Table 3

Explanations and examples of the main risks of bias within the health service accreditation literature

Main risks of biasExplanations of main risks of biasExemplifying references
Non-randomised comparisonDirect causal relationships between accreditation programmes and measures of health service quality and safety cannot be easily inferred using descriptive study designs (eg, cross-sectional, matched cohort and case control studies)41–43
Detection biasA variety of indicators are often used within studies to evaluate the impact of accreditation on health service processes of care, patient outcomes and other aspects of quality, such as financial sustainability. The validity and reliability of employed measures is infrequently justified and often debatable28 44–46
Performance biasProposed accreditation effects may be due to other factors, such as leadership or quality management activities concurrently undertaken by health services. Evaluation studies seldom account for factors that may potentially confound the relationship between accreditation programmes and outcomes measured6 24 27 38
Limited reporting and use of rigorous qualitative methodologiesInsufficient reporting of employed data collection and analysis methodologies impacts the credibility of qualitative studies. There is also limited use of rigorous methodologies, such as triangulation, to enhance the credibility and complexity of findings47–50