Abstract Background Methods to enhance the accuracy of the depression diagnosis continues to be of relevance to clinicians. The primary aim of this study was to compare the diagnostic precision of two different diagnostic strategies using the Mini International Neuropsychiatric Interview (MINI) as a reference standard. A secondary aim was to evaluate accordance between depression severity found via MINI and mean Major Depression Inventory (MDI) sum-scores presented at referral. Methods This study was a two-armed, cluster-randomized superiority trial embedded in the Collabri trials investigating collaborative care in Danish general practices. GPs performing case-finding were instructed always to use MDI when suspecting depression. GPs performing usual clinical assessment were instructed to detect depression as they would normally do. According to guidelines, GPs would use MDI if they had a clinical suspicion, and patients responded positively to two or three core symptoms of depression. We compared the positive predictive value (PPV) in the two groups. Results Fifty-one GP clusters were randomized. In total, 244 participants were recruited in the case-finding group from a total of 19 GP clusters, and 256 participants were recruited in the usual clinical assessment group from a total of 19 GP clusters. The PPV of the GP diagnosis, when based on case-finding, was 0.83 (95% CI 0.78–0.88) and 0.93 (95% CI 0.89–0.96) when based on usual clinical assessment. The mean MDI sum-scores for each depression severity group indicated higher scores than suggested cut-offs. Conclusions In this trial, systematic use of MDI on clinical suspicion of depression did not improve the diagnostic precision compared with the usual clinical assessment of depression. Trial registration The trial was retrospectively registered on 07/02/2016 at ClinicalTrials.gov. No. NCT02678845.