Individual level predictors of treatment use: A disaggregation of within- and between-person associations.

OBJECTIVE: Studies showing associations between impaired clinical functioning and treatment use are predominantly based on cross-sectional comparisons or a limited number of longitudinal assessments. Using these designs, it is unknown whether individuals engage in treatment on occasions of within-person symptom increases or social support reductions. This study examined the relations between depressive symptoms, family and friend support, and self-reported stressors and treatment use. METHOD: Participants (N = 1417) completed up to seven annual assessments assessing the study constructs. Multilevel modeling was used to examine the relations between fluctuations in depressive symptoms, family and friend support, and perceived life events and treatment use concurrently and prospectively. RESULTS: We found that within-person fluctuations in depressive symptoms were associated with treatment use concurrently, as well as at the next annual assessment, controlling for family and friend support and self-reported stressors. Fluctuations in family and friend support and self-reported stressors were inconsistently associated with treatment use controlling for depressive symptoms. CONCLUSIONS: Consistent with prevailing expectations, we found that fluctuations in depressive symptoms were most consistently associated with concurrent and future treatment use. Thus, individuals are more likely to use treatment when they are experiencing higher levels of symptoms than they usually experience.