Lower weight is usually considered advantageous to health, yet weight loss has been associated with increased mortality. An explanation for this paradox might be that the benefits of weight loss may depend on whether the loss is intentional or unintentional. The authors investigated whether intentional and unintentional weight loss differed in their associations with known risk factors for morbidity and mortality in a nationally representative sample of the US population. The sample consisted of 9,144 persons, aged 45 years and older, who answered questions regarding 1-year weight change in the diabetes risk factor Current Health Topic of the 1989 National Health Interview Survey (NHIS). Statistical analyses incorporated the sample weights and characteristics of the survey design. Relative to a common referent group, the factors associated with weight loss differed depending on whether the loss was defined as intentional loss, as unintentional loss, or regardless of intention. Restricting analysis to the 1,999 persons who lost weight, unintentional relative to intentional weight loss was significantly (p < 0.05) associated with older age, poorer health status, smoking, lower body mass index, and, in men only, widowhood and less education. Thus, unintentional weight loss may serve as a marker for factors that characterize persons at greater risk of mortality than persons undergoing intentional weight loss. Also, intention to lose weight may help clarify the relation between weight loss and mortality that, to this point, has shown counterintuitive results. Studies of the relation between weight loss and mortality should incorporate intention as a factor in the analysis.

Elevated activities of serum aminotransferases are a common sign of liver disease and are observed more frequently among diabetics than in the general population. Whether this association is due to confounding factors is unknown. The authors investigated whether diabetes was significantly associated with elevated serum activity of alanine aminotransferase (ALT) after adjustment for factors common to both diabetes and raised ALT. Data from 2,999 men and women aged 20-74 years representative of the Mexican American population of the southwestern United States were obtained from the Hispanic Health and Nutrition Examination Survey (1982-1984). Approximately 6% of men and 2% of women had elevated serum ALT activity (>43 IU/liter). The odds ratio for diabetes as a predictor of elevated ALT was 4.1 (95% confidence interval 2.3-7.6) adjusted for age and sex, which decreased to 3.0 (95% confidence interval 0.92-9.74) after adjustment for age, sex, body mass index, alcohol consumption, and other factors. In addition to diabetes, body mass index was also significantly (p < 0.05) associated with elevated ALT activity. Heavier alcohol consumption and male sex increased the likelihood of elevated ALT, whereas coffee consumption reduced it. Diabetes and liver injury appear to be associated, even with control for factors in common.

No abstract