Summary

This paper explores the scientific justification for the key action message “ensure skilled attendance at delivery.” Many governments and other provider agencies in poor countries will need to commit additional health resources in order to respond to this message, and opportunity costs will be incurred. Achieving targets will take time, and benefits in terms of maternal mortality may not be detected for several years. It is therefore crucial to review the basis for prioritising skilled attendance. This paper examines the historical and epidemiological evidence at both the individual and population levels of analysis.

The lack of a clear definition has been, and continues to be, the cause of much confusion over the role, and thus the potential, of skilled attendants. Recent initiatives to specify minimum and additional skills have improved understanding not only of training requirements but also of the wider environment which is required for skilled attendants to function effectively. This paper proposes that skilled attendance be conceived as encompassing 1) a partnership of skilled attendants (health professionals with the skills to provide care for normal and/or complicated deliveries), AND 2) an enabling environment of equipment, supplies, drugs and transport for referral.

At the individual level, there are sound clinical reasons for believing that the risk of maternal death can be reduced by skilled attendance, particularly as the causal pathways can be specified. However, this theory of how skilled attendance could work has not been rigorously tested, and the available empirical evidence - both historical and epidemiological, is flawed, either owing to weak study designs which fail to control for key confounding factors and/or inadequate power. Insights from modeling can be used to complement an empirical approach and in this paper a preliminary model is presented. This estimates that around 16% to 33% of all maternal deaths may be avoided through the primary or secondary prevention of four main complications (obstructed labour, eclampsia, puerperal sepsis and obstetric haemorrhage) by skilled attendance at delivery. The model highlights the importance of considering the potential of skilled attendance to impact not only on maternal mortality but also morbidity, and emphasises their primary prevention role through effective and appropriate management of normal labour and delivery. At the population or aggregate level, correlational analysis has been the major stimulus for prioritising skilled attendance. The paper discusses two drawbacks to this - the intrinsic inability of this type of analysis to make causal connections, and the problems of the data - its varying reliability and the limitations of the independent and dependent variables correlated, such as the institutional delivery rate and the maternal mortality ratio. In particular, the reliance on the crude indicator “percentage of deliveries with health professionals” which groups together doctors, midwives and nurses, is challenged and an alternative independent variable - the Partnership Ratio - proposed. Correlational analysis highlights the inconsistencies in the postulated link between maternal mortality and skilled attendance, and emphasises the importance of timely access to quality maternity care. In particular, insights from the Partnership Ratio and the modeling approach suggest there is an optimal professional mix for skilled attendance to be effective in different country and service settings.