Although the hospital is succeeding in providing regular training in venepuncture, this is only one aspect of quality. The process and outcome also need interventions.

Background Venepuncture involves the introduction of a needle into a vein to collect a representative blood sample for laboratory testing. In the pre‐analytical phase, haemolysis (the rupturing of erythrocytes and release of their contents into the extracellular compartment) has safety, quality and cost implications. Training in correct venepuncture practice has the potential to reduce in vitro haemolysis rates, but the evidence for this notion has yet to be synthesised. Design Systematic review (PRISMA Checklist). Methods Published studies on the effectiveness of venepuncture training on haemolysis rates were searched in relevant databases. The McMaster critical appraisal tool was used to assess methodological quality. The GRADE tool was used to evaluate the body of evidence in relation to the research questions. Implementation fidelity was also scrutinised in each study. Results Eight out of 437 retrieved studies met the inclusion criteria. None were randomised controlled trials (RCT). Between‐study heterogeneity in design, intervention characteristics and the biochemical threshold for haemolysis precluded a meta‐analysis. Post‐training reductions in haemolysis rates of between 0.4%–19.8% were reported in four of the studies, which developed their intervention according to a clear evidence base and included mentoring in the intervention. Rises in haemolysis rates of between 1.3%–1.9% were reported in two studies, while the intervention effect was inconsistent within two other studies. Conclusion There are no RCTS on the effectiveness of venepuncture training for reducing haemolysis rates, and findings from the existing uncontrolled studies are unclear. For a more robust evidence base, we recommend more RCTs with standardisation of haemolysis thresholds and training‐related factors. Relevance to clinical practice While venepuncture training is an important factor influencing quality of blood sample in clinical practice, more robust evidence is needed to make specific recommendations about training content for reduction of haemolysis rates. Standardisation of haemolysis thresholds would also enable future meta‐analyses.