OBJECTIVES: In the past decade, there was a marked increase in the number of tramadol-related deaths in the UK, and these fatalities occurred in persistent tramadol users prescribed antidepressants concomitantly. However, the prevalence of such combinations has not been explored in the UK. This study took a prescription sequence symmetry analysis approach to explore the prevalence of antidepressants prescription before and after the initiation of tramadol. METHODS: This retrospective cohort study used the Clinical Practice Research Datalink. Patients who were prescribed tramadol from 2000 to 2011 were identified and prescriptions of tramadol and antidepressants were followed 9 years before and after the date of the first tramadol prescription. Patients who were issued at least one tramadol prescription consecutively in each follow-up patient year were defined as persistent users. The proportions of tramadol users combined with antidepressants before and after tramadol initiation were compared by Chi square test and stratified by non-persistent and persistent tramadol users. RESULTS: Of the 526,624 included tramadol users, 22,003 (4.2%) were persistent users. The proportion of co-prescribing antidepressants in tramadol users significantly increased from 44.1% to 49.3% after the initiation of tramadol (P<0.001). Of those co-prescribed with antidepressants, the proportions of co-prescribing tricyclic antidepressants and selective serotonin reuptake inhibitors (SSRIs) were higher than other antidepressants. After tramadol initiation, proportions of co-prescribing SSRIs changed +0.2% and -1% in persistent and non-persistent groups, respectively. Comparing to persistent group, the prevalence of co-prescribing SSRIs was higher in non-persistent groups (56.2% vs. 49.9%) after tramadol initiation. CONCLUSIONS: The increased prevalence of co-prescription of antidepressants after initiation of tramadol shows that adverse drug-drug interactions might be underestimated by prescribers. Further studies are needed to evaluate the impact of such co-prescription on clinical outcomes and resource utilisation.