Objectives To evaluate the effect of psychotherapy for depression in patients receiving disability benefits.

Methods Using administrative data from a large Canadian, private, disability insurer, we evaluated the association between the provision of psychotherapy and other potentially predictive factors with time to long-term disability (LTD) claim closure.

Results We analysed 10,338 LTD claims in which depression was the primary disabling complaint. Depression management included psychotherapy in 1580 (15.3%) LTD claims. In our adjusted analyses, receipt of psychotherapy was associated with faster claim closure (hazard ratio [HR] = 1.42; 95% confidence interval [CI] = 1.33 to 1.52). Older age per decade (0.83 [0.80 to 0.85] respectively), a primary diagnosis of recurrent depression (0.80 [0.74 to 0.87]) versus major depression, a secondary psychological (0.77 [0.72 to 0.81]), or non-psychological diagnosis (0.66 [0.61 to 0.71]), a longer time to claim approval (0.995 [0.992 to 0.998], and an administrative services only policy (0.87 [0.78 to 0.96] or refund policy (0.73 [0.69 to 0.77]) versus non-refund policy were associated with longer time to claim closure. Residing in the Prairies (1.46 [1.35 to 1.57]) and Quebec (1.93 [1.82 to 2.05]) versus Ontario were associated with faster LTD claim closure.

Conclusions We found multiple factors, including psychotherapy, which were predictive of time to LTD claim closure. Our findings may however be influenced by selection bias and other biases that present challenges to the analysis and interpretation of administrative data, and highlight the need for well-designed prospective studies.