Evidence-Based Policy

Dear Editor (in reply):

All decision making should be based on the best available evidence. This must encompass both decision making for an individual's clinical care, and the policies put in place for whole nations. The editorial referred to by Chan and Webster did focus on individual clinical decision making, but Chan and Webster are correct to observe that the conclusions apply equally to developing health policy.^1,2

Every innovation in health care including whole-of-system change should be rigorously evaluated before implementation. By definition, every innovation will have positive and negative impacts, which need to be reconciled in a defined net benefit. Understanding in detail how new programs are likely to perform is crucial for engaging funders and the community more widely. It also creates a baseline against which the performance of a new program can be evaluated.

It is often far easier to evaluate new therapies for patients than it is to evaluate health systems innovation, although such studies in palliative care now date back more than 25 years. ³ Evaluation of something as ubiquitous as the end-of-life care pathway requires the evaluation of an individual's clinical outcomes and the consequences of widespread uptake across the health system. Assuming that the performance of the pathway when used in a palliative care unit will be the same as its interpretation and use in a tertiary hospital is an assumption that needs to be evaluated carefully and prospectively. Such an evaluation would seek to demonstrate improved individual patient outcomes and improved care delivery across the health system. Although at first such research may seem costly and complex, given the enormity of changes in care generated by uptake of end-of-life care pathways, such an investment is all but mandatory. Such studies are feasible and necessary if people are to be guaranteed the very best standard of care right through until death.

If for some reason it is not possible to evaluate a new systems-wide program before initiating it, then, at a minimum, there must be careful prospective evaluation of the new program. Post hoc evaluation is not sufficient given that such an evaluation is likely to miss early signals of problems if a program is not delivering the outcomes that were hoped for.

The other question that sits implicitly behind Chan and Webster's important observations is that in hospice and palliative care there has been unprecedented (ultra-rapid) uptake of the end-of-life care pathway by so many services. ⁴ Given the tardiness with which many evidence-based clinical changes are taken into practice, the reason for the contrast needs exploration. What is it about the end-of-life care pathway that has generated a frenzy of uptake well outside its original scope without due evaluation and despite significant and sustained concerns at the way that it is being implemented in health care more broadly? ⁵ There is a need to understand the reasons behind this in order to aid the uptake of new knowledge where strong evidence supports change in clinical practice or health policy.