Sinsky CA, Dugdale DC. Medicare Payment for Cognitive vs Procedural CareMinding the Gap. JAMA Intern Med. 2013;173(18):1733-1737. doi:10.1001/jamainternmed.2013.9257
Health care costs in the United States are rising rapidly, and consensus exists that we are not achieving sufficient value for this investment. Historically, US physicians have been paid more for performing costly procedures that drive up spending and less for cognitive services that may conserve costs and promote population health.
To quantify the Medicare payment gap between representative cognitive and procedural services, each requiring similar amounts of physician time.
Observational analytical study comparing the hourly revenue generated by a physician performing cognitive services (Current Procedural Terminology [CPT] code 99214) and billing by time with that generated by physicians performing screening colonoscopy (Healthcare Common Procedure Coding System code G0121) or cataract extraction (CPT code 66984) for Medicare beneficiaries.
Outpatient medical practice.
Medical care providers of outpatient services.
Work relative-value unit assigned to physician services.
Main Outcome and Measures
Payment for physician services.
The revenue for physician time spent on 2 common procedures (colonoscopy and cataract extraction) was 368% and 486%, respectively, of the revenue for a similar amount of physician time spent on cognitive care.
Conclusions and Relevance
Our analysis indicates that Medicare reimburses physicians 3 to 5 times more for common procedural care than for cognitive care and illustrates the financial pressures that may contribute to the US health care system’s emphasis on procedural care. We demonstrate that 2 common specialty procedures can generate more revenue in 1 to 2 hours of total time than a primary care physician receives for an entire day’s work.
The medical literature has highlighted the decline in the number of physicians entering cognitive specialties, with accompanying warnings about the impending collapse of primary care.1 The number of physicians in training who choose the primary care field of internal medicine dropped by 50% from 1998 to 2003,2 and primary care physicians (PCPs) in practice are leaving at a faster rate than other specialties.3 Increased workloads, administrative hassles, demanding time commitments, and low compensation relative to other specialties4 are major contributing factors.
Fewer physicians are choosing primary care fields while the needs of an aging population with multiple chronic diseases are projected to require an increase in the supply of primary care by at least one-third.5 This mismatch between supply and demand for PCPs has serious implications for the future of US health care. Health care costs in the United States are among the highest in the world and continue to rise. The quality of health care that Americans receive has been questioned.6,7 Worldwide and within the United States, health care costs are lower and quality is higher in regions with more PCPs. In addition, the quality of care is higher and costs are lower for patients whose first contact with the medical system is with a PCP.8
The US health care reimbursement system rewards procedural services while providing financial disincentives for physicians to spend time on cognitive care, the main professional activity of PCPs and other nonprocedural specialists. In a comparison of international health payment systems, Wilson9(p475) concluded that “the current system in the United States offers little incentive for PCPs to provide the kind of care coordination that is known to improve health quality.”
Several proposed or implemented changes, including an increase in the relative value of evaluation and management (E&M) codes,10 pay-for-performance programs,11 and primary care adjustment,12,13 may modestly address this issue, but each of these policy changes is projected to increase PCP compensation by only 1% to 10%.11,13,14 Herein we identify the magnitude of the payment gap for physician time spent on common procedures vs cognitive tasks.
We compared the hourly revenue generated by a physician performing exclusively cognitive services for established patients and billing by time using code 99214 from the Current Procedural Terminology (CPT) (a code commensurate with typical chronic illness care) with that generated by physicians performing screening colonoscopy (code G0121 of the Healthcare Common Procedure Coding System) or cataract extraction (CPT code 66984). These codes are among the top 40 services ranked by charges submitted to Medicare: the E&M CPT code 99214 is 2nd, cataract extraction is 4th, and screening colonoscopy is 36th.15
The payment for a physician service is determined by the total number of relative-value units (RVU) it is assigned. The work RVU (wRVU) for a service is intended to reflect the value of the physician-specific work. In this report we have limited our analysis to differences in payment for the physician work reflected in wRVU; we have not addressed differences in the other components of total RVU.
The 2013 Physician Fee Schedule wRVU for CPT code 99214 is 1.50; for screening colonoscopy, 3.69; and for cataract extraction, 8.52.16 The total wRVU for cataract extraction is a bundled payment and includes 2 CPT code 99212 visits (combined, 0.96 wRVU), 2 CPT code 99213 visits (total, 1.94 wRVU), and one-half of a CPT code 99238 discharge service (0.64 wRVU). Thus, the wRVU for the cataract extraction itself is 4.98. Codes 99212 and 99213 are E&M codes used for less complicated or shorter care than code 99214. Code 99238 is used to report 30 minutes or less of hospital discharge management.
We determined the physician time associated with each service using definitions from the Centers for Medicare & Medicaid Services (CMS)17 or literature estimates.18- 21 Careful history taking, clinical assessment, care coordination, and the management of multiple chronic conditions, commonly termed cognitive care, are core activities of primary care. When the billing of CPT code 99214 service is based on time, the CMS requires that 25 minutes be spent face-to-face with the patient, with more than half of that time spent counseling or coordinating care for the patient’s conditions.
Literature estimates are 13.5 minutes for colonoscopy procedural time18 and 14 minutes for cataract extraction.21 In our calculations, we estimated an additional 10 minutes per patient for each service, including same-day non–face-to-face activities considered by CMS to be integral to providing the service. These bundled activities include reviewing the patient’s medical record, documenting the visit, speaking with family members, and coordination of care with other health care providers. We calculated physicians’ hourly revenue for each of these services by multiplying the number of services performed per hour by the wRVU assigned to that service and then by the 2013 CMS conversion factor (CF) of $34.02 per RVU.22
Using the 2013 RVU schedule16 and the CMS CF,22 cognitive care (CPT code 99214) generates an hourly revenue of $87, whereas procedural care generates an hourly revenue of $320 (screening colonoscopy) or $423 (cataract extraction) (Table 1). We evaluated the effect of differing the duration of cognitive services by analyzing E&M visits of varying length. Each visit length is associated with a different level of service and wRVU per visit. At each visit length, the hourly revenue for cognitive care is markedly lower than the revenue for procedural services (Table 2).
We also evaluated the effect of varying lengths of procedural care times using literature-reported times. If even the longest published times for performing these procedures are used, procedural care generates more than 200% of the revenue that cognitive care generates per unit of physician time (Table 3).
Last, we compared data from our model with data from the Relative Value Scale Update Committee (RUC) (Table 4).23 Analysis of the revenue using RUC data yields a smaller difference between the revenue for cognitive and procedural services owing to substantially longer times assigned by the RUC to the procedural services.
The times used by the RUC (Table 4) show a cognitive-procedural differential that is considerably less than our calculations using published data. The RUC has 31 members, 22 of whom are appointed by national medical specialty societies, 5 of whom are appointed by the American Medical Association or its representative bodies, and 4 of whom are 2-year rotating members, with one of these positions reserved for a primary care representative. Thus, although PCPs constitute nearly half the physician population, they represent only 16% of the voting members of the RUC.4,24 A detailed review of the RUC process is beyond the scope of this report, but that process has significant methodological limitations.4,25,26 The RUC method uses self-reported times for services in which the purpose of the survey (to establish payment rates) is known to the respondents4,25 who stand to benefit by inflated estimates. McCall et al27 found that RUC survey data overestimated the procedural time by an average of 31 minutes across 60 procedures when compared with objective data from operative logs.
We analyzed complex chronic care based on CPT code 99214 as billed by time. The E&M services may also be billed on the basis of content (“intensity”), that is, the complexity of the history, examination, and decision making. A significant difference in same-day time between these 2 methods for the level of service determination would affect our analysis. Available data for office visits in primary care demonstrate concordance between observed intraservice time and the Medicare requirement when billing by time.27,28
Our analysis identifies the difference in hourly revenue for procedural vs cognitive services. The absolute difference in annual compensation for PCPs, gastroenterologists, and ophthalmologists will vary based on the blends of cognitive and procedural services within each specialty. Our analysis may underestimate the differential in revenue available to physicians for cognitive care vs procedural or surgical care in several important respects.
First, to the degree that same-day previsit and postvisit work (eg, reviewing hospital records and other physicians’ notes, responding to postappointment laboratory findings) is greater for cognitive care for complex chronic illness than for screening colonoscopy or uncomplicated cataract extraction, our findings underestimate the payment gap between these cognitive and procedural services. Limited data about physician work intensity support this possibility.29,30
Second, considerable interappointment work without additional revenue occurs during cognitive care. For example, a patient with diabetes and atrial fibrillation who is receiving anticoagulation and insulin therapy typically requires significant between-visit care, which CMS considers a bundled service and thus not a separately reimbursed service. To the extent that interappointment physician work for cognitive care exceeds that for procedural care, our findings underestimate the payment gap. Limited data suggest that surgeons spend less time on work without the patient present than do cognitive fields such as family medicine and internal medicine.29
Third, our model assumes a constant CF for physician services, which is true for Medicare. However, some commercial insurers use a higher CF for procedural services than for cognitive services.25 For example, in 2009, Wellmark of Iowa used a CF of $47.75/RVU for E&M services and $65/RVU for screening colonoscopy (Colleen Olson-Davis, FAHM, e-mail communication, May 12, 2009). We are not aware of any insurance plan that applies a larger CF to E&M services.
We created a model to examine how Medicare values cognitive vs procedural physician work. Depending on the comparison, Medicare values physician time performing procedures 3- to 5-fold more than physician time performing cognitive services. Our model demonstrates that an ophthalmologist will receive more revenue from Medicare for 4 cataract extractions, typically requiring 1 to 2 hours of time, than a PCP will receive for an entire day of delivering complex care for chronic illness to Medicare patients.
Medicare payment policy creates incentives for physicians who perform procedures to avoid providing cognitive care in favor of performing procedures. According to our model, a gastroenterologist is compensated almost 4 times more for time spent performing a procedure than for time spent evaluating the need for the procedure itself or for time spent managing complex illnesses within gastroenterology that do not require a procedure.
Public payers, including Medicare, directly determine approximately one-third of all physician income31 and are thus major contributors to the cognitive-procedural payment gap. Because most commercial payers use the same payment method as Medicare, payment disparities that originate with Medicare policy are extended to most physician payment policies.
The Medicare Payment Advisory Commission (MedPAC)32(p113) has stated that “beneficiary access to undervalued services may be threatened if providers are confronted with incentives to avoid furnishing them relative to more profitable services.” We believe that the overvaluing of procedural care relative to cognitive care may have contributed to an increase in consumption of procedural services and is worthy of further evaluation. Others, including the National Commission on Physician Payment Reform, have come to a similar conclusion.33
Medicare reimburses physicians for procedural care at 368% (screening colonoscopy) and 486% (cataract extraction) of the rate of cognitive care. This relative overvaluing of physician time spent on procedures has multiple effects on the health care system. This value discrepancy is a major contributor to the decline in the number of physicians choosing primary care careers. Such a discrepancy may also contribute to an excess of expensive procedural care. We believe the strong financial incentives described compromise access to primary care and ultimately contribute to the lower quality and higher costs experienced in the United States compared with other developed countries.1,8
Accepted for Publication: May 31, 2013.
Corresponding Author: Christine A. Sinsky, MD, Department of Internal Medicine, Medical Associates Clinic PC, 1000 Langworthy Dr, Dubuque, IA 52001 (firstname.lastname@example.org).
Published Online: August 12, 2013. doi:10.1001/jamainternmed.2013.9257.
Author Contributions:Study concept and design: All authors.
Acquisition of data: All authors.
Analysis and interpretation of data: All authors.
Drafting of the manuscript: All authors.
Critical revision of the manuscript for important intellectual content: All authors.
Statistical analysis: All authors.
Study supervision: Dugdale.
Conflict of Interest Disclosures: None reported.