Cover Image
close this bookEarthquakes and People's Health (WHO - OMS, 1997, 296 p.)
View the documentThe epidemiology of earthquakes: implications for vulnerability reduction, mitigation and relief
View the documentSeismological forecasting: prospects within the International Decade for Natural Disaster Reduction
View the documentHealth implications of earthquakes: physical and emotional injuries during and after the Northridge earthquake1
View the documentAn overview of the Earthquake Insurance Programme in Japan
View the documentSummary

The epidemiology of earthquakes: implications for vulnerability reduction, mitigation and relief

E.K. Noji1

1E.K. Noji M.D., M.P.H, is Chief of the International Emergency and Refugee Health Programs, Centers for Disease Control and Prevention (CDC), Atlanta, GA, USA.

Better epidemiological knowledge of the causes of death and types of injuries and illnesses caused by earthquakes is clearly essential for determining what relief supplies are appropriate and what equipment and personnel are needed to respond effectively to such situations, as well as to improve preparedness and reduce vulnerability to the effects of future earthquakes. The overall objective of the epidemiology of disasters is to scientifically measure and describe their health effects and the factors that contribute to these effects, with the goals of assessing the needs of disaster-affected populations, efficiently matching resources to needs, preventing further adverse health effects, evaluating programme effectiveness, and carrying out contingency planning.

This presentation focuses on the medical and public health impact of earthquakes and outlines a number of important areas where the science of epidemiology can reduce overall vulnerability to earthquakes and can contribute to improved disaster preparedness and mitigation.

Major components of vulnerability

Until earthquake prevention and control measures are adopted and mitigation actions implemented throughout the world, a single severe earthquake can cause tens of thousands of deaths and serious injuries and enormous economic losses.

During the past 20 years, earthquakes alone have caused more than a million deaths worldwide (1). Nine countries account for more than 80% of all fatalities this century and almost half of all earthquake deaths in the world during this period have occurred in just one country - China (Fig. 1.1). On 28 July 1976, at 3.42 a.m., an earthquake of magnitude 7.8 occurred in Tangshan in the northeastern part of China. In a matter of seconds, an industrial city of a million people was reduced to rubble and more than 240 000 people were killed (2). Recent accelerated urbanization in other seismically active parts of the world where population densities reach 20 000-60 000 inhabitants per square kilometre underscores the vulnerability of such areas to similar catastrophic numbers of earthquake-related deaths and injuries. In the past 10 years the world has witnessed four catastrophic earthquakes resulting in great loss of life: in Mexico City in 1985 (10 000 deaths); in Armenia in 1988 (25 000 deaths); in Iran in 1990 (40 000 deaths); and in India in 1993 (10 000 deaths) (Table 1.1). The United States has been relatively fortunate in terms of earthquake-related casualties so far (3). Only an estimated 1600 deaths in the USA have been attributed to earthquakes since colonial times, with over 60% of these having been recorded in California. The most serious earthquake in terms of loss of life was the 1906 San Francisco earthquake and fire that killed an estimated 700 people.

FIGURE 1.1. Location of earthquake deaths across the world.

Figure adapted from: Coburn AW, Pomonis A, Sakai S. Assessing strategies to reduce fatalities in earthquakes. In: Proceedings of the International Workshop on Earthquake Injury Epidemiology for Mitigation and Response, 12 July, 1989. Baltimore, Maryland. Baltimore, Johns Hopkins University, 1989:112.

Fire risks

One of the most severe secondary disasters that can follow earthquakes is fire (4). Severe shaking may cause overturning of stoves, heating appliances, lights, and other items that can ignite materials into flame. Earthquakes in Japan that trigger urban fires cause 10 times as many deaths as those that do not (4). The Tokyo earthquake of 1923, which killed more than 140 000 people, is a classic example of the potential of fires to produce enormous numbers of casualties following earthquakes. Similarly, the large fire that occurred after the 1906 San Francisco earthquake was responsible for much of the death toll following that event. More recently, the 1994 Northridge earthquake in southern California showed that strong vibrations may sever underground fuel lines or gas connection points, causing spills of volatile or explosive mixtures and resultant fires (5,6). Similarly, during the first seven hours following the 1989 Loma Prieta earthquake in northern California, San Francisco had 27 structural fires and more than 500 reported incidents of fire (7). Furthermore, the city water supply was disrupted, significantly impairing the ability to fight these fires (8).

Perhaps the most vulnerable areas of all are the informal housing sectors on the periphery of many rapidly growing cities in developing countries (so-called "squatter housing" or "shanty town" settlements). Many of these have the potential for catastrophic conflagrations following an earthquake.

Table 1.1. Earthquakes in the 20th century causing more than 10 000 deaths


Location (magnitude)

No. killed


Mexico City, Mexico (M 8.1 and 7.3)

10 000


India (M 6.4)

10 000


Agadir, Morocco (M 5.9)

12 000


Dasht-I-Biyaz, Iran (M 7.3)

12 000


Buyin Zhara, Iran (M 7.3)

12 225


Indonesia (M 7.0+

15 000


Tabas, Iran (M 7.7)

18 200


Kangra, India (M 8.6)

19 000


Ashkabad, USSR (M 7.3)

19 800


China (M 6.8)

20 000


Guatemala City (M 7.5)

23 000


Armenia, USSR (M 6.9)

25 000


Quetta, Pakistan (M 7.5)

25 000


Concepcion, Chile (M 8.3)

25 000


Chilián, Chile (M 8.3)

28 000


Avezzano, Italy (M 7.5)

32 610


Erzincan, Turkey (M 8.0)

32 700


Iran (M 7.7)

40 000


Tsinghai, China (M 8.0)

40 912


Messina, Italy (M 7.5)

58 000


Ankash, Peru (M 8.3)

66 794


Kanto, Japan (M 8.3)

142 807


Kansu, China (M 8.5)

200 000


Tangshan, China (M 7.8)

242 000


Approximately 1 500 000


Dams may also fail, threatening communities downstream. A standard procedure after any sizeable earthquake should be an immediate damage inspection of all dams in the vicinity and a rapid reduction of water levels in reservoirs behind any dam suspected of having incurred structural damage.

Structural factors

Trauma caused by partial or complete collapse of man-made structures is overwhelmingly the most common cause of death and injury in most earthquakes (1). About 75% of fatalities attributed to earthquakes this century were caused by the collapse of buildings that were not adequately designed for earthquake resistance, were built with inadequate materials, or were poorly constructed (9). Results of field surveys following earthquakes have demonstrated that different building types and structural systems deteriorate in different ways when subjected to strong earthquake ground-motion vibration. There is also evidence that different types of buildings inflict injuries in different ways and to different degrees of severity when they collapse (10,11,12).

Glass was one of the first in 1976 to apply epidemiology to the study of building collapse (13). He identified the type of housing construction as a major risk factor for injuries. Those living in the newer-style adobe houses were at highest risk for injury or death, while those living in the traditional mud-and-stick houses were at least risk. Figure 1.2 shows the breakdown of earthquake fatalities by cause for each half of this century. By far the greatest proportion of victims died in the collapse of reinforced masonry buildings (e.g. adobe, rubble stone or rammed earth) or unreinforced fired-brick and concrete-block masonry buildings that can collapse even at low intensities of ground-shaking and will collapse very rapidly at high intensities. Adobe structures in many highly seismic parts of the world (e.g. Iran, Pakistan, eastern Turkey, Latin America) not only have collapse-prone walls but also very heavy roofs (13,14). When they collapse, these heavy walls and roofs tend to kill many of the people inside (15,16). Unreinforced masonry buildings abound throughout earthquake-prone regions of the central United States (e.g., the New Madrid seismic zone). Most of these unreinforced masonry buildings are not equivalent to ancient construction (like Roman masonry) and remain essentially without retrofits and adequate seismic safety.

FIGURE 1.2. Fatalities attributed to earthquake by cause (Share of 795 000 fatalities)

FIGURE 1.2. Fatalities attributed to earthquake by cause (Share of 583 000 fatalities)

Figure from: Coburn A, Spence R, Earthquake protection. Chichester, John Wiley and Sons

Concrete-frame houses are generally safer (i.e. less likely to collapse), but they are also vulnerable and, when they do collapse, they are considerably more lethal and kill a higher percentage of occupants than do masonry buildings. In the latter half of this century, most of the earthquakes that have struck urban centres have involved the collapse of reinforced concrete buildings, and the proportion of deaths due to the collapse of concrete buildings is significantly greater than it was earlier in the century (Fig. 1.2).

Implications for vulnerability reduction, mitigation and relief

Prevention and control efforts need to be multidisciplinary and should include public education programmes, as well as better building design and improved quality of construction in those areas most likely to suffer an earthquake. The problem of earthquake casualties involves questions of seismology, the engineering of the built environment, the nature of both the physical and the sociological environments, aspects of personal and group psychology and behaviour, short-term and long-term economic issues, and many aspects of planning and preparedness.

Because of the multisectoral nature of these influences, public health and disaster response officials need to work together in the effort to develop and maintain effective seismic safety planning and earthquake mitigation programmes.

Planning scenarios for earthquakes

Relative chaos is likely to prevail immediately after a major earthquake. The area's residents, cut off from the outside, will initially have to help themselves and their neighbours. They can best do this if they have already planned their responses to the most likely earthquake scenarios and practiced the necessary skills. Medical preparedness plans can be built around similar earthquake scenario calculations based on the types of building likely to be affected, the population densities and settlement patterns, the size and characteristics of earthquakes expected in the region, and the medical facilities available in any given area. Such a regional hazard assessment, including "casualty scenarios", permits the development of specific training programmes for medical and rescue personnel as well as the appropriate deployment of medical and rescue equipment in advance of an earthquake disaster.

Because there never are enough rescuers or medical providers in major disasters, communities vulnerable to earthquakes should establish ongoing programmes to teach the public what to do when an earthquake occurs. These would include first aid education, basic rescue training and fire drills. Simulation exercises can be carried out jointly by volunteer groups, local fire brigades and hospitals. This training also may help to improve bystanders' responses during more common emergencies.

Early rapid assessment of the earthquake's impact

Rapid rescue of trapped victims and prompt treatment of those with life-threatening injuries can improve their outcome. Therefore early rapid assessment of the extent of damage and injuries is necessary to help mobilize resources and direct them to where they are most needed. Unfortunately, the very factors likely to cause large numbers of injuries are also likely to disrupt communications and transport and to damage medical care facilities. Public health officials need to establish in advance the procedures for surveying the affected areas.

Search and rescue

People trapped in the rubble will die if they are not rescued and given medical treatment. To maximize trapped victims' chances of survival, search-and-rescue teams must respond rapidly after a building collapses. Studies of the 1976 earthquake in Tangshan, China, the 1980 earthquake in Campania-Irpinia, Italy, the 1988 earthquake in Armenia, and the 1990 earthquake in the Philippines show that the proportion of trapped people found alive declined as the duration of entrapment increased. In the Italian study, a survey of 3619 survivors showed that 93% of those who were trapped and survived were extricated within the first 24 hours and that 95% of the deaths recorded occurred while the victims were still trapped in rubble. Estimates of the survivability of victims buried under collapsed earthen buildings in China and Turkey indicate that, within 2-6 hours, less than 50% of those buried are still alive. Although we cannot determine whether a trapped person died immediately or survived for some time under the debris, we can safely assume that more people would be saved if they were extricated sooner. As suggested by these data, teams with specialized expertise in areas such as search and rescue, on-site resuscitation and medical first aid arriving more than a couple of days after an earthquake's impact are unlikely to make much difference to the overall death toll of a large earthquake.

With the exception of personnel from countries in close geographical proximity, foreign assistance usually arrives after the local community has already carried out much of the rescue activity. For example, in southern Italy in 1980, 90% of the survivors of an earthquake were extricated by untrained, uninjured survivors who used their bare hands and simple tools such as shovels and axes. Following the 1976 Tangshan earthquake, some 200 000-300 000 trapped people crawled out of the debris on their own and went on to rescue others. They became the backbone of the rescue teams, and it was to their credit that more than 80% of those buried under the debris were rescued. Thus, life saving efforts depend heavily on the capabilities of relatively uninjured survivors and untrained volunteers, as well as those of local firefighters and other relevant professionals. This does not mean that people who were dead when they were extricated could not have been saved by a skilled team with sophisticated resources. However, people from the community can clearly play the most important role in rescue efforts if they are appropriately prepared.

Medical treatment

Just as speed is required for effective search and rescue, it is also essential for effective emergency medical services. The greatest demand occurs within the first 24 hours. Ideally, "disaster medicine" (medical care for victims of disaster) would include immediate life-supporting first aid, advanced trauma life support, resuscitative surgery, field analgesia and anesthesia, resuscitative engineering (search and rescue technology), and intensive care. Unconscious patients with either upper airway obstruction or inhalation injury or patients with correctable hypovolemia resulting from haemorrhage or bums would be especially likely to benefit from early medical intervention. Safar, studying the 1980 earthquake in Italy, concluded that 25-50% of victims who were injured and died slowly could have been saved if life-saving first aid had been rendered immediately.

Data from the 1976 earthquake in Guatemala, the 1985 Mexico City earthquake, the 1988 Armenian earthquake and the 1992 earthquake in Egypt showed that injured people usually seek emergency medical attention only during the first 3-5 days following the earthquake, after which hospital case patterns return almost to normal. From the sixth day onward, the need for emergency medical attention declined rapidly and the majority of the injured required only ambulatory medical attention, indicating that specialized field hospitals that arrive one week or more after an earthquake are generally too late to help during the emergency phase. Following the 1992 earthquake in Egypt, nearly 70% of all patients with earthquake-related injuries were admitted within the first 36 hours.

The medical and public health impact of a severe earthquake may well be compounded by significant damage to medical facilities, hospitals, clinics and supply stores within the affected area. In the worst-case scenario, a hospital building may itself be damaged by the earthquake, and the hospital staff may have to continue emergency treatment without using the buildings. For example, on 17 January 1994, at 4:31 a.m. Pacific Standard Time, an earthquake registering 6.8 on the Richter scale occurred in a previously unrecognized fault in Los Angeles County's San Fernando Valley, killing at least 60 people. The earthquake caused considerable damage to health facilities and significant health service disruption. Immediately after the shaking stopped, structural and nonstructural damage forced several hospitals to evacuate patients and move operations outside. Structural damage forced several older hospitals and medical buildings to cease or reduce operations. During the 1985 Mexico City earthquake, which killed an estimated 7000 people, a total of 4397 hospital beds were lost (about one in four of those available in the metropolitan Mexico City area). Hospital emergency plans in earthquake areas should provide for the contingency of evacuating patients from the wards; safely removing critical equipment from operating theaters, radiology departments, and other parts of the hospital; and re-establishing routine services for patient care.


A major earthquake in a major urban area ranks as the largest potential natural disaster in highly seismic parts of the world. Most of what can be done to mitigate injuries must be done before the earthquake occurs. Researchers have identified a number of potentially important risk factors for injuries associated (either directly or indirectly) with earthquakes. Because structural collapse is the single greatest risk factor, priority should be given to seismic safety both in planning land use and in the design and construction of safer buildings.

The integration of epidemiological studies with those of other disciplines such as engineering, architecture, social science and medical sciences is essential for improved understanding of the injuries that follow earthquakes. Better epidemiological knowledge of the risk factors for death and the type of injuries and illnesses caused by earthquakes is clearly an essential requirement for determining what relief supplies, equipment, and personnel are needed to respond effectively.

Strengthening communities' self-reliance in disaster preparedness is the most fruitful way to improve the effectiveness of relief operations. In disaster-prone areas, training and education in basic first aid and rescue methods should be an integral part of any community preparedness programme. Unfortunately, because of the relatively long periods between major earthquakes, the public health community faces a special challenge in effectively communicating the hazards posed by potential earthquakes and the need to plan and take action before an earthquake occurs.


1. Coburn A, Spence R, Earthquake protection. Chichester, John Wiley and Sons Ltd., 1992:2-12, 74-80, 277-284.

2. Chen Y, Tsoi KL, Chen F, et al. The Great Tangshan Earthquake of 1976: an anatomy of disaster. Oxford, Pergamon Press, 1988.

3. U.S. Department of Commerce, National Oceanic and Atmospheric Administration, and U.S. Department of Interior, Geological Survey. Boulder, CO, 1982, (Pub. no. 41-1). Earthquake history of the United States revised edition with supplement for 197140.

4. Coburn A, Murakami HO, Ohta Y. Factors affecting fatalities and injury in earthquakes. Internal Report, Engineering Seismology and Earthquake Disaster Prevention Planning. Hokkaido, Hokkaido University, 1987.

5. Goltz JD. The Northridge, California, earthquake of January 17, 1994: general resconnaissance report. Buffalo, NY, National Centre for Earthquake Engineering Research, 1994 (Technical Report NCEER 94-0005).

6. Hall JF. The January 17, 1994 Northridge, California earthquake: an EQE summary report. San Francisco, EQE International, 1994.

7. Benuska L, (ed.). Loma Prieta earthquake reconnaissance report. Earthquake Spectra, 1990,6 (Suppl.): 1-448.

8. EQE Engineering. The October 17, 1989 Loma Prieta earthquake: a quick look report. San Francisco, EQE Engineering, 1989.

9. Coburn A, Spence RJS, Pomonis A. Factors determining human casualty levels in earthquakes: mortality prediction in building collapse. In: Proceedings of the First International Forum on Earthquake-Related Casualties, Madrid, Spain, July 1992. Reston, VA, U.S. Geological Survey, 1992.

10. Noji EK, Kelen GD, Armenian HK, et al. The 1988 earthquake in Soviet Armenia: a case study. Annals of Emergency Medicine, 1990,19:891-897.

11. Armenian HK, Noji EK, Organessian AP. Case control study of injuries due to the earthquake in Soviet Armenia. Bull. World Health Organ. 1992,70:251-257.

12. Roces MC, White ME, Dayrit MM, Durkin ME. Risk factors for injuries due to the 1990 earthquake in Luzon, Philippines, Bull. World Health Organ. 1992,70:509-514.

13. Glass RI, Urrutia JJ, Sibony S, et al. Earthquake injuries related to housing in a Guatemalan village. Science 1977,197:638-643.

14. Mitchell WA, Wolniewicz R, Kolars JF. Predicting casualties and damages caused by earthquakes in Turkey. A preliminary report. Colorado Springs, CO, U.S. Air Force Academy, 1983.

15. Mehrain M. A reconnaissance report on the Iran earthquake. National Centre for Earthquake Engineering Research Bulletin 1991,5:1-4.

16. Coburn A, Petrovski J, Ristic D, et al. Mission report and technical review of the impact of the earthquake of 21 June 1990 in the provinces of Gilan and Zanjan. Earthquake reconstruction program formulation mission to the Islamic Republic of Iran. Geneva, United Nations Disaster Relief Office, 1990.

17. Ceciliano N, Pretto E, Watoh Y, et al. The earthquake in Turkey in 1992: a mortality study. Prehospital and Disaster Medicine, 1993,8:S 139.