluded from cover: everyone contributes, everyone is covered. The main issues for ARVs and national insurance schemes/social security schemes are therefore:

· can schemes afford to provide cover for ARV treatments? (see section 1)

· if schemes fund ARVs, will HIV-seronegative individuals be affected by reduced access to other services?

· if existing funds are unlikely to be sufficient, can premiums (the contributions made by employers/government/individuals) be increased to provide the required funding? Will this be acceptable to those who do not require ARVs but who will be expected to pay higher premiums?

· is provision of ARV treatments/increased premiums compatible with health sector reform initiatives related to health sector financing? For example, HSR may include an emphasis on cost containment and reduced government contributions to schemes.

Conceptually these issues are identical to those faces by publicly financed systems, the only difference is the population that is covered by each. These four questions will need to be addressed in any country or institution considering the funding of ARVs through social insurance. It is likely that in countries where HIV seroprevalence remains relatively low, where social insurance is widespread, and where incomes are comparatively high, social insurance does have significant potential to fund ARVs. In Latin America, where these three characteristics exist, social insurance may therefore be an important facilitator of ARV treatments. Both Costa Rica and Mexico, for example, are funding ARVs through this mechanism.

It is much less likely to have significant potential in Asia and Africa, simply because social insurance is very limited in these regions.

Where only part of the population is covered by social insurance, these programmes often exacerbate inequities in access between the insured and uninsured.

5.3 Private health insurance

Private insurance is likely to be of limited value. Companies can be expected to try to eliminate bad risks: evidence from the USA shows that some insurance companies use HIV testing to eliminate poor risks from pay pools, and cost-conscious employers have tried to exclude AIDS patients from group insurance. In South Africa, individual companies are offering lower premiums to people with low HIV/AIDS risk, and in order to remain competitive, other companies are likely to be forced to do the same. This means that HIV-infected people will find it hard, if not impossible, to obtain coverage; and it is increasingly common for policies to specifically exclude cover for HIV/AIDS-related health problems. It is worth noting that the majority of HIV/AIDS care is financed through the public sector even in the USA, where total public sector funding of health care is about 40% of total health spending. In other Industrialised countries public sector health funding is a much higher proportion of total health expenditure. It is therefore unlikely that private insurance has much potential to be a major source of funding for ARV treatments elsewhere.

An exception to the tendency of private insurance companies to exclude coverage for HIV/AIDS, however, is a company in South Africa which is offering insurance to employers rather than to individuals. This specifically covers the health care needs of HIV infected people, including provision of ARV therapy. Replication may depend on the ability of governments to put in place, and enforce, appropriate regulation on the private insurance market (see section 6).

5.4 User fees and other out-of-pocket payments

ARV treatments represent a large proportion of income at the level of the individual/household. This means that full cost-recovery through user fees for on-going treatment of people with HIV infection is unrealistic, except for a small minority of patients with much higher than average incomes. However, short-courses for PEP and MTCT may be more affordable (see section 1. for how this can be assessed), and partial cost-recovery may be possible in some settings. For example, partial cost-recovery for ARV treatments for AIDS patients is planned in Thailand. For these sorts of cost-sharing approaches to succeed, initial subsidies may be required. This is because sufficient confidence in government-run health services will have to be generated if individuals/households are to be willing to pay what may be a significant portion of disposable income for ARV treatment costs.

A potential difficulty which it is worth noting is that as partial and interrupted therapy encourages the development of resistant strains, legislation to restrict ARV provision to specialist centres and approved prescribers is likely to be required. This will limit opportunities for out-of-pocket purchases.

5.6 Community-based financing and voluntary/charitable funding

Experience with community financing schemes indicates that their revenue potential is not large, and mobilising finance for costly medical treatment that does not benefit all members of a community presents obvious challenges. Solidarity funds with NGOs and community organisations may be a potential strategy, especially if strong PLHA North-South alliances are developed. Such relationships can mobilise drug donations, but may be relatively unstable in the long-term and only benefit small numbers of individuals. Furthermore, these initiatives often rely on highly motivated individuals, and high quality medical supervision, regulation and quality control may be difficult to ensure. Even when such schemes exist solely for supplying drugs, they are not useful or usually not sustainable, unless they have a broader purpose and are linked to the overall health system.

5.7 Donor financing and development loans

Donor financing and development loans may be a partial source of funding for the most cost-effective interventions in the near future in some low-income countries. These interventions would be for prevention of MTCT and perhaps PEP for health staff. However, convincing arguments will have to be put forward to secure significant levels of funding, and the reliability of long-term donor financing for such programmes cannot always be assured.

5.8 Public/private co-financing arrangements

Semi-private financing exists in all countries. It is often employment-based and reserved for employees and their dependants working in specific areas. It includes occupational health services and company hospitals (e.g. in the mines, banks or large agricultural concerns), and services for government employees, the police and the military. These may offer specific opportunities for government/private sector co-financing opportunities, especially for the more highly trained employees whose health and productivity is of significant value to the company.

Such opportunities need to be explored on a case-by-case basis. Care needs to be taken to avoid this becoming a mechanism for indirectly subsidising an urban elite in relatively high-status and well-paid employment.

In conclusion, while partial cost-recovery from patients may be feasible in middle and high-income countries, the public sector or social insurance schemes will have to finance the majority of ARV treatment costs. This means that unless the public sector can afford to provide a substantial portion of the funding required for ARV therapy, or unless social insurance is widespread, it is unlikely that provision to anything but a small minority of patients is feasible.

6. Provision

In recent years there has been increased recognition that financing and provision of health care can be treated as distinct issues. Even if public funding finances health care, this does not mean that public provision is required: for example, government can contract with NGOs and the private sector, thus funding care but not taking direct responsibility for its provision. In addition, where the public sector does not fund health care, this does not mean that it should not be involved in provision, for example through regulation of the private sector or through setting the overall policy framework within which the health sector operates. In the area of ARV treatments, the private sector will have an important role, especially where public funding is non-existent or limited. NGOs, which have frequently been instigators in other areas of HIV/AIDS care, are also likely to be involved. In addition, the pharmaceutical industry is also of fundamental importance, since pricing policies on ARVs are arguably the single biggest influence on whether or not ARV treatments can be widely provided. This section therefore discusses, in general terms, what role each can play in provision of ARV treatments, highlighting important issues to be considered in each case. These roles are largely those of direct provision, regulation, negotiation, facilitation and price-setting.

6.1 The private sector

The private sector is likely to become a major player in provision of treatment for both opportunistic infections and ARV therapy. This has already occurred in the USA and, for the urban elite, in many parts of sub-Saharan Africa. Serious consideration should therefore be given to the sector, which includes private practitioners/clinics/ hospitals, the NGO sector, and semi-private health providers (company clinics).

The main concerns with the private sector are:

· few doctors have had any training or experience of using ARV treatment;

· the necessary infrastructure for proper use of ARVs and management of patients may not be available;

· exploitation of patients may occur - the profit motive rather than the welfare of the patient may be the main motivation for what treatment is provided. Desperate people living with HIV/AIDS may be prepared to pay for inappropriate or partial therapy if this is all they can afford, as there is a widespread belief that some drug is better than none and will have a worthwhile effect;

· incorrect use of ARVs may lead to drug resistance: if private sector providers use ARVs inappropriately, there is a very real danger that drug resistance will rapidly emerge.

These concerns mean that government has an important role to play in provision of training and guidelines, and in regulation (see below).

Theoretically, an advantage of collaborating with and supporting private sector providers is that those who can afford to pay for treatment may be persuaded to use them, thus ‘freeing up’ more public sector resources for those in more genuine need of subsidised ARV treatments. However, such ‘freeing up’ of resources has never been shown empirically, and the likely magnitude of any such gains would be quite small.

6.2 The pharmaceutical industry

The importance of the pharmaceutical companies who research and develop, manufacture, market and sell ARVs cannot be underestimated. Their pricing policies and willingness to develop special - perhaps unique - partnerships in the area of ARV drugs will be critical in determining what level of ARV provision is feasible. Without substantial reductions in drug prices, access to ARVs is likely to remain restricted. Some positive developments have already occurred, with the price of some drugs being substantially reduced in early 1998. Collaborative partnerships among companies, international organisations and governments may facilitate further progress.

6.3 Government/public sector

The concerns with private sector provision and drug pricing mean that government, and the public sector as a whole, have a key role in ARV treatment provision, even if the public sector is not a direct provider of care. This is particularly the case in countries where the private sector is very extensive, such as India. Partnerships with private sector providers need to be developed to facilitate both training and regulation, and constructive partnerships with pharmaceutical companies are essential.

An important way of improving the quality of private sector provision is to offer training and instruction in the best ways to use ARV treatments. Supervisory bodies such as national Medical Councils may be especially important partners in such initiatives. Moreover, the public sector has the mandate and resources to conduct regular in-service training, and the skills to supervise and assure the quality of such post-graduate education. A requirement to complete training in use of ARVs (perhaps through the award of a certificate of competence on completion of a training workshop) would ensure good standards of clinical practice, and could be used as a way of demonstrating to the public that private practitioners have attained a certain competence in the use of ARV drugs. Such accreditation schemes are particularly useful because:

· they give providers an incentive to undertake training: with accreditation, providers may be able to attract fee-paying patients away from untrained competitors, and avoid loss of existing patients to those with accreditation;

· they provide consumers with more knowledge about which providers will provide a good service.

A free or subsidised training scheme, if widely advertised and correctly administered, could therefore be used to develop a regulatory system based on those who were registered or licensed, and approved for ARV prescribing.

The use of voucher systems could also help to assure quality. If some public funding is available for ARV treatments but public provision is not considered a viable option, the government could issue vouchers which patients could only use at a range of licensed private practitioners. This would ensure that patients received quality care, and private providers would be able to redeem the vouchers in the form of cash (or perhaps drugs) at the local health service. Such services have been piloted in several different regions for a variety of non-HIV-related problems.

Government’s role may be to provide the necessary in-service training and guidelines for ARV treatments, including monitoring and adherence (in whatever programmes are to be implemented); and then to supply regularly, at cost or with subsidy, a pre-fixed amount of ARV drugs for that NGO’s use (some NGOs may even be able to bear some of the costs of setting up the necessary services by raising money from their overseas partners). Government may also contract with NGOs to provide defined services in their catchment areas (see below).

Maintaining the integrity of the drug supply is likely to be a key responsibility of government. The high cost of the drugs, and their probable high resale value, means that the likelihood of the development of a black market in counterfeit and high quality drugs is high. Government is the appropriate agency to develop mechanisms (e.g. laws with appropriate punishments for offences) to protect the integrity of the drug supply, and should regularly monitor and evaluate their performance (see Module 8 for further details).

Government may also have a role to play in monitoring and regulation of the private insurance market. This is particularly the case where innovative schemes are being developed, since this will facilitate an early response to problems which arise, allow important lessons to be learned, and enable public awareness of the advantages and pitfalls of these initiatives to be raised.

Government is the appropriate body to negotiate with pharmaceutical companies, and to facilitate the development of constructive partnerships with them. Much can be achieved in this way, as the drug reductions negotiated by the Uruguayan government shown in section 1 illustrate.

6.4 NGOs

NGOs are significant health care providers in many low and middle income countries and have a number of strengths. They often have important financial links with sponsors and charitable partners in industrialised countries, and donors frequently channel considerable budgets through NGOs because of their good reputation for financial probity. Some NGOs and CBOs may have also have links with PLHA networks in industrialised countries, and be capable of accessing additional ARV supplies and perhaps specific technical support and training for their optimal use. Moreover, NGO services are usually run independently from any government services and there is little need for management and supervision.

These strengths may imply that NGOs have an important role to play in ARV treatment provision. However, some caution is needed. The sustainability of NGO provision of ARV treatments in the medium and long-term is uncertain: NGOs are often not self-supporting and receive explicit subsidies from governments and/or staff secondments, which may be withdrawn. Links with PLHA networks also depend on many factors beyond the control of individuals working for the NGO, so they may not be reliable in the long-term. In addition, NGOs may have an ambiguous relationship with government health services. They may be poorly regulated and there may be a need for NGO providers to be brought within the overall policy framework of the publicly managed health service to ensure that a parallel service, or a service catering for a minority population, is not given preferential funding.

NGOs will require training and guidelines on ARV treatments from government (see above), and may be suitable contracting partners for government. For example, the government could fund an NGO to provide a defined ARV service in their catchment area. However, NGOs are still likely to require support for specific capital items such as the building of secure new dispensaries and storage facilities for high-value ARV drugs.

The role and place of NGOs and CBOs in specialist therapeutic care such as provision of ARV treatments, where specialist expertise is required, is an issue which needs to be examined further.

6.5 International organisations

International organisations are unlikely to become involved in extensive direct provision of ARVs. Their major role in provision is likely to be in helping to facilitate and negotiate constructive partnerships between governments and pharmaceutical companies, in dissemination of lessons of experience from a wide variety of countries, and in provision of technical assistance.

7. ARVs, economics, and health sector reforms

Many governments have been in the process of reform of their health sectors since the mid-late 1980s. This section therefore briefly discusses the main reform initiatives that have been observed, many of which have an economic aspect to them, and suggests their implications for the introduction of ARV treatments into health systems. The common reform themes include:

· decentralisation of decision-making to district health systems;

· essential health care packages, in which the most cost-effective interventions have first call on public sector resources, are emphasised;

· clear separation of the bodies who have responsibility for financing and provision of care;

· cost control linked to both insurance and user fees;

· increased role for non-taxation based financing mechanisms such as insurance schemes and user fees, with reform of insurance in particular linked to the need for effective cost control;

· a focus on performance-based accountability, to replace top-down administration;

· promotion of the role of the private sector and the importance of collaborative partnerships between the private and public sectors; and

· greater prominence for issues of equity and quality.

Wherever decentralisation of decision-making to district level (particularly of decisions related to budget allocation) is occurring , the issues discussed in sections 1 to 6 of this module will need to be considered at district as well as national level - national level remaining relevant because even with such decentralisation, the MoH is usually still responsible for setting the overall policy framework within which the entire sector operates. It is therefore important that managers at district level are provided with information (for example this module) which will assist them to make the decisions for which they are now being made responsible. This is particularly necessary given the emphasis on essential health packages, in which cost effective interventions are supposed to have first call on public resources. Wherever district managers are to be responsible for decisions regarding whether or not to fund ARVs from their budgets, it is critical that district health teams be provided with information (and possibly training) concerning evidence on cost effectiveness, how to assess cost effectiveness, the problems that exist with cost effectiveness analyses, and how difficulties with cost effectiveness analysis can be minimised. Section 2 of this module is therefore very relevant in the context of HSR. Section 3 is also relevant, because decentralisation of decision-making should provide managers with greater autonomy in how they can allocate and mobilise resources. Indeed, decentralisation of such decisions may be necessary if the resources required for ARV treatments are to be put in place in a timely way, since decentralisation should provide more flexibility in deployment of resources and facilitate more rapid decision-making related to recruitment/purchase of new inputs.

The main implication of the separation of the financing and provision functions in the health sector for ARV treatment is the increased potential for the private sector to play a role in provision, even when public funds are used to provide the necessary finance. This also ties in with the general emphasis on promotion of public/private partnerships in HSR, and suggests that the discussion of the role of different players in provision of ARVs in section 6. is of especial relevance. In addition, the focus on quality assurance in HSR indicates a key role for the public sector in regulation and monitoring of ARV provision in both the public and private sectors - again discussed in section 6.

The emphasis on non-taxation based financing mechanisms in HSR means that the discussion of financing options in section 5 needs to be read with that overall context in mind. This section suggested that user fees and private insurance are unlikely to be viable financing options for ARVs, and that the most promising options if provision is to extend beyond a small minority are public funding and social insurance. Yet HSR tends to play down the role of public funding, and to emphasise cost containment (and sometimes extension of coverage, as in Colombia) within insurance schemes. Limits on public funding, and cost containment in insurance schemes, make ARV financing through increased funding from these sources less likely. Instead, diversion of resources from other services, or improvements in efficiency in other areas of health care delivery, are likely to be the main ways in which insurance schemes fund ARVs. This has equity implications, since in the absence of improved efficiency and increased resources, HIV sero-negative people are likely to be penalised. Certain aspects of HSR therefore have potential to be in conflict with some of the options for financing ARVs.

Module 3. ARV Treatments: planning and integration into health services

The following module provides guidance for planners, managers and HIV care providers in resource constraints settings who may be introducing antiretroviral (ARV) treatments in their services or in situations where ARV treatments are currently available, such as the private sector.

Before setting up ARV services key planning questions need to be addressed. A framework for these questions is provided in Table 1, which summarise decision options. The organisation of interventions with ARV treatments is discussed in section II, quality management in III, access to ARV treatment in IV, and the creation of information systems is discussed in section V. Section VI addresses the need for partnerships to ensure sustainability and section VII focuses on monitoring and reporting.

I. Introduction

The incorporation of ARVs into existing heath care services, in a safe and effective manner, is challenging; as is securing sustainable financing of such interventions.

Key issues to be addressed before introducing ARV treatment into existing health systems include: establishing laboratory capacity, as well as mechanisms to control drug supply; assuring quality management; informing the public about the benefits and implications of ARV interventions; providing training and assuring an adequate number of health care and support staff.

Characteristics, specific to ARV interventions, require that these issues be resolved promptly:

· ARV drugs are particularly high-value items; therefore, regulatory mechanisms must be secured to avoid misuse or misappropriation.

· ARV treatments are complex to use and monitor.

· ARVs can easily be inappropriately prescribed or misused with little health gain and potentially adverse public health consequences, such as the development of drug resistance.

· Interruption of the drug supply may have serious repercussions for patient care and overall programme efficiency.

· Additional staff time and technical training is required for patient counselling on adherence to the complex tablet regimens.

When introducing ARV treatments into health care services the following approaches may be adopted:

1. Attach a new ARV treatment service, specially organised as a parallel or vertical programme, to private or public health services.

2. Develop a phased approach where existing health care services for people with HIV are adapted and strengthened to incorporate ARV treatments.

3. Integrate ARV treatments into general health services by training existing staff and modifying the current infrastructure for ARV intervention.

Decisions on the implementation of ARV interventions are dependent upon the local and regional context. Critical factors to consider include:

· Seroprevalence of HIV in the country/community
· Attitudes toward HIV in the country/community
· Existing health system and support services
· Commitment to ARV interventions
· Maturity of the epidemic and stages of development of associated support services
· Financing options

Effectively initiating ARV interventions require general health care system strengthening. However, specific programmes, such as the implementation of mother-to-child transmission of HIV (MTCT) prevention in antenatal care and maternity services, and providing antiretrovirals for post exposure prophylaxis (PEP) of health care staff may also need to be considered.

The decision to expand pilot initiatives to national programmes should only be taken following a careful analysis of the requirements for the intervention and an evaluation of pilot programmes on the clinical, operational and social effects. Sections of health service infrastructure may need to be improved before a larger scale intervention can begin. Assumptions made in the pilot phase with regard to support services will need to be analysed to ensure project feasibility.

Table 1: Questions options and criteria for decision making when considering planning an ARV intervention

I. Organisation of ARV Treatments into HIV Care and General Health Services

Methods of incorporating ARV interventions into existing HIV care services need to be determined before project onset. Questions regarding the implementation of ARV interventions need to be answered such as:

· Should the management of drugs, project monitoring, and follow up be created as a separate vertically organised and funded service or integrated within existing services?

· How will HIV/AIDS clinical care fit in with a district health service that is in the process of decentralisation or currently decentralised?

Integrating ARV treatments with other services

The method for introducing ARV into existing health services is dependent upon local resources. Although integration is likely to be the final goal, initially, vertical approaches may be more feasible. A national authoritative committee to guide ARV implementation should be established to serve as the key mechanism for achieving consensus on programme initiation.

Experience from other interventions and programmes, which initially began vertically, but are now being integrated in the general health services, may provide insight into effective programme implementation and management.

Management at the national level

The national committee should be specifically organised to guide ARV interventions. This may be either a new committee with sole responsibility for the ARV intervention and other HIV drugs, or an existing drug regulatory body. A multidisciplinary approach is best, and maintaining links with key partners including "care focal points" from HIV/AIDS programmes, experts (pharmacists, physicians, nurses and counsellors), consumer organisations (PLHA support groups) and drug control officers (pharmacists, administrators, private sector representatives) is critical. Progress of the ARV initiatives will ultimately depend on how well this central body performs its duties.

The tasks of the national committee should include the introduction, phasing, implementation and evaluation of this intervention, as well as the following:

· Arrangement of financing for the intervention, be it from public and/or private sources.

· Establishment of a regular distribution system for ARV drugs.

· Identification of sites for ARV interventions.

· Development of process indicators and evaluation tools.

· Provision or dissemination of public information on the advantages and implications of ARV treatments.

To generate broad consensus, immediate tasks for a national committee include:

1. National policy formulation on ARV, including financing, pricing mechanisms and phasing.

2. Development of country specific practical guidelines.

3. Capacity building for clinicians, counsellors and drug administrators at identified sites.

4. Estimation or assessment of realistic drug needs.

5. Strengthening and development of a storage and distribution system; ensuring regularity of supplies and distribution.

6. Selection of clinics/sites suitable for the provision of ARV treatment.

Organisational options for start up at site level

Most countries affected by the epidemic provide HIV/AIDS care in a variety of settings, such as:

· Special units in major hospitals (district, mission, provincial and national referral institutions) where HIV/AIDS clinical care is often linked to an NGO to assure comprehensive care needs (Patient Support Units and HIV clinics).

· Private physicians' clinics.

· Clinics within the workplace.

The decision as to which site should initiate ARV interventions should be based on site fulfilment of the minimal requirement standard, as described in Table 2.

It is best to first pilot the ARV intervention at one or a few sites meeting the eligibility criteria, and monitor the programme for information regarding effective ways of integrating ARV treatment into existing HIV care before expanding the initiative.

Limiting the intervention to sites that meet the minimum requirements will likely raise questions of equity and accessibility. Some physicians may complain that their professional integrity is being questioned if they cannot freely prescribe. Therefore, it is critical that the national committee develops a communication strategy with professionals to inform them of the reasons for selecting their approach, and include the private sector in programme initiation discussions.

Table 2: Implementation steps to consider for ARV program introduction

Organisational Requirements:

1. Training and support of health care staff

Training in ARV interventions is a prerequisite for staff at the unit where ARV will be implemented. Moreover, information sharing about ARVs must be made available to a wide cross-section of clinic staff. Continued training and updating of correct information/knowledge of health care workers should also be built into the system.

Health care staff undertaking new duties as a result of the ARV intervention may feel overworked and possibly suffer "burnout". Support for health care staff should be planned to minimise staff stress.

2. Motivation and time

A positive work environment is important in maintaining staff morale and commitment. Extra time should be allocated for staff to adequately explain to patients:

· Complex drug taking schedules, timing, and practice.

· Monitoring the effectiveness of ARV therapy, adverse effects and drug interactions.

· Counselling those who do not respond positively to therapy.

· Updating on new drug developments.

Staff who can effectively interact with and assist patients, their needs and reactions, and provide complex drug counselling should be recruited and supported.

1. Space

All programmes will require that suitable space be arranged for intensive drug counselling. Rooms where confidential sessions can be held and where local PLHA support groups can meet to provide peer support should be arranged. Space requirements for confidential counselling can be significant be should not be underestimated.

2. Laboratory support

In limited-resource settings, relatively inexpensive, simple methods for CD4 counts and standard methods for HIV diagnosis, as well as monitoring of ARV side effects, and diagnosis of opportunistic infections will require additional supplies, staff and space. This may be organised with the support of one central laboratory responsible for quality control and training. Where possible, facilities for monitoring ARV treatment response should include measurement of viral load. However, this may not be available in many developing countries. Laboratory requirements are discussed in detail in module 5.

From vertical to horizontal service delivery

ARV interventions initiated at specialised centres should be monitored, and if successful, included as part of an integrated follow-up service at peripheral clinics or primary health centres where adherence can be reinforced and other care and support requirements can be met. This continuum of care for patients receiving ARVs, as well as cross-referral between hospitals and the community is critical and will require extensive collaboration between clinic staff and care providers in the community.

Boundaries between services are often difficult to define. For example, questions concerning limiting the role of counsellors to ARV-MTCT programmes (VCT, ARV follow up, infant feeding and family planning options) or incorporating counselling services for entire health facilities are often debated. Decisions regarding the role of counsellors should be determined in the local context. However, different solutions maybe adopted at varying stages in the implementation process.

For ARV interventions to be safe and effective, they must be well resourced and staffed. Tensions between services in the same facility may result as disparities with other clinical services. Pressure to utilise ARV facilities for other equally compelling clinical problems may result. This type of operational difficulty has been witnessed with STI treatment kits, where broad-spectrum antibiotics are not otherwise available, but urgently required and administered to particularly sick patients.

I. Quality Management

Quality management and quality assurance are monitoring mechanisms used to ensure that minimal standards are met for the components of ARV interventions, including the inputs required for the intervention, the process of implementing the intervention, staff performance, job satisfaction and outcomes. The use of indicators for these various components is essential and must be incorporated into all ARV interventions or programmes from their initiation.

After the preconditions for the safe and effective use of ARV are met (see module 4), decisions regarding the implementation of the ARV intervention will be based on the assumption that a certain level of efficiency and efficacy will be delivered by the service. Staff performance, drug supply and clinical practical guidelines should all include quality control. The quality of the ARV service provided must be maintained and regularly reviewed to ensure that expected outputs are being delivered. Otherwise, failure to maintain efficient and effective services may jeopardise the future of ARV interventions.

Standards setting and the monitoring of quality and performance for drug counselling services may be a new process in many settings. Simple process indicators can be easily set up for this purpose, including items such as:

· The number of patients recollecting drug dosage schemes
· Pill collection from the pharmacy
· The length and type of each counselling session
· The number of patients who returned to discuss drug related issues
· The number of patients able to involve a "significant other" or a family member in treatment issues

Clinical monitoring is similarly carried out by reviewing process indicators such as:

· Number of patients attended
· Number of patients found eligible for the intervention
· Number of patients accepting intervention
· Number of patients with reduced incidences of:

HIV-associated morbidity
opportunistic infections
hospital admissions

· Number of patients with adverse reactions
· Number of patients with drug interactions
· Patient mortality

Monitoring adherence is critical as problems of drug resistance may result if adherence is inadequate. Adherence may be monitored through semi-structured or qualitative interviewing of patients or "significant others" to explore difficulties in adherence of drug, or laboratory surveillance of drug resistance in the community.

Feedback from the results of the monitoring processes should be provided to health care staff, thus allowing early detection and rectification of problems in service provision. Staff discussions should be encouraged to address issues which may improve motivation and performance.

Quality assurance from the patient or "user" perspective is complex to define, assess and monitor. Interviewing patients as they exit the service may be subject to bias, although important breakdowns in the quality of the service can be quickly identified in this manner. More sophisticated methods of determining patient satisfaction include focus group discussions and household surveys. Although these are difficult and timely to conduct, they may provide critical process information. Partnerships with PLHA support groups or community-based organisations may prove helpful in assessing the quality of ARV services delivered, as perceived by the patient.

I. Access to ARV Treatment

Technical complexity and costs of ARV therapy make it likely that ARV programmes will be piloted before widespread implementation occurs. Often, the intervention will be introduced in a limited number of facilities with the capacity to deliver an effective service. However, it must be noted that in many settings, the prospect of offering universal coverage with ARV programmes is remote, even if the goal is to reduce MTCT.

The issue of who will benefit from "limited access programmes" may result in serious operational problems in ARV programme implementation. In many countries, access will be limited to those who can afford to pay for drugs, even if subsidised, or to those who live close to specialist centres or district referral points. A few middle-income countries such as Brazil, ensure universal access and free ARVs as a legal right, but in doing so face managerial, quality assurance and sustainability challenges.

An effective referral system, with set criteria for eligibility, will be required at the introduction of ARV treatments. These criteria will have to consider issues of equity in determining access. Deciding who will partake in designing these criteria, as well as making information about the services widely available, is central to ensuring that the referral system operates effectively.

Equity in the provision of ARV treatments

The potential for ARV treatment services to be highly inequitable exists, particularly in high-prevalence countries with limited resources. In certain low-prevalence, middle-income countries, those in need may be able to access ARV treatment for all indications. Nonetheless, the rationing or subsidising of ARV treatment will need to be considered to avoid that the sole criteria for treatment access become financial.

Issues concerning equity of access to treatment and care for PLHA will differ for geographic regions and social groups. ARV provision within the public health system is likely to be centralised in cities, as this is where capacity is most likely to exist within the health system to implement quality ARV interventions. This places those patients from the periphery or from rural areas in a disadvantaged position. Referral arrangements for supporting patients from other areas will entail high familial costs, such as transport, loss of income, and accommodation.

Uptake will increase if services are more accessible, but if capacity and finances are limited, services may then be overloaded. Demand may exceed supply when ARV services are introduced and the initial need may often be to ration access rather than to broaden it, with the issues of distributive justice being important (see module 9). This will focus the ARV programme on the need to design "eligibility criteria", including clinical, psychological, economic and support criteria.

Guidelines for eligibility

National authorities must determine ways of obtaining broad consensus on ARV treatment interventions. Guidelines for eligibility will have to be drafted by national authorities based on these broad public consultations. Transparency in establishing criteria for entry into the ARV intervention programme is critical. These guidelines should also be made public and widely disseminated.

Guideline recommendations should be country and context specific, decided upon only after considerable debate. The debate itself, as a consultation exercise, is an important part of the ARV introduction process. It should be steered by the national committee guiding the ARV intervention programme, and will require that broad political support be obtained. It should be noted that this might prove to be a very time consuming process, therefore, this should be factored into the timetable of programme introduction.

Once these "guidelines for eligibility" have been determined, the process of implementation needs to be clearly delineated. Consistent monitoring of the intervention will be required.

V. Information Systems

There will be great public interest, expectation and confusion about any ARV programme as it is being planned. Many misunderstandings may result at the onset concerning ARV treatment. For example, it is often not understood that ARV does not cure HIV. It is critical that accurate and accessible information be regularly and widely disseminated about ARV interventions to ensure public knowledge. Messages should also be consistent with information provided from other HIV/AIDS activities.

Service provision and access

Clear and simple messages about ARV interventions must be prepared and widely disseminated once a final decision regarding ARV interventions has been determined. Public information campaigns should be released prior to programme implementation. A budget should be allocated for this purpose. The information should address:

· Who will be eligible for the interventions
· How ARV interventions may be accessed
· Advantages and disadvantages of ARVs
· ARV interactions with other common treatments
· Realistic expectations of ARV therapy (emphasising that ARV treatments are not curative)
· Specific services provided the benefits of:

Reducing the number of infants infected with HIV born to seropositive mothers
Protecting staff from accidents that may transmit HIV

Certain countries have established HIV treatment information services via the Internet. Information can continually be updated as new treatments and new drug interactions become know adverse effects are discovered. Pharmaceutical companies may collaborate in this endeavour, although monitoring to prevent bias will be necessary.

It may also be important to inform the public about the costs and financing of ARV programmes. This may result in considerable debate about fund allocation. Nonetheless, this process should be encouraged particularly where large public subsidising has occurred.

Information regarding the involvement of partners should be shared with the public. In situations where the private sector is collaborating with ARV programmes, information on which private practitioners have been licensed to implement the intervention, as well as which NGOs have been sub-contracted should be disseminated to the public.

Regardless of whether ARVs are available through the public health service or not, accurate information must be provided to the public in order to avoid misuse and/or misunderstanding of ARVs.

Prevention messages: integration and reinforcement

The need to inform the public about ARV treatment and related care services affords the opportunity to reinforce HIV awareness and HIV preventive messages. Prevention messages must be widely and repeatedly stated to guard against the idea that one can be less careful. Safer sex practices should continue to be used by those on ARVs and widely promoted.

VI. Partnerships for providing ARV services

The advantages of collaboration between various care providers at the institutional level (counsellors, physicians, social workers and nurses), welfare and community based organisations, social and spiritual services, and support groups has been witnessed in the provision of care for PLHA. Examples of successful continuum of care initiatives exist in several sub-Saharan African countries and Thailand. The introduction of ARVs allows for the opportunity to build on existing networks as well as expand or initiate comprehensive support services. Methods for collaborating with the private sector and NGOs have been discussed in module 2.

Table 6: Partners in ARV interventions

PLHA support groups

PLHA groups can be involved in the introduction of ARV treatments in a variety of ways. Furthermore, experience frorm several countries have shown that prevention and care is enhanced by including input from PLHA. Several ARV planning steps could benefit from input from PLHA groups including: Representation in the formation of the national committee guiding ARV interventions Guidelines for eligibility criteria Advocacy to increase access to ARVs Peer advocates for prevention and general HIV care Peer counselling and support to improve adherence to ARVs Evaluation of patient assessment of service quality

Community based organisations (CBO) and NGO

Many CBOs are active in HIV prevention and care. Moreover, /AIDS field, and a significant number have a particular focus on the care and support of PLHA people living with the disease. Like PLHA support groups, these CBOs play an very important advocacy role on behalf of their clients and patients. Therefore, they will need to be represented in the national debate on eligibility and entry and hopefully will have some firm and robust ideas to promote.

CBOs often focus on counselling and social support services that can be attached to HIV care units in institutions such as "Patient Support Units" to ensure comprehensive care and necessary linkages with community structures and homes to ensure continuum care.

Community focus of these organisations is extremely beneficial asstrenghts they can provide realistic information about ARV sinterventions and ensure that a clear understanding of what is involved in the intervention: how to access it; what the programme involves; the likely benefits; and, possible problems that may exist, and so on. Counselling provided by CBOs may also have the benefit of strengtheningmonitoring adherence to ARV drug regimens a very important possible problem in many countries where compliance for something like tuberculosis chemotherapy is often very low. However because relatively few have significant clinical capacity they are unlikely to be able to assist in ARV distribution.

CBOs may be able to help in reducing certain access barriers and, thus, minimising the inequity of most programmes for those living in rural areas or in the periphery of large cities, a long way away from ARV treatment centres. CBOs may also be able to provide transport services to individuals to access the ARV intervention programmes either through providing bus-fares or even by making available local accomodation. This sort of initiative needs to be discussed because a successful project to increase access may end up recruiting too many patients and clients for the ARV intervention to cope with.

CBOs and religious groups often provide emotional and social support for PLHA and their families. close to home. If As ARVs are not a cure for HIV, services must include comprehensive treatment of opportunistic and associated infections. If ARV interventions are not adhered to, or are not effective, palliative and supportive care will be particularly necessary. Establishing links between CBOs and ARV services is critical for ensuring that PLHA are provided a continuum of care should treatment be unsuccessful. Often, CBOs also involved in TB care (, the DOTS approach) and care for other chronic illnesses; therefore, there is great potential that HIV/AIDS may begin to be or can be de-stigmatised at the community level.

The private for profit sector

In many countries of both the developing and industrialised world, the private sector will play an increasingly important role in promoting and prescribing ARVs. An important The main initial issue in issue in developing a partnership between the national ARV initiative andwith the private sector will probably be to regulate the existing use and delivery of ARV drugstreatments. ARVs are often widely available through legitimate sources and/or donations from abroad. It is important to develop a partnership between the national ARV initiative and the private sector to regulate the existing use and delivery of ARV drugs to avoid inadequate prescription and/or patient care., and some form of could beis thought to be a widespread practice now in many resource-poor countries One worry is exploitation of poorly informed patients; the other is limited value for money - and the ever present spectre of poor use leading to high drug resistance rates.

Regulation and licencing can be promoted by having the government services running a series of training workshops (free or at subsidised rates) after which a certificate of training is issued. These could then be used by the private practitioner to show to the public his/her competence as opposed to competitors down the road wh did not have the certificate. Voucher schemes are also a possibility. The need is somehow to control or at least regulate the private market for ARVs to limit improper use and exploitation.

Partnerships rather than sanctions will be more likely to work.

There is a possibility of organising an extensive network of private practitioners across most countries for the delivery of ARV therapies. This model has already been implemented in some middle income countries. However this seems a limited option at least initially in many countries. Although manyost private practitioners maywill not have access to the necessary counselling, monitoring and laboratory facilities needed rigorously to supervise ARV treatment to set them up has considerable cost implications and this may generate significant duplication of equipment and resources strengthening links with hospitals where counselling and laboratory services could complement the private practice may be feasible.

There are several advantages of using however the private sector in the delivery of do quality-controlled ARV services, then, particularly if the public sector health service capacity itself is somewhat limited. Several possible ways can be envisaged whereby private firms or companies contract a private health care worker to provide ARV services with drug procurement from the public sector at an agreed price.

Caution is needed in some partnerships with the private for profit sector. It is of great importance that all those who can afford private ARV treatment (usually out of the pocket rather than any insurance programme) end up paying for it and liberating the subsidised space in the government services for those most in need. Or that those organisations who can afford to pay for their employees to receive treatment contribute fully to the drug and treatment costs.

The private not for profit NGO clinical services

In many countries a significant proportion of the population have access to, and use, NGO-managed clinical services. It therefore is obvious that iIf ARVs are to be scaled up in implementation beyond just a few specialist clinics in the capital city, then NGO-run/supported hospitals maywill be important sites in which to incorporate an expanded programme. With financial and technicaloverseas support from various sourcesand links, these NGO-run hospitals (including "mission hospitals") often have better infrastructureinfrasutucture, drug supplies and even resources than their government equivalent; and can be relied upon to provide relatively good-quality service. It is assumed that aAudit processes are becoming standard practice in NGO services; therefore,and the expected quality assuranceand audit components may also needs to be carried out.

The advantages of this are similar to any partnership with the private for profit sector. Government need not to be involved in the day-to-day management and supervision of such an intervention ; and can take advantage of the extensive NGO clinical capacity. Indeed f In terms of increasing access and equity, incorporating NGOs is vital as they may be the sole provider of services in a particular locality.

VII. Monitoring andMonitoring and reporting

7. Evaluation of the intervention

1. Data collection

Given the cost and complexity of any ARV intervention, and the often marginal(likely) tight cost-benefit and cost-effectiveness profiles, it is important that services are provided as effectively and efficiently as possible. SFrom the outsetpecial attention must needs be given to the monitoring and evaluation process, and. Important considerations include:

· in particular what data needs regularly to be collected;

· what system of reporting and sending data to a central surveillancesurveillence/monitoring unit will be utilised;

· how will data be analysed;

· and what use will be made of the summary results

Routine methods for collecting monitoring data shouldregularly must be established and properly implemented. A standard report form for, usable by all centres delivering the ARV intervention shouldshould be designed and piloted. This can then be modified for more wide-spread use, and then scaled up for general use. Pro-formas should be used in different settings and clinics for comprehensive evaluation. Staff will likely require ust have been properly trained and prepared training for the regular auditauditing purposes. Evaluation and audit need require active promotion promotion and ownership with staff, as the staff may initially find the process threatening and intimidating.

clinical efficacy as defined by outcome (eg reduction in vertical transmission; improvement in surrogate disease markers like viral load and CD4 count; etc)

amount of drug supplied and prescribed

integrity of drug supply

rates of HIV drug resistance (where appropriate)

behavioural aspects, especially HIV prevention activities

in-service training and refresher courses provided

specific problems which have arisen and affected service delivery (eg lack of drugs, problems in the laboratory; personnell issues)

Routine ways to collect such data regularly must be set up and properly implemented. A standard report form, usable by all centres delivering the ARV intervention, should be designed and piloted, and then scaled up for general use. Pro-formasThe same forms should be used in different settings and clinics for comprehensive evaluation. Staff must have been properly trained and prepared for the regular audit, otherwise this task will be poorly carried out - and thus lose much of its impact. Evaluation and audit It may be important torequire active promotion ewith the value of audit and evaluation as some staff as they may initially find the process threatening and intimidating.

2. Rreporting to a central unit

Once collected, monitoring data need regularly to be sent to a central unit which will process and analyse the information. It is probably best if dData should be are routinely reported every few months to the central unit. A clear chain must be set up, and when reports do not appear, a tracking device and default procedure organisedin play to ensure that the problem can be identified, the data collected, and steps taken to rectify the difficulties immediately implemented. It is best if, so they can be held accountable for delays and lack of data - or be rewarded for regular and efficient performance.

Such at the central unit may best be located at the national level, perhaps within the Ministry of Health. This to some extent However, this decision depends on whether a vertically or horizontally managed service is implemented. Although, even within a decentralised health service managed at the district level, a necessary role for central monitoring remains, particularly because of the cost and central purchasing and supply of ARVs.

Table 7: Data for monitoring ARV interventions

3. Acting on the data received

The final step in monitoring and evaluation is the use to which the data, analyses and results are put to. It is not enough just to collect the information, however efficiently this is done. TheThe data must be regularly processed and analysed. It may be best if a particular officer is identified for this responsibility. for the speed and accuracy with which such analysis is carried out and the appropriate evaluation and monitoring results are generated.

After this regular Regular review needs to be carried out of both the aggregated national and the local results. Regular audit should identify any problems with the integrity of drug supply. Cost-effectiveness and cost-benefit data could also regularly be generatedgererated if economic analysis and review is available.

Problems identified by "front-line" staffFrom the point of view of the staff at the local care facility, must be attended in the "front-line" as it were, it is very important that any problems identified - or generally recognised and reported upwards for some sort of action - toare actuall speedily and effectivelyy dealt with in a speedy fashion, otherwise. If not confidence in the system will deteriorate, and possible negative repercussions may include minimal effort being exerted into collecting the data, in the first place, further undermining the process.

Module 4. Safe and Effective Use of Antiretrovirals

Acknowledgements

The World Health Organization and UNAIDS would like to thank Dr Robert Colebunders, Institute of Tropical Medicine, Antwerp, Belgium for preparing the text of this module;

Dr David Cooper, University of South Wales, Sydney, Australia, Dr Danielle Mercey, University College London Medical School, London, UK, Dr Mary Ann De Groote, University of Colorado, Denver, Colorado, USA, Dr Kevin De Cock, Centers for Disease Control and Prevention, Atlanta, USA, Professeur Jean-Pierre Coulaud, Hal Bichat-Claude Bernard, Paris, France, Dr Badara Samb, Institut de Mcine et Epidologie, Hal Claude Bernard, Abidjan, Cd’Ivoire, Dr Mary Couper, Division of Drug Management and Policies, WHO, and Dr Pedro Cahn, Fernandez Hospital, Buenos Aires, Argentina, for their most thoughtful reviews of the draft text; and

Dr Rachel Baggaley, Dr Susan Fernyak, Ms Alison Martin Katz, Ms Marguerite Nguyen, and Dr Eric van Praag, Office of HIV/AIDS and Sexually Transmitted Diseases, WHO, and Dr Joseph Perri, Policy, Strategy and Research (PSR), UNAIDS, for coordinating the project.

©World Health Organization and Joint United Nations Programme on HIV/AIDS, 1998

This document is not a formal publication of the World Health Organization (WHO) and Joint United Nations Programme on HIV/AIDS (UNAIDS), and all rights are reserved by them. The document may, however, be freely reviewed, abstracted, reproduced, or translated in part, but not for sale or for use in conjunction with commercial purposes. It may also be reproduced in full by non-commercial entities for information or for educational purposes with prior permission from WHO and UNAIDS.

For authorization to translate or reproduce the work in full, and for any use by commercial entities, applications and enquiries should be addressed to the Office of HIV/AIDS and Sexually Transmitted Diseases, World Health Organization, 1211 Geneva 27, Switzerland, which will be glad to provide the latest information on any changes made to the text, plans for new editions, and the reprints and translations that are already available.

The mention of specific companies or of certain manufacturers’ products does not imply that they are endorsed or recommended by either WHO or UNAIDS in preference to others of a similar nature that are not mentioned. Errors and omissions excepted, the names of proprietary products are distinguished by initial capital letters.

The views expressed in these documents by named authors are solely the responsibility of those authors.

This module provides guidance primarily to clinicians, both doctors and nurses, counsellors, and managers of clinical services. Policy makers, people living with HIV/AIDS (PLHA), and mid-level decision makers in major and district hospitals, and in training institutions, may also find this guidance useful.

1. Introduction

In recent years powerful antiretroviral (ARV) treatment regimens have become available. So far, with a few exceptions, only people living with HIV/AIDS from industrialised countries have benefited from these treatments. In a proportion of patients, triple combination ARV therapies reduce viral load plasma levels to undetectable levels and have a beneficial clinical effect. HIV-related symptoms may disappear, the incidence of opportunistic infections is reduced and quality of life improves. Triple therapy has decreased dramatically the number of hospitalisations for HIV-related illness in industrialised, and some middle income, countries and prolonged the lives of many people with HIV.

However, these treatments are not a cure for AIDS and their long term effectiveness is not yet known. Resistance may develop even in patients who adhere strictly to the regimen. Combination therapy including protease inhibitors is most effective in patients who have never been treated with ARVs. In patients who have already taken ARVs however, up to 40% treatment failure has been observed after one year of treatment. HIV may persist in “sanctuary sites” where the drugs cannot suppress viral replication. Studies have shown that after 20-30 months of successful combination ARV treatment, HIV can still be isolated despite patients having undetectable HIV RNA plasma levels. This suggests that ARV treatment cannot be stopped; it is a treatment for “life” unless or until it fails, or until current regimes are superseded by new drug combinations.

Current ARV therapy is far from ideal. The regimens are complicated, requiring between 2 and 20 pills a day, they may cause severe side effects, they require close and often expensive laboratory monitoring, and there may be interactions with other drugs. Finally, the drugs themselves are very costly. In May 1998, the annual cost of triple therapy including a protease inhibitor amounts to US$ 10,000 - 15,000. ARV treatment guidelines change rapidly and differ between countries. It is expected that more powerful drugs and easier regimens will become available in the near future.

Currently available ARVs belong to two major classes of drugs: reverse transcriptase inhibitors (RTIs) and Protease Inhibitors (PIs). RTIs are further divided into Nucleoside Reverse Transcriptase Inhibitors (NRTI) and Non-Nucleoside Reverse Transcriptase Inhibitors (NNRTI).

All these drugs target enzymes which are important for RNA replication and viral functioning. Once the HIV retrovirus has invaded a macrophage or T-lymphocyte, the enzyme HIV reverse transcriptase converts the viral RNA genome into a DNA copy, which is then integrated into the host chromosome by the enzyme integrase. A third enzyme, protease, contributes to viral functioning by catalysing the cleavage of viral core proteins for the final assembly of viable virions. Once HIV is integrated into the host DNA, the virus multiplies rapidly, creating several billion new copies a day. This inevitably results in mutations of the viral genome. Some of the mutations may confer resistance to one specific agent or to a whole class of ARVs.

In most industrialised countries eleven antiretroviral agents have been approved for treatment. Other classes of drugs are under development. (See section on specific drugs including Tables 9-14 for detailed information on each of the drugs.)

Good adherence to the ARV regimen is essential to avoid development of resistant strains. However, the regimens are complicated and adherence is difficult even in the most favourable circumstances; a very supportive environment, materially and socially, is required.

Combination therapy requires taking drugs twice or three times a day at regular time intervals. There are often adverse effects, and the drugs have to be taken for life. One of the protease inhibitors (Ritonavir) requires refrigeration and other ARVs must be taken with large quantities of water (see drug data sheets at end of the document). Patients may also have to take other drugs to prevent or to treat opportunistic infections. Adherence to such a complicated treatment for a person who has little knowledge about the disease, has many other problems, eats only when food is available and has no watch, is not an easy task. A physician who has only a few minutes per patient cannot adequately explain the regimen and support the patient’s adherence. Moreover, because of the stigma associated with HIV infection, individuals may be reluctant to reveal their seropositivity to others. Therefore, except for the health care staff, no one will be able to help them to adhere to their ARV regimen.

Many clinicians do not have enough time to provide adequate counselling about HIV. If counsellors are available, they will need special training in ARVs and will work closely with the prescribing clinician. In many settings clinicians do not have enough time per consultation to explore all the client’s concerns and questions about ARVs and adherence. The client may in any case feel uncomfortable discussing “non-medical” issues with his or her doctor and may prefer talking to a counsellor. The physician will have to make extra time to counsel patients on ARV treatments, and he/she may need extra training to improve counselling skills or may choose to work closely with a counsellor.

2. What needs to be in place before initiating ARV therapy

Due to their high cost , the complexity of regimens and need for careful monitoring, certain services and facilities must be in place before considering the use of ARVs in any situation.

¹See Module 8 on drug regulatory mechanisms and safeguards for storage, distribution and prescription for further details.

²See Module 5 laboratory requirements for safe and effective use of ARVs for further details.

Advantages and disadvantages of ARV treatment

There are advantages and disadvantages to using ARVs at any stage of HIV disease and it is important for the physician and patient to consider these before embarking on difficult and costly treatment.

3. When to initiate antiretroviral therapy

As yet it is not clear when is the best time to start combination therapy. The recommendations for ARV treatment in industrialised countries are often based on virological criteria, but not on clinical outcome or quality of life data (proven clinical efficacy).

Table 1 Suggested priority to be accorded to ARV treatment according to disease stage and available laboratory facilities

*Depending on available resources, CD4 count and/or viral load results, bitherapy, or ideally tritherapy, could be considered.

**Depending on available resources bitherapy, or ideally tritherapy could be considered.

***Start ARV treatment based on CD4 count and/or viral load results. With only a small budget it is probably best not to start too early with ARV treatment to allow the continuation of this treatment once it has been started. A maximum benefit on the quality of life of patients is expected in patients with CD4 counts < 200/mm³ and/or a viral load > 30.000-100.000 copies/ml plasma.

3.1 Acute HIV illness

In industrialised countries ARV treatment is sometimes recommended in the acute phase of the illness. The data currently available however suggest that even with early, highly active antiretroviral therapy (HAART), eradication of HIV is not possible. Starting ARV treatment during the acute phase of the illness is very expensive as it must continue for life. Moreover, an acute HIV illness is generally not recognised in many developing countries because such episodes are difficult to distinguish clinically from other infections prevalent in these parts of the world, such as malaria or typhoid fever.

When resources are limited, there is therefore a strong case against using ARVs for this indication.

3.2 Asymptomatic patients

In most industrialised countries ARV treatment is recommended in asymptomatic patients with plasma HIV RNA levels greater than 5,000 to 10,000 copies per ml regardless of the CD4 cell count. However, even in industrialised countries, many asymptomatic patients choose not to take ARVs, because although there may be theoretical reasons to take ARVs, the complex and unpleasant regimes may outweigh the benefits at this stage. Furthermore many people are deciding to wait rather and remain “ARV naive” so that when more powerful, simpler and less costly drugs become available they can still benefit. ARVs often cause side effects, the regimens are complex and the long term benefits have not been demonstrated for this specific group of patients.

Disadvantages of “early combination ARV treatment” in asymptomatic HIV disease

Treatment of asymptomatic people in resource poor countries is problematic. Firstly, it has to be determined whether an asymptomatic patient really is HIV infected. The HIV test result should therefore be reliable and accompanied by adequate counselling. Recently the World Health Organisation and UNAIDS have developed new guidelines for HIV testing (Revised recommendations for the selection and use of HIV antibody tests, 1997). A positive HIV test (rapid or ELISA) should be confirmed by at least one other test (also rapid or ELISA). Resources should be available to implement these guidelines.

Secondly, to decide when to start ARV treatment in an asymptomatic patient, it is recommended that a reliable CD4 lymphocyte count and viral load test be performed. Both tests require a sophisticated laboratory and trained personnel. At the time of writing a viral load test costs about US$ 100 per test. Some commercially available viral load tests perform less well with certain HIV genotypes that are frequently seen in poor countries, though the most recent tests have largely overcome this problem.

Thirdly, in many countries the stigma associated with HIV remains high. Women are often unable to tell their partner that they are HIV seropositive because they risk being abandoned. A person may lose his/her job when it is known that he/she is HIV seropositive. People may be reluctant to accept ARV treatment because having to take so many pills is difficult to hide and may reveal their seropositivity to others. Furthermore, in poor countries with limited ARV availability, access to ARVs may be limited to symptomatic patients, as the priority group. (See Modules 2 and 3 for further details)

Thus in countries where resources for ARV treatments are limited, it may be difficult to argue for treatment for asymptomatic individuals.

3.3 Symptomatic patients

ARVs are certainly beneficial for symptomatic patients. Opportunistic infections occur less often and patients may even become completely asymptomatic for a substantial period of time. Current recommendations are however, to continue with prophylaxis against certain OIs, (such as cotrimoxazole for pneumocystis carinii pneumonia in industrialised countries and for pulmonary infections and septicaemia in developing countries, and isoniazid prophylaxis for TB in HIV-infected people at high risk of developing TB, especially in developing countries) even if CD4 count rises. There are many case reports of patients with CMV, cryptosporidiosis, wasting or chronic diarrhoea whose symptoms have resolved after starting triple therapy.

Disadvantages of ARVs in symptomatic patients in poor resource settings.

Despite these possible disadvantages, treatment of symptomatic patients is the norm in industrialised countries. In developing countries where resources are available for ARV treatment, treatment of symptomatic patients should take priority over treatment of asymptomatic patients.

3.4 Terminal HIV and palliative care

Patients with terminal HIV disease, such as widespread malignancy and patients who are severely immunocompromised (CD4 <10) will rarely respond to ARV therapy and initiating therapy at this stage may be a waste of limited resources and a burden to the patient. As stated earlier, ARV therapy should be provided in the context of a continuum of care for PLHA. If treatment fails or patients have terminal HIV and do not respond to ARV therapy, adequate palliative care and emotional and spiritual support should be provided. Referral to a home based care team, if available, for management of pain and symptomatic relief is important if the patient chooses not to remain hospitalised. It is also important to have bereavement counselling available for carers and family members.

4. ARV treatment in clinical practice: the steps to follow

Table 2 lists the steps that should be followed by the clinician when considering initiating ARV therapy at any stage of HIV disease.

Table 2

5. Which antiretroviral regimen to initiate

5.1 Triple combination therapy

The treatment of choice is triple combination therapy including at least one protease inhibitor with potent in vivo activity (indinavir, ritonavir, nelfinavir and the new formulation saquinavir). An alternative may be triple combination therapy with ritonavir, saquinavir and an NRTI. Less potent regimens are 2 NRTIs + 1 NNRTI. It is generally recommended that the combination should include a drug that penetrates well into the brain such as zidovudine or stavudine.

Nucleoside analogue combinations that could be used in triple therapy are:

· zidovudine (ZDV) + didanosine (ddI)
· zidovudine (ZDV)+ lamivudine (3TC)
· stavudine (d4T) + lamivudine (3TC)
· stavudine (d4T) + didanosine (ddI)

The above are roughly equivalent. Zidovudine (ZDV) and zalcitabine (ddC) in combination are less active and are regarded as second tier.

Certain combinations are not advisable, because of overlapping toxicity:

· didanosine (ddI) + zalcitabine (ddC)
· stavudine (d4T) + zalcitabine(ddC)

or because of an antagonistic effect due to overlapping intracellular phosphorylation pathways:

· zalcitabine(ddC) + lamivudine(3TC)
· stavudine(d4T) + zidovudine (ZDV)

Table 3 Examples of triple therapy regimens including a protease inhibitor

*PI with potent in vivo activity: indinavir, ritonavir, nelfinavir

**The best alternative PI after failure on an initial PI containing regimen is unknown.
Cross-resistance between indinavir and ritonavir is almost total.

5.2 Double combination therapy

Double combination therapy (2 NRTIs) produces clinical and virological improvement, but is inferior to triple therapy . However, many patients in industrialised countries continue to take bitherapy either because they do not want to take more complicated tritherapy regimes or because they do not want to “use up all their options.” Double therapy could also be considered if patients do not tolerate triple combination regimens or if there are contraindications for the use of protease inhibitors. Double therapy can also be used where protease inhibitors are not available or affordable. Double combination therapy with 2 NRTIs is generally well tolerated and is easier to monitor than triple combination therapy.

When double therapy with two nucleoside analogues is failing, adding only a protease inhibitor without changing the NRTIs (because alternative ARVs drugs are not available) may lead to rapid resistance against protease inhibitors since the protease inhibitor then behaves as monotherapy. Double combination of one NRTI and one protease inhibitor has a stronger antiviral effect but also leads to resistance against the protease inhibitor.

6. How to monitor ARV treatment

It is advisable for patients on triple therapy to be seen monthly; particularly at the start of treatment. Once stabilised, patients may then be seen every three months. At each visit, side effects and adherence to the treatment should be discussed in depth. At the start of treatment, routine haematological and biochemical tests should be done monthly to detect potential side effects. Careful monitoring is needed particularly in patients who abuse alcohol or are infected with hepatitis B or C, who have abnormal liver function, and in patients with abnormal renal function. The appearance of certain symptoms such as nausea, anorexia or abdominal pain is a reason for repeating certain laboratory tests.

Ideally, if available, viral load and /or CD4 count should be measured regularly, e.g., in untreated patients every 6 - 12 months and in patients on ARVs every 3 - 6 months.

The results of viral load testing should be interpreted with caution. Certain viral strains that are particularly frequent in developing countries may be difficult to detect with commercially available testing methods; and viral load levels may vary according to the technique that has been used, the laboratory where the test has been done, the time and the way the sample was transferred to the laboratory. Levels may increase after a recent infection or vaccination. An increasing viral load does not necessarily mean that the ARV treatment is not effective. It may be that the patient is not adhering to the treatment. There may also be problems in measuring CD4 counts in developing country settings (for further details on laboratory monitoring see Module 5).

Treatment decisions should take into account these possible variations and the clinical condition of the patient, information about adherence and availability of ARVs to which he/she could switch. Ideally a viral load measurement should be done before starting and before changing ARV treatment; and 1 to 2 months after the start of the initial or new treatment. Where viral load testing is not available CD4 counts can be used instead to guide initiation and monitoring of ARV treatment. CD4 lymphocyte counts are also used to decide whether the patient should receive prophylactic treatment against opportunistic infections. Ideally, in countries where ARVs are introduced, a reference laboratory should be established where these tests could be performed (see Module 5 for further details).

In developing countries where CD4 counts or viral loads are not routinely available, clinical indicators such as weight and total lymphocyte count will give some indication of disease progression and treatment response.

Table 4 Laboratory monitoring during ARV treatment

* Advisable, but not essential

**Highly recommended, but not essential

7. Changing antiretroviral therapy

7.1 When to change?

Reasons for changing ARV treatment are: treatment failure, toxic effects, intolerance, non-adherence, and current use of a sub-optimal treatment regimen (Table 5). In industrialised countries definitions of treatment failure are based on viral load measurements - a rising plasma viral RNA level or failure to achieve the desired reduction in plasma viral load. A declining CD4 count is also an indication to change treatment (see Module 5 for further details). In settings where it is not possible to do viral load tests and CD4 counts, weight loss and other clinical changes may provide a good indication of the need to change treatment.

Asymptomatic patients currently taking double NRTI combination treatment with undetectable plasma HIV RNA levels should be closely monitored. If plasma HIV RNA levels increase above 5,000 to 10,000 copies per ml, the treatment regimen should be changed. In the absence of laboratory facilities to measure viral load, decisions to change treatments should be based on results of CD4 lymphocyte counts and/or clinical condition, including weight of the patient.

Factors other than viral resistance that can lead to loss of viral suppression are: non-adherence to treatment and intercurrent illnesses. To evaluate the efficacy of an ARV regimen, viral load should not be measured in the days following a vaccination. If during ARV treatment a patient develops side effects, the offending drug should be replaced by another drug with a different toxicity profile. If the cause of the side effects is not clear, a brief and complete interruption of the full therapeutic regimen is generally preferred. Because ARVs may have to be changed it is not ideal to start ARV treatment when only a limited number of ARVs are available - often the case in developing countries. If the ARV drug is stopped because of intolerance or side effects and no alternative treatment is available, resistance may develop rapidly.

Table 5 Indications for changing therapy

7.2 What to change to?

When treatment failure occurs all drugs in the regimen should be changed if possible, because it is currently impossible to determine which of the combination’s drugs is the one to which resistance has developed. Changing both drugs in a double NRTI combination regime or changing a protease inhibitor and a NRTI in a three-drug combination is advised. Simply adding a protease inhibitor or any other drug to a failing regime is never recommended.

Table 6 Predicting the effectiveness of switching therapy

*There is a great deal of cross resistance between the protease inhibitors, therefore switching protease inhibitors in cases of treatment failure is rarely successful.

When changing because of non-adherence, the reasons for non-adherence must be explored. For example, if a patient does not like to take three times daily (tds) regimen, a twice daily (bd) regimen may be proposed. If a change in therapy is needed because of drug toxicity, only the drug causing the toxicity should be replaced.

8. Special considerations

8.1 ARV treatment in women of childbearing age

Before considering ARV treatment in a woman, a pregnancy test may be required. Data on ARVs in pregnancy is limited, particularly in the first trimester. A study in the USA is currently assessing potential foetal risks. So far we know that zidovudine given in the second or third trimester does not cause congenital malformations. The long-term effect of zidovudine on the child’s health remains unknown. As regards the other ARVs, very little is known about their potential toxicity for the foetus (see Module 6 for more details).

Triple therapy, therefore, should probably not be prescribed for women during early pregnancy or for women at risk of pregnancy. Contraception is recommended for all women considering combination ARV treatment. When a woman treated with bi- or triple combination therapy becomes pregnant, the risks of such treatment for the foetus should be discussed. The physician or counsellor should discuss family planning with all women before starting ARVs.

If a woman becomes pregnant whilst taking ARVs the counsellor will have to help her decide whether to continue with the pregnancy or whether to have a termination, if this is available. The desire to have a child for many women is of great importance and may override medical concerns. Once the decision has been made to have a baby or to continue with a pregnancy this decision should be supported by the counsellor and ARV therapy may have to be modified during this period. The possible detrimental effects to the mother of altering her ARV treatment during pregnancy must also be discussed.

8.2 ARV treatment in children with HIV (see Module 6 for further details on ARVs to reduce mother-to-child transmssion).

In infected children in industrialised countries ARV treatment is continued after birth particularly in children with a poor prognosis (high viral load, symptomatic). If a child becomes infected despite zidovudine prophylaxis, ideally the zidovudine treatment should be replaced by a new ARV treatment regimen. In poor resource countries PCR testing is generally not available to confirm an early diagnosis of HIV infection. ARV treatment for children should only be considered if a definite diagnosis of HIV is made.

Experience with triple therapy including a protease inhibitor in children remains limited. However early reports are encouraging and ARVs may be indicated, particularly for symptomatic children. So far only a few licensed ARVs are available in paediatric formulation: ritonavir syrup which is difficult to tolerate and nelfinavir powder which is not widely available. For the moment no indinavir syrup is commercially available. Obtaining good compliance with ARVs may be particularly difficult with children. Monitoring ARV treatment in children < 1 year old requires extra expertise. CD4 lymphocyte counts in children are much higher than in infected adults. Viral load is also generally higher in children.

8.3 HIV-2 infection

All the ARVs are effective against HIV-2 infection except the NNRTIs. ARV treatment with potentially toxic drugs is probably not appropriate in asymptomatic people with HIV-2 infection as HIV-2 infection progresses more slowly than HIV-1 infection. The monitoring of ARV treatment in people with HIV-2 infection is also complicated by the fact that there are no commercially available tests to measure viral load.

8.4 ARVs in patients with tuberculosis

In HIV infected patients being treated for tuberculosis, protease inhibitors increase the serum levels of rifampicin. On the other hand, rifampicin decreases the level of protease inhibitors. Therefore HIV infected tuberculosis patients should have double therapy without a protease inhibitor or no ARV therapy until TB therapy is complete. (Table 7). Before considering ARV treatment, patients should be screened for active tuberculosis infection, with at least a sputum examination and a chest X-ray.

TB preventive therapy (TBPT) has been shown to reduce the incidence of TB in people with HIV. The most commonly used regimen, isoniazid, is not contraindicated if patients are taking ARVs. Rifampicin containing TBPT should, however, be avoided.

Table 7

8.5 What should not be done

9. Counselling for ARV therapy

The psychosocial aspects associated with deciding whether to start and then continue on ARV treatment are as important as the medical aspects. Counselling for ARVs is often time consuming and physicians may choose to work with a trained counsellor with specialised skills. If appropriate and informed decisions are to be made, and treatment is to be provided safely and effectively, counselling services must be available to people living with HIV/AIDS (PLHA). Those who make the decision to start ARVs will need information, support and encouragement on a regular and long term basis in order to maintain adherence to the regimen.

Where ARVs are readily available, the possibility of obtaining these treatments is an incentive to seek out counselling and testing. It has been shown that one of the main barriers to knowing one’s HIV status, particularly in high prevalence areas, is the lack of perceived benefit for those who fear they might be seropositive.

Counsellors and clients will be dealing with very different psychosocial and material problems depending on the setting.³ Where ARVs are not available or are prohibitively expensive, knowledge of these treatments may create additional stresses, for both the client and the counsellor. In such situations, counsellors will be helping clients cope with the anguish of knowing that a treatment exists which is not accessible to them. Where ARVs are provided routinely, counselling will focus on encouraging correct use, coping with a return (albeit temporary) to a more healthy and productive life, and overall emotional well-being. In settings where the client and his/her family will be investing large sums of money in the drugs, problems relating to this use of scarce resources will have to be discussed.

³for additional information see UNAIDS Technical Update November 1997 on Counselling and HIV/AIDS

ARVs have received a large amount of favourable coverage in the popular press. Even in low income countries many people with HIV are knowledgeable about ARVs and have unrealistic expectations about their availability and their efficacy. It is very important for the counsellor to inform the client about the likely availability of ARVs - in both pre- and post test counselling.

After receiving a positive result, the client will need emotional support from his or her counsellor, in addition to information about many aspects of treatment and prevention. Discovering one is HIV seropositive is traumatic and marks the beginning of a difficult psychological adjustment for the client. Psychological distress and depression may be alleviated by sensitive counselling. Several counselling sessions may be needed initially for the client to cope with their positive status in addition to understanding the implications of ARV therapy and its monitoring. Even in countries where ARVs are readily available many patients testing seropositive will not be at a stage of disease to warrant or desire ARVs and this will need careful discussion.

Whatever decision is made about taking antiretroviral drugs, comprehensive care and social support need to be ensured on a long term basis, to identify, prevent and treat opportunistic infections and to deal with social and family problems. This may require collaboration between medical and social services in public, private and NGO sectors.

9.1 Confidentiality and sharing HIV results

The disruption of life style brought about by complicated ARV regimens should not be underestimated. Involving a partner or significant other in ARV treatment will make taking ARVs much simpler. The counsellor should encourage sharing HIV results with a partner or close relative so that the burden of the drug taking schedule can be understood and shared. The counsellor should stress the importance of informing partners of the risk of HIV transmission and taking protective action.

In many countries, women risk being abandoned if their positive status is discovered; it is frequently the partner who has infected the woman and overall it may be better not to disclose her status. The clients’ own perception of the risks and benefits of the particular situation should be respected. However, many health workers feel they have a duty to inform and thereby protect a possibly uninfected partner and that this duty is greater than the duty to protect confidentiality of the client. These are ethical issues which remain unresolved. Certainly in countries where the infected and abandoned partner will not be left destitute and where people can easily take preventive measures, some people would feel that there would appear to be no justification for protecting confidentiality over risk of transmission of a fatal infection.

9.2 Financial commitment

Discussing how the drugs are going to be paid for, is very important. In many developing countries the only option is for the patient or his/her relatives to pay for treatment. They may think that they need only take ARVs for a few months, or intermittently when they have funds. If relatives are paying, they may have to be involved - again to make decisions about willingness to pay for treatments indefinitely and the need for a consistent supply of drugs. These decisions may be very painful and difficult, but need to be thought through before embarking on treatment.

9.3 Particular challenges

Complicated drug taking schedule:

The counsellor will have to deal with many of the problems associated with complex life long therapy. A “drug time table” helps clients to organise their drug taking schedule.

Table 8 Example of a daily drug regimen

Not a cure: The counsellor must ensure that the client is fully aware that ARV therapy is not a cure, and that the drugs must be taken for the remaining time the client has to live, or until they are too sick to take them any longer. He/she should ensure that the client understands that ARVs are new drugs and no one knows whether their short term benefits, which are often great, will continue in the long term. The counsellor needs to provide accurate up to date information about the possible positive and negative outcomes of treatment, taking into account all personal and contextual factors which may influence the outcome.

Adverse effects: Adverse effects are common in the first few weeks of ARV treatment. The counsellor should be aware of these and reassure the client that these initial adverse effects will usually lessen with time. The counsellor should also be able to help the client distinguish between potentially serious and non-serious signs and symptoms.

Disruption of life style: Dietary changes may need to be made as many of the drugs are affected by the presence and types of food eaten. Meals often have to be planned carefully around the drug regimen. This can be inconvenient and disrupt family and social life. If the counsellor can involve the family in discussions about these issues, it will help them to understand the importance of timing meals and changing routines. The counsellor may have to take time to work out a "meal and drug taking time table" that fits in with the client’s and the family’s life style. Client and counsellor should also discuss what to do if a dose is missed or a meal time rearranged.

Coping with "health”: Taking ARVs will be for life, and once stabilised on a particular regime, the client may feel very well. This may be particularly true for people who were symptomatic prior to ARV therapy. They may be well enough to resume more of their previous lifestyle - going back to work and taking up previous hobbies or activities that were impossible before. Paradoxically, some people find the adjustment to "health" after a prolonged period of ill health difficult. Friends and family may continue to treat them as a patient despite them feeling well. The new situation and perspective will require understanding and adjustment on the part of the patient and the family.

Guilt: Some people describe feelings of guilt about taking ARVs. They may have lost partners or friends because ARVs were unavailable or could not be afforded. Many people are painfully aware of the enormous sacrifices that their friends and/or family are making to provide ARVs for them.

Coping with a "medicalised" life style: People with HIV who are asymptomatic when they start taking ARVs may resent the constraint that taking the drugs imposes on their lives. This has to be acknowledged and explored when starting ARVs. People who feel unable to embark on the strict regime that ARVs will impose on them may do better to postpone treatment.

Coping with treatment failure: Although the short term follow up of people on ARVs often results in remarkable improvement in symptomatic patients, this improvement is not universal. Furthermore, symptomatic improvement may not last long, particularly if resistance to ARVs develops, or the patient has previously been on sub-optimal ARV therapy. In developing countries, some people with HIV will put off starting ARV therapy, because of the prohibitive cost, until the last possible moment. They may by this stage be severely immunocompromised. People who have very low CD4 counts, or are terminally ill, when they start on ARV therapy tend to respond less well.

The counsellor will have to support them through the disappointment of treatment failure and balance optimism and realistic caution. Depression and despair are common when CD4 counts do not go up or weight is not gained as has been predicted or read about. This may be made worse when the client is aware of the draining of his or her financial resources into a treatment that may be futile. There may be a time when the client and counsellor will have to discuss ceasing treatment and preparing for death. There is no place for antiretroviral drugs in palliative care for severely ill patients.

9.4 Adherence issues

Combination therapy involves taking between four and 20 capsules/tablets a day depending on the regimen prescribed. Understanding how to take the drugs and the importance of not missing doses should be explained by the counsellor. The information about HIV is complicated and often confusing for clients and should be followed up with written, or clearly illustrated, instructions for future reference.

There should be open access to a counsellor for information as the need arises, and if possible, involvement of a partner or significant other, to help remind the patient about taking drugs and watching out for adverse effects. It is also important that the client understands the need for careful monitoring by a doctor and regular medical check ups.

How to improve adherence to ARV treatment?

9.5 ARVs must not detract from HIV prevention messages

People who take ARVs will understand that the aim of this medication is to lower the amount of virus in their blood, often to undetectable levels. They may conclude that if their viral load is very low they do not need to use protective measures to prevent HIV transmission. There have been anecdotal reports of people abandoning condom use when on ARVs. The counsellor should stress that the virus can still be transmitted and that safer sex should continue.

Alcohol and recreational drugs

There is limited information on how any of the ARVs interact with alcohol or recreational drugs. A study looking at factors that influenced HIV disease progression in long term survivors found that cannabis use, diet, and alcohol consumption, did not significantly influence disease progression. Therefore counsellors should advise that moderate alcohol intake and cannabis smoking are not contraindicated. People taking ARVs should be encouraged to lead as normal a life as possible. However, there is evidence that many recreational drugs (for example “ecstasy”, amphetamines (speed) and diazepam (Valium)) interact with PIs (especially ritonavir). Any client taking recreational drugs should be encouraged to discuss this with his/her counsellor.

To the extent, however, that alcohol and any other drugs may make people less alert and responsible and therefore less able to adhere to the regimen, excessive use should be discouraged.

Support groups

Many people with HIV have found mutual support in groups of people living with HIV, very helpful. When embarking on ARV therapy it may be useful to be in contact with other people who are having similar treatments, share experiences and discuss overcoming problems associated with taking ARVs. Support groups for other medical conditions such as breast cancer have been shown to significantly improve prognosis. Clinicians and counsellors need to be aware of support groups in their area and proactively refer patients.

Counsellor support

Counselling people with HIV can be very stressful for the counsellor, especially when looking after clients and their families when a treatment fails. Regular supervision and a counsellor support group to share complicated or distressing cases provide valuable support.

Antiretroviral drugs

Nucleoside Reverse Transcriptase Inhibitors (NRTI)

(Tables 9 and 10)

Zidovudine (ZDV) often called AZT.

(Trade name Retrovir)

Zidovudine is an analogue of the nucleoside thymidine. After the drug is phosphorylated by cellular enzymes it competitively inhibits the incorporation of thymidine into the proviral DNA. The drug inhibits the prolongation of the DNA chain by preventing the addition of further nucleosides. Penetration into the cerebrospinal fluid is about 50 - 70 % of plasma levels.

Dosage and administration

250 or 300 mg bd orally

200mg tds orally

In patients with AIDS dementia complex a dosage up to 1 g a day can be considered.

Zidovudine can be taken with food or on an empty stomach but taking it with food may reduce feelings of sickness.

Adverse effects

Nausea, headache, muscle pain, skin rashes, insomnia and fatigue occur relatively frequently when starting zidovudine treatment. In most patients these complaints disappear within the first six weeks of therapy.

Haematological toxicity may develop within one month of the start of the zidovudine treatment, but generally anaemia occurs in patients with advanced disease. In patients without symptoms treated with 500-600 mg a day, haematological toxicity is rare (severe anaemia in 2% of the patients after 18 months of treatment). The mean corpuscular volume of red blood cells will increase in all patients receiving zidovudine. Long term use of zidovudine may lead to myopathy which may be preceded by an increase in creatine phosphokinase. In most patients, myopathy is reversible within 8 weeks of stopping the zidovudine therapy. Fatty liver and lactic acidosis have occasionally been reported: zidovudine should be stopped in patients with progressive hepatomegaly or if amino transferase levels rapidly increase. Bluish pigmentation of nails and mucosa is another uncommon adverse reaction.

Drug interactions

Zidovudine should not be given with stavudine because of viral antagonism. Zidovudine may cause bone marrow suppression and must be used with caution with other drugs commonly used in HIV disease such as cotrimoxazole (septrin) and gancyclovir that also cause neutropenia and anaemia.

Contraindications/precautions⁴

Precautions when prescribing all ARVs during the first trimester of pregnancy (patient must be fully informed of possible risks compared to benefits). See Module 6 for details of prescribing ARVs during pregnancy.

Do not start zidovudine in the presence of severe anaemia (haemoglobin < 7g/dl) neutropenia or leucopenia (neutrophils/leukocytes < 750/mm³).

Combination with other potentially haematotoxic drugs such as cotrimoxazole (septrin), pyrimethamine and ganciclovir should be avoided.

Stop zidovudine if haemoglobin levels decrease below 7 g/dl, if neutrophils/leukocytes < 750/mm³ or if persistent myalgia.

Didanosine (ddI)

(Trade name Videx)

Didanosine is an adenosine analogue. In contrast with zidovudine and stavudine, didanosine (as well as zalcitabine and lamivudine) has a greater activity in resting peripheral mononuclear cells infected with HIV than in activated cells.

Didanosine is rapidly destroyed by exposure to acid. Tablets therefore contain an antacid to prevent drug degradation and to maximise bio-availability.

Dosage and administration

200 mg bd orally for a person ³ 60 kg
125 mg bd orally for a person < 60 kg
400mg od/250-300 mg od <60 kg

Tablets should be taken on an empty stomach. No food should be taken for 1 hour afterwards The tablets must be dispersed in water, chewed or crushed thoroughly before swallowing. The flavour of the dissolved tablets can be improved by adding clear apple juice.

Adverse effects

No haematological side effects.

Minor side effects such as nausea, vomiting, diarrhoea, skin rashes, headache, itch and lethargy have been reported but these usually settle after the first 6 weeks of treatment.

The major toxicity is a dose related, predominantly sensory, symmetric polyneuropathy. This neuropathy is generally reversible within weeks of stopping treatment. If the didanosine however is continued too long the polyneuropathy may become irreversible (the same is true for the other neurotoxic ARVs)

Another potentially serious toxicity is pancreatitis. Fatalities from ddI induced pancreatitis have been reported. Xerostomia has been reported in up to 10 % of patients. Other adverse reactions include gastrointestinal intolerance, hepatitis, hyperuricaemia and gout.

Drug interactions

The antacid in the didanosine tablets can reduce the absorption of ketoconazole and dapsone. These drugs should therefore be taken with food and several hours apart from the didanosine. Oral ganciclovir may increase the absorption of didanosine and therefore increase the risk of didanosine toxicity. ddI seriously reduces the serum level of ganciclovir. For these reasons it is best not to prescribe didanosine with ganciclovir. The use of quinolones is also contraindicated because the divalent cations in the buffer can interfere with absorption.

Didanosine may also decrease the absorption of protease inhibitors. Therefore PIs should be taken at least 2 hours before taking didanosine.

Contraindications/precautions

Caution is needed when didanosine is given to patients with a history of peripheral neuropathy or to patients already taking neurotoxic drugs.

Patients with advanced HIV disease, with a history of pancreatitis or who are taking other drugs or excessive alcohol that may also cause pancreatitis are at an increased risk. If the patient develops pancreas specific serum amylase levels greater than twice the normal level, stopping the didanosine should be considered.

Zalcitabine (ddC)
(Trade name Hivid)
Zalcitabine is a cytidine analogue.

Dosage and administration
0.75 mg tds orally for a person ³ 40 kg mg tds orally for a person < 40 kg
The absorption of zalcitabine is slightly better when taken with a meal.

Adverse effects

Polyneuropathy is more common with zalcitabine than with didanosine. It is also a dose dependent sensory, symmetric peripheral neuropathy, usually involving the lower limbs initially. The neuropathy is generally reversible after stopping treatment. If, however, the drug is not stopped in time, symptoms of polyneuropathy may persist.

Zalcitabine may cause stomatitis, mouth ulcers, nausea and vomiting, gastro-intestinal upsets, headaches and anorexia.

Pancreatitis is less often observed than with didanosine.

Contraindications/precautions

Caution is needed when zalcitabine is given to patients with a history of peripheral neuropathy or to patients already taking neurotoxic drugs. Some drugs used in HIV disease can increase the risk of developing peripheral neuropathy. These include stavudine (d4T), dapsone and isoniazid, and therefore should be avoided in combination.

Caution is also needed for patients with a history of pancreatitis or who are taking other drugs or excessive alcohol that may cause pancreatitis. If a patient develops pancreas specific serum amylase levels greater than twice the normal level, stopping the zalcitabine should be considered.

Stavudine (d4T)
(Trade name Zerit)
Stavudine is a thymidine analogue. Stavudine also penetrates well into the CSF.

Dosage and administration
40 mg bd orally for a person ³ 60 kg
30 mg bd orally for a person < 60 kg
The absorption of stavudine is slightly better when taken with a meal but it can be taken at any time.

Adverse effects

Similar peripheral neuropathy as with didanosine and zalcitabine is seen. Pancreatitis and hepatitis have also been described. There is an increased risk of pancreatitis in patients treated with pentamidine or ganciclovir. Other side effects include nausea and vomiting, fever, headache, skin rashes, diarrhoea and lethargy.

Contraindications/precautions

Caution is needed when stavudine is given to patients with a history of peripheral neuropathy or patients already taking neurotoxic drugs, in particular ddC, dapsone and isoniazid.

Caution is also needed for patients with a history of pancreatitis or who are taking other drugs or excessive alcohol that may cause pancreatitis. If a patient develops pancreas specific serum amylase levels greater than twice the normal level, stopping the stavudine should be considered.

Zidovudine should not be given with stavudine because of mutual antagonism.

Lamivudine (3TC)
(Trade name Epivir)
Lamivudine is a cytidine analogue. Rapid development of resistance is seen when used as monotherapy (causes a mutation in the HIV polymerase gene at codon 184, but lamivudine resistant strains with this mutation remain sensitive to zidovudine).

Dosage and administration
150 mg bd orally
The absorption is slightly better when lamivudine is taken with a meal.

Adverse effects

Very well tolerated. Occasionally headache, fatigue, insomnia, nausea, abdominal pain and peripheral neuropathy.

Pancreatitis has been reported in children.

Contraindications/precautions

No drug interactions have been described

Table 9: Nucleoside reverse transcriptase inhibitors

T: tablet, A: ampoule, S: solution, C: capsule, P: powder

*costs at time of writing

Table 10

Interactions between nucleoside analogues and other drugs: prevention, monitoring and reducing adverse effects

Non-nucleoside Reverse Transcriptase Inhibitors (NNRTIs)

(Table 11)

These compounds inhibit the HIV reverse transcriptase by binding to it non-competitively. Because they interact with a specific binding site on the enzyme any slight variation because of a point mutation may cause resistance. Therefore high level resistance develops quickly with this group of compounds if they are not given with other antiretroviral classes. The NNRTIs are inactive against HIV-2. It is currently believed that NNRTIs are only effective when used in 3 drug combinations where each of the other 2 drugs are also new to the patient.

Nevirapine
(Trade name Viramune)
Nevirapine is a dipyrido diazepinone derivative. Synergism has been shown in three drug regimens including zidovudine and lamivudine or didanosine. Nevirapine levels in the cerebrospinal fluid reach 45% of plasma concentrations.

Dosage and administration
Nevirapine can be taken with or between meals.
To evaluate whether a patient might develop an adverse reaction, it is recommended to start with 200 mg orally once a day for 2 weeks. Starting nevirapine at this lower dose supposedly reduces the chance of an adverse event. Then the dosage is increased to 200 mg orally bd.

Adverse effects
Rash, which may be accompanied by fever. Several cases of Stevens-Johnson syndrome have been reported.

Abnormal liver function tests.

Drug interactions
Nevirapine is metabolised via the cytochrome p450. Nevirapine induces cytochrome activity and may therefore decrease, serum levels of protease inhibitors.
The effectiveness of the oral contraceptive pill may be reduced when taking nevirapine.
Rifampicin may decrease nevirapine absorption.

Delavirdine
(Trade name Rescriptor)
Delavirdine is a bis (heteraryl) piperazine derivative. Synergism has been shown with zidovudine, didanosine, zalcitabine or lamivudine.

Dosage and administration
400 mg tds orally.
The drug can be given with or without food. Antacids should not be administered within at least one hour of taking delavirdine.

Adverse effects
Rash (in approximately 40% of patients, usually mild). The rash usually occurs in the first 6 weeks of treatment and will disappear in 2-6 weeks if treatment is continued, however Stevens-Johnson syndrome has been reported. Other side effects include mild headache, nausea, diarrhoea, insomnia, vivid dreams and fatigue.
Abnormal liver function is observed occasionally.

Drug interactions
Delavirdine is an inhibitor of the cytochrome P450 and may therefore increase serum levels of protease inhibitors.
It should not be given at the same time as didanosine (separate administration by at least 1 hour).
Delavirdine increases serum levels of non-sedating antihistamines (terfenadine, astemizole & loratadine) and there is an increased risk of arrhythmias. Many other drugs may decrease absorption of delavirdine including sedative hypnotics (barbiturates), anti-arrhythmics, calcium-channel blockers, ergot-alkaloid preparations, vitamins, H2 antagonists, rifampicin and cisapride.
Antifungals and erithromycin may increase the absorption of delavirdine.

Contraindications/precautions
Do not co-administer with rifampicin, rifabutin, phenytoin, phenobarbital carbamazepine.

Table 11

Non-nucleoside reverse transcriptase inhibitors

T: tablet

Protease inhibitors
(Tables 12,13,14)

The enzyme protease is responsible for the cleavage of certain viral proteins. This cleavage is needed for the production of new infectious virus particles. The currently approved protease inhibitors bind to the active site of the enzyme. Protease inhibitors are metabolised by the cytochrome P450 system in the liver and many interactions are therefore possible between protease inhibitors and other drugs.

Indinavir
(trade name Crixivan)
Indinavir is a very powerful protease inhibitor. Indinavir has to be taken 3 times daily on an empty stomach. Virus isolates that are resistant to indinavir are generally cross-resistant to ritonavir. Indinavir can be stored at room temperature but in the tightly closed original container of the manufacturer because the capsules are sensitive to moisture.

Dosage and administration
800 mg orally tds. If nevirapine is associated with indinavir the dosage of indinavir should be increased to 1000 mg orally tds. Indinavir has to be taken on an empty stomach, 1 hour before food or at least 2 hours after a meal. If this is difficult - for example when taken in combination with ddI which is taken at a different time of day but also on an empty stomach - indinavir tablets can be taken with a low fat light meal.

Adverse effects
Nephrolithiasis is observed in 1-5 % of the patients. Nephrolithiasis may be a particularly frequent complication in patients treated in warm climates because of insufficient hydration.
Other side effects include: nausea, vomiting, indigestion, diarrhoea, rash, pruritis, dry skin, dry mouth, taste alteration, vivid dreams and lethargy.
Hyperbilirubinaemia occurs frequently but does not cause symptoms. Increased liver enzymes and hyperglycaemia are also seen.
A redistribution of subcutaneous body fat has also been noted (lipodystrophy) and a case of coronary artery disease has been described. Delavirdine increases indinavir levels two fold.

Drug interactions
Medications that should not be given with indinavir include rifampicin, astemizole, terfenadine, cisapride, midazolam and triazolam, antifungals and anti-epileptics. Didanosine reduces the absorption of indinavir unless taken > 2 hours apart. Antacids will also decrease absorption of indinavir.

Contraindications/precautions
Precautions when prescribed during the first trimester of pregnancy (informed discussion of the possible risks compared to the benefits). Avoid with severe liver abnormalities. Patients must drink at least one and a half litres of water a day to prevent kidney stones.

Saquinavir
(Trade name Invirase (SQV-HGC) and new formulation Fortovase (SQV-SGC) soft gel capsules)
Saquinavir in its new formulation has increased bioavailability making it a much more effective protease inhibitor.

Dosage and administration
Old preparation saquinavir (SQV-HGC):
The official recommended dosage is 2400 mg tds. Saquinavir should be taken with food.
Saquinavir in this preparation has low bioavailibility. It is now used in association with ritonavir or nelfinavir. (Saquinavir 2 x 400mg ritonavir 2 x 400mg) Taking the capsules with grapefruit juice increases their absorption.

New preparation saquinavir (SQV-SGC):
Has greatly enhanced bioavailability. Soft gel dosage is 1200 mg orally tds, when given as the only protease inhibitor.
When given in combination with nelfinavir 750mg orally tds the saquinavir SGC dosage should be 800 mg orally tds.

Adverse effects
Saquinavir is generally well tolerated. Spontaneous bleeding episodes have been observed in patients with HIV infection with haemophilia. Hyperglycaemia has also been reported.
Minor side effects such as diarrhoea, abdominal pain, headaches, skin rashes and parasthesia may occur.

Drug interactions
Rifampicin reduces saqinavir absorption. Anti-epileptic drugs (phenytoin and carbamazipine) and steroids (dexamethasone) also decrease saquinavir absorption.
Antihistamines such as astemizole and terfenadine should not be given with saquinavir as there is an increased risk of cardiac arrhythmia.
Antifungal drugs (ketoconazole, fluconazole, miconazole, itraconazole) may increase saquinavir absorption.
Hypoglycaemia (with some fatalities) has been reported.

Contraindications/precautions
Saquinavir is contraindicated in people with severe liver test abnormalities.

Ritonavir
(trade name Norvir)
Ritonavir is a very powerful protease inhibitor but has many side effects. When ritonavir is given as the only protease inhibitor in a combination with NRTIs, the optimal treatment dose is 600 mg orally bd. Ideally the drug has to be kept refrigerated at a temperature of 3 - 8 °C. However ritonavir can remain unrefrigerated if kept below 25 °C and used within 30 days.

Dosage and administration
The starting dose is 300 mg orally bd. This dose should be increased progressively over 10-14 days to reach an optimal dose of 600 mg orally bd. Ritonavir has to be taken with food.

Adverse effects
Nausea is very common particularly when starting treatment. Other side effects include vomiting, diarrhoea, anorexia, perioral paresthesias, peripheral paresthesias including a burning sensation of the skin, dizziness, insomnia, changes of taste and tiredness. Some of these side effects are mainly observed during the beginning of the treatment and are dose dependent. They may be avoided to a certain degree by progressively increasing the dose. Other side effects include an increase in serum triglyceride levels (> 200%), uric acid and hepatic enzymes. There is an increased risk of liver toxicity in patients with underlying hepatitis B or C infection. Hyperglycaemia has been reported in patients with and without a known history of diabetes. There are reports of ritonavir causing epilepsy, orthostatic hypotension and renal insufficiency, lipodystrophies and increased bleeding in patients with haemophilia.
As ritonavir may cause dizziness and fatigue it may interfere with some patients’ ability to drive.

Drug interactions
Since ritonavir is a very potent inhibitor of the cytochrome P450 system a long list of drugs are contraindicated as concomitant treatment.
Ritonavir increases plasma saquinavir levels 50-100 fold. Therefore the combination of ritonavir 400 mg bd and saquinavir 400 mg bd is useful. This combination is generally well tolerated and the twice daily dosage is an advantage. The most frequently observed side effect is diarrhoea.
Ritonavir reduces the effectiveness of the oral contraceptive pill.

Contraindications/precautions
Severe liver function abnormalities.

Nelfinavir
(trade name Viracept)
Nelfinavir has a good bioavailability and is highly potent. The powder formulation of nelfinavir is being investigated for the treatment of children with HIV infection.

Dosage and administration
750 mg tds. If nevirapine is used in combination with nelfinavir the nelfinavir dosage should be increased to 1000 mg tds. Nelfinavir has to be taken with food.

Adverse effects
The most commonly reported side effect of nelfinavir is diarrhoea. Other reported side effects include: nausea, elevated transaminase enzymes, hyperglycaemia and lipodystrophy.

Drug interactions
Medications that should not be given with nelfinavir include rifampicin, astemizole, terfenadine and cisapride. Rifabutin levels increase three fold with nelfinavir, so if prescribing together rifabutin levels should be reduced to half. Saquinavir increases nelfinavir levels by about 20%.

Contraindications/precautions
Avoid when patient has severe liver function abnormalities.

Future developments

Once daily preparations
There are several newer and better versions of some of the currently available drugs, which may require less frequent dosing. (For example once daily ddI and twice daily indinavir).

Combined ARVs
Preliminary data shows that the combined formulation produces similar clinical results to the conventional separate drug regimen. It is hoped that combined formulations will make adherence easier. (For example lamivudine (3TC) 150mg + zidovudine (ZDV) 300mg: trade name Combivir)

New drugs
There are currently many new ARVs undergoing clinical trials and under development. Table 12

Protease inhibitors

T: tablet, A: ampoule, S: solution, C: capsule, P: powder

Table 13

Medications that should not be used with protease inhibitors

Table 14

Important interactions between protease inhibitors (PI) and other drugs (OD)

*Rit = Ritonavir
*Ind = Indinavir

Tremendous optimism has been generated by the recent development of new antiretrovirals, particularly the triple combination therapies including one protease inhibitor, which promise a longer and better life for people living with HIV/AIDS (PLHA). In response to requests for the treatments from PLHA, and for policy and technical guidance from health professionals and governments, WHO, in collaboration with UNAIDS, held an Informal Consultation on the Implications of Antiretroviral Treatments with particular reference to low and middle income countries, in April 1997.

Participants at the consultation, ministries of health, health professionals, PLHA, donors and NGOs working in HIV/AIDS have all called for technical and policy guidance for health planners and policy makers, and decision makers in training institutions, central and district hospitals on antiretroviral treatments, as an immediate follow up to the consultation.

In collaboration with UNAIDS, WHO has produced a set of nine guidance modules on the following aspects of antiretroviral treatments:

1. Introduction to antiretroviral treatments
2. Introducing antiretroviral treatments into national health systems: economic considerations
3. ARV treatments: planning and integration into health services
4. Safe and effective use of antiretrovirals
5. Laboratory requirements for the safe and effective use of antiretrovirals
6. The use of antiretroviral drugs to reduce mother to child transmission of HIV
7. Treatments following exposure to HIV
8. Antiretrovirals: regulation, distribution and control
9. Ethical and societal issues relating to antiretroviral treatments

Module 5. Laboratory Requirements for the Safe and Effective Use of Antiretrovirals

Acknowledgements

The World Health Organization and UNAIDS would like to thank Professor Gunnel Biberfeld, Swedish Institute for Infectious Disease Control, Stockholm, Sweden, for preparing the text of this module;

Dr Gaby Vercauteren, Blood Safety Unit, WHO, Dr Frank De Wolf, University of Amsterdam, the Netherlands and Professor Nkandu Luo, University Teaching Hospital, Lusaka, Zambia, for their most thoughtful reviews of the draft text; and

Dr Rachel Baggaley, Dr Susan Fernyak, Ms Alison Martin Katz, Ms Marguerite Nguyen and Dr Eric van Praag, Office of HIV/AIDS and Sexually Transmitted Diseases, WHO, and Dr Joseph Perri, Policy, Strategy and Research, UNAIDS, for coordinating the project.

©World Health Organization and Joint United Nations Programme on HIV/AIDS, 1998

This document is not a formal publication of the World Health Organization (WHO) and Joint United Nations Programme on HIV/AIDS (UNAIDS), and all rights are reserved by them. The document may, however, be freely reviewed, abstracted, reproduced, or translated in part, but not for sale or for use in conjunction with commercial purposes. It may also be reproduced in full by non-commercial entities for information or for educational purposes with prior permission from WHO and UNAIDS.

For authorization to translate or reproduce the work in full, and for any use by commercial entities, applications and enquiries should be addressed to the Office of HIV/AIDS and Sexually Transmitted Diseases, World Health Organization, 1211 Geneva 27, Switzerland, which will be glad to provide the latest information on any changes made to the text, plans for new editions, and the reprints and translations that are already available.

The mention of specific companies or of certain manufacturers’ products does not imply that they are endorsed or recommended by either WHO or UNAIDS in preference to others of a similar nature that are not mentioned. Errors and omissions excepted, the names of proprietary products are distinguished by initial capital letters.

The views expressed in these documents by named authors are solely the responsibility of those authors.

Introduction

The introduction of combination antiretroviral treatment (ARV) for HIV infection has dramatically changed the clinical perspective for people living with HIV/AIDS, for health care professionals responsible for their care and for policy makers. This module reviews the laboratory requirements for safe and effective use of ARVs including:

· Diagnosis of HIV infection through the detection of antibodies or viral genome by polymerase chain reaction (PCR)

· Monitoring of adverse reactions to ARV drugs

· Diagnosis of opportunistic infections

· CD4+ T-cell determinations

· Viral load monitoring

· Monitoring of resistance of HIV

The importance of reliable and accurate laboratory support for ARV treatment is stressed and basic information provided on how to achieve a high standard of performance. Well trained staff is a prerequisite for a high standard of performance. Basic training and regular refresher training should be routine in the laboratories set up for ARV services. Cost estimates provided are based on commercial prices given by the manufacturers. Costs may vary between geographical areas and are often negotiable. Labour costs have not been included.

HIV testing

The standard procedure for laboratory diagnosis of infection with the human immunodeficiency virus (HIV) usually includes screening for virus specific antibodies with an enzyme-linked immunosorbent assay (ELISA) or a rapid simple test, followed by confirmatory testing of screened positive samples.¹

¹Joint United Nations Programme on HIV/AIDS (UNAIDS)/WHO. Revised recommendations for the selection and use of HIV antibody tests. Weekly Epidemiological Record 1997;72:81-88.

The most widely used screening tests are ELISAs which comprise a number of variants based on different principles including indirect, competitive, sandwich and capture assays, all of which may detect HIV-1 as well as HIV-2 antibodies. Most of the first generation ELISA tests were based on viral lysate antigens. Later variants utilise recombinant proteins and/or synthetic peptides. The latter are generally more sensitive and specific as there are fewer contaminating products such as cellular components in the viral lysate. Through a WHO initiative, extensive evaluations of HIV antibody assays have been performed.

The antibody screening assays, though very sensitive, specific and reproducible, are not completely free from false reactions. The most common testing strategy for detection of HIV antibodies is therefore to perform a second more specific supplementary (confirmatory) assay on sera repeatedly reactive on a screening assay. Western blot (WB) and/or line immunoassays (LIA, RIBA) are tests used specifically for serological confirmation.

WB tests are expensive, time consuming and have technical disadvantages, such as lack of standardisation in performance and interpretation, and indeterminate reactions. Therefore, alternative strategies for confirmation of HIV antibody positive samples that can replace conventional WB testing have been evaluated. WHO has issued guidelines on testing strategies according to the purpose of testing. The general principle is to use a combination of cheaper and less sophisticated tests, e.g. two or three ELISAs or simple/rapid tests, in such a way that sensitivity and specificity will be comparable to the more commonly used strategies including confirmation by WB. Ideally, an alternative strategy for confirmation of HIV infection should consist of a combination of assays of different test principles and different antigen preparations. It is also possible to design an alternative strategy for confirmation to include discrimination between HIV-1 and HIV-2. Furthermore, the complete test procedure (from receiving the sample to sending out the final result), using the alternative confirmatory strategy, saves time. Positive samples may be confirmed and results submitted to the referring doctors the same day as they have been screened.

Apart from standard laboratory equipment, ELISA testing requires a spectrophotometer at a cost of around US$ 10,000. Simple/rapid test kits are often supplied with all the reagents needed while basic utensils such as serum tubes, pipettes and a simple centrifuge for separation of serum have to be provided by the laboratory. For storage of test kits a refrigerator is required. Serum can be kept in a refrigerator for some time but for longer storage they should be frozen at -20oC at least. ELISA tests for HIV can be purchased from around US$1 per test, while simple/rapid tests are usually more expensive (from around US$3). Western blot tests cost US$ 20 - 30. Many HIV antibody tests can be obtained through WHO bulk purchase at a reduced rate.

Apart from HIV antibody measurements, detection of virus genome by nucleic acid amplification techniques (NAT), such as the polymerase chain reaction (PCR) has a place in certain cases for confirmation of HIV infection. NAT are particularly useful for further characterisation of virus strains, such as subtyping, monitoring of resistance to ARV drugs, and measurement of viral load. PCR is also the main mode of diagnosis of HIV in infants. In settings with limited laboratory facilities, modified heat-denatured antigen detection may be used as an alternative to PCR for the early diagnosis of HIV infection in infants. While antigen detection is mainly based on the same principles as antibody testing, NAT are sensitive and complicated methods requiring special laboratory facilities, instruments and skills, and careful quality assurance. For further information on PCR and related NAT, see the sections on viral load monitoring and monitoring resistance of HIV.

Finally, it should be noted that when HIV infection has been detected for the first time in a person, the diagnosis should always be reconfirmed on a second blood sample, in order to exclude laboratory and administrative mistakes.

In conclusion, laboratory diagnosis of HIV infection is usually in the initial phase done by the detection of HIV specific antibodies in serum or plasma. Antibody detection can be done with simple/rapid tests which require a minimum of laboratory equipment and material. Testing strategies based on simple/rapid assays are most suitable for small numbers of specimens. For larger numbers of specimens it is more practical and economic to use ELISAs. These assays require larger investments in equipment and a regular supply of electricity and water. Detection of the virus itself, through antigen detection or NAT is usually used for specific purposes, e.g. during early stages of infection when antibody levels may be low, for diagnosis in infants born to mothers with HIV,for subtyping of virus strains, or for determination of virus load.

Monitoring of patients for adverse reactions to ARVs

Like most potent medical drugs, ARVs may cause adverse reactions which have to be weighed against the potential benefits. Adverse reactions vary from one drug to another and between individual patients. Some are of a general character and can be monitored clinically, e.g. skin rashes, nausea, headache and polyneuropathy. Some adverse reactions decline after time. Other adverse reactions can be monitored with laboratory tests. The major ARV drugs, their most important side effects and the modes of detection and follow-up of these adverse reactions are summarised in Table 1.

Among nucleoside reverse transcriptase inhibitors (NRTIs), zidovudine may cause anaemia and neutropenia. Lamivudine has relatively few side effects while the other NRT inhibitors have caused pancreatitis in a few cases. Thus, regular blood counts and analyses of amylase in serum (or symptomatic patients) are required. Among the protease inhibitors, indinavir may cause elevated bilirubin levels and clinical nephrolithiasis with pain in the flank and hematuria (reported in 3-5% of all patients). The most common adverse events reported for the other protease inhibitors are diarrhoea and other gastrointestinal disturbances. Protease inhibitors have also been reported to cause liver enzyme elevations and diabetes mellitus in a few cases.

Timing of check-ups depends on the status of the patient. If the patient tolerates the treatment well and is in a stable clinical condition, the interval between check-ups can be extended to 3-4 months. Otherwise the intervals should be shorter according to need, as decided by the doctor in charge.

In conclusion, regular clinical and laboratory monitoring of patients on ARVs is necessary. The laboratory tests to be performed depend on which ARV drugs are being used, but generally comprise blood counts including differential, as well as bilirubin and liver enzymes, amylase, urate and glucose (simple dip tests of the urine are usually sufficient). It must be emphasised that baseline laboratory analyses should be performed before ARV treatment is started.

Table 1: Antiretroviral drugs, main adverse reactions and follow-up action/tests

IDENTIFICATION AND DIAGNOSIS OF OPPORTUNISTIC INFECTIONS

Opportunistic infections are significant clinical features of HIV/AIDS and their detection and treatment is of considerable importance in the care of HIV infected people. The incidence of opportunistic infections usually decreases with the introduction of ARVs. However, in cases of non-optimal compliance with ARV treatment or the development of resistance to the ARV drugs used, opportunistic infections may reappear.

A summary of the main opportunistic infections and complicating syndromes in HIV disease including the mode of diagnosis and basic equipment and reagents needed, is provided in Table 2. (For further reading see Reference 9.)

Some of the opportunistic infections, like oral or oesophageal candidiasis, cytomegalovirus retinitis and cutaneous Kaposi’s sarcoma (KS) may require only a clinical examination for diagnosis and follow-up (KS is often confirmed by histopathological examination of tissue specimens). Many of the parasitic infections can be diagnosed by microscopy of relevant specimens, such as stool and tissue preparations. Lymphomas will also be diagnosed and typed through histopathological examination by microscopy of tissue specimens. A pulmonary X-ray in combination with microscopy of a sputum sample is often sufficient for confirmation of pulmonary tuberculosis or pneumocystis carinii infection. Computed tomography (CT) or magnetic resonance imaging (MRI) are valuable tools in the of brain manifestations of HIV disease, such as toxoplasmosis of the brain, lymphomas and progressive multifocal leukoencephalopathy (PML). For proper confirmation of bacterial infections it is necessary to perform cultures. Tuberculosis and virus cultures require special laboratory facilities and reagents.

Thus, with a good knowledge and clinical experience, it is possible to adequately monitor many opportunistic infections with clinical examinations and basic technical facilities, such as light microscopy and chest X-ray only. Diagnostic possibilities can be extended with the addition of a bacteriological laboratory. Ideally, facilities for TB and virus cultures, CT and/or MRI scans should be available. The level of sophistication will be determined locally according to needs, available resources and existing technical facilities.

Table 2. Summary of main opportunistic infections and complicating syndromes in HIV infection, mode of diagnosis, basic equipment and reagents

a) Clinical examination also includes clinical history and, ideally, BAL, Bronchoalveolar lavage routine blood samples such as full blood counts and inflammatory parameters.
CT, Computed Tomography
b)* Minimum (basic) requirement MRI, Magnetic Resonance Imaging
(*) Additional requirement PCR, Polymerase Chain Reaction
c) *** common
** less common CSF, Cerebrospinal fluid
* rare

(N.B. May vary by geographical region or population group)

DETERMINATION OF CD4+ T-CELL COUNTS

Determination of CD4+ T-cell counts has for many years been the conventional method for assessing the immune status of HIV infected persons. Low levels of CD4+ T-cells (<200 cells/cmm) predict decreased survival and increased risk of acquiring opportunistic infections. More recently the introduction of methods for assessment of plasma viral RNA levels has provided an alternative and complement to CD4+ T-cell measurements. Since plasma viral load is also a good predictor of clinical progression of the HIV disease, the need for regular CD4+ T-cell measurements may decrease. However, determination of CD4+ T-cell counts is still widely used, often in parallel with plasma viral load determinations. The extent to which CD4+ T-cell counts and plasma viral load measurements should be used and how the two methods can complement each other, is still to be determined.

In industrialised countries, the standard method for performing CD4+ T-cell counts in HIV-infected people is by flow cytometry. The method for flow cytometric immunophenotyping of peripheral blood samples using two-, three-, or four-colour monoclonal antibody panels is highly standardised and results are generally accurate and reliable. Most laboratories measure absolute CD4+ counts combining the results of two (or three) different laboratory techniques: the proportion of CD4+ T-lymphocytes is determined by flow cytometric analysis whereas the lymphocyte count (or the white blood cell count and the differential) is measured by haematological testing. Thus, each of the techniques used contributes to the accuracy of the final results.

Recently, several methods have been developed to derive absolute CD4+ T-cell counts directly from flow cytometric analysis. Such methods may yield less variable results but the accuracy must be validated against the laboratory’s currently used conventional method. Laboratories performing CD4+ T-cell counts should follow published guidelines as well as manufacturers’ instructions, and participate in performance evaluation programmes (11).

Flow cytometric instruments can today be purchased from three different manufacturers at a cost of around US$ 40,000 - 80,000. These instruments require highly trained personnel and are thus unsuitable for routine use in laboratories with limited facilities. In such places, specimens may be transported to a central laboratory if they can be maintained at room temperature during transport and if they can be processed within acceptable time limits. A specimen for flow cytometric immunophenotyping must be processed within 48 hours of sample collection. Haematological testing must be performed within the time limit for which the haematology instrument has been validated, usually varying from less than 6 to 30 hours after specimen collection. Immunophenotyping and haematological testing can thus be carried out at different laboratories. If specimens cannot be transported to laboratories at required temperatures or within acceptable time limits for flow cytometric analysis, e g in developing countries, an alternative method for CD4+ T-cell enumeration may be used.

Alternative methods for monitoring CD4+/CD8+ T-lymphocytes are summarised in Table 3. Alternative assays were reviewed in a WHO workshop in 1992. Results of a WHO collaborative multicentre field evaluation of alternative CD4 quantification techniques and other studies have demonstrated the reliability of these techniques and their potential for use in settings with limited facilities. A variety of assay principles exist, including commercial kits as well as in-house systems based on immunological reagents also available commercially. Some assays are designed to detect CD4⁺ cells only whereas others also detect CD3⁺ and CD8⁺ cells. The various assay principles include detection of lymphocytes through labelling with monoclonal antibodies (mab) and then visualisation by colour reagents, either directly conjugated to the MAB (FACScount, Cytosphere, the immunoalkaline phosphatase technique) or indirectly via lysis of the mab labelled cells (Dynabeads). In the latter case the cells are visualised by trypan blue staining of the cell nuclei remaining after lysis. Alternatively, the CD4 and/or CD8 molecules may be detected by ELISA techniques, either directly on the cells (Capcellia, Zymmune) or after releasing of the cell surface molecules through solubilization treatment (TRAx CD4). Apart from FACSCount, which requires a special instrument, these techniques can be applied without sophisticated instrumentation; Cytosphere, Dynabeads and the immunoalkaline phosphatase technique are read in a light microscope while Capcellia, Zymmune and TRAx CD4 require instruments for ELISA analysis. The correlation between the alternative methods and flow cytometry is generally high, although in some studies variable correlation coefficients have been observed particularly for the ELISA based techniques when testing African samples (Table 1). When comparing the various methods with flow cytometry the correlation coefficients may be high but the values obtained may be consistently too low or too high. It is, thus, important to evaluate and standardise the method chosen, in the local environment.

The choice of method will vary from one setting to another since each meets different laboratory requirements (Table 3). For example, methods based on the microtitre enzyme immunoassay technique would be suitable for laboratories which handle a large number of specimens but which have few staff. Such assays are also suitable for specimens collected from remote sites since they allow samples to be frozen and shipped in batches for testing. Methods based on cell counting using a microscope would be of use in laboratories which process few samples and have few staff. In general, the alternative methods require lower initial investments and are cheaper to perform than conventional flow cytometry. The price of the reagents needed to test one sample using an alternative method is about US$ 5-10, with the exception of FACSCount and Capcellia which are more expensive. The corresponding cost using a flow cytometric method is US$ 10-15. However, the alternative assays are less flexible and often more time consuming which is a constraint in large scale investigations.

When using the alternative methods for follow-up of patients it should be noted that results are likely to be more consistent if the same method is used each time. It is therefore recommended that laboratories use one type of assay. If a change is necessary it should be evaluated locally in parallel with the previous assay in order to make correct comparisons of the results over time. Quality control programmes should also be set up and form part of the assay protocol.

In conclusion, the conventional and most reliable method for accurate CD4+ T-cell counts, especially when carried out large-scale, is flow cytometry. If flow cytometric analysis is not available, a number of alternative methods are available. These alternative methods meet different laboratory requirements and suit different settings. The initial investment and the cost of maintenance of equipment are substantially higher for flow cytometry testing as compared to alternative methods. These costs and subsequent expenditures for reagents for each tested sample are thus dependent on what method is used but do not vary much between different countries. Requirements for trained personnel and labour intensity largely depend on the method and thus, the labour costs will vary between different countries. The accuracy of the assay used has to be monitored and this is especially important when an alternative method is chosen.

Table 3. Summary of main characteristics of seven alternative methods for CD4/CD8 lymphocyte determination and flow cytometry.