Performance Measurement In Healthcare Essay

Introduction

Health insurance entities such as Medicare and other private health plans have shifted and are continuing to shift their payment structure from volume to value. This shift has been largely driven by a myriad of performance measures developed in the last three decades – these measures have been developed and/or endorsed by various payers, accreditation agencies and non-profit organizations such as Centers for Medicaid and Medicare Services (CMS), National Quality Forum (NQF), Joint Commission (JC) and others. Performance measurement in healthcare is largely focused on an organizational level such as hospitals, skilled nursing facilities, hospice etc. for a variety of purposes such as public reporting, accreditation, payments and penalties etc. A recent report found that out of 127 distinct uses of performance measures, more than half of them were at the hospital level, and physician practices were measured at a lower rate.1  The higher uptake at a hospital level is mainly driven by external factors such as the health care reform, public reporting, expectations from insurance companies, and consumers. Nevertheless, one can argue that it has also be driven by the hospitals’ commitment to patient safety and quality in order to attract more patient lives and positively impact their bottom line(s). Measuring performance at any level is challenging and in general, there has been a lot of pushback for various reasons such as costs, burden, feasibility, and lack of data.1-3  Despite the challenges, it is still more established at the hospital level than at the physician level since value-based purchasing initiatives for hospitals have been established for several years.

Hospitals continue to be under immense pressure to align with physicians in this quest to collect, report, and use performance measures to meet regulations for payment, but also to improve quality of care, reduce costs, and attract more patients. Initially, physicians were unclear as to how their reimbursement and practice(s) will be impacted under the new payment reform.  However, in the last five years, payment models are making individual physicians more accountable for quality and the cost of care that they provide.3  In 2015, CMS introduced the “Value-Based Payment Modifier” which was calculated based on the Physician Quality Reporting System (PQRS). More recently, under Medicare Access and CHIP Reauthorization Act of 2015 (MACRA) the CMS has embedded “Merit-based Incentive Programs (MIPS)” and “Alternative Payment Model (APM)” as the value-based payment programs for physicians.4 Therefore, it is paramount that physicians get more comfortable with performance measurement and its use as it directly impacts their reimbursement.

Unlike other facility level measures, measuring physician performance as it relates to quality and cost has been more challenging.5  Studies show that patients and families are most interested in getting information regarding quality of care from their specific provider.6,7  However, performance measurement and its use at the individual provider level has been subject to a lot of criticism since its development and deployment. With several thousand measures available to measure physician performance and numerous value-based plans, it has been unclear as to how these measures can be used effectively. The objectives of this paper are to discuss how physicians can use performance measures at a large academic medical center, identify potential barriers to uptake of performance measurement by physicians and discuss some possible solutions, and subsequently suggest a plan to improve physician measurement.

Using Physician Performance Measures in a Hospital Setting

The overarching goal of developing and employing healthcare performance measures has been to improve care and reduce costs. Therefore, performance measures should be used as enablers to meet organizational and individual provider goals while keeping quality of care and patient satisfaction at its core. Physicians and organizations can use performance measures in several ways that may or may not be interrelated -

Clinical Decision Making

Clinical decision support (CDS) are tools that are being widely used by organizations to enable better decision making at the point of care. The ubiquity of electronic medical record (EMR) and healthcare informational technology systems can allow these measures to be available for use to make better choices. In conjunction with evidence based clinical information and patient specific data, physician measures can be integrated in the clinical work flow to provide assistance during care delivery.8

Physician Profiling

The original purpose of physician profiling was to track resource utilization and curb costs. However, if developed properly, they can be effectively used to improve quality of care by incentivizing physicians to change behavior.9  Internally, these profiles have the ability to provide valuable data for root-cause analysis, quality improvement cycles, and physician performance when assessed over time.10  Externally, profiling data is helpful to consumers, payers, and accrediting organizations. Consumers are already using data collected by payers and other independent organizations to choose their providers. CMS introduced the Physician Compare website under the MACRA act that provides MIPS performance scores on their website.11  Independent websites such as Health Grades12  report patient self-reported satisfaction measures. While physicians might feel overwhelmed with the amount of profiling, it is important that they actively participate in this process so that it truly reflects the care they provide.

Pay-for Performance

Similar to value-based payment programs for hospitals, there are several new payment plans based on the similar principle that provide financial incentive to physicians based on a set of performance measures. Since the healthcare reform, organizations are shifting their strategy for physician compensation. This means they are not just focusing on relative value units (RVU) but also focus on other quality, outcome, and patient satisfaction measures. Such models that use physician performance measures have not only been adopted by managed care organizations but are also being used by large payers such as the CMS and other private insurers. Despite the success of such programs in other professions, in regard to physicians, studies have shown marginal improvement in quality unless the incentives are larger, or, perhaps, implemented in another way.13 

Academic Detailing

Academic detailing can be based on performance measures. While each physician has the best intention for their patient, several factors could lead to inconsistencies in care. The use of measures will allow them to monitor and benchmark their performance against internal goals. By utilizing this quality improvement technique, physicians can implement practice changes with the help of their “academic detailer” who can be a fellow physician who might have faced the same challenge.14

Barriers to Physician Performance Measures

Improving quality of care, efficiency, and affordability of care has always been central to using performance measures. However, most physicians and hospitals find it challenging to adopt physician level measures given skepticism and lack of trust on whether they can positively impact quality of care. The potential barriers to measuring physician performance are discussed below –

Validity of Measures

Studies have shown that approximately a third of the measures used for the MIPS program were valid, meaningful or both.2  The same assessment revealed that 35% were not valid and almost 28% had uncertain validity. However, it was seen that the National Committee for Quality Assurance (NCQA) and NQF endorsed measures were valid at-least 50% of the time relative to non-endorsed measures. This study also revealed that validity assessments of physicians’ measures conducted by various stakeholders such as the CMS, NDF, and the American College of Physicians did not concur.2,15  These inconsistencies have not only led to mistrust and confusion amongst physicians but also a change in care practices since these measures could potentially harm patients and/or increase unnecessary administrative costs.

Applicability of Measures

The new MIPS system depends on CMS approved measures which scores each provider based on a set of measures from various data sources. However, the set of measures depends on the provider’s selection, a flexible process which was introduced to increase participation. This flexibility comes with several limitations such as lack of clinical relevance or applicability, problems with inadequate distribution of a particular measure, biased scores. This is exaggerated in case of specialists as there are very few measures specifically available for that specialty.15

Data Sources used for Calculation

In general, the biggest barrier to using performance measures continues to be the access to good quality data.1  The data source and calculation of performance measures have often been questioned. For instance, several measures are calculated using administrative claims data because they do not require much effort to collect. However, they lack clinical depth and information. For instance, to identify if a diagnostic test was unnecessary it would require merging of clinical information from the medical record.16  Additionally, when using administrative claims, it is important to merge data for an individual provider for all the insurance providers to provide an accurate evaluation of physician performance.

Sample Size Limitations

One of the most significant limitations of appropriately using physician performance measures is small sample size(s). A single provider is likely to have very few patients with a specific disease within the same health plan.17  For most measures, a minimum threshold of 30 patients is required to make any valid assessments.18,19 A study that aimed to assimilate data from several insurance plans to benchmark physicians on 10 quality measures found that the episodes per physicians were not enough to make valid comparisons. Moreover, less than 20 percent of physicians had sufficient events to calculate composite measures.20

Gaps in Measurement, Relevance, and Training

Despite the presence of several measures, there are gaps in measurement in the area of care transitions, safety, palliative care. Although measures are risk adjusted, they are criticized for not considering all the risk factors applicable to the older adult population. Additionally, post-acute care settings that are known to significantly impact patient outcomes are not often considered. As per the Donabedian framework, measures can be differentiated into structure, process, and outcome measures. Most value-based care programs rely on process rather than outcome measures.16  Even though the reflect best practices, they are not always associated with good outcomes in the short-term.9  Moreover, they also tend to generalize these care practices for all populations such as the elderly where they may or may not be applicable. It may also provide an imperfect picture of the services provided by the provider. Lastly, providers do not have the adequate training to pull these measures for themselves and have to rely on what is available to them via their organization.1

Lack of Measure Prioritization and Alignment

Another challenge that prohibits the use to physician measures is the lack of prioritization and the inability to align measures to regulations at the national, state, local level and/or accreditation requirements.1  The “measure fatigue” is augmented when it is challenging to gather all resources just to make sure the construct requirements are met.1  Most providers and organizations will likely opt for measures that are endorsed by NQF and/or are used in payment or reporting programs. However, often there are certain measures that may be non-endorsed but more applicable to their needs.1  The number of measures and reporting requirements often overwhelms the provider leading to poor prioritization and alignment.

Opportunities/Solutions to Improve Physician Performance Measurement

Despite the shortcoming of existing measures, it is unlikely that performance measures will be eliminated. The limitations of measures are being discussed and evaluated at the national level. The quest to improve measure development, calculation, and use will continue as payers and other organizations aim to incentivize better care and/or modify payments for physicians. Some opportunities to improve uptake of physician performance measurement are discussed below -

Participate in Development and Review of Measures

Physicians are the subject of measurement and the most proximate to the patient during the care delivery process. Therefore, it is important for physicians to participate in measure development and/or selection at the organization level.21  If they are involved in the decision-making process, they are more likely to trust and use these measures in order to make changes to their individual practice.

Prioritize and Align Measures

Given the sheer number of measures, organizations should invest in measures strategically and make sure they are prioritized and aligned with existing external reporting requirements.1,16 Organizations/physicians should adopt outcome and process measures that have high return on investment.1  Creating a priority list of measures that align with hospital level measures and external reporting requirements will reduce the burden of collection and reporting.

Invest in Health Information Technology Systems

The reliance on limited data sources such as administrative claims data does not fully capture the performance of a physician. Organizations should invest in EMR capabilities and other Health Information technology tools that allow for better risk adjustment and improve construct validity of measures.1  As discussed previously, these may not only help with the measurement and calculation process but also enable use of CDS tools.

Evaluate Impact of Measures and Use Strategically

One of the criticisms of pay-for-performance programs for profiling has been that it lacks a long-term impact, and that organizations should build in evaluation plans within their performance measurement portfolio to justify its continued use.1,16

Provide and Build Support

It is critical that organizations provide training and support to physicians on how to extract reports, use measures and/or provide suggestions to improve performance.1,3 Additionally, an organization should invest in tools and talent that allow for continuous refinement and improvement in the performance measurement process.

Plan to Improve Physician Measurement

Measure Selection

Measures for physician performance should be selected based on the entire care continuum offered by the organization. In this case, as an academic medical center, measures should include inpatient, outpatient, and ambulatory care providers (primary care and specialists).9  In order to get a holistic picture of physician’s performance the organization should select a combination of outcome, process and satisfaction measures. It is important that measures are evidence based and have a sound methodology, endorsed by NQF or other major payers (when possible), aligned with external reporting and internal needs, inclusive of all types physicians, and selected on the basis of buy-in from all stakeholders i.e., physician, patients, and the leadership. Collecting and reporting of performance measures is an expensive process.22  Based on the “ideal principles”9 , here is how and why the organization should select the measures –

• Select a set of core outcome, process, satisfaction, resource/cost measures that are applicable to all providers.9  This will yield a larger sample size for benchmarking.
• The first round of selection could be based on what is collected routinely9  through the EMR or existing patient survey(s) but ultimately be scaled to include multiple data sources to reduce bias. These core measures should have the capability of risk adjustment.1,15,16
• When selecting process measures, only opt for measures that are clearly associated with outcomes to maximize its impact.9,16
• Next, based on existing priorities and external reporting requirements one must select measures from the NQF website. The quality and performance measurement team can organize their list of metrics23,24   in the following manner in order to effectively align and prioritize measure selection –

• After the measures have been shortlisted, physician leaders from each specialty should review the measure selection in order to get buy-in as they are more likely to trust the plan when they have participated in the process.
• Once the measures have been finalized, the organization should review the master measure list and identify if they could choose one or more of the major physician measurements sets that encompass the measures that have been finalized. Major performance measure sets are Physician Consortium for Performance Improvement (PCPI), Healthcare Effectiveness Data and Information Set (HEDIS), Consumer Assessment of Healthcare Providers and Systems (CAHPS), Promoting Healthy Development Survey (PHDS). Each of these sets have measures that are partly endorsed by NQF.6 

Type(s) of Reports

Organizations should provide individual-level physician dashboards as well as have the capability of organizing data by group. This will enable the use of performance measures not just at the physician level but also allow for quality improvement initiatives at the group level. The reports should be designed such that they do more than just measure (e.g., provide suggestions or tactics for improvement).25  The reports should ideally be in the form of insightful dashboards that don’t just display numbers but also provide suggestions and action steps. These can be anonymous but allow physicians to benchmark and monitor their progress versus peers. These reports can focus on a core set of measures picked by the organization but also have the ability to customize according to specialty and strategic focus areas. Reports should be easy to interpret, understand, and visually appealing so that physicians can quickly assess their performance.26  The delivery of the feedback reports/dashboards is as important as its content and design. Studies have shown that these reports are likely to be impactful when it has been verbally handed-off by a trusted supervisor, applicable and routinely conducted for all physicians, and is embedded within quality improvement procedures within the organization.25

Introducing the Initiative into the Organization

A successful performance measurement plan and quality improvement project hinges fundamentally on buy-in from end-users / subjects of interest. As seen with other programs, a successful program should be introduced in a formal setting, clearly delineating all the benefits and considerations both at the macro (i.e., organization-level) and at a micro (i.e., physician) level.27  Prior to rolling-out, measurement selection and reporting format should be piloted with physicians to drive adoption rate(s) and more broadly, physician level buy-in. The analytic methods underpinning each measure and the risk adjustment strategy should be transparent and available for review in order to minimize skepticism and increase trust. Finally, as given that this is a physician performance measurement plan, the initiative should be introduced with a summary of the literature and/or case studies delineating successful outcomes, both at an organization and at a specialty level.

Conclusion

A holistic evaluation of physician performance is challenging but given that American health care system is still adapting to the nuances underpinning the health care reform, it is critical to participate and use performance measurement to make better clinical decisions, leverage good performance for financial benefit, attract new patients, and improve care practice performance. Existing measures for physicians may not be perfect due to concerns regarding validity, data sources, sample size limitations, and overall gaps and intent of each measure. Despite that, there are several ways organizations and physicians can leverage measures to improve individual performance as well as align them with organizational strategic goals. Also, fundamentally, a successful plan to improve physician measurement should be based on the organization’s patient, payer, and provider mix. Organizations must provide physicians with options wherein they have an opportunity to monitor these measures on a regular basis as it directly impacts their compensation, rating, and preferred status. Newer models of health care delivery such as concierge medicine are being employed to increase access, reduce costs and improve patient satisfaction. While providers are trying to adjust to these newer roles, organizations must offer physicians with the resources they need to ultimately do their job and use these measures in a way they were originally intended to - improve quality of care and shift from volume to value.

References

1. Damberg CL, Sorbero ME, Lovejoy SL, et al. An evaluation of the use of performance measures in health care. . 2011.
2. MacLean CH, Kerr EA, Qaseem A. Time out—charting a path for improving performance measurement. N Engl J Med. 2018;378(19):1757-1761.
3. Miller TP, Brennan TA, Milstein A. How can we make more progress in measuring physicians’ performance to improve the value of care? Health Aff. 2009;28(5):1429-1437.
4. Centers for Medicaid and Medicare Services. What are the value-based programs? https://www.cms.gov/Medicare/Quality-Initiatives-Patient-Assessment-Instruments/Value-Based-Programs/Value-Based-Programs.html. Accessed July 15, 2019.
5. Young RA, Roberts RG, Holden RJ. The challenges of measuring, improving, and reporting quality in primary care. The Annals of Family Medicine. 2017;15(2):175-182.
6. Agency for Healthcare Research and Quality, Rockville, MD. Major physician measurement sets. content last reviewed december 2018. https://Www.ahrq.gov/talkingquality/measures/setting/physician/measurement-sets.html. https://www.ahrq.gov/talkingquality/measures/setting/physician/measurement-sets.html. Accessed July 10, 2019.
7. Ranganathan M, Hibbard J, Rodday AMC, et al. Motivating public use of physician-level performance data: An experiment on the effects of message and mode. Medical care research and review. 2009;66(1):68-81.
8. Quality D. Performance Measurement—A foundation for clinical decision support. A consensus report. 2010.
9. Stason WB, Auerbach B, Bloomberg M. Principles for profiling physician performance. The Health Report. 1999.
10. Welch HG, Miller ME, Welch WP. Physician profiling--an analysis of inpatient practice patterns in florida and oregon. N Engl J Med. 1994;330(9):607-612.
11. Centers for Medicaid and Medicare Services. Physician compare. https://www.medicare.gov/physiciancompare/ Website.
12. Heathgrades. www.healthgrades.com Website.
13. Damberg CL, Raube K, Teleki SS, dela Cruz E. Taking stock of pay-for-performance: A candid assessment from the front lines. Health Aff. 2009;28(2):517-525.
14. Agency for Healthcare Research and Quality, Rockville, MD. Module 10. academic detailing as a quality improvement tool. content last reviewed may 2013.  https://www.ahrq.gov/professionals/prevention-chronic-care/improve/system/pfhandbook/mod10.html. Accessed July 20, 2018.
15. Rathi VK, McWilliams JM. First-year report cards from the merit-based incentive payment system (MIPS): What will be learned and what next? JAMA. 2019;321(12):1157-1158.
16. Berenson RA, Pronovost PJ, Krumholz HM. Achieving the potential of health care performance measures. Timely Anal Immed Health Pol. 2013(2013):2.
17. Scholle SH, Roski J, Dunn DL, et al. Availability of data for measuring physician quality performance. Am J Manag Care. 2009;15(1):67.
18. National Committee for Quality Assurance. HEDIS 2007 technical specifications for physician measurement. . 2007.
19. Kaplan SH, Griffith JL, Price LL, Pawlson LG, Greenfield S. Improving the reliability of physician performance assessment: Identifying the “physician effect” on quality and creating composite measures. Med Care. 2009;47(4):378-387.
20. Scholle SH, Roski J, Adams JL, et al. Benchmarking physician performance: Reliability of individual and composite measures. Am J Manag Care. 2008;14(12):833.
21. Sandy LG, Haltson H, Metfessel BA, Reese C. Measuring physician quality and efficiency in an era of practice transformation: PCMH as a case study. The Annals of Family Medicine. 2015;13(3):264-268.
22. Casalino LP, Gans D, Weber R, et al. US physician practices spend more than $15.4 billion annually to report quality measures. Health Aff. 2016;35(3):401-406.
23. Analytics and Implementation Support, Advisory Board Company. Physician performance improvement (PI) metric pick list. https://www.advisory.com/research/physician-executive-council/tools/2009/physician-performance-improvement-metric-pick-list. Updated 2009. Accessed July 15, 2019.
24. Karol G. Wicker. Meeting measurement requirements: The health system and hospital
    perspective. 2018.
25. McNamara P, Shaller D, De La Mare D, Ivers N. Confidential physician feedback reports: Designing for optimal impact on performance. Confidential physician feedback reports: Designing for optimal impact on performance.Rockville, MD: Agency for Healthcare Research and Quality; 2016. 2017.
26. The Advisory Board Company. Building actionable performance dashboards. . 2012.
27. Kazandjian VA, Lawthers J, Cernak CM, Pipesh FC. Relating outcomes to processes of care: The maryland hospital association’s quality indicator project (QI project®). Jt Comm J Qual Improv. 1993;19(11):530-538.