Do Hospital Competition Measures Matter in Investigating the Patient Safety Culture in Hospitals?

Objective: The purpose of this study was to examine the relationship between hospitals neighborhood effect and safety culture.

Methods: The data were collected from the Taiwan patient safety culture survey. The survey consisted 26 core questions for 5 major aspects: teamwork climate, safety climate, job satisfaction, perception of management, and working conditions was used to measure patient safety culture. The questionnaire adopted the Likert 5-point scale: disagree strongly, disagree slightly, neutral, agree slightly and agree strongly. To ensure the quality of data, questionnaires with all identical answers and missing data by aspects were eliminated. For calculating hospital competition indices, we adopted geopolitical boundaries-based and radius-based approaches to define market area, the hospital density and Herfindahl-Hirschman Index (HHI) was used for measuring competition intensity. In addition, the average number of discharges from each sample hospital from 2005 to 2007 was used to calculate HHI. A total of 12 competition indices were produced. The Pearson's correlation and general linear model were used for model fit test. All competition indices were calculated by employing geographic information system (ArcGIS), and he SAS software Ver.9.1.3 was used for statistical analysis. Results: The safety culture survey questionnaire was distributed to 60,000 hospital staff in 200 hospitals in Taiwan in 2008. 45,242 hospitals staff responded to this survey with response rate of 78%. According to Pearson's correlation, all competition indices were significant positive correlated with SAQ score except for 4 geopolitical boundaries-based measures. We also put hospital accreditation level, ownership and size as control variables into baseline model, while these 12 competition indices were put into the baseline model respectively, as the results of Pearson's correlation, except for boundaries-based indices, the likelihood ratios were significant improved, especially the Herfindahl-Hirschman Index at 10 km. Conclusions: Our study findings suggested that social processing of market information between socially equivalents was a factor motivating patient safety culture. However, the findings show that hospitals may react to their surrounding neighbors to whom they are not necessarily socially equivalent. The structural feature of market environment is also an important factor for understanding patient safety culture.