Caring for the Homebound
Almost 2 million older adults are completely or mostly homebound, and another 5.5 million are dependent on others or have difficulty leaving their homes. Persons who are limited to the home tend to have multiple chronic conditions, functional challenges, and a high prevalence of cognitive impairment. These individuals constitute a high-risk, high-cost population who can be better cared for in the home. As home-based medical care physicians, we came to appreciate the issues faced by these patients in their everyday lives.
Our patients were often bed- or chair-bound, contending with multiple complex medical problems and a high symptom burden. Many had dementia, Parkinson’s disease, or other neurodegenerative disorders, and were tended to by heroic and worn-out caregivers. Their multitude of complex medical problems made prognostication about their life course and life expectancy difficult; we anticipated life-threatening illness lurking around the bend. For some of our patients, we observed limited social support networks and, when these networks failed, the patients often ended up in the emergency department. In every instance, we saw patients whose clinical, functional, and social profiles were different from the patients in our ambulatory clinics and needed a type of care that focused more on function and quality of life than on screening and prevention.
Unfortunately, home-based medical care providers were not available for many of these patients, and incentives for providers to enter the field were limited. While the Veterans Health Administration has led the way in the development of comprehensive home-based medical practices for homebound adults within that system, homebound adults in the fee-for-service world are less likely to find a home-based medical care provider.
Increasing recognition of the potential value of home-based primary care at the national level began in 2012, when the Centers for Medicare and Medicaid (CMS) Independence at Home Demonstration was signed into law as part of the Affordable Care Act. The purpose of the Independence at Home Demonstration was to determine whether a shared-savings payment mechanism predicated on the provision of comprehensive medical care in the home could (1) confirm the positive clinical and economic outcomes seen in previous studies and systematic reviews of home-based medical care and (2) provide a sustainable economic model for providers.
After the first year of the Independence at Home Demonstration, CMS reported that the practices saved over $25 million, an average of $3,070 per participating Medicare beneficiary. Savings continued to be accrued in the second year, with participants having fewer hospital readmissions and emergency department visits, better follow-up care after hospitalization, better medication reconciliation, and better documentation of care preferences.
The Homebound Have Been Left Out of the National Conversation on Care Quality
While we were thrilled with the rollout of the Independence at Home Demonstration, we still had some concerns. First, there was no existing framework for high-quality medical care in the home setting. Furthermore, we knew from our clinical experience that applying existing measures that were not appropriate for homebound persons and the home setting had the potential to cause harm. Finally, we were disappointed with the existing quality measures that were used for performance measurement in other CMS programs (e.g. the Heathcare Effectiveness Data and Information set [HEDIS], accountable care organization [ACO] measures, etc.) as well as in the Independence at Home Demonstration.
Our Path to Building Quality into Home-Based Medical Care
To address our concerns, we developed a strategy focused on the quality of home-based medical care in order to make known the existence of these heretofore mostly invisible patients and the providers who serve them. Our goals were also to ensure the provision of high-quality care and to raise the field of home-care medicine as a whole. We had no idea how fraught with challenges and obstacles this process would be.
In 2015, we performed a national survey of home-based medical practices (excluding those within the Veterans Health Administration) to get a better grasp of the home-based medical care practice environment. Our survey revealed a wide array of practices, ranging from solo and small group practices caring for a few patients to corporate practices extending across multiple states and caring for tens of thousands of patients (median practice size, 100 patients; range, 1 to 30,972 patients).
Most practices operated under a fee-for-service revenue model, whereas others engaged more in value-based models. One practice provided care for homebound patients in a Medicare Accountable Care Organization that was the fourth most successful ACO in the U.S. in 2017 in terms of savings generated. Over the last few years, as the savings associated with home-based medical practices became more established, venture capital funds started to enter the market and assisted with the launching of for-profit practices such as Landmark, Principium, and Aspire. With the growth of home-based medical care, the timing seemed right to get a better handle on what constituted quality medical care in the home setting.
We started by bringing together a stakeholder network consisting of representatives from exemplar home-based medical care practices, relevant medical professional societies, and patient advocacy groups who understood the field as well as the population and its needs. We performed qualitative research, asking our stakeholders and patients and caregivers about what constituted good-quality home-based medical care and developed a quality framework consisting of primary domains (e.g., “quality of life”) and subdomains (e.g., “reducing treatment burden”).
We then used the National Quality Forum Measures, Reports & Tools website to look for matches between our quality framework and existing quality indicators. We found many gap areas for which we could find no existing quality measures. Gap areas included domains such as “provider competency” and subdomains such as “assessment and reduction of treatment burden,” “optimizing comfort and safety in the home environment,” and “interdisciplinary care team assessment of patients.” We then developed an initial set of 20 quality measures that included functional assessment, cognitive assessment, management of delirium, interprofessional care, and others.
Mainstreaming Our Quality-of-Care Framework and Measures for the Homebound
In 2015, the Department of Health and Human Services (HHS) made clear its intention to move Medicare toward a value-based care framework. Furthermore, the Medicare Access and CHIP Reauthorization Act of 2015 (MACRA) instituted a new Merit-Based Incentive Payment System (MIPS), which evaluated health care providers on the basis of quality, resource use, and clinical practice improvement activities.
As HHS imperatives accelerated the shift to value-based care, the need to make our quality measures more available to home-based medical practices became more urgent. The best tactic to increase the availability of these measures, however, was not clear. The traditional approach to develop and disseminate quality measures has been through the National Quality Forum endorsement process. This process requires the use of measures by numerous providers along with rigorous and expensive testing.
An alternative approach that seemed particularly attractive with the passage of MACRA was to make home-based medical care measures available through a qualified clinical data registry (QCDR). QCDRs are CMS-approved entities that collect clinical data to foster improvement in the quality of care provided to patients.
Novel measures that are not National Quality Forum–endorsed may be included in a QCDR with CMS approval. Thus, providers are not constrained by the quality measures currently available in the CMS Physician Quality Reporting System (PQRS) library, most of which are not appropriate or relevant to home-based medical care. Practices participating in QCDRs have the opportunity both to engage in quality improvement and to avoid negative payment adjustments that are part of MACRA. The development of a QCDR was particularly attractive because it offered an infrastructure for reporting, benchmarking, quality improvement, research, and the use and testing of multiple measures.
The Many Challenges
In collaboration with the West Health Institute and The John A. Hartford Foundation, we developed the National Home-Based Primary Care and Palliative Care Registry. While the QCDR delivered on its promise in terms of improved reporting and benchmarking, the development process was not without difficulty.
- One challenge associated with QCDRs (or any clinical practice-facing registry) is the need to navigate multiple electronic medical record systems, all with varying capabilities of providing easy access to the data needed to determine whether a quality measure has been met.
- Another challenge was that, for CMS-approved QCDRs, each measure still undergoes review. As a result of variations in the understanding of home-based medical care on the part of CMS, we found that a large part of the measure-nomination process required educating CMS on what constitutes home-based medical care and on the special needs of the homebound population.
- Many home-based medical practices are small and do not have robust administrative or information technology resources, which are critical for capturing electronic medical record–based data, mapping data to a registry, and ensuring data security.
- Home-based medical care measures are entering a measure landscape already crowded by star ratings, ACO Quality Measures, and other payer or health system–mandated metrics.
- CMS has made a strong push to harmonize measures across settings. While we see an underlying rationale for such an approach, unrelenting harmonization fails to recognize that certain patients and care settings do have unique features that need to be addressed in the context of quality measurement.
- MIPS performance payment formulas grade “on a curve” rather than on a quality threshold. In a situation in which measure denominators are restricted to particular populations with a small community of health care providers, such as the homebound population, this system sets up a dynamic in which there will be winners and losers for performance payment purposes, even if all providers are performing at a high level. This approach creates strong disincentives for providers to care for this complex and costly population. Challenges associated with MIPS were recognized by the Medicare Payment Advisory Commission when they recently advised scrapping the MIPS program.
The Current State
Our work continues and is progressing well. In 2018, we used our QCDR for a pilot quality-of-care learning collaborative among practices at U.S. Medical Management, one of the largest home-based medical care practices in the United States. In January 2019, we launched a national learning collaborative for a select cadre of home-based medical care practices. In addition, we are building a national data repository for the field.
The path ahead remains uncertain. Our hope is that this work will serve to highlight the existence and unique needs of homebound persons and will support clinicians and our health care system in providing value-based care in all settings.