I am a big fan of the guys over at Power to Persuade and recently wrote a blog post on Social Impact Bonds (SIBs). Of late I have been a little concerned about the fanfare with which SIBs are being met, particularly given the lack of evidence there is to suggest that they are effective and given some of the potential risks that are involved with these mechanisms.
The Grattan Institute recently published a report on the provision of questionable health care and set out suggestions for how the Australian health system might avoid ineffective treatments. In this they argue that health professionals continue to deliver treatments that are unnecessary and in some cases even where there is evidence that they are ineffective. They conclude that ‘Australia urgently needs a system to identify these outlier hospitals and make sure they are not putting patients at risk’ and go on to outline a possible solution to this problem. While this is an interesting proposition, implementation of this might find that it compounds, rather than alleviates, some of the issues faced.
Internationally efforts have been made to identify health care interventions that deliver only marginal benefit. The Grattan report analyses the variation of practice in terms of a number of treatments that either don’t work or which should not be given routinely and the findings seem to be consistent with other studies of low-value health practices in an Australian context. It is important to note, that this work focuses on the presence of particular practices and not broader issues of variation in terms of outcomes of health care services where we know there are significant inequalities.
It is true that there is significant variation in clinical practice (and patient outcomes) across Australia. A recent OECD study analysed Australian clinical practice, finding variation in all interventions they studies with amounts ranging from 1.6 to 7.4-fold variation in primary care and up to 11-fold differences in the hospital sector. Variation alone is not indicative of poor practice, but in some places variation can mean ineffective care services and inappropriate use of scarce healthcare resources.
For many years there have been studies seeking to explore why it is that doctors don’t follow clinical practice guidelines and much of this discussion is set within broader concerns about degree to which evidence makes its way into policy and practice. One of the major problems with the Grattan study is that it doesn’t really engage with the issue of why doctors don’t use evidence or follow guidelines and as a result their solution for reducing ineffective care processes is likely to be limited in terms of impact.
The Grattan solution is for the Australian Commission on Safety and Quality in Health Care to publish a list of treatments that clinicians should not do, to identify hospitals that provide these more often than usual and analyse why these are being used. If on investigation use is found not to be warranted and there is no improvement over time, the State health department should provide consequences in terms of the hospital’s management and funding.
This is a fairly heavy-handed approach in the sense that it involves a high degree of external scrutiny of practice and the threat of sanctions where improvements are not made. It seems to assume that provided we have the appropriate ‘sticks’ then doctors can be made to adhere to the guidelines by hospitals via threat of sanctions from state government.
Yet, healthcare organisations, and the actions of doctors more generally are well known to often be rather impenetrable to traditional management approaches. Hospitals are prime examples of professional bureaucracies with highly skilled doctors who have a large degree of control over their everyday work practices. As such, ‘normal’ organisational power structures often have less influence on the practice of doctors, who are often more responsive to peer influence. International evidence shows that where doctors (and other professionals) are engaged in their organisations this leads to better quality services and patient outcomes. The challenge of managing health organisations are further compounded by the expansion of numbers of Visiting Medical Officers, where doctors seek to combine practice across a variety of different organisational settings and therefore do not hold full time or continuing posts across any.
Within the broader literature on adherence to practice guidelines it suggests that there are a variety of different factors which have an influence including: awareness and accessibility of guidelines, agreement with guidelines and the degree to which they are too rigid or challenge autonomy, doctors being unable to follow guidelines, being unmotivated to do so, inability to reconcile guidelines with patients, presence of contradictory guidelines or evidence, lack of time, lack of resources, lack of alternative treatments and a whole host of other types of factors. Focusing only on the financial, regulatory and policy context as the Grattan’s solution does, fails to address more than only a small proportion of the reported barriers to guideline adherence.
Of course it is difficult to develop a strategy that speaks to each different organisational and individual context, but starting with macro-level issues without first considering those around the workforce and practice may potentially have unintended longer-term impacts on clinical practice. By seeking to identify areas of inappropriate variation and penalising this we may find that this has a negative impact on professional engagement and also patient outcomes.
Variation exists across all areas of health care practice (and other areas of public service delivery) for both good and bad reasons. A conversation on this topic is long overdue but this doesn’t mean that we should leap straight to ‘hard’ levers in an attempt to bring about change rather than thinking through what some of the other possibilities may be. Without engaging professionals in this process we might find that this solution exacerbates issues of variation in practice and ultimately has the opposite impact on clinical performance that it is intended to.