Potential errors resulting from sex and age difference in assessing family history of diabetes.

Abstract

Diabetes mellitus occurs nearly exponentially with aging and its occurrence differs between men and women in adulthood. Therefore, the sex and age of family members should be considered in assessing the family history. In this report the effects of sex and age on the positivity of family history were estimated numerically. Sex- and age-specific proportion of a positive history of diabetes mellitus among 24,273 family members was obtained from a questionnaire survey of 2,316 high school students in Japan. By analyzing the sex- and age-specific proportion with the logistic regression model, odds ratios were estimated which indicated potential bias or misclassification resulting from sex and age differences. The odds ratios were 1.97 (95% confidence interval, 1.74-2.23) for the sex difference and 1.05 (95% confidence interval, 1.04-1.05) for an age difference of 1 year. This indicated that a male family member had a 1.97 times higher chance of having a positive history than a female member and that a positive history increased by (1.05)y, where y was age difference in years. A control for sex and age of family members will be required in assessing the family history of diabetes mellitus as a risk factor.