Predictors of response rates of safety culture questionnaires in healthcare: a systematic review and analysis

Abstract

Objectives Safety culture questionnaires are widely used in healthcare to understand how staff feel at work, their attitudes and views, and the influence this has on safe and high-quality patient care. A known challenge of safety culture questionnaires is achieving high response rates to enhance reliability and validity. This review examines predictors of response rates, explores reasons provided for low response rates and identifies strategies to improve those rates.Methods Four bibliometric databases were searched for studies assessing safety culture in hospitals from January 2008 to May 2022. Data were extracted from 893 studies that included a safety culture questionnaire conducted in the hospital setting. Regression was used to predict response rate based on recruitment and data collection methods (incentives, reminders, method of survey administration) and country income classification. Reasons for low response were thematically analysed and used to identify strategies to improve response rates.Results Of the 893 studies that used a questionnaire to assess safety culture in hospitals, 75.6% reported a response rate. Response rates varied from 4.2% to 100%, with a mean response rate of 66.5% (SD=21.0). Regression analyses showed that safety culture questionnaires conducted in low-income and middle-income countries were significantly more likely to yield a higher response rate compared with high-income countries. Mode of administration, questionnaire length and timing of administration were identified as key factors. Remote methods of questionnaire data collection (electronic and by post) were significantly less likely to result in a higher response rate. Reasons provided for low response included: survey fatigue, time constraints, and lack of resources.Conclusion This review presents strategies for researchers and hospital staff to implement when conducting safety culture questionnaires: principally, distribute questionnaires in-person, during meetings or team training sessions; use a local champion; and consider the appropriate time of distribution to manage survey fatigue.