Patient involvement in quality improvement: a survey comparing naturalistic and reflective approaches

Background This study investigates reflective and naturalistic approaches to patient involvement in quality improvement. The reflective approach, using, for example, interviews, provides insights into patient needs and demands to support an established improvement agenda. The naturalistic approach, for example, observations, is used to discover practical problems and opportunities that professionals are currently unaware of. Methods We assessed the use of naturalistic and reflective approaches in quality improvement to see whether they differed in their impact on patient needs, financial improvements and improved patient flows. Four possible combinations were used as a starting point: restrictive (low reflective–low naturalistic), in situ (low reflective–high naturalistic), retrospective (high reflective–low naturalistic) and blended (high reflective–high naturalistic). Data were collected through an online cross-sectional survey using a web-based survey tool. The original sample was based on a list of 472 participants enrolled in courses on improvement science in three Swedish regions. The response rate was 34%. Descriptives and ANOVA (Analysis of Variance) in SPSS V.23 were used for the statistical analysis. Results The sample consisted of 16 projects characterised as restrictive, 61 as retrospective and 63 as blended. No projects were characterised as in situ. There was a significant effect of patient involvement approaches on patient flows and patient needs at the p<0.05 level (patient flows, (F(2, 128)=5.198, p=0.007) and patient needs (F(2, 127)=13.228, p=0.000)). No significant effect was found for financial results. Conclusions Moving beyond restrictive patient involvement is important to meet new patient needs and improve patient flows. This can be done either by increasing the use of a reflective approach or by increasing the use of both reflective and naturalistic approaches. A blended approach with high levels of both is likely to produce better results in addressing new patient needs and improving patient flows.