Abstract
Purpose: The quality of health care is often measured using quality indicators, which can be utilized to compare the performance of health-care providers. Conducting comparisons in a meaningful and fair way requires the quality indicators to be adjusted for patient characteristics and other individual-level factors. The aims of the study were to develop and test a case-mix adjustment model for quality indicators based on patient-experience surveys among inpatients receiving interdisciplinary treatment for substance dependence, and to establish whether the quality indicators discriminate between health care providers.
Patients and methods: Data were collected through two national surveys involving inpatients receiving residential treatment in Norway in 2013 and 2014. The same questionnaire was used in both surveys, and comprised three patient-experience scales. The scales are reported as national quality indicators, and associations between the scales and patient characteristics were tested through multilevel modeling to establish a case-mix model. The intraclass correlation coefficient was computed to assess the amount of variation at the hospital-trust level.
Results: The intraclass correlation coefficient for the patient-reported experience scales varied from 2.3% for “treatment and personnel” to 8.1% for “milieu”. Multivariate multilevel regression analyses showed that alcohol reported as the most frequently used substance, gender and age were significantly associated with two of the three scales. The length of stay at the institution, pressure to be admitted for treatment, and self-perceived health were significantly related to all three scales. Explained variance at the individual level was approximately 7% for all three scales.
Conclusion: This study identified several important case-mix variables for the patient-based quality indicators and systematic variations at the hospital-trust level. Future research should assess the association between patient-based quality indicators and other quality indicators, and the predictive validity of patient-experience indicators based on on-site measurements.
Acknowledgments
We thank Marit Seljevik Skarpaas for performing data collection and management, and Inger Opedal Paulsrud and Linda Selje Sunde for administrative help with the data collection. We also thank Olaf Holmboe for providing the case-mix adjusted indicator scores using R software. We are grateful to the contact persons and project-management professionals in the departments, institutions, and health regions concerned. We also thank the patients who participated in the surveys. The Norwegian Knowledge Centre for the Health Services (now part of the NIPH) was responsible for the planning and execution of the data collections that are described in this manuscript. The NIPH is responsible for the current study. The Norwegian Directorate of Health funded the national surveys.
Ethics approval and informed consent
Data were collected anonymously, with no registration of the patients being surveyed. The project was run as part of the national program and was an anonymous quality assurance project. According to the Norwegian Regional Committees for Medical and Health Research Ethics, research approval is not required for quality assurance projects. The Norwegian Social Science Data Services states if the information used are anonymous, the project is not subject to notification (http://www.nsd.uib.no/personvern/en/notification_duty/meldeskjema?eng). Hence, no ethics approval was needed in this project. Patients were informed that participation was voluntary and that they would remain anonymous. In accordance with all the patient surveys in the national program, health professionals at the institutions could exclude individual patients for special ethical considerations. Since no notification or ethics approval was needed, the NIPH obtained signed agreements with all the participating institutions, describing the project and both the institutions’ and NIPH’s responsibility in data collecting, handling, analyzing and reporting. Previously established guidelines concerning consent through a returned questionnaire were applied.
Data availability
The datasets generated and analyzed during the current study are not publicly available due to this being one part of an ongoing PhD-project at the NIPH and the University of Oslo. Request for permission to use the data must be directed to the NIPH.
Author contributions
All authors made substantial contributions to conception and design, acquisition of data, or analysis and interpretation of data; took part in drafting the article or revising it critically for important intellectual content; gave final approval of the version to be published; and agree to be accountable for all aspects of the work.
Disclosure
The authors report no conflicts of interest with this work.