NYAPRS Note: A value-based payment system must, above all, measure and value the impact of service interventions on healthcare beneficiaries over simple cost and savings analyses. One powerful tool to capture this is via Patient Reported Outcomes: see https://www.qualityforum.org/Publications/2012/12/Patient-Reported_Outcomes_in_Performance_Measurement.aspx. PROS have been a prominent topic under discussion by New York’s Advocacy and Engagement subcommittee of our Value Based Payment work group.
Value-Based Payments Require Valuing What Matters to Patients
Joanne Lynn, MD, MA, MS1; Aaron McKethan, PhD2; Ashish K. Jha, MD, MPH3
JAMA. 2015;314(14):1445-1446. doi:10.1001/jama.2015.8909
Sylvia Burwell, Secretary of Health and Human Services, recently announced the department’s intention to tie most Medicare fee-for-service payments to value by 2018.1 Most commercial insurers already incentivize quality to some degree and encourage beneficiaries to consider quality and cost.2,3 Having payers aim for value should improve health system performance, certainly when compared with traditional incentives for the volume of services, which have failed to deliver the kind of care that is possible.4
Paying for value, though, requires measuring what actually matters to patients. Yet almost all current quality metrics reflect professional standards: eg, medications after myocardial infarctions, cancer screening according to guidelines, or glycated hemoglobin A1c levels being under control for patients with diabetes.5These metrics are relatively straightforward to calculate with available data, and patients’ interests usually align with professional standards—people want medical services to help them live longer, prevent or cure illnesses, limit the likelihood of and morbidity from disease and injury, and avoid or effectively manage symptoms. Although there are instances when professional standards seem to diverge from what some patients want (eg, when patients request antibiotics for upper respiratory tract infections), in most situations, there is congruence between professional standards and patient needs.
However, serious, life-altering, and ultimately life-ending chronic conditions, often in old age, pose a particular challenge for the health care system because traditional professional standards may not effectively address what an individual most wants.6 Individuals confronting grave illnesses, severe pain or impairment, and mortality must manage their hopes and fears and consider critical factors such as being comfortable, controlling finances, having food and shelter, being connected with others, honoring their family and social role, and being right with their spiritual commitments. Patients and physicians are often confused by this unfamiliar situation. Usually, a process of careful discussion and engagement can help patients and their families identify deeply held values and personal perspectives. These are difficult and intensely personal conversations resulting in identification of patients’ goals—goals that the current approach to measuring quality undervalues and therefore fails to integrate. Although professional standards are important, they can fail to capture what matters most to each individual.
A century ago, these aspects of care would have been of little importance. Historically, people died within hours or days, or maybe a few weeks following becoming ill, after appearing to be fairly healthy. Now most people accumulate chronic conditions in old age. The typical 70-year-old person will need daily help from another person for an average of 2.7 years before dying, and this just to accomplish activities of daily living, including eating, dressing, and toileting.7 Service delivery arrangements have neither adjusted to this new demographic reality, nor have measures of quality. People known to be dying soon are often included in the denominator for metrics like cancer screening, diabetes management, or hypertension control. Only a few of the hundreds of quality measures that Medicare now uses are particularly relevant to people living with frailty or advanced illnesses, measures such as screening for depression and prevention and treatment of pressure ulcers. Even fewer may be meaningful to younger disabled persons.
So when it comes to older or disabled people, what should be measured? Two categories are important: priorities that matter to most frail or disabled people and those that are important to the specific individual. Among frail older persons, population health priorities could include maximizing physical comfort, avoiding delirium, receiving treatment at home, having meaningful interactions with family and friends, maintaining as much independence as possible, having a trustworthy care system that is always available, avoiding impoverishment, and reducing the challenges facing family caregivers. Researchers could use established methods to develop and implement measures of these indicators, and Medicare should validate the priority domains and then fund the development of the measures and their implementation strategies.
Younger seriously disabled persons may require a different set of priorities: restoring ability to function in daily activities, returning to work, getting married, earning a living, and being in control of one’s own life. Both categories of priorities are measurable and important.
How can a care system be structured to deeply respect the myriad differences among patients when disabilities or advanced age makes those differences especially important? The answer is that the delivery system must proactively help affected people articulate their priorities and goals. In addition, the delivery system should be informed by honest and accurate predictions of patients’ likely futures with various treatment and support strategies. Each individual’s situation, preferences, and experiences amount to a unique lifetime, and each person’s priorities differ. One person may focus on the well-being of an intellectually impaired adult son, for example, and prioritize preserving assets to support him; another person with the same medical situation might have no family or dependents but an overwhelming fear of pain. A high-value health care system would identify each individual’s priorities and measure the fidelity with which those priorities are met.
These issues of identifying measures that matter are important to all payers but particularly salient for Medicare, which finances health care coverage for older and disabled persons. Medicare needs to approach value by understanding and serving the actual priorities and preferences of its beneficiaries. How can Medicare do this? Every older patient should have comprehensive assessments and care plans that include their goals. A key measure will probably be patients’ confidence that they can obtain timely care when conditions change. The tendency of the delivery system to drain personal assets and to rely too much on family caregivers should also be part of the measurement scheme. Feedback from downstream experiences to guide improvements in care planning will be important for system improvement. Patients, family members, and friends should be able to report whether the patient received what he or she most needed and wanted. Length of life is often an important component—sometimes, the most important. However, that decision should be made by the affected patients whenever possible.
Implementing such a measurement scheme might seem challenging, especially since clinicians may feel inadequately equipped to ascertain personal preferences and values. Yet with practice, it need not be any more difficult than many other measures. Electronic records could maintain documentation of the person’s priorities, goals, and the person-driven comprehensive care plan. Patients and caregivers could report the degree to which they find the care plan and team to be pursuing what matters most. Financial costs to patients and families could be tallied, and burdens on family and caregivers could be assessed. Quality measurement systems in many health care organizations already monitor complex issues, such as appropriate medication use, and often take patient preferences into account. Documenting patients’ priorities and how well the health care system meets them should not be appreciably more difficult. In fact, such an approach will engage clinicians because it aligns naturally with how they view their professional roles: to assess the needs and priorities of each patient and customize care accordingly. While posing new challenges, it is certainly possible to move toward a quality measurement scheme that reflects the effectiveness with which clinicians and health systems are able to achieve patients’ goals.
As Medicare moves to implement value-based payment initiatives tied only to current quality measures, the values of large populations of disabled and frail persons, whose care is the most costly and most concentrated in Medicare, are at high risk for being seriously neglected. These people will receive the medications and treatments that would be mostly correct if they were healthier, but their actual priorities will be neither noticed nor met. If the United States intends to pay on the basis of value, it is essential to ask patients what they value, and then deliver on those priorities.
REFERENCES
1 Cutler DM. Payment reform is about to become a reality. JAMA. 2015;313(16):1606-1607.
2 Rosenthal MB, Dudley RA. Pay-for-performance: will the latest payment trend improve care? JAMA. 2007;297(7):740-744.
3 Robinson JC. Reinvention of health insurance in the consumer era. JAMA. 2004;291(15):1880-1886.
4 McKethan A, Jha AK. Designing smarter pay-for-performance programs. JAMA. 2014;312(24):2617-2618.
5 Rosenthal MB, Frank RG, Li Z, Epstein AM. Early experience with pay-for-performance: from concept to practice. JAMA. 2005;294(14):1788-1793.
6 Naci H, Ioannidis JP. Evaluation of wellness determinants and interventions by citizen scientists. JAMA. 2015;314(2):121-122.
7 Crimmins EM, Hayward MD, Hagedorn A, Saito Y, Brouard N. Change in disability-free life expectancy for Americans 70-years-old and older. Demography. 2009;46(3):627-646.