BACKGROUND: As many as 47% of adults over age 50 discontinue treatment with antidepressants after redeeming only one prescription. The study aim was to assess the risk of suicide in adults aged 50+ who discontinue antidepressants at an early stage of treatment. METHOD: Case control study of all individuals aged 50+ living in Denmark and who initiated antidepressant treatment between July 1st 1995 and December 31st 2000 (N=217,123). Hazard ratios were calculated using Cox regression analyses, propensity score matching techniques, and marginal structural models. RESULTS: During the study period, 78,594 men and 138,529 women aged 50+ began treatment with an antidepressant medication, of whom 309 men and 229 women died by suicide. Men aged 50+ who discontinued treatment early had a suicide rate of 167 per 100,000 compared with 175 per 100,000 in those who continued refilling prescriptions; hazard ratio=0.98 [CI-95%: 0.78-1.23]. The suicide rate in women who discontinued treatment was 52 per 100,000 compared with 74 per 100,000 in those who continued refilling; hazard ratio=0.72 [CI-95%: 0.55-0.94]. Although people with previous psychiatric hospitalizations had greater risk of suicide than those without past hospital admissions, the difference was not significant in the adjusted model. LIMITATIONS: Prescriptions redeemed at pharmacies are our only indicator of treatment adherence. Also, information on severity of depression was not available. CONCLUSIONS: We did not find a lower suicide risk among people over age 50 who seemingly follow treatment in comparison with those who discontinued treatment with antidepressants at an early stage.