Osteoarthritis (OA) is a common and costly disease affecting 27 million adults in the United States.1 In 2013, OA accounted for $73 billion, or 2.5% of US healthcare spending.2,3 Total knee arthroplasty (TKA), which is performed predominantly to treat OA, accounted for $28.5 billion, or 1.1% of all healthcare spending and (along with total hip arthroplasty) was the single largest expenditure for the Centers for Medicare and Medicaid Services (CMS).4 As a result of an increasingly large cohort of increasingly overweight/obese senior citizens, utilization of TKA is expected to grow over the next 15 years,5 with Medicare facing estimated costs of $50 billion.6
Viewed within the context of the US healthcare system, which is generally recognized as spending more money than other developed nations,7 treatment for OA generally, and TKA specifically, is a bright target for cost-reduction initiatives. Given the many documented opportunities to reduce healthcare costs in the United States generally, opportunities certainly exist to reduce the cost of care delivered for OA by working to eliminate unsafe, unnecessary, and duplicative care.
The narrative surrounding cost reduction for OA, however, is incomplete without consideration of the impact on health that is achieved by the health care we purchase to treat OA. What improvements are achieved for the $73 billion spent to treat OA and for the $28.5 billion spent on TKA? Did patients achieve reductions in pain and improvements in function? If so, how much? Were the improvements achieved big enough to justify the money spent to achieve them?
Prompted by the following aims to improve the patient experience of care (including quality and satisfaction), improve the health of populations, and reduce the per capita cost of health care,8 the US healthcare system is embarking upon a new era in care delivery that seeks to optimize healthcare value. “Value,” which is the consideration of quality relative to cost, can be increased by improving quality, reducing cost, or doing both. For OA, patient-reported outcomes, especially of pain and function, figure prominently in understanding the quality and hence value.
Healthcare Quality Defined
Healthcare quality has been defined by the Institute of Medicine 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.”9 Donabedian10 describes quality of care as the application of medical science and technology in such a way as to maximize health benefits without increasing health risks.10 It is important to note that in both definitions, quality is defined not by absolute health outcomes or changes in health outcomes, but rather in terms of the probability of achieving maximal health benefit as a result of the medical care provided.
The quality of health care can be assessed by measuring the structures, processes, and outcomes of health care. Structure refers to the tools, resources, organization, and financing used in the provision of health care. Structural measures assess whether infrastructure elements are in place that are likely to increase or decrease the probability of good health outcomes. Structures relevant to OA include qualified staffing to manage OA, hospital programs to track and report surgical complications, and participation in quality-improvement activities such as registries. Quality structures related to patient-reported outcome measures (PROMs) might include having the infrastructure to collect PROMs or participation in a registry or another quality activity into which PROMs are reported and feedback on performance is provided.
Process of care describes what healthcare providers do for patients and includes taking a health history, prescribing medications, and performing tests or procedures. Care processes relevant to OA include treating pain and functional limitations with appropriate therapies, including TKA, and the use of therapies to prevent complications, such as prophylaxis to prevent venous thromboembolism. Quality processes related to PROMs include collection of data on pain, functional status, and/or quality of life using validated tools.
Health outcomes include both health status and discrete events such as recovery from an illness, death, and complications of medical care. Patient-reported outcomes are health outcomes that patients can feel for themselves and are best reported by the patient because (1) only the patients know whether they experience it, or (2) only the individual patients can provide details about the level of difficulty that they experience or assign context/importance to the activity. Outcomes relevant to OA include patient-reported pain and functional status and complications of care, such as gastrointestinal bleeding, infections, and venous thromboembolic events.
As discussed elsewhere in this issue,11 a number of validated instruments are available to assess patient-reported outcomes for OA, generally, and for knee OA, specifically. The two main constructs assessed by these instruments are pain and functional status. Several instruments also assess the ability to work and quality of life, both of which can be impacted by pain and functional status, and by the health effects (positive and negative) of therapies for OA.
Measuring Healthcare Quality
We measure quality to drive improvements in health. Measurement and reporting of quality drives population's health improvement in the following two ways: it identifies for providers' areas of deficiency or “care gaps” that can be addressed to produce better health outcomes, and it identifies for patients and other purchasers of health care the quality of different providers, allowing them to choose ones who have the highest likelihood of maximizing their personal health outcomes or the health outcomes of populations.
Quality measures quantify quality concepts. They do this by precisely defining the following: the structure, process, or outcome of interest; the population to which they are relevant; and the specific clinical circumstances in which they matter. Quality measures facilitate quantitative measurement of quality within a defined population and allow for comparison over time and across measured units. For example, does quality differ across geographies, insurance companies, or healthcare providers?
Characteristics of Good Quality Measures
The National Quality Forum standards for measure endorsement12 lay out at a high level the characteristics of good healthcare quality measures (Table 1). These standards require measures to be important, scientifically acceptable, usable, and feasible. As a foundation, good measures should address something important to health that can be meaningfully impacted by health care. The primary health effects of OA that matter to patients are pain, functional limitation, and impacts on quality of life, each of which can be meaningfully impacted by health care. Hence, measures of these outcomes or of structures and processes that lead to these outcomes would be meaningful.
Good measures must also be scientifically acceptable such that they produce valid and reliable results. Valid measures actually measure what they are intended to measure. Reliable measures will produce the same result with repeated measurement. The validity and reliability of numerous PROMs for OA have been demonstrated.13–18
Scientific acceptability also requires appropriate risk adjustment. Measures for which performance would vary based on factors other than the health care delivered need to be risk-adjusted to understand the magnitude of quality and/or to compare it across measured populations that might be differentially affected by those factors. Given the known impacts of baseline functional status, contralateral joint disease, psychosocial, and socioeconomic status19–22 on patient-reported outcomes of pain and physical function in OA, use of any such PROMs to assess quality would require risk adjustment.
Usability refers to whether the results of a measure can be used by healthcare providers, patients, or purchasers to impact the quality of care delivered to populations or individuals. Good measures should be actionable by healthcare providers to improve the care they deliver, to patients to inform therapeutic choices including provider, and to purchasers to drive quality for their beneficiary population through quality-based incentives and network design.
Feasibility refers to the practicality of measurement. Currently in the United States, most quality measures being used by health plans and accreditation and other quality-reporting organizations are based on administrative data because these data are readily available and far less costly to obtain than measures based on clinical data that require medical record abstraction or measures based on data from patient's self-reporting. Quality measures based on patient-reported outcomes have been largely regarded as not feasible because they are not routinely collected in clinical practice, and the infrastructure to collect PROM data for quality measurement largely does not currently exist. Increasing recognition of the importance of PROMs in understanding health outcomes and quality, especially for musculoskeletal disease, has led to the development of quality measures based on PROMs.
Characteristics of Good Quality Measurement
To promote health improvement and value, we not only need good quality measures but also to make sure those measures are used well. Specifically, the scoring of quality measures needs to reflect meaningful clinical standards, report statistically valid estimates, and be transparent in the methods used to calculate reported point estimates and thresholds. Clinical standards should inform determination of the level of performance that represents good quality or a meaningful care gap. However, this is not the way quality is scored for many programs. Rather, quality is often defined based on performance relative to the population mean, regardless of whether it is a clinically rational quality threshold.
Given that the only reason TKA is performed is to improve either pain or function, PROMs that measure these are logical candidates for quality assessment. However, to use such measures to drive quality and value, the translation of the results into meaningful levels of quality performance must be determined. Is there a certain absolute threshold or a change score for pain and function that should be achieved to define a high-quality result? Meaningful scores will no doubt require risk adjustment. In addition, given the range of clinical scenarios and patient goals that drive decisions to perform TKA, it is unlikely that percentiles of change scores or simple observed expected ratios will identify high versus low quality.
Current Uses of Patient-Reported Outcome Measures to Assess Quality for Total Knee Arthroplasty
Probably, the most robust effort to use PROMs to assess the quality surrounding TKA is the Patient-Reported Outcome Measures Initiative of England's National Health Service (NHS), which reports, collects, and publicly reports hospital-level risk-adjusted Oxford Knee Scores (OKSs).23 In this program, preoperative, postoperative, and change scores are reported relative to the national average. A similar program recently launched by MN Community Measurement likewise publicly reports hospital-level OKSs for patients who have undergone TKA as an observed rate relative to an expected rate that is adjusted for presurgical score and the type of health insurance.24 MN Community Measurement also deploys a process measure to assess the proportion of patients who had an OKS collected both preoperatively and postoperatively; the statewide average is 24%. Similarly, the application for the Blue Distinction Centers for Knee and Hip Replacement and the Spine Surgery Program requests information on the percentage of patients with knee or hip replacement who have undergone both preoperative and postoperative functional assessment at least 6 months after surgery but does not report it publicly.25
Working in conjunction with a multistakeholder group, CMS has defined a specific set of PROMs for collection as part of their comprehensive care for joint replacement (CJR) model (Table 2).26 Reporting of these PROMs to CMS is voluntary, although hospitals that report earn points for their overall quality scorecard. Hence, these PROMs are being used as process measures, that is, the hospital passes the measure if it completes pre- and postsurgical PROMs for a specified number or proportion of patients who undergo total joint replacement.
Use of Patient-Reported Outcome Measures in Value-Based Purchasing
Broadly speaking, “value-based purchasing” (VBP) refers to any health payment system that purchases based on value or seeks to use payment mechanisms to drive high-value care. Opportunities to purchase or create value exist across the spectrum of care aggregation from the individual unit to the procedure, the episode, and the population. Opportunities exist to create value at each of these levels with greater opportunity as the aggregation increases (Figure 1). The specific opportunity differs at each level, but at every level, the opportunity entails either improving quality or reducing cost, or both; the bigger the unit of aggregation, the greater the opportunity to create value. The value efforts across these levels of aggregation should roll up to create the best outcomes available at the lowest cost.
Because the primary treatment objectives for OA are improvement of pain, functional status, and quality of life, PROMs have an important role in determining whether the care delivered creates the health we seek to create and direct care processes to achieve those goals. In a value-driven marketplace, PROMs can serve as benchmarks against which reimbursements can be determined and providers of specific procedures such as TKA can be selected by patients or payers. Perhaps, more importantly, they can be used as guideposts to drive value improvement by either improving quality or reducing cost, or both.
The Patient-Reported Outcome Measures Initiative of England's National Health Service has used routine collection and reporting of PROMs to promote quality improvement through hospital-level transparency and financial rewards to high-performing hospitals in several areas including hip and knee arthroplasty.27,28 Availability of preoperative and postoperative PROM data to the hospitals has facilitated a variety of thoughtful, clinician-led analyses, which in turn have led to the changes in care processes that have improved outcomes.29
The use of PROMs in VBP programs in the United States is currently limited. At the time of this writing, the CJR program is the only large-scale VBP program for TKA in the United States using PROMs as part of a quality scorecard that is linked to reimbursement. Comprehensive care for joint replacement is a bundled payment program for total joint arthroplasty of the lower extremity, which includes most health care delivered during the 90 days after the procedure. If the actual episode cost comes in below the set target price, the hospital can keep the difference within specified rules if it meets the program quality standards.26 Hospitals that report specified preoperative and postoperative PROMs to CMS can earn two points on the program scorecard. For hospitals near a threshold, those points could boost the hospital to the next quality tier. Depending on which threshold is crossed and the year of the program, the hospital could become eligible for a reconciliation or quality incentive payment, or see a 0.5% to 1.0% reduction in the effective discount for reconciliation payment or repayment amount. It is anticipated that CMS will eventually publicly report facility-level performance on PROMs. The manner in which these will be reported, used to define quality, and/or incorporated into quality-based payments remains unknown.
Future of Patient-Reported Outcome Measures in Value-Based Purchasing
Realization of the full potential of PROMs as tools to create value in the United States will require the following: implementation of infrastructure to routinely collect and report them, analytics to develop appropriate risk adjustment, thoughtful consideration to develop clinically meaningful quality thresholds, and financial incentives that promote quality improvement within individual hospitals and collaboration across hospitals.
1. Lawrence RC, Felson DT, Helmick CG, et al. Estimates of the prevalence of arthritis and other rheumatic conditions in the United States: Part II. Arthritis Rheum 2008;58:26–35.
2. Soni A, Mitchell E: Expenditures for Commonly Treated Conditions Among Adults Age 18 and Older in the U.S. Civilian Noninstitutionalized Population, 2013. Statistical Brief #487. Rockville, MD, Agency for Healthcare Research and Quality. http://www.meps.ahrq.gov/mepsweb/data_files/publications/st487/stat487.pdf.
Accessed May 1, 2016.
4. Murphy L, Helmick CG: The impact of osteoarthritis in the United States: A population-health perspective. Am J Nurs 2012;112:S13.
5. Kurtz S, Ong K, Lau E, Mowat F, Halpern M: Projections of primary and revision hip and knee arthroplasty in the United States from 2005 to 2030. J Bone Joint Surg Am 2007;89:780–785.
6. Wilson NA, Schneller ES, Montgomery K, et al. Hip and knee implants: Current trends and policy considerations. Health Aff 2008;27:1587–1598.
7. Squires D, Anderson C: U.S. Health Care From a Global Perspective: Spending, Use of Services, Prices, and Health in 13 Countries. The Commonwealth Fund, 2015.
8. Berwick DM, Nolan TW, Whittington J: The triple aim: Care, health, and cost. Health Aff 2008;27:759–769.
9. Lohr K, Schroeder S: A strategy for quality assurance in medicare. N Engl J Med 1990;322:10.
10. Donabedian A: The Definition of Quality and Approaches to Its Assessment. Ann Arbor, MI, Health Administration Press, 1980.
11. Lyman S, Yin KL: Patient-reported outcome measurement for patients with total knee arthroplasty. J Amer Acad Orthop Surg 2017;25(suppl 1):S44–S47.
12. National Quality Forum: National Quality Forum Criteria for Measurement Endorsement. http://www.qualityforum.org/Measuring_Performance/ABCs/What_NQF_Endorsement_Means.aspx
. Accessed May 1, 2016.
13. Roos EM, Lohmander LS: Knee injury and Osteoarthritis Outcome Score (KOOS): From joint injury to osteoarthritis. Health Qual Life Outcomes 2003;1:64.
14. Rodriguez-Merchan EC: Knee instruments and rating scales designed to measure outcomes. J Orthop Traumatol 2012;13:1–6.
15. Collins NJ, Misra D, Felson DT, Crossley KM, Roos EM: Measures of knee function: International Knee Documentation Committee (IKDC) Subjective Knee Evaluation Form, Knee Injury and Osteoarthritis Outcome Score (KOOS), Knee Injury and Osteoarthritis Outcome Score Physical Function Short Form (KOOS-PS), Knee Outcome Survey Activities of Daily Living Scale (KOS-ADL), Lysholm Knee Scoring Scale, Oxford Knee Score (OKS), Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC), Activity Rating Scale (ARS), and Tegner Activity Score (TAS). Arthritis Care Res 2011;63(suppl 11):S208–S228.
16. Alviar MJ, Olver J, Brand C, Hale T, Khan F: Do patient-reported outcome measures used in assessing outcomes in rehabilitation after hip and knee arthroplasty capture issues relevant to patients? Results of a systematic review and ICF linking process. J Rehabil Med 2011;43:374–381.
17. Garratt AM, Brealey S, Gillespie WJ: Patient-assessed health instruments for the knee: A structured review. Rheumatology 2004;43:1414–1423.
18. Collins NJ, Roos EM: Patient-reported outcomes for total hip and knee arthroplasty: Commonly used instruments and attributes of a “good” measure. Clin Geriatr Med 2012;28:367–394.
19. Franklin PD, Li W, Ayers DC: The Chitranjan Ranawat Award: Functional outcome after total knee replacement varies with patient attributes. Clin Orthop Relat Res 2008;466:2597–2604.
20. Ayers DC, Li W, Oatis C, Rosal MC, Franklin PD: Patient-reported outcomes after total knee replacement vary on the basis of preoperative coexisting disease in the lumbar spine and other nonoperatively treated joints: The need for a musculoskeletal comorbidity index. J Bone Joint Surg Am 2013;95:1833–1837.
21. Sharma L, Sinacore J, Daugherty C, et al. Prognostic factors for functional outcome of total knee replacement: a prospective study. J Gerontol A Biol Sci Med Sci 1996;51:M152–M157.
22. Barrack RL, Ruh EL, Chen J, et al. Impact of socioeconomic factors on outcome of total knee arthroplasty. Clin Orthop Relat Res 2014;472:86–97.
23. PROMs at Health and Social Care Information Center. http://www.hscic.gov.uk/proms
. Accessed May 1, 2016.
24. MN Community Measurement Health Care Quality Report. Available at: http://mncm.org/wp-content/uploads/2015/11/2015-Health-Care-Quality-Report-Final3.pdf
. Accessed May 1, 2016.
26. Centers for Medicare & Medicaid Services: Medicare program; comprehensive care for joint replacement payment model for acute care hospitals furnishing lower extremity joint replacement services. Fed Regist 2015;80:73272–73554.
27. Appleby J, Devlin N: Measuring Success in the NHS: Using Patient Assessed Health Outcomes to Manage the Performance of Health Care Providers, London, United Kingdom, Dr Foster, 2004.
28. Monitor & NHS England. 2014/15 National Tariff Payment System—Annex 4A: Additional Information on Currencies With National Prices. London, United Kingdom, Monitor & NHS England, 2013.