It’s been a controversial couple of weeks for those involved in the review of the Medicare Benefits Schedule. Support for the MBS Review by the Federal Opposition and the AMA has waned amidst concerns it is simply a cost-cutting exercise, and a report on ABC Four Corners about the dangers of waste in healthcare polarised health, medical and consumer groups.
The Review continues, with input sought following the release of two consultation papers on 27 September. Below, health economist, Professor Jon Karnon writes that, if the MBS Review is to be an an agent of improving quality and value in healthcare (not just cutting costs), we need to develop robust methods of evaluating the patient outcomes of any changes made.
****
Jon Karnon writes:
There is increasing evidence and discussion globally, around the issue of value in health care. As a fee-for-service system, with limited tools and resources to audit and act, Australia’s Medicare Benefits Scheme (MBS) is at high risk of funding low value health care.
The MBS review team is tasked with identifying low value health care that is being funded by the scheme, and making recommendations to the Minister to reduce such care.
Identifying low value health care
Technologies or services that provide little benefit for any patient represent low value health care, but much more commonly, the term refers to the use of technologies or services that provide meaningful benefit in some situations, but little benefit in others.
Examples cited by the Minister in her piece for the Drum included knee arthroscopy for patients with osteoporosis, and patients undergoing multiple colonoscopies within a short timeframe.
Areas in which low value health care is suspected can be split into two broad categories:
- There is robust evidence that the health care of interest provides little benefit for specific groups of patients who are currently receiving it.
- Evidence is lacking on the effects of the health care of interest for some or all of the patients who are currently receiving it.
Lack of evidence of effect, is not evidence of no effect, but equally we should not assume positive effects just because a form of health care has been used for some time, especially if its use has expanded over time.
The MBS review team will no doubt prioritise the identified items for which the provision of some form of low value health care is suspected.
The prioritisation should reflect the burden of the suspected low value health care (both financial and patient effects); the level of certainty that the health care is low value; and the expected ease with which relevant MBS items can be refined.
Unintended effects of refinements to MBS items include failure to reduce the use of the targeted item in the targeted patient group(s); substitution of the targeted item with other potentially low value items; and reduced use of the item in patient groups for whom evidence of positive effects is strong (for example, if extra paperwork and procedures are required to demonstrate eligibility).
For every proposed refinement, a corresponding research plan
Given the uncertainty around both the identification of MBS items that reflect low value health care, and the outcomes of any refinements made, it is important to evaluate any actions taken, using robust research methods that focus on patient outcomes.
Establishing processes for ongoing assessment, with the potential to update and amend refinements to MBS items, will provide assurance to stakeholders (health professionals and patients) that the review is concerned with improving value and not just cutting costs. For every proposed refinement to an MBS item, a corresponding research plan should be developed.
A similar process is starting to be applied to new pharmaceuticals for which significant uncertainty remains around their effects, which are listed on the Pharmaceuticals Benefits Schedule (PBS) through the Managed Entry scheme.
The most robust way of evaluating refinements to the MBS would be via randomised controlled studies, e.g. randomising States and Territories to standard and refined versions of MBS items. If randomised study designs are not politically viable, observational study designs could also produce robust evidence, but would require careful consideration of the potential for bias and confounding.
The government should be planning such evaluation now. Two areas in which planning should be focussed are:
-
Improving and developing structures to support the collection of relevant data.
Data should describe include meaningful patient outcomes and the use of non-MBS funded health care, in particular services provided by public hospitals.
It is now possible to link MBS and PBS data with public hospital and mortality data, which can identify changes in patient pathways of care, and in some cases, identify relevant patient outcomes. For example, admissions for major cardiovascular events could be used as a measure of effect following refinements to a cardiovascular related MBS item.
In other areas, hospital data is unlikely to be sensitive to relevant patient outcomes, for example for high volume low impact conditions. To evaluate such areas, primary data could be collected from consenting patients, but such study designs are costly and can be prone to important biases, in particular selection bias.
An alternative approach is to leverage off the data routinely collected in general practice, using advances in information technology that facilitate the extraction of de-identified data from electronic medical records. Some Primary Health Networks (PHNs) are already undertaking this activity to inform their operations.
It may also be possible to incentivise the collection of data that is most relevant to the assessment of patient outcomes, for example by establishing Practice Improvement Payments to reward the recording of relevant Patient Reported Outcome Measures (PROMs).
Indeed, routine collection of PROMs and access to de-identified general practice data (as implemented in England) would inform a wide range of valuable health services research to improve health services generally.
-
Data collection options and appropriate research methods should be informed by collaborative engagement of all stakeholders, including government, health professionals, researchers, and consumers.
A series of research workshops could be held to correspond with key time points in the work of the MBS review team, with an initial workshop addressing the feasibility and process of collecting the data needed to evaluate the refinements to MBS items. Subsequent workshops might focus on appropriate study designs to evaluate specific items identified for refinement.
The objective of the MBS review is to improve quality of care, not simply to cut costs, and careful evaluation will be needed along the way to make sure it is achieving these aims. To support quality evaluation, we need to act now to establish the necessary frameworks and methods that can inform the continuous improvement of this important pillar of public health care in Australia.
Recent Croakey articles on this topic:
Taming the tsunami of too much medicine
Spending in the spotlight amid wrangling over Medicare review
Reports of waste as patient health spend grows
Review chair: it’s all about injecting some evidence into Medicare
Jon Karnon is a Professor in Health Economics at the University of Adelaide, and president of the Health Services Research Association of Australia & New Zealand.
Follow him on Twitter @jonkarnon
