Abstract Background District health managers play a key role in the effectiveness of decentralized health systems in low- and middle-income countries. Inadequate management and leadership skills often hamper their ability to improve quality of care and effectiveness of health service delivery. Nevertheless, significant investments have been made in capacity-building programmes based on site-based training, mentoring, and operational research. This systematic review aims to review the effectiveness of site-based training, mentoring, and operational research (or action research) on the improvement of district health system management and leadership. Our secondary objectives are to assess whether variations in composition or intensity of the intervention influence its effectiveness and to identify enabling and constraining contexts and underlying mechanisms. Methods We will search the following databases: MEDLINE, PsycInfo, Cochrane Library, CRD database (DARE), Cochrane Effective Practice and Organisation of Care (EPOC) group, ISI Web of Science, Health Evidence.org, PDQ-Evidence, ERIC, EMBASE, and TRIP. Complementary search will be performed (hand-searching journals and citation and reference tracking). Studies that meet the following PICO (Population, Intervention, Comparison, Outcome) criteria will be included: P: professionals working at district health management level; I: site-based training with or without mentoring, or operational research; C: normal institutional arrangements; and O: district health management functions. We will include cluster randomized controlled trials, controlled before-and-after studies, interrupted time series analysis, quasi-experimental designs, and cohort and longitudinal studies. Qualitative research will be included to contextualize findings and identify barriers and facilitators. Primary outcomes that will be reported are district health management and leadership functions. We will assess risk of bias with the Cochrane Collaborationâ s tools for randomized controlled trials (RCT) and non RCT studies and Critical Appraisal Skills Programme checklists for qualitative studies. We will assess strength of recommendations with the GRADE tool for quantitative studies, and the CERQual approach for qualitative studies. Synthesis of quantitative studies will be performed through meta-analysis when appropriate. Best fit framework synthesis will be used to synthesize qualitative studies. Discussion This protocol paper describes a systematic review assessing the effectiveness of site-based training (with or without mentoring programmes or operational research) on the improvement of district health system management and leadership. Systematic review registration PROSPERO CRD42015032351