Key Points
Question
Since the 2021 Medicare fee schedule updates, are patient demographics associated with work relative value units (wRVUs) generated during outpatient dermatology encounters?
Findings
This cross-sectional study found that across 47 607 encounters before and 42 049 encounters after the update more wRVUs were generated during encounters with White, older, and male patients relative to patients who were Asian or Black, younger, and female. The wRVU differences by race decreased after 2021, with remaining differences largely mediated by premalignant destructions and biopsies.
Meaning
After the 2021 Medicare fee schedule updates, there was a persistent, albeit lessened, gap between wRVU productivity in outpatient dermatology visits for Asian and Black compared with White patients.
This cross-sectional study of outpatient dermatology encounters examined the association between dermatology encounter work relative value units and patient age, sex, and race and ethnicity since the 2021 Medicare physician fee schedule update.
Abstract
Importance
Clinical productivity measures may incentivize clinical care to specific patient populations and thus perpetuate inequitable care. Before the 2021 Medicare physician fee schedule changes, outpatient dermatology encounters for patients who were younger, female, and races other than White systematically generated fewer work relative value units (wRVUs).
Objective
To examine the association of patient race, age, and sex with wRVUs generated by outpatient dermatology encounters after 2021.
Design, Setting, and Participants
This multi-institutional cross-sectional study evaluated demographic and billing data for outpatient dermatology encounters across 3 academic dermatology practices. The study compared wRVUs generated by outpatient general dermatology encounters in 6-month periods before and after the 2021 fee schedule updates (March 1 to August 31, 2019, and March 1 to August 31, 2021). Eligibility required an age of 18 years or older and available age, race, and sex data. Data analysis was performed from September 2022 to March 2024.
Main Outcomes and Measures
The primary outcome was wRVUs generated per encounter.
Results
This study included 89 656 encounters (47 607 before the 2021 Medicare physician fee schedule update and 42 049 after the update). Across all encounters, the mean (SD) patient age was 56.3 (17.8) years; 55 460 encounters (61.9%) were with female patients and 34 196 (38.1%) were with male patients; and 3457 encounters (3.9%) were with Asian patients, 10 478 (11.7%) with Black patients, 72 894 (81.3%) with White patients, and 2287 (3.2%) with patients of other race or ethnicity (Latino and multiracial). The mean (SD) wRVUs per outpatient dermatology encounter was 1.44 (0.88) before the update and 1.80 (0.99) after (P < .001). After 2021, adjusted analyses demonstrated significantly fewer wRVUs per encounter for female (β, −0.11; 95% CI, −0.13 to −0.10) compared with male patients, and for younger (β, 0.04 [95% CI, 0.04 to 0.05] per 10-year increase in age) compared with older patients. After the update, compared with White patients, visits with Asian patients generated fewer wRVUs (β, −0.12; 95% CI, −0.17 to −0.08) as did visits with Black patients (β, −0.14; 95% CI, −0.17 to −0.11), both statistically significant reductions compared with prior comparisons (P < .001 for both). After 2021, mediation analysis identified that premalignant destructions and biopsies mediated many of the remaining differences in wRVU generation by patient age, race, and sex.
Conclusions and Relevance
This study found that after the 2021 Medicare fee schedule updates, there was a persistent, albeit reduced, gap between wRVU productivity in outpatient dermatology visits for Asian and Black compared with White patients. These persisting differences were attributable to skin biopsies and cryotherapy of premalignant lesions.
Introduction
Work relative value units (wRVUs) are assigned by the Centers for Medicare & Medicaid Services as an estimate of physician time and effort for rendering medical services. wRVUs are linked to payment for medical services and are used by many private and public health systems to calculate physician productivity and compensation. In outpatient dermatology, encounters for patients who are female, pediatric, or of races other than White systematically generate fewer wRVUs compared with encounters for older White men.1,2,3 Black patients have reported difficulty accessing dermatologic care, are prescribed different treatments, and experience worse outcomes across common dermatologic conditions, such as acne, atopic dermatitis, and melanoma.4,5,6 A myriad of potential causes unrelated to billing policies, such as health insurance–related barriers to access to care, may underlie these disparities, and no causal mechanism between billing policies and sociodemographic disparities has been established. Nonetheless, wRVUs influence physician compensation in many health systems, which may influence recruitment and retention.7 Thus, differences in wRVU generation by patient population may incentivize dermatologists to practice in locations that care for certain patient populations or pursue some subspecialty training pathways over others, which may contribute to structural inequities.8,9,10
To simplify billing processes, reduce documentation burden, and improve compensation for medical decision-making, the Centers for Medicare & Medicaid Services enacted a physician fee schedule change on January 1, 2021, which updated guidance and valuation for outpatient evaluation and management visits. The impacts of the 2021 physician fee schedule change on wRVUs in outpatient dermatology practices have not yet been well characterized. Immediately after implementation of the new guidelines, level 4 and 5 visits increased among ambulatory practices across the US.11 In dermatology, level 4 billing increased from 21.5% to 33.5% of evaluation and management visits.11
This study examined the association between dermatology encounter wRVUs and patient age, sex, and race since the 2021 Medicare physician fee schedule update and evaluated whether there has been a reduction in previously observed differences in encounter wRVUs by patient sociodemographic characteristics. We secondarily aimed to determine the extent to which skin biopsies and premalignant destructions explained any wRVU differences that persisted after the fee schedule change.
Methods
Study Population
Billing data from 3 academic outpatient dermatology practices in Atlanta, Georgia; Durham, North Carolina; and Boston, Massachusetts were retrospectively evaluated in a 6-month period after the fee schedule update (March 1 to August 31, 2021) and in a 6-month control period before the fee schedule update (March 1 to August 31, 2019). Data from 2019 were used instead of data from 2020 due to the impact of the COVID-19 pandemic on practice volumes. This cross-sectional study aimed to compare wRVU generation among adult general outpatient dermatology encounters and therefore excluded visits with patients younger than 18 years, dermatologic surgery visits (visits with any Current Procedural Terminology codes for Mohs surgery, excisions, or malignant destructions), inpatient consultations, pathology or laboratory services, and visits that generated no wRVUs, such as nursing visits, postoperative visits, phototherapy, and cosmetic procedures. The study was reviewed and approved by the Emory University, Duke University, and Mass General Brigham institutional review boards. Patient consent was waived by all 3 institutional review boards, which determined that the study was minimal risk. The study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guidelines.
Study Design
Encounter characteristics, including wRVUs generated, encounter diagnoses, procedures rendered, and patient sociodemographics, were stratified by period (before or after the Medicare patient fee schedule change) and clinic site and then analyzed using descriptive statistics. Encounters with missing data were excluded. The primary analysis examined changes in the association between the outcome of wRVUs per outpatient dermatology encounter and exposures, including participant age, self-reported race and ethnicity (Asian, Black, White, and other [including Latino and multiracial]), and sex, before and after the 2021 Medicare physician fee schedule update. Insurance type was not included in this analysis because wRVUs are uniform across insurance types, and a prior analysis did not demonstrate a meaningful association between insurance type and wRVUs generated in outpatient dermatology encounters.1
Statistical Analysis
For each study site and each study period (before and after the 2021 fee schedule update), we assessed separate multivariable fixed-effects linear regressions with the outcome of wRVUs per encounter and exposures, including participant age, race and ethnicity, and sex. For each study period, we also generated estimates that combined all sites by performing mixed-effects linear regression models with the outcome of wRVUs per encounter; fixed-effects exposures of participant age, race and ethnicity, and sex; and random effects of study site.
For the primary analysis assessing changes in the association of age, race and ethnicity, and sex with wRVUs per encounter between the preupdate and postupdate study periods, we combined the datasets from before and after the 2021 Medicare physician fee schedule update and ran a mixed-effects multivariable linear regression with the outcome of wRVUs per encounter; fixed-effects exposures, including patient age, race and ethnicity, sex, and billing period (before or after fee-schedule changes); and interaction terms between the billing period and each sociodemographic variable, with study site as the random effect. Significant P values in the β-coefficients for the interaction terms indicated a statistically significant change between the preupdate and postupdate periods for the influence of the sociodemographic variable on wRVUs per encounter.
In another prespecified analysis, mediation analysis using the difference technique was performed to evaluate the contribution of biopsies and premalignant destructions to the observed wRVU differences by patient age, race, and sex in the period after the 2021 fee schedule update.12 This method examines changes in the estimated coefficients for the influence of age, race and ethnicity, and sex on wRVUs per encounter when mediators (biopsies and premalignant destructions) are added to the regression model. This approach yields estimates for (1) the average causal mediation effect (ie, the change in outcome [wRVUs] attributable to the mediators [biopsies and premalignant destruction]), (2) the average direct effect (ie, the change in outcome [wRVUs] attributable only to the exposure [eg, age, race, or sex] and not attributable to the mediators), and (3) the proportion mediated (ie, the average causal mediation effect divided by the total effect of the exposure seen in the model without the mediators). We performed bootstrapping for 1000 cycles with replacement to estimate 95% CIs for these analyses. Two-sided P < .05 was considered significant in 2-tailed tests. Statistical analyses were performed using SAS software, version 9.4 (SAS Institute Inc) and R, version 4.2.2 (R Foundation for Statistical Computing). Data analysis was performed from September 2022 to March 2024.
Results
The final study sample included a total of 89 656 outpatient medical dermatology encounters (47 607 before the 2021 Medicare physician fee schedule update and 42 049 encounters after). Across all encounters, the mean (SD) patient age was 56.3 (17.8) years; 55 460 encounters (61.9%) were with female patients and 34 196 (38.1%) were with male patients; and 3457 encounters (3.9%) were with Asian patients, 10 478 (11.7%) with Black patients, 72 894 (81.3%) with White patients, and 2287 (3.2%) with patients of other race or ethnicity (Latino and multiracial). Across all 3 sites, the mean (SD) wRVUs generated per outpatient dermatology encounters increased from 1.44 (0.88) before the updates to 1.80 (0.99) after the updates (P < .001). Encounter characteristics and patient demographics are detailed in Table 1.
Table 1. Visit Characteristics for Adult Outpatient Dermatology Encounters in 3 Academic Dermatology Practices, March 1 to August 31, 2019, and March 1 to August 31, 2021a.
| Characteristic | No. (%) of encountersb | |||||
|---|---|---|---|---|---|---|
| Site 1 | Site 2 | Site 3 | ||||
| Before updates (n = 21 412) | After updates (n = 18 880) | Before updates (n = 18 570) | After updates (n = 15 797) | Before updates (n = 7625) | After updates (n = 7372) | |
| wRVUs per encounter, mean (SD) | 1.34 (0.95) | 1.63 (1.12) | 1.58 (0.87) | 2.03 (0.90) | 1.40 (0.61) | 1.72 (0.72) |
| Sex of patient | ||||||
| Male | 7814 (36.5) | 6391 (33.9) | 7691 (41.4) | 6219 (39.4) | 3186 (41.8) | 2895 (39.3) |
| Female | 13 598 (63.5) | 12 489 (66.1) | 10 879 (58.6) | 9578 (60.6) | 4439 (58.2) | 4477 (60.7) |
| Age, mean (SD), y | 56.8 (17.1) | 53.1 (17.9) | 57.8 (17.7) | 57.3 (18.1) | 57.1 (17.9) | 56.4 (17.8) |
| Race and ethnicity | ||||||
| Asian | 806 (3.8) | 902 (4.8) | 539 (2.9) | 579 (3.7) | 316 (4.1) | 315 (4.3) |
| Black | 1231 (5.8) | 1314 (7.0) | 2228 (12.0) | 2043 (12.9) | 1735 (22.8) | 1927 (26.1) |
| White | 18 420 (86.0) | 15 737 (83.4) | 15 384 (82.8) | 12 848 (81.3) | 5478 (71.8) | 5027 (68.2) |
| Otherc | 955 (4.5) | 927 (4.9) | 419 (2.3) | 327 (2.1) | 96 (1.3) | 103 (1.4) |
| Insuranced | ||||||
| Commercial | NA | NA | 9547 (51.4) | 7951 (50.3) | 5739 (75.3) | 5257 (71.3) |
| Medicare | NA | NA | 7941 (42.8) | 6767 (42.8) | 2928 (38.4) | 2874 (39.0) |
| Medicaid | NA | NA | 461 (2.5) | 484 (3.1) | 167 (2.2) | 107 (1.5) |
| Self-pay | NA | NA | 470 (2.5) | 500 (3.2) | 84 (1.1) | 38 (0.5) |
| Other | NA | NA | 308 (1.7) | 286 (1.8) | 144 (1.9) | 136 (1.8) |
| Procedures rendered | ||||||
| Premalignant destruction | 3210 (15.0) | 2310 (12.2) | 3685 (19.8) | 3143 (19.9) | 1065 (14.0) | 899 (12.2) |
| Biopsy | 2550 (11.9) | 2158 (11.4) | 3250 (17.5) | 2968 (18.8) | 1262 (16.6) | 1206 (16.4) |
Abbreviations: NA, not applicable; wRVU, work relative value unit.
All characteristics in this table are analyzed at the encounter level, not at the level of unique patients.
Unless otherwise indicated.
Other races included Latino and multiracial.
Totals for insurance type exceed the total number of encounters in the cohort because more than 1 insurance type was billed in some encounters.
Before the Medicare physician fee schedule update, outpatient dermatology encounters with patients who were White, male, and older generated more wRVUs than encounters for patients who were Asian, Black, or of other race and ethnicity; female; and younger (Table 2). In 2019, fewer wRVUs were generated for visits with Asian patients (β, −0.24; 95% CI, −0.28 to −0.19), Black patients (β, −0.22; 95% CI, −0.24 to −0.19), and patients of other race or ethnicity (β, −0.14; 95% CI, −0.19 to −0.10) compared with White patients. Encounters for female patients generated fewer wRVUs than encounters for male patients (β, −0.12; 95% CI, −0.13 to −0.10). There was also an increase in wRVUs for every 10-year increase in patient age (β, 0.05; 95% CI, 0.05-0.06).
Table 2. Factors Associated With wRVUs Generated by Outpatient Dermatology Encounters Across 3 Academic Dermatology Practices, Excluding Dermatologic Surgery Visits.
| Factor | wRVUs per encounter, mean (SD) | Adjusted β (95% CI)a | |||
|---|---|---|---|---|---|
| Before update | After update | Before update | After update | P valueb | |
| Age (per 10 y) | NA | NA | 0.05 (0.05 to 0.06) | 0.04 (0.04 to 0.05) | .08 |
| Site 1 | NA | NA | 0.07 (0.06 to 0.08) | 0.04 (0.04 to 0.05) | <.001 |
| Site 2 | NA | NA | 0.03 (0.02 to 0.03) | 0.04 (0.03 to 0.05) | .02 |
| Site 3 | NA | NA | 0.06 (0.05 to 0.07) | 0.04 (0.04 to 0.05) | .004 |
| Sex | |||||
| Male | 1.54 (0.94) | 1.90 (1.04) | 0 [Reference] | 0 [Reference] | NA |
| Site 1 | 1.45 (1.04) | 1.73 (1.19) | 0 [Reference] | 0 [Reference] | NA |
| Site 2 | 1.65 (0.91) | 2.12 (0.93) | 0 [Reference] | 0 [Reference] | NA |
| Site 3 | 1.48 (0.67) | 1.81 (0.79) | 0 [Reference] | 0 [Reference] | NA |
| Female | 1.38 (0.84) | 1.74 (0.96) | −0.12 (−0.13 to −0.10) | −0.11 (−0.13 to −0.10) | .76 |
| Site 1 | 1.28 (0.89) | 1.58 (1.08) | −0.14 (−0.17 to −0.12) | −0.13 (−0.16 to −0.09) | .49 |
| Site 2 | 1.53 (0.84) | 1.97 (0.87) | −0.10 (−0.12 to −0.07) | −0.11 (−0.14 to −0.08) | .65 |
| Site 3 | 1.34 (0.56) | 1.67 (0.67) | −0.09 (−0.12 to −0.07) | −0.10 (−0.14 to −0.07) | .68 |
| Race and ethnicity | |||||
| Asian | 1.16 (0.67) | 1.62 (0.89) | −0.24 (−0.28 to −0.19) | −0.12 (−0.17 to −0.08) | <.001 |
| Site 1 | 1.11 (0.75) | 1.52 (0.99) | −0.17 (−0.23 to −0.10) | −0.07 (−0.15 to 0.00) | .07 |
| Site 2 | 1.24 (0.64) | 1.79 (0.81) | −0.34 (−0.42 to −0.27) | −0.22 (−0.30 to −0.15) | .03 |
| Site 3 | 1.17 (0.43) | 1.63 (0.64) | −0.23 (−0.30 to −0.16) | −0.07 (−0.15 to 0.01) | .003 |
| Black | 1.24 (0.68) | 1.67 (0.84) | −0.22 (−0.24 to −0.19) | −0.14 (−0.17 to −0.11) | <.001 |
| Site 1 | 1.09 (0.75) | 1.49 (1.01) | −0.20 (−0.26 to −0.15) | −0.12 (−0.18 to −0.05) | .04 |
| Site 2 | 1.38 (0.76) | 1.81 (0.81) | −0.21 (−0.25 to −0.17) | −0.20 (−0.25 to −0.16) | .88 |
| Site 3 | 1.17 (0.46) | 1.64 (0.71) | −0.25 (−0.29 to −0.16) | −0.08 (−0.12 to −0.05) | <.001 |
| White | 1.49 (0.91) | 1.84 (1.02) | 0 [Reference] | 0 [Reference] | NA |
| Site 1 | 1.38 (0.97) | 1.66 (1.14) | 0 [Reference] | 0 [Reference] | NA |
| Site 2 | 1.63 (0.89) | 2.08 (0.91) | 0 [Reference] | 0 [Reference] | NA |
| Site 3 | 1.48 (0.64) | 1.76 (0.73) | 0 [Reference] | 0 [Reference] | NA |
| Otherc | 1.25 (0.78) | 1.64 (0.99) | −0.14 (−0.19 to −0.10) | −0.07 (−0.12 to −0.02) | .10 |
| Site 1 | 1.18 (0.84) | 1.54 (1.03) | −0.13 (−0.19 to −0.07) | −0.06 (−0.14 to 0.01) | .16 |
| Site 2 | 1.41 (0.65) | 1.91 (0.92) | −0.16 (−0.25 to −0.08) | −0.09 (−0.19 to 0.01) | .27 |
| Site 3 | 1.28 (0.55) | 1.67 (0.68) | −0.13 (−0.24 to −0.01) | −0.02 (−0.16 to 0.12) | .28 |
Abbreviations: NA, not applicable; wRVUs, work relative value units.
β Estimates give the change in wRVUs billed per encounter by patient characteristic.
Based on a separate linear regression with wRVUs as the outcome and independent variables including age, sex, race, period (before or after update), and interaction terms between period and each other independent variable. The P value listed in each row is the P value for the interaction term between period and that independent variable (eTable in Supplement 1).
Includes Latino and multiracial.
After the Medicare physician fee schedule update, outpatient dermatology encounters with patients who were White, male, and older continued to generate more wRVUs than encounters with non-White, female, and younger patients. In 2021, significantly fewer wRVUs were generated for Asian patients (β, −0.12; 95% CI, −0.17 to −0.08), Black patients (β, −0.14; 95% CI, −0.17 to −0.11), and patients of other race or ethnicity (β, −0.07; 95% CI, −0.12 to −0.02) compared with White patients. Encounters for female patients in this period generated fewer wRVUs than encounters for male patients (β, −0.11; 95% CI, −0.13 to −0.10). Younger patients also had significantly fewer wRVUs than older patients (β, 0.04 [95% CI, 0.04-0.05] per 10-year increase in age).
Comparing before and after the fee schedule update, there was a statistically significant reduction in the difference between wRVU generation for visits with Asian and Black compared with White patients (Table 2; eTable in Supplement 1). The attenuation in wRVU differences for Black vs White patients was more pronounced at site 3 (β = −0.25 [95% CI, −0.29 to −0.16] before vs −0.08 [95% CI, −0.12 to −0.05] after; P < .001) compared with site 1 (β = −0.20 [95% CI, −0.26 to −0.15] before vs −0.12 [95% CI, −0.18 to −0.05] after; P = .04), and there was no statistically significant change at site 2 (β = −0.21 [95% CI, −0.25 to −0.17] before vs −0.20 [95% CI, −0.25 to −0.16] after; P = .88). The 2021 fee schedule update was not associated with any change in wRVUs generated according to patient age or sex. After the 2021 Medicare physician fee schedule update, mediation analysis identified that skin biopsies and premalignant destructions mediated at least 90% of the observed differences in wRVU generation by patient age, race and ethnicity, and sex when estimated across all 3 sites (Table 3).
Table 3. wRVU Differences Mediated by Biopsies and Premalignant Destructions in a General Dermatology Practice After the 2021 Medicare Physician Fee Schedule Update, March 1 to August 31, 2021.
| Site | Estimate (95% CI)a | ||||
|---|---|---|---|---|---|
| Age per 10-y increase | Female sexb | Black raceb | Asian raceb | Other race or ethnicityb,c | |
| ACME | |||||
| All sitesd,e | 0.066 (0.064 to 0.069) | −0.12 (−0.12 to −0.11) | −0.18 (−0.19 to −0.17) | −0.15 (−0.16 to −0.13) | −0.12 (−0.13 to −0.10) |
| Site 1 | 0.075 (0.071 to 0.080) | −0.11 (−0.13 to −0.09) | −0.18 (−0.20 to −0.16) | −0.14 (−0.16 to −0.11) | −0.10 (−0.13 to −0.08) |
| Site 2 | 0.057 (0.053 to 0.060) | −0.12 (−0.13 to −0.10) | −0.18 (−0.20 to −0.17) | −0.16 (−0.19 to −0.14) | −0.11 (−0.15 to −0.08) |
| Site 3 | 0.042 (0.038 to 0.047) | −0.07 (−0.09 to −0.06) | −0.15 (−0.16 to −0.13) | −0.11 (−0.14 to −0.09) | −0.11 (−0.15 to −0.06) |
| Average direct effect | |||||
| All sitesd,f | −0.014 (−0.018 to −0.009) | −0.01 (−0.03 to 0.01) | 0.07 (0.05 to 0.10) | 0.02 (−0.02 to 0.06) | 0.00 (−0.05 to 0.05) |
| Site 1 | −0.031 (−0.039 to −0.022) | −0.02 (−0.05 to 0.01) | 0.06 (0.01 to 0.12) | 0.06 (−0.01 to 0.13) | 0.04 (−0.03 to 0.10) |
| Site 2 | −0.016 (−0.023 to −0.009) | 0.01 (−0.02 to 0.03) | −0.02 (−0.06 to 0.02) | −0.06 (−0.13 to 0.01) | 0.02 (−0.08 to 0.12) |
| Site 3 | 0.002 (−0.007 to 0.011) | −0.03 (−0.06 to 0.00) | 0.06 (0.02 to 0.10) | 0.04 (−0.03 to 0.11) | 0.08 (−0.05 to 0.22) |
| Proportion mediated, % | |||||
| All sitesd,g | 126 (115 to 138) | 90 (79 to 104) | 168 (137 to 219) | 116 (87 to 176) | 99 (70 to 172) |
| Site 1 | 169 (142 to 210) | 87 (69 to 112) | 155 (105 to 303) | 184 (86 to 1059) | 152 (−996 to 989) |
| Site 2 | 140 (119 to 170) | 108 (87 to 140) | 88 (74 to 110) | 73 (55 to 106) | 120 (−446 to 837) |
| Site 3 | 95 (80 to 118) | 72 (53 to 106) | 175 (120 to 314) | 150 (−354 to 998) | 108 (−1818 to 1815) |
Abbreviations: ACME, average causal mediation effect; wRVU, work relative value unit.
The 95% CIs were calculated through bootstrap, consisting of 1000 replicate samples.
Male sex and White race were used as reference groups.
Includes Latino and multiracial.
Based on multivariable fixed-effects model, not accounting for clustering by site.
ACME refers to the change in outcome (wRVUs) that is attributable to the mediators (biopsies and premalignant destructions). For example, when comparing encounters for Black and White patients, biopsies and premalignant destructions explained a 0.18 (95% CI, 0.17-0.19) reduction in wRVUs between encounters for Black patients and White patients.
Average direct effect refers to the change in outcome (wRVUs) that is attributable only to the exposure (ie, age, race, or sex) and not attributable to the mediators (biopsies and premalignant destructions). For example, if not for the mediation effect of biopsies and premalignant destructions, encounters for patients who are Black would generate 0.07 wRVUs more than encounters for White patients.
Proportion mediated is calculated as (ACME / Total Effect), where total effect refers to the β value given in Table 2. In cases where the average causal mediation effect exceeds the total effect, the proportion mediated is greater than 100%. For example, biopsies and cryotherapy mediated a 0.18-wRVU reduction per encounter for Black compared with White patients, although the total observed effect was that encounters for Black patients generated 0.14 fewer wRVUs than encounters for White patients.
Discussion
This multisite, retrospective, cross-sectional study demonstrated a meaningful reduction in the wRVU productivity gap for outpatient dermatology visits for Asian and Black compared with White patients after the 2021 Medicare physician fee schedule update. This reduction in wRVU differences suggests that updates to Medicare policy may be associated with reduced differences in wRVU incentives by patient race. Nonetheless, differences in wRVUs generated by patient age, race and ethnicity, and sex persisted. Visits for patients who were younger, female, and who identified as Asian, Black, or other race or ethnicity continued to generate significantly fewer wRVUs compared with those for White, older, or male patients even after the 2021 fee schedule updates, consistent with prior studies.1,3 After these updates, we found that a higher proportion of the differences in wRVU productivity by patient demographics were mediated by skin biopsies and premalignant destructions compared with what was found in a prior study.1 Across all 3 sites, wRVUs per encounter increased after the 2021 Medicare physician fee schedule update. This finding is consistent with prior studies that demonstrated higher-level evaluation and management coding11 and increasing time required per patient encounter in dermatology.13
Although intended to signify physician effort, wRVUs under the Medicare physician fee schedule have long placed lower value on cognitive work compared with procedures. One prior study suggested that physicians generate 3 to 5 times more revenue for time spent performing cataract extraction or colonoscopy than providing cognitive care.14 Within dermatology, the prevalence of skin cancers varies by patient age, sex, and race and ethnicity. Appropriate compensation for skin cancer treatment remains important to ensure optimal outcomes for this population. However, undervaluing cognitive services may systematically disincentivize dermatologic care of patients who are least likely to develop skin cancers—those who are young, are female, and have more skin pigmentation. Additional work is needed to determine whether the wRVU valuation for cognitive dermatologic services compared with procedures such as biopsies and cryotherapy accurately reflects dermatologist effort, as a mismatch between effort and wRVU valuation has the potential to disincentivize care for patients least likely to experience skin cancers and perpetuate structural inequities in care.
Limitations
This study has several limitations. Data were drawn from 3 academic practices located in the Eastern US; therefore, findings may not be generalizable to other practice settings. Data collection began 2 months after the 2021 Medicare physician fee schedule updates went live and continued for 6 months. It is possible that coding practices evolved over time and may have changed since this initial phase.
It is difficult to precisely convert differences in wRVU generation to differences in revenue because revenue depends on many additional factors, including insurance type, local claims collection practices, and other local factors. In 2019 and 2021, the Medicare conversion factors were $36.04 and $34.89 per wRVU,15 respectively, meaning that the observed wRVU differences between encounters for Black and White patients with Medicare insurance result in a difference of $7.93 per encounter before the 2021 Medicare physician fee schedule update and $4.88 per encounter afterward.
Additionally, our study excluded some visits that may influence revenue generation. For example, cosmetic procedures and patch testing may yield revenue without generating wRVUs. We also excluded dermatologic operations. In prior work at a single site,1 the inclusion of dermatologic operations exacerbated differences in wRVU generation by race, sex, and age, most likely due to high wRVUs for Mohs surgery for nonmelanoma skin cancers that disproportionately affect older, White men. However, this approach also excludes operations such as those for hidradenitis suppurativa that disproportionately affect Black women.
Across the 3 study sites, there were some notable differences in the association between period and wRVU generation by patient demographic. For instance, the wRVU gap between Black and White patients was not reduced across periods at site 2, although sites 1 and 3 demonstrated large reductions in the wRVU gap between Black and White patients after 2021. This variation across sites may reflect differences in the mix between new and established patient visits, differences in patients’ chief concerns across sites, or local differences in implementation of the new coding guidelines. Data collection across more dermatology practices with different care models could enhance the external validity of these findings and clarify the reasons for the observed differences.
Clinician time-motion data were unavailable, limiting our ability to evaluate whether the differences observed in this study represent inequitable valuation of the care provided to women, younger patients, and patients of minoritized race and ethnicity or true differences in clinician time due to procedures rendered. However, a prior study suggests that clinician time per encounter is not significantly different by patient race.13
Conclusions
This study found that after the 2021 Medicare patient fee schedule updates, there was a persistent, albeit lessened, gap between wRVU productivity in outpatient dermatology visits for Asian and Black patients compared with White patients. These persisting differences were attributable to skin biopsies and cryotherapy of premalignant lesions. Although the impact of these differences remains unknown, undervaluing cognitive services within dermatology could disincentivize dermatologic care of patients who are least likely to develop skin cancers. Additional work is needed to determine whether the wRVU valuation for cognitive dermatologic services compared with procedures accurately reflects practitioner effort, as a mismatch has the potential to disincentivize care for patients least likely to experience skin cancers and perpetuate structural inequities in dermatologic care.
eTable. Changes in Factors Associated With wRVUs Generated by Outpatient Dermatology Encounters Across Three Academic Dermatology Practices – A Mixed Effects Multivariable Linear Regression
Data Sharing Statement
References
- 1.Orenstein LAV, Nelson MM, Wolner Z, et al. Differences in outpatient dermatology encounter work relative value units and net payments by patient race, sex, and age. JAMA Dermatol. 2021;157(4):406-412. doi: 10.1001/jamadermatol.2020.5823 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Supapannachart KJ, Wolner Z, Miller AE, et al. Comparison of work relative value units for outpatient pediatric and adult dermatology encounters. JAMA Dermatol. 2022;158(11):1300-1303. doi: 10.1001/jamadermatol.2022.3609 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Supapannachart KJ, Wolner Z, Swerlick RA, Orenstein LAV. Outpatient dermatology clinic half days with more women, racial minority, and younger patients generate fewer work relative value units. Int J Womens Dermatol. 2022;8(3):e048. doi: 10.1097/JW9.0000000000000048 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Barbieri JS, Shin DB, Wang S, Margolis DJ, Takeshita J. Association of race/ethnicity and sex with differences in health care use and treatment for acne. JAMA Dermatol. 2020;156(3):312-319. doi: 10.1001/jamadermatol.2019.4818 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Abuabara K, You Y, Margolis DJ, Hoffmann TJ, Risch N, Jorgenson E. Genetic ancestry does not explain increased atopic dermatitis susceptibility or worse disease control among African American subjects in 2 large US cohorts. J Allergy Clin Immunol. 2020;145(1):192-198.e11. doi: 10.1016/j.jaci.2019.06.044 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Cormier JN, Xing Y, Ding M, et al. Ethnic differences among patients with cutaneous melanoma. Arch Intern Med. 2006;166(17):1907-1914. doi: 10.1001/archinte.166.17.1907 [DOI] [PubMed] [Google Scholar]
- 7.Witt J. Physician recruitment and retention in Manitoba: results from a survey of physicians’ preferences for rural jobs. Can J Rural Med. 2017;22(2):43-53. [PubMed] [Google Scholar]
- 8.Morra DJ, Regehr G, Ginsburg S. Medical students, money, and career selection: students’ perception of financial factors and remuneration in family medicine. Fam Med. 2009;41(2):105-110. [PubMed] [Google Scholar]
- 9.Pisaniello MS, Asahina AT, Bacchi S, et al. Effect of medical student debt on mental health, academic performance and specialty choice: a systematic review. BMJ Open. 2019;9(7):e029980. doi: 10.1136/bmjopen-2019-029980 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Tripathi R, Knusel KD, Ezaldein HH, Scott JF, Bordeaux JS. Association of demographic and socioeconomic characteristics with differences in use of outpatient dermatology services in the United States. JAMA Dermatol. 2018;154(11):1286-1291. doi: 10.1001/jamadermatol.2018.3114 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Apathy NC, Hare AJ, Fendrich S, Cross DA. Early changes in billing and notes after evaluation and management guideline change. Ann Intern Med. 2022;175(4):499-504. doi: 10.7326/M21-4402 [DOI] [PubMed] [Google Scholar]
- 12.VanderWeele TJ. Mediation analysis: a practitioner’s guide. Annu Rev Public Health. 2016;37:17-32. doi: 10.1146/annurev-publhealth-032315-021402 [DOI] [PubMed] [Google Scholar]
- 13.Roster K, Barbieri JS. Temporal trends and sociodemographic differences in face-to-face time spent with dermatologists, 2006-2016. Arch Dermatol Res. 2023;315(9):2741-2744. doi: 10.1007/s00403-023-02686-3 [DOI] [PubMed] [Google Scholar]
- 14.Sinsky CA, Dugdale DC. Medicare payment for cognitive vs procedural care: minding the gap. JAMA Intern Med. 2013;173(18):1733-1737. doi: 10.1001/jamainternmed.2013.9257 [DOI] [PubMed] [Google Scholar]
- 15.History of Medicare conversion factors. American Medical Association. https://www.ama-assn.org/system/files/cf-history.pdf
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
eTable. Changes in Factors Associated With wRVUs Generated by Outpatient Dermatology Encounters Across Three Academic Dermatology Practices – A Mixed Effects Multivariable Linear Regression
Data Sharing Statement