PubMed1990
In September 1983, we selected 30 villages in four rural counties of northern Haiti for a partial census and malaria prevalence survey. A cohort of 1,577 persons was enumerated in this census. Survey teams revisited the same houses in these villages in September 1984 and updated the previous census, inquiring about all listed family members. We administered an additional questionnaire to each household concerning the occurrence over the past year of deaths, births, pregnancies, and migration. Among the 1,218 persons who had been followed for one year, there were 21 deaths (crude mortality rate: 17 per 1,000 population; 95% confidence interval [95% CI]: 12 per 1,000, 25 per 1,000) and 35 births (birth rate: 29 per 1,000 population year; 95% CI: 19 per 1,000, 38 per 1,000). The infant mortality rate was 171 per 1,000 live births (95% CI: 81 per 1,000, 315 per 1,000), and the mortality rate for children less than 5 years old was 36 per 1,000 (95% CI: 24 per 1,000, 56 per 1,000). Although small sample surveys are subject to limitations of precision dependent on sample size, they can be a simple method by which researchers may estimate vital statistics for rural areas of less-developed countries.