For organizations looking to measure value from the patient’s perspective, as well as from an all-inclusive systemwide view, here’s a model to use as a form of a cost-benefit evaluation.
ABSTRACT: For organizations looking to measure value from the patient’s perspective, as well as from an all-inclusive systemwide view, here’s one model that can be used as a form of a cost-benefit evaluation. It outlines the thought process for decision-makers to consider in an industry that reimburses for outcomes and quality over volume. Organizations need a unifying and clear understanding of their priorities; the equation proposed here provides a detailed perspective on value creation to meet the need. When used as a yardstick against measured operational decisions, it aligns an organization’s goals with its patients’ goals.
Population health refers to the distribution of health outcomes within a population, the health determinants that influence that distribution and the policies and interventions that impact the determinants.1,2 Its key components include focusing on an entire population, health system-based approaches, application of behavioral principles, community-level public health factors, and a focus on policy, legislative activities and patient quality of life.3
Total population health strives to improve the health care experience, quality and costs for the populations for which an organization is responsible. These populations can be defined by geography, employer, payer, readmission risk, comorbid conditions or other similar boundaries.4 The shift from a volume- and revenue-generation measurement to the evaluation of value is a central theme of population health.
The notion of value is typically considered from the perspective of the customer in most industries. In health care, this is more complicated because there are several potential “customers” to be considered. The most important perspective from which to measure value is that of the patient through patient-centered outcomes. The inputs of patient-reported outcomes, quality, process, morbidity, utilization and cost all influence how value is achieved.5
We propose a framework for measuring value from the patient’s perspective as well as from an all-inclusive, health systemwide view. When thinking of population health in an accountable care organization or a patient-centered medical home, the appraisal of success or failure cannot hinge on a purely financial outcome for the health system. The basic at-risk financial model driving an ACO or PCMH intends to decrease total costs to the payer, such as the Centers for Medicare and Medicaid Services or a state health department. The provider delivery system accepts a reimbursement methodology that typically yields less than current total revenue or expected future revenue. The provider’s goal is to lower total expenditures and generate a return in the form of shared savings or margin in a capitated or global-budget arrangement.6 These financial considerations might ignore the patient-centric value equation we’re outlining in this proposal. The patient-centric value and the payer-provider financial benefits can be mutually supporting and potentially synergistic, although this is not obvious in some ACO or PCMH organizations.
Measuring value for a population health program presents methodological challenges. Episodes of care- or diagnosis-related procedure bundles do not fit the typical delivery of care, which is grounded in the primary-care setting and de-livered by an interdisciplinary team. Acute-care episodes and hospital admissions have beginning and end dates that permit evaluation of the services delivered to the patient. Preventive health services, chronic disease management and ongoing care coordination for complex medical conditions do not allow for simple calculations of value. Services provided for chronic conditions affect important outcomes, but often these positive impacts are more apparent in the longer term, creating a challenge for short-term value measurement.
Each population circumscribed for health management exhibits distinct characteristics. These can be influenced by geography, age, gender, occupation or innumerable other population demographics.7 The consideration of geography introduces a dimension for intervention design. Depending on the outcome goal, population health interventions can be taken at the national, state, local or community level. The geographic unit is a key component of the population health landscape.8 These two perspectives of medical utilization and geographic units highlight challenges for a rigid value equation. A value equation structured to allow for the insertion of unique measures can provide flexibility for adaptation to the unique characteristics of a population for analysis.
Outcome measures for population health are largely determined by policy, such as those imposed by CMS or accrediting bodies such as the National Center for Quality Assurance. Recommended outcome measures from the National Quality Forum and payer-directed incentives also have influence. Through such regulation, multiple sets of quality metrics are used by payers to determine payment to providers and health systems. Health systems might report 1,000 or more measures to various regulating bodies. Many of these are duplicative, as each payer defines the metric in a slightly different fashion or limits the sample measure to only their beneficiaries. Arguably, each health system should define its own set of measures to determine the value of the care provided to the populations for which they are responsible. Table 1 proposes a conceptual basis for such measures.9
The National Academy of Medicine (formerly the Institute of Medicine) has identified six key dimensions within health care quality: safety, effectiveness, timeliness, patient-centeredness, equity and efficiency. Overall, the academy defines health care quality as “the degree to which health services for individuals and populations increase the likelihood of desired health outcomes and are consistent with current professional knowledge.”10
The well-known Triple Aim serves as a convenient frame-work to organize the value relationships within population health.4 Its three goals:
- Improving the individual experience of care.
- Improving the health of populations.
- Reducing the per capita costs of care for populations.
Our proposal includes variables to account for each arm of the Triple Aim. The suggested value model is a form of a cost-benefit evaluation, where benefits are divided by costs. Cost effectiveness is a consideration within this value equation, but without the patient-specific outcomes and other pieces proposed below, it would be insufficient to assess actual value. The pure financial study of population health must be synchronized with the value model.
A systemwide view of population health described by the Institute of Healthcare Improvement outlines the relationship and pathways among determinants and outcomes. Our proposed value equation is designed to capture key elements in the model.11 Health equity is derived from medical care and multiple factors collectively referred to as “prevention” and “health promotion.” Socioeconomic issues and the physical environment are the upstream factors that influence five individual factors: genetic endowment, spirituality, resilience, physical factors and behavioral factors. Collectively, they influence disease and injury development across a lifespan. Health and function, well-being and mortality are the final states of health and quality-of-life outcomes.12
So how do we measure value? Other proposals have used distributional cost-effectiveness analysis and interquartile differences between populations.13 Ours is designed to act as a blueprint for measuring value. This is not an equation meant to be solved for a specific integer or used to compare the relative value of one service or product over another. Instead, this value equation is meant as a representation of the thought process necessary for an organization or provider to consider how to operate in a health care industry that reimburses outcomes and quality instead of volume. It might be possible to solve the equation for an integer to compare similar approaches to managing a specific condition or comparing interventions in order to set standards or create guidance for clinicians treating ACO or PCMH patients.
The value equation illustrated in Figure 1 is presented as a straightforward macro view of the value equation, and is expanded14 for consideration in Figure 2. The inclusion of key variables with the larger sets of appropriate care, outcomes and costs are illustrated in Figure 3.
For organizations, the benefit of adopting this value equation is the clear articulation of the purpose and goals of the health system. For organizations focused on maximizing the value they provide in health care delivery, this equation makes explicit the factors an organization will weigh in determining how to approach decisions involving clinical operations. Physicians, nurses, ancillary staff, administrators, patients and everyone else involved in health care can see clearly how adjusting certain variables affects the value of the care provided.
Separating “Appropriate Care” as its own variable makes explicit the necessity to provide only the right care. Inappropriate, or even marginally indicated, medical care delivered at a very low price will never have the same value as the most appropriate care at the best cost achievable.
Our expansion of the typical “Outcome” variable to include quality indicators and patient-focused outcomes allows for a deeper appreciation of all aspects of value creation. The lack of modifying cofactors in this equation is not to suggest each variable is of equal importance, but reflects the weighting of each variable that must be considered for each patient or organization, depending on the context in which the equation is used. Differential weighting of clinical-outcome measures and patient-focused outcomes must be considered from the patient’s or organization’s perspective when using this equation to consider different strategies.
The overarching equation previously outlined might best be viewed as a construct or thematic approach to evaluating value in population health endeavors. A reductionist approach to translate this construct to a functional equation is a potential future step in assessing value.
The decision underlying population health is how to allocate resources in the population and maximize health outcomes in a way that is consistent with the preferences of the population. We have achieved an optimal allocation if we cannot improve individual health at cost to another individual’s health, which is paraphrased as Pareto optimality.15 For example, if $1 million is invested in a certain number of resource units, and the health or the cost outcomes are maximized, then we have reached Pareto optimality. Restated, this is the expression of a value equation in which costs or resources are aligned to deliver maximum quality. A quantitative approach is useful when we want to compare allocation choices and develop decisions. This empirical model is limited because the decisions are not framed and the preferences of the individuals affected by decisions are not explicitly framed. The medical professionalism charter published by the American Board of Internal Medicine provides guidance applicable to population health definitions:
While meeting the needs of individual patients, physicians are required to provide health care that is based on the wise and cost-effective management of limited clinical resources. The provision of unnecessary services not only exposes patients to avoidable harm and expense but also diminishes the resources available for others.16
The equation solutions are intended to produce a product that allows a decision-maker (physician, patient, plan administrator) to develop an informed choice between alternatives.
The next proposed step is including outcomes in the formula. One challenge of population health is finding the right balance between prevention and treatment, given limited resources. These resources are consumed as goods. Individual health may be considered the benefit or product within our framework.17 These health goods are consumed in order to produce the product of health. Health maintenance is an objective at one extreme of the continuum, with lifesaving regimens at the other. Benefits from any resource consumption or goods occur at subsequent times in varying lengths. Pain medication might allow for virtually immediate relief to the individual compared to daily consumption of fruits and vegetables with a much longer timespan to influenced outcomes. This time-sequence difference makes measuring the benefits to a population of a mix of preventive and therapeutic interventions complex and challenging.
An additional consideration is the effectiveness of any intervention in combination with the cost of stimulating the desired behavior.13,17 Losing 7 percent body fat might drive a certain level of health improvement in an individual with a diagnosis of diabetes. The calculated value of this behavior change is probably high. A population health decision-maker will need to assess the cost and the effectiveness of an intervention program to stimulate obese individuals with Type 2 diabetes to engage in a body-fat reduction program. The cost effectiveness and product of the program might be less or even negative when compared to the cost effectiveness of individual behavior. These two value considerations are separate calculations and should be considered in combination when assessing value and program decisions at the population level.
A reductionist approach to measuring value can be challenging. The numerical product derived by solving the mathematics of the proposed value equation has limited utility as a discrete value. A value output plotted over time might indicate a favorable or negative trend. Variables within the value equation can be designed to capture specific characteristics of interest to the decision-maker or value-equation consumer, allowing for the comparison of multiple improvement interventions within a population.
The variables within the proposed equation can be modified by the user. Each variable can be designed to capture a construct of interest. Performance data can be compared to normative data, a benchmark, a local goal or any target as a frame of reference. The performance frame of reference selected will shape the output. As an example, normative data could be calculated from a national average of all adults, a regional average or a comparison average with patients diagnosed with the condition of interest. The value outcome should be tailored to capture key performance aspects in the population. The same performance data such as medication adherence will produce different factors depending on the comparative data.
Table 2 includes examples of the variable calculations for a population segment of individuals associated with a diabetes mellitus diagnosis. The variables from Table 2 are then inserted into the illustrated equation depicted in Figure 3. Value-equation calculations and the methodology to create each output are unique. The purpose of a value equation is to assist a decision-maker in the decision formulating process. The value output has the greatest utility as a point of reference. As performance increases, value will increase. The exact numerical value output has low utility as a discrete number. Trending value outputs and comparing values between different disease states or population segments using similar methods can yield insights for population health decision-makers.
Creating value in health care is a complicated endeavor. Organizations need a unifying and clear understanding of their priorities; the modified value equation proposed here is meant to provide a sufficiently detailed perspective on value creation to meet that need.
Adopting the value equation as a strategic priority, and a yardstick against which operational decisions are measured, aligns organization goals with patient goals in seeking treatment and payers’ goals as well.
Philip A. Smeltzer, PhD, is an assistant professor in the population health department at the Medical University of South Carolina.
Timothy Peterson, MD, MBA, FACEP, is an assistant professor in the emergency medicine department at the University of Michigan Medical School. He also is executive director of the Physician Organization of Michigan, an accountable care organization.
Martha Sylvia, PhD, MBA, RN, is an associate professor in the College of Nursing at the Medical University of South Carolina.
- Kindig DA (2007). Understanding population health terminology. Milbank Q, 85(1), 139-161. doi:10.1111/j.1468-0009.2007.00479.x
- Nash DB, Reifsnyder J, Fabius RJ and Pracilio VP (2011). Executive Summary Population Health: Creating a Culture of Wellness. Sudbury, MA: Jones & Bartlett Learning.
- Glasgow RE, Wagner EH, Kaplan RM, Vinicor F, Smith L and Norman J. (1999). If diabetes is a public health problem, why not treat it as one? A population-based approach to chronic illness. Annals of Behavioral Medicine, 21(2), 12.
- Berwick DM, Nolan TW and Whittington J. (2008). The triple aim: care, health, and cost. Health Aff (Millwood), 27(3), 759-769. doi:10.1377/hlthaff.27.3.759
- Porter ME (2010). What is value health care? New England Journal of Medicine, 363(26), 5.
- Smithback EL, Spector JM, Dieguez G and Mirkin DP (2012). Accountable care organizations: financial, clinical, and implementation considerations for academic medical centers. Milliman white paper commissioned by UHC.
- Brundage JF, Johnson KE, Lange JL and Rubertone MV (2006). Comparing the population health impacts of medical conditions using routinely collected health care utilization data: nature and sources of variability. Military Medicine, 171, 10.
- Baehr A, Holland T, Bials K, Margolis GS, Wiebe DJ and Carr BG (2016). Describing total population health: a review and critique of existing units. Population Health Management, 19(5), 9.
- Porter ME, Pabo EA and Lee TH (2013). Redesigning primary care: a strategic vision to improve value by organizing around patients' needs. Health Aff (Millwood), 32(3), 516-525. doi:10.1377/hlthaff.2012.0961
- Institute of Medicine Committee on Quality of Health Care in America (2001). Crossing the Quality Chasm: A New Health System for the 21st Century. Washington (DC): National Academies Press (U.S.).
- Evans RG and Stoddart GL (1990). Producing health, consuming health care. Social Science Medicine, 31(12), 17.
- Steifel M and Nolan K (2012). A Guide to Measuring the Triple Aim: Population Health, Experience of Care, and Per Capita Cost. IHI Innovation Series, 36. Retrieved from A Guide to Measuring the Triple Aim: Population Health, Experience of Care, and Per Capita Cost.
- Asaria M, Griffin S and Cookson R. (2016). Distributional cost-effectiveness analysis: a tutorial. Medical Decision Making, 36, 12.
- Duffy PL (2014). Real value: a strategy for interventional cardiologists to lead health care reform. Catheter Cardiovasc Interv, 84(2), 188-191. doi:10.1002/ccd.25159
- Barr N (2012). Economics of the Welfare State (5th ed.). Oxford, UK: Oxford University Press.
- Sox H (2002). Medical professionalism in the new millennium: a physician charter. Ann Intern Med, 136(3), 243-246.
- Woolf SH, Husten CG, Lewin LS, Marks JS, Fielding JE and Sanchez EJ (2009). The economic argument for disease prevention: distinguishing between value and savings. Partnership for Prevention policy paper. Retrieved from prevent.org/data/files/initiatives/economicargumentfordiseaseprevention.pdf, July 6, 2017.