Hospital survey on patient safety culture (HSOPS): variability of scoring strategies.


Objective To assess the variability of safety culture dimension scores and their associated rankings depending on three different scoring strategies using the Hospital Survey On Patient Safety Culture (HSOPS). Design Cross-sectional study using a self-administered questionnaire. Setting The study was conducted in an 1836-bed acute-care French university hospital with an annual volume of 135 999 stays, between April 2013 and November 2014. Participants All caregivers and technical-administrative staff with at least 6 months of employment, spending at least half of their working time in the hospital, were asked to participate. Intervention None. Main outcome measure The variability of the HSOPS results using three different scoring methods: the percentage of positive responses recommended by the Agency for Healthcare Research and Quality, the averaged individual means and the averaged individual sums. Results The response rate was 78.6% (n = 3978). The percentage of positive responses resulted in lower scores compared to averaged individual means and averaged individual sums in the six least developed dimensions, and gave more widely spread scores and greater 95CIs in the six most developed dimensions. Department rankings also varied greatly depending on the scoring methods. Conclusion The values of the HSOPS scores and their corresponding rankings greatly depended on the computation method. This finding shows how important it is to agree on the use of the same scoring strategies, before broadly comparing results within and across organizations.

DOI: 10.1093/intqhc/mzx086

Cite this paper

@article{Giai2017HospitalSO, title={Hospital survey on patient safety culture (HSOPS): variability of scoring strategies.}, author={Joris Giai and Bastien Boussat and Pauline Occelli and Gerald Gandon and Arnaud Seigneurin and Philippe Michel and Patrice François}, journal={International journal for quality in health care : journal of the International Society for Quality in Health Care}, year={2017}, volume={29 5}, pages={685-692} }