Numerous studies have investigated the relationship between depression and temporomandibular disorders (TMD), but the conclusions remain vague. The aim of this study was to examine the causal effect between depression and TMD.The reporting of this study conforms to the STROBE statement. In this retrospective cohort study, all samples were recruited from a representative subdataset of 1 million insured persons for the year 2005 Longitudinal Health Insurance Database, who were randomly selected from all beneficiaries enrolled in the National Health Insurance program of Taiwan. We used a propensity score and stratified 926,560 patients into 2 groups (propensity1 = 588,429 and propensity2 = 338,131) and 4 cohorts (propensity1 with depression = 18,038, propensity1 without depression = 570,391, propensity2 with depression = 38,656, propensity2 without depression = 299,475) to detect the development of TMD among the depressive and nondepressive patients between 2004 and 2013.The positive correlative factors of TMD included female, total number of times seeking medical advice (TTSMA) for anxiety state, TTSMA for generalized anxiety disorder, TTSMA for mandible fracture, and TTSMA for unspecified anomaly of jaw size. The propensity2 group was represented by elder and female-predominant patients who used more psychiatric health resources. Among 3 types of depression, only dysthymia (so-called chronic depression) had a causal impact on TMD in the propensity 2 group. In the propensity 2 group, the hazard ratio of dysthymia for TMD measured by Cox's regression was 1.64 (95% confidence interval 1.28–2.09), after adjusting for demographic factors, psychiatric comorbidities, and maxillofacial confounders. The first-onset mean time of TMD as the consequence of dysthymia was 3.56 years (sd = 2.74, min = 0.08, median = 2.99, max = 9.73).This study demonstrates that dysthymia increases the risk of TMD in elderly and female-predominant patients who use more psychiatric health resources.