American journal of epidemiology
-
The authors have conducted an analysis to validate a computer model that uses age-specific human immunodeficiency virus (HIV) prevalence data to estimate age-specific HIV incidence rates. Data for the analysis are from a cohort study of volunteer male workers in Bujumbura, the capital city of Burundi. Testing for HIV prevalence was conducted at baseline, and HIV-negative subjects were retested annually from 1990 to 1993 to determine rates of seroconversion. ⋯ Expected numbers of cases within age intervals did not differ significantly from observed number of cases. The authors conclude that the model proved to be successful in approximating observed incidence rates and that it is a useful tool, particularly in countries where prevalence data are available and where HIV prevalence has stabilized, which is when the underlying assumptions in the model are best met. The model provides crucial information about incidence rates that might not be evident from prevalence data alone.
-
Review
A critical look at methods for handling missing covariates in epidemiologic regression analyses.
Epidemiologic studies often encounter missing covariate values. While simple methods such as stratification on missing-data status, conditional-mean imputation, and complete-subject analysis are commonly employed for handling this problem, several studies have shown that these methods can be biased under reasonable circumstances. The authors review these results in the context of logistic regression and present simulation experiments showing the limitations of the methods. ⋯ While these methods are superior to simple methods, they are not commonly used in epidemiology, no doubt due to their complexity and the lack of packaged software to apply these methods. The authors contrast the results of multiple imputation to simple methods in the analysis of a case-control study of endometrial cancer, and they find a meaningful difference in results for age at menarche. In general, the authors recommend that epidemiologists avoid using the missing-indicator method and use more sophisticated methods whenever a large proportion of data are missing.
-
Biography Historical Article
A new perspective on John Snow's communicable disease theory.
When John Snow undertook the studies of the cholera epidemic of 1854 in London, he was testing his theory of communicable disease, which had been enunciated in an oration delivered at the 80th anniversary of the Medical Society of London. Snow had been elected orator of the year for 1853 and, according to his biographer, had spent the better part of a year in preparation. ⋯ Although Snow's theory is similar to Jacob Henle's formulations of a decade earlier, it is more precise, more comprehensive, and more explicit. On the basis of this work alone, Snow deserves broader recognition than he has received.
-
The effects of marital status and change in marital status on mortality among middle-aged British men were examined in a prospective cohort study, the British Regional Heart Study. This is a nationally representative cohort of men selected at random from general medical practices in 24 towns in England, Wales, and Scotland. It comprises 7,735 men aged 40-59 recruited in 1978-1980 and followed up for 11.5 years. ⋯ It is unlikely that selection bias, chance, or artifact is responsible for the general relation between marital status and mortality. Variable and incomplete control for confounding by socioeconomic status and risk factors for common diseases may explain some of the inconsistencies observed between studies and between different categories of unmarried men (i.e., never-married, widowed, and divorced). It is possible that the social support offered by marriage exerts a protective effect for some men.