The Relationship Between Pain Severity, Insomnia, and Suicide Attempts Among a National Veteran Sample Initiating Pain Care.

OBJECTIVE:We assessed the longitudinal association of suicide attempts by moderate to severe pain and insomnia before and after the initiation of pain services among veterans. METHODS:A cohort of 221,817 veterans initiating pain care was divided into four subgroups: a) no/mild pain + no insomnia (LowPain-NoINS), b) no/mild pain + insomnia (LowPain-INS), c) moderate/severe pain + no insomnia (HighPain-NoINS), and d) moderate/severe pain + insomnia (HighPain-INS). Data on diagnoses, pain severity, demographics, medications, and suicide attempts were extracted from Veterans Health Administration data sets. RESULTS:Overall, there were 2227 (1.0%) suicide attempts before initiating pain services and 1655 (0.8%) after initiating pain services. Cox proportional hazard models accounting for key covariates revealed that patients in the HighPain-INS group were significantly more likely to attempt suicide in the year after the initiation of pain services relative to all subgroups (versus LowPain-NoINS: hazard ratio [HR] = 1.44, 95% confidence interval [CI] = 1.21-1.72; versus LowPain-INS: HR = 1.71, 95% CI = 1.23-2.38; versus HighPain-NoINS: HR = 1.17, 95% CI = 1.01-1.34) even after accounting for prior attempts. Adjusted logistic regression analyses found that patients with moderate/severe pain and insomnia had higher odds of attempting suicide in the year before initiating pain services compared to all subgroups (versus LowPain-NoINS: HR = 1.75, 95% CI = 1.50-2.05; versus LowPain-INS: HR = 1.41, 95% CI = 1.09-1.82; versus HighPain-NoINS: HR = 1.21, 95% CI = 1.07-1.37). CONCLUSIONS:These results suggest that those with both moderate/severe pain and insomnia are more likely to have a history of suicide attempts and are at greater risk of a suicide attempt relative to those with insomnia with low/mild pain and those with moderate/severe pain with no insomnia. Suicide prevention efforts for chronic pain and insomnia could address pain and insomnia within the same intervention or in parallel.