Introduction by Croakey: In March 2016, the then Health Minister, Sussan Ley, issued a media statement announcing a trial of Health Care Homes in general practices and other primary healthcare services, as the centrepiece of a “revolutionary” reform package.
The trial finished in mid-2021, and an evaluation was recently published. While it leaves some important questions unanswered, the evaluation’s findings will be useful reading for the Federal Government and its new Strengthening Medicare Taskforce, reports Associate Professor Lesley Russell.
Lesley Russell writes:
The final evaluation report of the Health Care Homes (HCH) trial is now available on the Department of Health (DoH) website. It makes for fascinating reading and there is much to learn about how to implement future health reform initiatives.
In this context, Health Care Homes (HCH) are general practices or Aboriginal Community Controlled Health Services that aim to provide better coordinated and more flexible care for Australians with chronic and complex illnesses. The trial lasted for four years but involved only a limited number of practices and patients.
The final evaluation report was produced by Health Policy Analysis, a consultancy used regularly by the DoH and which did the two previous interim reports, with input from the University of New South Wales Centre for Big Data Research in Health and the Centre for Health Economics Research and Evaluation at the University of Technology Sydney. The time frame provided on the first page of the report indicates that it was subject to several revisions by the Department.
It’s no surprise to anyone who has followed this initiative that it failed to deliver any of the promised outcomes. The report attributed this to a failure to faithfully implement the model for HCH as articulated by the Primary Health Care Advisory Group (PHCAG), to low levels of participation by general practitioners (GPs) and patients, and to an implementation timeframe that was too short.
The report concludes:
The HCH trial highlighted there is appetite for changing the focus of primary care towards the principles articulated by the PHCAG for a medical home model, but that there are variable capacities amongst practices to undertake and manage significant change within their practice.”
As I outline below, my conclusions are that, at least in some circumstances (ie, in those practices that were committed to the HCH ideal and were flexible enough to implement the necessary changes), the trial worked well enough to justify its continuance (ie, it delivered value for money) and indeed that some practices would be happy to continue to work this way (ie, it was economically viable and enhanced the care general practitioners were able to provide to their patients, thus delivering satisfaction to both providers and patients).
That the trial delivered no measurable changes in the desired clinical outcomes (specifically in rates of avoidable hospitalisations) may be a consequence of its limited lifetime, or may mean that the focus of care and the required changes in care were not those likely to deliver improved outcomes. It could also mean that the method for selecting patients for enrolment did not assess those most likely to benefit.
Ironically, although the care provided for enrolled patients is detailed in the report, there was no evaluation of the extent to which these patients received better coordinated and more flexible care.
The reasons for the failure of the trial are obvious (frustratingly, they were obvious from the very beginning).
They include: a lack of timely involvement from the Department of Health (DoH) and the Health Minister (while the trial was announced by Sussan Ley, it was conducted during the tenure of Greg Hunt); the lack of leadership from organised medicine; the failure of some practices and some GPs to recognise that the way they were working was not compliant with the model; and the paucity of the efforts around patient education and involvement.
These issues were recognised in the interim evaluation reports, although these came late in the implementation of the trial.
The 2019 interim evaluation (released May 2020) stated that “The implementation of the program has identified many areas that need greater attention [ahead of further rollouts]” and the 2020 interim evaluation (released March 2021) stated “.. further time is needed for practices to continue with changes they have started to make, and for outcome data to be available before a decision about the future of the program is made.”
Background
The HCH trial was a key recommendation of the 2015 report from the PHCAG that was tasked with investigating options into the reform of primary care to support patients with chronic and complex illness. (And here we are in 2022 back looking again at the same issues – let’s hope the Strengthening Medicare Taskforce reads this evaluation very diligently!).
It was announced in March 2016 and commenced in July 2017. It was initially planned to run until June 2019 but, due to problems enrolling and retaining both practices and patients, was extended to June 2021.
About 200 practices from 10 Primary Health Networks (PHNs) were initially recruited, with additional practices recruited later (there are more than 8,000 general practices in Australia). At the end of the trial only 106 practices remained and 121 had withdrawn; a total of 11,332 patients were enrolled and at the end 7,742 remained (this is well short of the plan for 65,000 enrolled patients). Half the practices that withdrew had enrolled 10 or fewer patients.
The practices that remained to June 2021 had an average of 89 patients enrolled but 19 percent of practices had less than 20 patients enrolled. Of these active practices, 17 percent were Aboriginal Medical Services and 41 percent were in areas classified as “most disadvantaged”.
Key features of the HCH trial were:
- Voluntary patient enrolment with general practices
- Patients were stratified (by their GP) into three tiers based on their risk of hospitalisation
- Bundled payments for every enrolled patient based on their risk of hospitalisation (see Table 1)
- Facilitation for practices to transform provided by PHNs and a $10,000 grant per practice
- Use of electronic planning and shared care tools.
The bundled payment was to cover the costs of care related to a patient’s chronic health conditions. Practices could still bill Medicare for HCH enrollees for other services related to the patient’s acute conditions and certain other items.
Practices had to develop a care plan with each enrolled patient, and update this regularly. They also had to install and use shared care planning software to develop the care plan and share it with the patient’s other health care providers outside of the practice as well as with the patient (and where relevant, their carer/ family).
Initially patients were required to have a MyHealthRecord (MHR) but this requirement was lifted in late 2018.
It is not clear how patients were assessed as belonging to the three tiers.
The estimated average annual cost to Medicare for patients with a chronic condition is $862 (2016 prices). The DoH estimated that these payments under the trial were 10 percent more than the annual payments under fee-for-service.
In August 2018 the HCH trial was expanded to include the Community Pharmacy in Health Care Homes trial for which $30 million was provided in the 6th Community Pharmacy Agreement. Under this additional provision, enrolled patients could be referred to community pharmacists for medication management services.
Overall, $84.7 million was spent on the HCH trial; of this $54.6 million was for clinical purposes. This does not include the pharmacy component of the trial.
Funding for the HCH trial was provided in the 2016–17 Budget with an initial $21.3 million allocated over four years. Additional funds were to come from re-directing payments for Chronic Disease Management Medicare items for patients participating in the trial. The Government later indicated that in total, some $100 million would be available for the trial.
Evaluation findings
The DoH developed a “conceptual” model of how the HCH trial was intended to work, and the evaluation was done against four elements.
1. Elements promoting transformation (this includes bundled payments)
- Advice and guidance on key elements of the program were delivered late and were not sufficiently comprehensive. In particular, more guidance was needed about which services were covered by the bundled payment and which could be charged separately.
- There were issues with software that impacted care planning, clinical management and interoperability.
- PHN’s knowledge could have been better leveraged. There was a high turnover of the Practice Facilitators provided by the PHNs so many practices did not get the support they needed.
- Better training needed in the use of Risk Stratification Tools (such as the Hospital Admissions Risk Program) along with more research and evaluation of these tools.
- Bundled payments were both a motivator and a deterrence for participation. The risk stratification levels and payments need to be finessed and more tiers added.
- The effect of HCH on the viability of the practice was seen as positive by 20 percent of all practices in the trial and by 44 percent of practices with 50 or more enrolled patients.
- 58 percent of practices with 50 or more enrolled patients would continue in a similar program; only 25 percent of practices with fewer than 50 patients enrolled would want to continue.
2. Structural change and transformation (includes collaboration with other providers)
- Whole-of-practice involvement is important.
- The most common changes reported were: improved care planning; more regular recalls of patients; enrolled patients were given priority access and/or had the ability to call the practice; more team meetings.
- Those practices that had more enrolled patients had greater flexibility to deploy resources.
- GPs’ lack of willingness to delegate care responsibilities was a barrier in many cases.
- Only 26 percent of practices employed additional staff as a result of implementing HCH; only 18 percent had formal arrangements for working with / in local hospitals; 52 percent relied on deputising services for after-hours care.
3. Process of care and patient engagement
- Practices reported that care planning was improved and patients had better access to GPs, practice nurses and allied health professionals outside of the practice.
- MHR must be enhanced to facilitate its use in HCH model.
- Patents got more tests and were more likely to get an annual flu vaccination.
- Need to develop the capacity of practices to engage patients, families and carers in designing and implementing change.
- Need to raise the awareness of HCH initiative with other providers (for example, hospitals, pharmacists, allied health services).
4. Outcomes
The measured outcomes were: patients’ improved experience of primary care; patients’ improved health-related quality of life; better control of healthcare costs (for both government and patients); reductions in avoidable hospitalisations.
Enrolled patients generally expressed satisfaction with their care, and many appreciated being able to interact with other providers (such as practice nurses) in the practice.
The quality of care and the timeliness of care improved.
There were no changes in a raft of key measures for enrolled patients, including:
- Hospital and Emergency Department use
- Entry into aged care facilities
- The numbers of specialist visits and imaging services
- Serious cardiovascular events and mortality
- The numbers of patients reaching targeted blood pressure and glycaemic control
- Health-related quality of life.
There was little impact on patients’ out-of-pocket (OOP) costs for Medicare services and no impact on OOP costs for PBS medicines.
The community pharmacy component of the trial had minimal impact and indeed was not designed to encourage pharmacist/GP collaboration. Although 689 pharmacies expressed interest in taking part, at the end of the trial only 95 pharmacies had undertaken a medication management consultation with at least one patient. In total only 1,531 patients had a medication review. Many practices were not aware of this program.
The COVID-19 pandemic was obviously a factor that impacted the latter stages of the trial and this issue is addressed in the evaluation report.
Doing better
What follows is my assessment, based on the evaluation report, of what might have been done better.
Role of the Department of Health and the Health Minister
The DoH and associated government agencies got so much wrong from the very inception of the trial. Despite having more than 12 months from announcement to commencement, the informational and training materials provided were late and inadequate.
No thought was given to the need to finesse the model as it was implemented, despite early warning signs that it was not proceeding well. There was a multi-tiered advisory structure with multiple expert groups, and it’s not possible to know what they were telling the Department and the Minister – who clearly were not listening.
There were some serious errors in DoH predictions about the trial. For example, it was initially proposed that it would involve some 200 practices and 65,000 people and cost around $100 million, when in fact it covered far fewer people and practices and cost $84.7 million. The high cost for fewer patients may reflect that more enrolled patients were in Tier 3 than anticipated (see Table 2).
In a scan of Health Minister Greg Hunt’s media statements, I could find only one statement from him on HCHs – at the 2018 National Conference of the Australian Medical Association, he said that the Government was working on lessons from the HCH trial. Perhaps his lack of interest lay in the fact that this was an idea from his predecessor Sussan Ley?
Role of Primary Health Networks
Little attention is paid in the evaluation report to the role (or the potential role) of PHNs in the trial. This is somewhat surprising given their stated role to “streamline health services – particularly for those at risk of poor health outcomes – and to better coordinate care so people receive the right care, in the right place, at the right time”.
The PHN Practice Facilitators had little guidance and often felt overwhelmed. Staff turnover problems also mean that they were less effective than they might have been. Apparently little opportunity was provided for collaboration across the ten PHNs involved in the trial.
Some PHNs are involved in the implementation of patient-centred medical homes (PCMH) initiatives that are not part of the HCH trial (see, for example, this description of a PCMH initiative in NSW). That may have provided useful information for the HCH trial – or may have been a confounding factor.
Role of General Practitioners and their practices
Evaluation reports are not noted for delivering very sharp observations, but it is clear that the GPs and the practices involved in this trial come in two distinct categories: those who were leaders and enthusiastic adopters and were able to involve their colleagues in the practice and share patient care through team work, and those who are unable or unwilling to change.
While the HCH trial had some “clinical champions”, it is not clear from the evaluation report how these worked with their colleagues. Clinical champions (un-named) are listed as attending a Health Care Homes Forum in 2019.
Many GPs and practices struggled with managing both enrolled patients and their usual patients and the dual funding systems involved, and with determining how the bundled payments should be allocated to doctors and the practice. Apparently no guidance was provided from the DoH or professional medical bodies on how this might be done.
One aspect of GP life that the HCH trial did not address – in fact it may have aggravated – is how time-poor many GPs are. Perhaps that explains why only five percent of participating GPs completed all the online training modules (although it might also say something about the perceived value of these).
The report refers to the fact that some practices believed they were already operating as a HCH and made no changes to their model of care – there are currently no standards against which to assess this statement. I found it shocking that some practices were reported as suggesting that the pandemic negated the need for HCHs because it increased that use of telehealth and eScripts. That is a ridiculously narrow interpretation of the HCH model.
One of the aims of the bundled payment was for new roles to be introduced into general practices. Of the 73 practices reporting (over 30 did not), 71 employed practice nurses, 44 employed allied health staff and 10 employed “other medical staff” (not specified). That does not seem to be an adequate response to the stated aim of expanding the primary care team.
The report also notes the lack of willingness of some GPs to delegate care responsibilities and GPs’ concerns about pharmacists doing medication management reviews working outside of their scope of practice. Such mindsets do not augur well for primary care team work.
Role of Medicare economics
As previously stated, bundled payments are both a motivator and a deterrence for participation. The way they were calculated, they represented a higher overall level of payment than would have been received under Medicare FFS.
But practices had different views about the adequacy of the payments and whether they were sufficient to cover otherwise unfunded work. The evaluation report did not explore how practices spent both the $10,000 initial payment and the bundled payments.
Incentive payments should meet the costs involved. These costs could be reduced by tackling issues like IT on a system-wide basis rather than for each practice. Likewise, the bundled payments should adequately cover the costs of the various tiers.
One clear recommendation from the evaluation report is that the risk stratifications and the payment rates need more finessing. There also needs to be some incentive to drive care that will see a patient move down a tier; currently this results in a financial loss for the practice.
The data show that bundled payments have a positive impact on the financial viability of many practices (specifically those with a significant number of patients enrolled). Only 20 percent of practices said that the financial impact of the HCH trial on the practice was negative.
Role of Aboriginal Medical Services and Aboriginal Community Controlled Health Services
The issues that go to an easy and economically viable implementation of the HCH model are exemplified in the primary care services that are specifically designed for Aboriginal and Torres Strait Islander people, and so it is no surprise to find that those involved in the trial were generally enthusiastic.
They saw bundled payments as a more viable, more appropriate payment approach that provided certainty of income and enabled staff to be paid for additional work.
Eighteen Aboriginal Community Controlled Health Services (ACCHSs), all in the Northern Territory, entered the trial and 14 (with 1,025 patients) continued to the end.
The key enablers were the existing operational structure of the ACCHSs, and the existing relationships between communities, clinical staff and patients.
The challenges for these primary care providers included: the transient nature of community populations, sub-optimal communications with other healthcare providers, the availability of staff to follow through on care plans, and that patients were largely unaware of Health Care Homes and the trial.
Role of community pharmacy
Community pharmacies make an important contribution to patient care, which includes the management of medications and the provision of triage services for a range of community health concerns. This is acknowledged by organised medicine (see this statement from the RACGP). Nevertheless, relationships between doctors and pharmacists remain fraught, as we saw on ABC TV’s 7.30 report last night.
Tacking the pharmacy trial onto the HCH trial in late 2018 was also going to be a wasted move. Presumably this was pushed by the Pharmacy Guild (PGA) as part of negotiations around the 6th Community Pharmacy Agreement (CPA) with Health Minister Hunt. The 2017-2018 Budget included an additional $825 million over three years for primary care services and pilot programs under the 6th CPA.
However, the uptake of these programs has been described as “alarmingly low”, and evaluations of a number of long-running pharmacy programs have failed to demonstrate their value. In most cases, there has been insufficient data to enable any assessment of the impact of these programs on health outcomes, although some of them have been operating through multiple CPAs.
The Community Pharmacy HCH programs now joins that long list of expensive programs unable to demonstrate any value.
Role of patients, carers and families
The majority of patients enrolled in the trial were aged between 45 and 84. Apparently there were particular difficulties in enrolling children, although there were some young participants.
The most common chronic conditions for enrolled patients were diabetes, renal failure, liver disease, cardiac conditions, and chronic respiratory conditions along with the need to manage dementia, falls, chronic pain and incontinence. Not surprisingly, they had high levels of lifestyle risk factors such as obesity, physical inactivity and polypharmacy.
Surprisingly, there is no mention of cancer or of mental health issues here.
The patients were much more at risk of hospitalisation than estimated by DoH (see Table 2).
Many practices struggled to create a coherent and convincing message about the value of enrolling in a HCH and there were difficulties with language, culture and homelessness in enrolling some patients.
What was missing in the model and evaluation?
The focus of the HCH trial was very narrow. There was little apparent focus on communication across the medical and social welfare neighbourhood, except where it was noted that current IT and medical information systems made this difficult.
I could find no mention of patients’ needs for mental health services in the trial – surely a major omission for the targeted patient group. There was also no mention of improved liaison and care coordination between GPs and specialists.
Every patient involved in this trial was required to have a care plan, developed in consultation with them, their healthcare providers and their careers and family. We know nothing about the adequacy of these plans, how accurately they were reflected in the care received, and whether recommended care was denied by cost or because it could not be accessed.
While the enrolled patients had improved access to allied health services, there is no detail about which services were provided and whether these were available within the practice.
There was no independent analysis of the financial impact on practices – the evaluation was reliant on the practices’ own assessments – and no was work done on how the $10,000 incentive and the bundled payments were spent to expand patient care.
The key issue now is what will be done going forward as a consequence of the $100+ million investment in improvements in primary care? And do we know what has happened to those practices and patients who participated – we must assume that they have reverted to fee-for-service normality, with none of the additional care management efforts?
We can only hope that the Albanese Government and the new Health Minister Mark Butler show some bravery and leadership in implementing new models of primary care delivery and funding – but necessarily after they have read, digested and discussed the HCH trial evaluation reports.
• Adjunct Associate Professor Lesley Russell is a health policy analyst; she is a member of Croakey Health Media and a contributing editor and columnist
See Croakey’s extensive archive of articles on primary healthcare
