This study examines the relationship between marital status, antidepressant treatment response, and the dynamics underlying differential responses between marital status groups. In a seven-year, double-blind, randomized clinical trial comparing nortriptyline and sertraline for the treatment of major depression at The New York State Psychiatric Institute, multiple sources of data were collected from the 118 adults in the sample. The data, which include measures of social support and network, health behaviors and status, and depression, were analyzed to gain a better understanding of the connection between marital status and treatment response.

Based on a prior analysis of the data that found the ever-married groups had a significantly greater improvement in their depression scores than the never-married group (Pesce, Tower, & Roose, 2004), this study examines measures of constructs relating to (a) social causation, (b) social selection, and (c) depression subtype, in an attempt to explain differences between those who marry and those who do not. The research questions examined were: (a) whether participants in the ever-married group have higher scores on measures of social support and behavioral control, perhaps indicating a permanent gain in protective health benefits through marriage; (b) whether participants in the ever-married group have higher baseline scores on measures of health status and lower baseline scores on measures of depression, perhaps indicating that those who marry may be healthier prior to marriage than those who never marry; and (c) whether the distribution of melancholic and non-melancholic participants varies by marital status group.

Results indicated that the marital status groups differed with respect to (a) social network size (the ever-married had larger social networks than the never-married); (b) baseline depression ratings (the never-married had lower baseline depression ratings than the ever-married); and (c) depression subtype (the ever-married were more likely to be in the melancholic group than the never-married). Additional analyses showed that marital status moderated the impact of melancholic depression in terms of response to medication--the never-marrieds had a poorer response to both types of medication when compared with the ever-marrieds. Clinical implications and suggestions for future research are discussed.