Implementation and data-related challenges in the Deepening our Understanding of Quality in Australia (DUQuA) study: implications for large-scale cross-sectional research. (5th February 2020)

• Record Type: Journal Article
• Authors: Arnolda, Gaston; Winata, Teresa; Ting, Hsuen P; Clay-Williams, Robyn; Taylor, Natalie; Tran, Yvonne; Braithwaite, Jeffrey
• Is Part Of: International journal for quality in health care. Volume 32(2020)Supplement 1
• Page Start: 75
• Page End: 83
• Publication Date: 2020-02-05
• Subjects: methods -- cross-sectional studies -- health services research -- quality of healthcare -- patient safety -- hospitals; Medical care -- Quality control -- Periodicals; 362.1068
• DOI: 10.1093/intqhc/mzz108
• Languages: English
• ISSNs: 1353-4505

Abstract

Abstract: Healthcare organisations vary in the degree to which they implement quality and safety systems and strategies. Large-scale cross-sectional studies have been implemented to explore whether this variation is associated with outcomes relevant at the patient level. The Deepening our Understanding of Quality in Australia (DUQuA) study draws from earlier research of this type, to examine these issues in 32 Australian hospitals. This paper outlines the key implementation and analysis challenges faced by DUQuA. Many of the logistical difficulties of implementing DUQuA derived from compliance with the administratively complex and time-consuming Australian ethics and governance system designed principally to protect patients involved in clinical trials, rather than for low-risk health services research. The complexity of these processes is compounded by a lack of organizational capacity for multi-site health services research; research is expected to be undertaken in addition to usual work, not as part of it. These issues likely contributed to a relatively low recruitment rate for hospitals (41% of eligible hospitals). Both sets of issues need to be addressed by health services researchers, policymakers and healthcare administrators, if health services research is to flourish. Large-scale research also inevitably involves multiple measurements. The timing for applying these measures needs to be coherent, to maximise the likelihood of finding real relationships between quality and safety systems and strategies, and patient outcomes; this timing was less than ideal in DUQuA, in part due to administrative delays. Other issues that affected our study include low response rates for measures requiring recruitment of clinicians and patients, missing data and a design that necessarily included multiple statistical comparisons. We discuss how these were addressed. Successful completion of these projects relies on mutual and ongoing commitment, and two-way communication between the research team and hospital staff at all levels. This will help to ensure that enthusiasm and engagement are established and maintained.