Psychosocial stressors are commonly classified according to such recent life events (RLE) as arguments and exits. Based on retrospective studies, exit events, e.g., deaths and marital separations, are generally believed to precede the onset of depression. We are reporting the first prospective study of RLE in 123 addicts, who were evaluated for depression using the Beck Inventory, then treated for drug abuse and reevaluated 6 months later for depression and the occurrence of intervening RLE. During the 6 months, only 31 per cent (38) of the addicts either remained (26) or became depressed (12), but 98 per cent (120) had RLE with a mean of 6.3 +/- 2.9. Both "normals" (1.1) and primary depressives (3.4) have fewer RLE. Among the 36 addicts with low numbers of RLE (zero to four), 25 per cent either remained or became depressed, while among the 26 addicts with high numbers of RLE (nine to 15), 50 per cent either remained or became depressed. When RLE were categorized into exit and argument events, arguments had a stronger relationship than exits to depression. Furthermore, only arguments showed the expected relationship to recovery from depression, that is, addicts who recovered from depression were less likely to have argument events than addicts who remained depressed. Thus, among the 58 addicts who had argument RLE, 45 per cent either remained or became depressed, 36 per cent recovered from depression, and 19 per cent remained asymptomatic; among the 60 addicts who had exit RLE, 36 per cent either remained or became depressed, 40 per cent recovered from depression, and 23 per cent remained asymptomatic. When the post-therapeutic analysis included several outcome events other than depression, argument RLE were most strongly related to depression and exit RLE to resumption of illicit drug use during treatment. We conclude that recovery from depression in addicts is related to negative RLE such as arguments and that treatment should focus on helping addicts control their response to these RLE to facilitate recovery from depression.