Social and economic consequences of injury in a developing nation

Thesis (Ph. D.)--University of Washington, 1997Efforts to study and prevent injury in the developing world are hampered by lack of adequate data sources. Vital statistics registries are often incomplete and health care data may be of limited value as many injured persons do not receive medical care. To better understand the incidence, characteristics, and outcomes of injury in a developing nation, the author undertook a population based survey of injury in Ghana. Using two stage cluster sampling with probability proportional to size, a sample of 21,105 persons in 431 separate clusters were selected. Information was sought on any injury which resulted in one or more days of loss of normal activity during the preceding year. A total of 1609 such injuries were reported. In the following dissertation, the author reports on: (1) methodologic issues concerning the effect of recall on estimation of incidence rates; (2) characteristics and outcomes of transportation related injuries; and (3) health care utilization patterns. It was found that: (1) There was notable memory decay during the one year recall period. The estimated annual nonfatal injury rate declined from 27.6/100/year for a one month recall period to 7.6/100/year for a 12 month recall period (72% decline). Memory decay was minimal for more severe injuries, those with disability times of 30 or more days. In this setting, shorter recall periods (1-3 months) should be used when estimating the overall injury rate. However, longer recall periods (12 months) may be safely used to study more severe injuries. (2) In urban areas, transportation injuries accounted for 16% of injuries, but were the most severe injuries in terms of disability time and treatment costs. Hence, these injuries should be a priority for prevention efforts. (3) Of the nonfatal injuries, 58% received formal medical care, with such care being less in the rural areas (51%) compared with the urban area (68%). These data indicate that health care records would underestimate the significance of injury as a health problem in this setting and show the importance of the use of population based surveys