Humanitarian settings in Pakistan face major gaps in mental health service delivery, with limited specialist resources and a high burden of untreated conditions. Though the World Health Organization’s mhGAP-Humanitarian Intervention Guide aims to strengthen the capacity of primary care physicians to identify and manage common mental health conditions, the evidence for multidimensional training outcomes remains limited. This study evaluated the effectiveness of mhGAP-HIG training outcomes in terms of knowledge, skills, attitudes and confidence among physicians across different humanitarian contexts in Pakistan. As part of implementing an evidence-based digital MHPSS service model, seven mhGAP-HIG workshops were held across two provinces and a federal area. We conducted a quasi-experimental, mixed-methods evaluation where a total of 149 physicians completed standardized assessments of knowledge (mhGAP-HIG), therapeutic skills (ENhancing Assessment of Common Therapeutic Factors), attitudes (Mental Illness: Clinician’s Attitudes Scale-v4) and confidence, before and after the training. Paired sample t-tests and ANCOVA were used to assess within-group and between-group changes, respectively. Quantitative feedback and qualitative reflections were also evaluated. We found significant improvements across the domains of knowledge, skills, confidence and a modest improvement in attitude. The participants from Khyber Pakhtunkhwa showed greater gains in knowledge, while participants from Baltistan showed greater gains in skills, as compared to other provinces. Participants expressed a high level of satisfaction, reporting improvements in knowledge, confidence and ability to identify and manage common mental health conditions. Systematic efforts to adapt and implement mhGAP-HIG may lead to significant improvements across multidimensional competencies of physicians.