Epidemiology
-
This paper focuses on some statistical considerations in the estimation of dose-response in case-control studies when the exposure variables are continuous measurements. The first point is that the effects of differential variability in the exposure distributions over cases and controls cannot be differentiated from a true quadratic risk model. ⋯ Failure to do so can lead to differential variability among cases and controls and the resulting confounding with a quadratic risk model. Both of these points are illustrated by an example.
-
We conducted a case-control study to determine whether adolescents and young adults who have been in a motor vehicle crash or hospitalized for unintentional and intentional injury are at greater risk for suicide. Cases were 700 Washington State residents age 16-35 with a driver's license who died of suicide during 1987-1989. Controls were 3,494 licensed drivers matched by age, sex, and zip code. ⋯ Many of these admissions were for suicide attempts [odds ratio (OR) = 56, 95% confidence interval (CI) = 27-120], but the risk of suicide was also higher among those hospitalized for unintentional injuries (OR = 5.0, 95% CI = 2.2-11.5) and assaults (OR = 4.5, 95% CI = 1.1-18). The relative risk for suicide was 2.7 (95% CI = 2.0-3.5) for those with prior injury as a driver in a motor vehicle crash and 2.9 (95% CI = 2.2-3.8) for those with involvement in a single vehicle crash. Many unintentional injury hospitalizations and a proportion of motor vehicle crashes in younger adults may represent unrecognized suicide attempts.
-
Ethical conflicts between moral principles and methodologic standards sometimes occur in epidemiologic research. When dilemmas are discerned, they may be analyzed using the ethical principles of beneficence, nonmaleficence, justice, and respect for the autonomy of persons. We argue that, in addition to scientific validity, the welfare and rights of research subjects should be taken into account in making decisions regarding all aspects of the design and conduct of epidemiologic studies, and that the commitment of epidemiologists to the advancement of scientific knowledge should not outweigh or override all other considerations.
-
Age at menarche shows a downward secular trend and differs according to socioeconomic conditions, presumably the result of dietary variations. We report two studies conducted in Quebec City in which the relation between diet and menarche was evaluated. In 1978-1980, baseline information for these studies was obtained on girls aged 9 through 15 and included body weight, height, fatfold thickness at six sites, percent of body fat (derived from underwater weighing), and a 3-day dietary record. ⋯ Body fat, however, was not strongly related to the onset of menstruation. In both studies, a higher dietary energy intake was associated with an earlier age at menarche. Dietary composition, and dietary fat in particular, was not related to menarche.
-
The effects of barrier and spermicidal methods of contraception on cervical cancer risk were examined by studying 479 cases of histologically confirmed invasive cervical cancer cases and 788 random digit dialing controls. In addition to a detailed history of contraceptive practices, information was available on numerous potential confounders, including demographic characteristics, sexual behavior, reproductive factors, Pap smear screening history, and smoking. ⋯ Effects were only seen among subjects of higher income and education levels, suggesting that patterns of usage may be important. The potential ability of spermicides to reduce cervical cancer risk by neutralizing viral agents warrants further attention.