In the third of a series of articles that Croakey is running on the Choosing Wisely campaign (see the other two here and here), Professor Jon Karnon highlights the role of health services research and argues that it has an essential contribution to make if this initiative is to be successful in improving the quality and safety of health care in Australia. He writes:
Choosing Wisely is a high profile quality improvement initiative, initiated in the US, which has spread to other countries, and is about to be launched in Australia. It aims to improve health care quality by reducing the use of technologies for which existing evidence shows they provide no benefit, and in some cases, lead to harm. In Australia, medical colleges are being asked to develop a list of ‘Five things providers and patients should question’.
Interventions to improve clinical decision-making, such as the Choosing Wisely initiative, are derived from health services research. Whilst medical (or clinical) research focuses on the development and evaluation of specific technologies, health services research is concerned with improving approaches to organising, financing, and managing health care systems and services.
Using Choosing Wisely as an example, this blog illustrates the capacity of health services research to contribute to a sustainable health care system that provides high quality health care to all Australians.
Identifying priority technologies for improvement
The first stage in the Choosing Wisely process is the identification of technologies that ‘providers and patients should question’. Initial priority might be given to high impact technologies (e.g. low cost, high frequency, or high cost, low frequency interventions), but also to technologies for which there are fewer barriers to changing current practice.
In the US, whilst the inclusion of clinical leaders has been recognized as an important step forward, there has been criticism around the variability, and potential for bias, of the methods used to select targets for improvement.
A transparent and robust health services research methodology is required. Initial qualitative stakeholder engagement, using appropriate survey methods, could identify a range of technologies that are perceived to have capacity for improved use. Stakeholders could include health professionals, patients, and the broader community.
The survey methods could include structured, or semi-structured questions regarding potential barriers and facilitators to change in proposed technologies. Such information may also inform the selection of priority technologies.
Subsequent reviews of the existing evidence will either confirm or reject the validity of the capacity for improvement in the use of proposed technologies. Alternatively, a review might identify significant uncertainty around the appropriate use of a technology, and highlight the value of targeted research to reduce uncertainty.
Evidence on the inappropriateness of a technology’s use would be complemented by the collection and analysis of relevant health care data to validate and quantify the magnitude of the perceived inappropriate use of specific technologies.
Examples of the use of health services research to identify priority areas for improvement include recent research undertaken in South Australia. Using data that are routinely collected when patients are treated in a hospital, significant variations in processes of care, patient outcomes, and health service costs were observed between patients presenting at four different emergency departments with chest pain. After controlling for relevant differences in the characteristics of presenting patients, at one hospital, an extra 2 in every 100 presenting patients experienced a subsequent cardiovascular-related inpatient admission or death within 12 months of presenting with chest pain.
The implication of these findings is that services and patient outcomes could be improved at one of the four hospitals. Further effort and resources are required to investigate and implement improved quality, but with the hospital with poorer outcomes treating over 1,500 patients per year, up to 30 significant clinical events could be avoided each year at this one hospital.
Designing and implementing effective interventions
The next phase involves the design and implementation of strategies to reduce inappropriate use of the prioritised technologies. One size is unlikely to fit all, and implementation strategies might need to vary from clinical group to clinical group, and from technology to technology.
Health services researchers have experience in many areas that could support such strategies, including the effective use of decision support systems, ‘shared decision making’ to support informed conversations between doctors and their patients, and stakeholder engagement to support the five principles underlying Choosing Wisely: physician led, patient focused, evidence based, multi-professional, and transparent.
Examples include a recent study of the cognitive determinants of hand hygiene decisions for health care workers. Thematic analysis of a qualitative data collected through focus group discussions identified a range of potential facilitators, including peer-based initiatives to increase shared responsibility, and a role for patients in encouraging compliance. Supply side issues included the lack of appropriate materials to reduce hand damage, and the finding that sensor taps were a barrier due to environmental concerns.
Evaluating intervention effectiveness and cost-effectiveness
For the Choosing Wisely initiative, the strategy to improve the use of targeted technology can be considered a separate intervention – each of which can be evaluated to identify successful strategies, which may inform subsequent interventions. Unanticipated consequences may also be identified, such as reduced use in patients for whom a targeted technology is considered to be appropriate.
The randomised controlled trial is often cited as the gold standard for evaluating intervention effects, though observational studies can be used when randomisation is not possible. Indeed, observational studies can be more useful as they often better reflect the effects of an intervention in routine practice.
The Choosing Wisely initiative will not be implemented nationally, and so a carefully designed observational study will be required. The capacity for high quality non-randomised evaluations of health service interventions has never been greater. A key driver has been more timely access to increasingly detailed and accurate, routinely collected health systems data.
Retrospective and prospective data describing the characteristics of patients receiving targeted technologies, the processes of care within which technologies are used, and patient outcomes will inform whether Choosing Wisely is changing practice appropriately, and the impact of that change on costs and patient outcomes.
Examples of health service evaluations using routinely collected data include a recent study of the effects of practice nurse involvement in the management of chronic disease in a primary care setting. This study collected routinely collected data from general practices, hospitals, and Medicare to quantify the important role practice nurses have to play in managing chronic disease in the community.
Prioritising health services research
The delivery of health care is a complex process, and variation in practice is inherent. The challenge is to identify is to identify areas with capacity for improvement, and to intervene to increase service quality. To meet this challenge, the recent Strategic Review of Health and Medical Research in Australia highlighted health services research as a priority area for investment.
This blog has described research options to inform the Choosing Wisely initiative, as well as providing examples of completed health services research: identifying variation in care pathways for patients experiencing chest pain, understanding determinants of hand washing among health care professionals, and quantifying the costs and benefits of increased practice nurse involvement in the management of chronic disease.
These examples illustrate the breadth of health services research, and the potential for such research to improve the quality of the health care system, and reduce inequities in the system, so that patients can be confident that they are receiving the best quality care, no matter where they live.
Jon Karnon is a Professor in the School of Population Health, University of Adelaide and President of the Health Services Research Association of Australia and New Zealand.