The relationship between age and depressive symptoms in two national surveys.

There is inconsistency in the literature on the relationship between age and depressive symptoms. Although a careful review shows that some of this inconsistency can be reconciled by recognizing the nonlinear relationship (Newmann, 1989), 2 additional issues remain unclear. One is that most previous studies used depression screening scales that contain somatic items that could introduce an age bias. The other is that most previous studies combined samples of men and women even though there is evidence that the sex difference in depressive symptoms varies with age. These 2 issues are addressed in this article, using analyzed data from 2 large national surveys. There is a consistent, but quite modest, nonlinear association between age, somatic, and nonsomatic depressive symptoms in both surveys. There is no significant sex difference in the age curves.