Association Between the Declining Conversion Factor and Access to Care for Medicare Beneficiaries

Abstract

CMS has maintained budget neutrality in the Medicare Physician Fee Schedule through decreases in the conversion factor, but reduced reimbursement may have secondary access-to-care consequences for Medicare beneficiaries. Hence, this study examined the association between inflation-adjusted conversion factor declines and changes in care utilization for various socioeconomically defined subgroups of Medicare beneficiaries. Using all claims from a nationally representative 5% sample of Medicare fee-for-service beneficiaries, we determined by year (2005-2023) whether beneficiaries used care. Care utilization was compared between subgroups: metropolitan versus small/rural communities, high-income versus low-income communities, and least-deprived versus most-deprived (Area Deprivation Index [ADI]) communities. Care-utilization rates were normalized to account for within and between year differences in subgroup characteristics using multivariable logistic regression. Care-utilization gaps were computed as the difference in normalized utilization values between subgroups. The temporal association between gap size and the inflation-adjusted conversion factor was assessed with linear regression. There were 32,764,305 beneficiary years (4,662,729 unique beneficiaries) that met the selection criteria. Over the 2005-2023 period, the care-utilization gaps for any physician care increased 1.76, 5.89, and 3.82 percentage points for the urbanicity, community-income, and ADI gaps, respectively. Gap growth was significantly associated with the inflation-adjusted conversion factor, which decreased 48.3% between 2005 and 2025. This inflation-adjusted conversion factor decrease was associated with care-utilization gap increases of 2.46 (urbanicity), 7.00 (community-income), and 5.31 (ADI) percentage points across all physicians. Percentage-point gap increases varied by specialty and were largest for family practice physicians: 21.31 (urbanicity), 14.02 (community-income), and 19.55 (ADI) percentage points. Hence, decreased reimbursement resulting from the Medicare budget neutrality mandate may exacerbate disparities in access to care for traditionally underserved Americans.

Highlights

What Do We Already Know About This Topic?

Budget neutrality in the Medicare Physician Fee Schedule has resulted in a substantially declining conversion factor in real (inflation-adjusted) dollars and individuals of lower socioeconomic status have less access to care than those of higher socioeconomic status.

How Does Your Research Contribute to the Field?

The degree to which inflation-adjusted conversion factor declines are associated with widening access-to-care disparities between lower and higher socioeconomic groups is unknown.

What Are Your Research’s Implications Toward Theory, Practice, or Policy?

We found that the declining inflation-adjusted conversion factor is associated with widening access-to-care disparities between lower and higher socioeconomic groups as measured by urbanicity, community income, and the Area Deprivation Index; hence, there is an association between reimbursement policy and the magnitude of access-to-care disparities.

Introduction

For care provided to Medicare fee-for-service beneficiaries, clinician reimbursement equals the conversion factor (a dollar amount) times the units of service (called relative value units [RVUs]) provided. Over the 2005-2021 period, the conversion factor declined 7.9% nominally, and 33.6% when adjusted for inflation. ¹ CMS uses conversion factor cuts to maintain budget neutrality in the Medicare Physician Fee Schedule (MPFS). Various studies have shown the impact of inflation-adjusted conversion factor declines on reimbursement for various physician specialties.^1-11 The specialty-specific impact depends on a combination of conversion factor changes and changes in the RVUs allocated to specific procedures. While such reimbursement declines have direct impacts on clinicians, practices, and health systems, they may also have indirect impacts on patients, which is the focus of this study.

Commercial reimbursement generally exceeds Medicare reimbursement, which generally exceeds Medicaid reimbursement.^12,13 Accordingly, commercial patients are economically more advantageous for providers than Medicare or Medicaid, as demonstrated by the increasing percentage of physicians who have opted out of Medicare^14,15 and studies finding that Medicaid patients have less success in getting appointments than commercial patients,^16-19 although some survey studies find comparable or better access for Medicare compared with commercial patients.^15,20 Similarly, studies show that increased Medicaid reimbursement relative to Medicare reimbursement was associated with increased care utilization, acceptance of new Medicaid patients, and appointment availability for Medicaid patients.^21-24 Hence, differential access to care reflects in part reimbursement differences across insurance types.^17,25-31

Access also varies across socioeconomic strata.^32-35 Given the value of improving health equity across various socioeconomic factors, the objective of this study was to examine the potential secondary association of inflation-adjusted conversion factor declines with changes in care utilization for Medicare fee-for-service beneficiaries by urbanicity, community income, and the area deprivation index (ADI).

Methods

Data and Study Population

The Advarra institutional review board deemed this study exempt from oversight. This retrospective study (2005-2023) was conducted using a nationally representative 5% sample of Medicare fee-for-service beneficiaries and their associated claims. For each year, beneficiaries who were enrolled in Part B for the entire year except for the year they turned 65 met the inclusion criteria. On this basis, the beneficiary-year was the unit of analysis. Beneficiary-years were excluded from the study if they had missing data for urbanicity or gender. Annual Medicare conversion factor values ³⁶ were adjusted for inflation using the Consumer Price Index. Medicare Data on Provider Practice and Specialty (MD-PPAS) were used to determine the number of physicians and non-physician practitioners (NPPs, nurse practitioners and physician assistants) in each year and U.S. Census Bureau data were used for yearly population.

Variables

The outcome variables were the normalized care-utilization gaps for beneficiaries in the following subgroup comparisons: (1) metropolitan versus small/rural areas, (2) communities with high (≥$50,000) versus low (<$25,000) per capita income, and (3) the least versus most deprived communities (best versus worst ADI quartile). For each comparison, we examined the association between changes in the conversion factor and the gap size each year over the 2005-2023 period. We stratified beneficiaries by urbanicity (metropolitan [50,000+ and associated commuting area], micropolitan [10,000-49,999 and associated commuting area], small/rural), per capita community, Area Deprivation Index quartiles, and year. Urbanicity, per capita community income, and ADI are based on each beneficiary’s ZIP code using the rural-urban commuting area codes, IRS income data, and the Neighborhood Atlas, respectively.^37-39 The inflation-adjusted conversion factor was computed using 2005 conversion factor ³⁶ increased by the Consumer Price Index through 2025.

Given that beneficiary characteristics differ by subgroups within and between years, it was necessary to statistically adjust values to reflect a constant set of beneficiary characteristics within subgroups and over time. Accordingly care-utilization gaps were normalized across the following covariates: gender, age group (<65, 65-74, 75-84, ≥85), race/ethnicity (non-Hispanic White, non-Hispanic Black, Hispanic, Asian/Pacific Islander, other/unknown), Charlson Comorbidity Index (CCI: 0, 1, 2, 3-4, ≥5), urbanicity, per capita community income, ADI, year, the state-level average percentage of beneficiaries with care, and the percentage of individuals in the state with commercial insurance (by year). Year (year effects) was included as a categorical variable to control for other factors or laws (such as COVID, the implementation of the Affordable Care Act and Medicare Access and CHIP Reauthorization Act, population growth, and growth in the number of physicians/NPPs), which may impact care utilization. The state-level average percentage of beneficiaries with care was included to control for state-level differences in utilization/access patterns. State-level averages were calculated separately for beneficiaries with care within the broad specialty groups as well as for the select physician specialties included in the stratified/subset analyses (as described in Statistical Analysis). The yearly state-level percentage of individuals with commercial insurance was included to control for the degree to which providers can chose between commercially insured and other patients. ⁴⁰ Population-to-clinician ratios for each year were computed as the ratio of population to the number of providers. This was done overall and by specialty group.

Statistical Analysis

To estimate statistically normalized care-utilization gaps, we used multivariable logistic regression to estimate the association between beneficiaries with care and the various covariates. This was done separately for beneficiaries residing in (1) metropolitan versus small/rural areas (urbanicity gap), (2) higher-income (≥$50,000) versus lower-income (<$25,000) communities (community-income gap), and (3) the least-deprived versus most-deprived ADI quartile communities (ADI gap). Using the coefficients from these logistic models, we predicted care-utilization such that the mix of beneficiaries was identical across years and across urbanicity, community income, and ADI. In this way, the predicated care-utilization gaps only reflected differences in estimated associations (i.e., coefficients) not differences in populations or changes in these populations over time. Through this approach, we estimated normalized care-utilization separately for beneficiaries residing in metropolitan areas, small/rural areas, high-income communities, low-income communities, best-ADI communities, and worst-ADI communities. The urbanicity care-utilization gap (measured in percentage points) was the difference between the care-utilization percentage for metropolitan and small/rural. Likewise, the care-utilization gap by community income (ADI) was computed as the difference between the high and low (best and worst) groups. This was done for each year from 2005 to 2023.

To estimate the association of the care-utilization gap with the inflation-adjusted conversion factor, we used linear regression to estimate the association between the annual care-utilization gap and the decline in the inflation-adjusted conversion factor controlling for the population-to-clinician ratio (19 data points from 2005 to 2023). This was done separately for urbanicity, community-income, and ADI care-utilization gaps.

We computed these care-utilization gaps (and their respective associations with the inflation-adjusted conversion factor) for receiving any physician care. In addition, we computed these gaps (and associations) for beneficiaries with care from specific types of clinicians as defined by CMS in MD-PPAS: mutually exclusive broad-specialty groups (primary care, medical specialties, surgical specialties, hospital-based specialties, obstetrics/gynecology, psychiatry, and NPPs); select specialties within these broad-specialty groups where specialty selection was based on care volume (primary care: internal medicine and family practice; medical: cardiology, hematology/oncology, rheumatology; surgical: ophthalmology, orthopedic/sports medicine, urology; hospital-based: radiology, pathology); and all physicians/NPPs combined. Statistical significance was assessed at p-value<0.05. All analyses were conducted using Stata 18 (StataCorp, College Station, Texas).

Results

Across the study period (2005-2023), 32,776,349 beneficiary years (4,663,782 beneficiaries) met the inclusion criteria. After exclusions for missing urbanicity (n=12,028) and gender (n=16), 32,764,305 beneficiary years (4,662,729 beneficiaries) were included in the analysis. These beneficiaries were predominantly female (57.2%), non-Hispanic White (66.8%), had no comorbidities as measured by the CCI (38.0%), and resided in metropolitan areas (76.5%) (Supplemental Table 1). Across all beneficiary-years, 87.9% had claims for physician or NPP care (87.1% for physicians only and 33.7% for NPPs only), with lower percentages by broad-specialty group ranging from 5.4% for psychiatry to 72.5% for primary care (Supplemental Table 2).

Over the 2005-2025 period, the conversion declined $5.55 or 14.6% from $37.8975 in 2005 to $32.3465 in 2025 (Figure 1). However, when inflation-adjusted, the 2005 conversion factor would have been $62.5149 in 2025 dollars, meaning there was an inflation-adjusted decline of $30.17 or 48.3% between 2005 and 2025. During the study period (2005-2023) the conversion factor declined 10.6% or 42.7% when inflation adjusted.

Figure 1.

Trend in conversion factor values, 2005-2025, nominal and inflation adjusted. Legend: The conversion factor was adjusted for inflation using the Consumer Price Index (CPI).

The urbanicity-care-utilization gap grew significantly over the study period. In 2005, the normalized care-utilization for any physician care was 90.29% for metropolitan areas compared with 90.03% in small/rural areas (Figure 2A). This reflected a small care-utilization gap of 0.26 percentage points in 2005. Care-utilization for any physician services declined between 2005 and 2023 for both metropolitan and small/rural areas. By 2023, care-utilization was 85.75% for metropolitan areas and 83.73% for small/rural areas reflecting 2.02-percentage-point care-utilization gap in 2023 or a 1.76-percentage-point increase (678%) from the 2005 gap (Table 1 and Figure 2B). The magnitude of this gap is correlated with the inflation-adjusted conversion factor. Figure 2B shows both the normalized urbanicity-care-utilization gap and this gap as statistically predicted by the decline in the inflation-adjusted conversion factor. Statistically, the association between the inflation-adjusted conversion factor and the gap is such that a $1 decline in the inflation-adjusted conversion factor is associated with a 0.08 (95% CI: 0.01, 0.15) percentage-point increase in the gap (Supplemental Figure 1). This modeled value was multiplied by the $30.17 decline in the inflation-adjusted conversion factor to estimate the associated effect on the urbanicity-care-utilization gap. Accordingly, this $30.17 decline was associated with a 2.46 (95% CI: 0.45, 4.47) percentage point increase in the urbanicity care-utilization gap (Figure 3).

Figure 2.

Trends in use of any physician care by year, and conversion-factor-associated gaps in physician care. Legend: Panels A, C, and E show the percentage of Medicare beneficiaries with any physician care by urbanicity, community-income-per-capita, and ADI. Panels B, D, and F show the actual care-utilization gaps and the care-utilization gap as predicted based on the inflation-adjusted conversion factor controlling for the population-to-clinician (physicians and NPPs) ratio. The gaps (solid blue lines) are normalized to reflect the same distribution of beneficiaries for all years. The modeled gaps (dash blue lines) are the linear prediction of the gap based on the inflation-adjusted conversion factor. ADI, Area Deprivation Index; NPP, nonphysician practitioner

Table 1.

Changes in Care-utilization Gaps by Physician Specialty and Urbanicity, Community-Income-per-Capita, and ADI. Values Reflect Changes in Care-utilization Gaps Between More and Less Disadvantaged Groups Between 2005 and 2023

Specialty	Percentage-point change in the care-utilization gap, 2005 to 2023
	Urbanicity gap	Community-income gap	ADI gap
All physicians	1.76	5.89	3.82
Primary care	13.32	10.34	12.71
Internal medicine	7.74	6.79	8.58
Family practice	12.73	5.51	9.51
Medical specialties	1.92	7.46	4.16
Cardiology	0.57	3.25	2.17
Hematology/oncology	0.96	1.12	1.23
Rheumatology	0.44	0.95	0.68
Surgical specialties	5.27	8.07	7.76
Ophthalmology	4.61	6.18	7.38
Orthopedics/sports	1.65	2.48	1.17
Urology	2.84	2.30	2.66
Hospital-based	1.60	6.14	4.76
Radiology	1.94	5.98	4.64
Pathology	3.18	5.13	4.16
Obstetrics/gynecology	0.20	-0.69	-0.42
Psychiatry	-0.22	0.38	-0.31
Any physician or NPP	0.84	4.37	2.29
NPP	3.25	2.38	-4.25

For 2005 and 2023, the care utilization gap is calculated as the difference in the percentage of beneficiaries with at least one healthcare claim from a physician/NPP in the noted specialties, with comparisons as follows: the urbanicity gap is for metropolitan versus small/rural; the community-income-per-capita gap is for ≥$50,000 versus <$25,000; the ADI gap is for the best versus worst ADI quartile. ADI, Area Deprivation Index; NPP, nonphysician practitioner.

Figure 3.

Projected change in the care-utilization gaps between more and less disadvantaged groups by urbanicity, community income per capita, and ADI. Legend: Change associated with the $30.17 decline in the inflation-adjusted conversion factor over the 2005-2025 period. Results are presented for physicians overall, broad specialty groups, any physician or NPP, NPPs, and for select specialties (largest care volume) within the broad specialty groups. The urbanicity gap is for metropolitan versus small/rural. The community-income-per-capita gap is for ≥$50,000 versus <$25,000. The ADI gap is for the best versus worst ADI quartile. ADI, Area Deprivation Index; CI, confidence interval; Ob/gyn, obstetrics/gynecology; NPP, non-physician practitioner; IM, internal medicine; FP, family practice; cardio, cardiology; hema/onc, hematology/oncology; rheum, rheumatology; ophth, ophthalmology; ortho/sports, orthopedics/sports medicine; radio, radiology; path, pathology

Income-based care-utilization gaps were larger than either urbanicity-based or ADI-based gaps (Figure 2). The comparison of lower-income versus higher-income communities found the care-utilization gap increased 218% from 2.70 to 8.59 percentage points (2005-2023). Statistically, a $1 decline in the inflation-adjusted conversion factor is associated with a 0.23 (95% CI: 0.14, 0.33) percentage-point increase in the care-utilization gap for low-income versus high-income communities (Supplemental Figure 1). Hence, for the $30.17 inflation-adjusted conversion factor decline, the associated increase in the community-income care-utilization gap was 7.00 (95% CI: 4.10, 9.91) percentage points (Figure 3).

ADI comparisons between the most-deprived versus least-deprived communities found care-utilization for the most-deprived communities was slightly higher than the least-deprived in 2005 (-0.59 percentage points). However, by 2023, an ADI care-utilization gap emerged (3.23 percentage points). The normalized, statistical result was a 0.18 (95% CI: 0.09, 0.26) percentage-point increase in the care-utilization gap per $1 decline in the inflation-adjusted conversion factor, and a total ADI care-utilization gap of 5.31 (95% CI: 2.76, 7.87) percentage points was associated with the $30.17 conversion factor decline.

In analyses focused on utilization of primary care physician specialties only, the care-utilization gap grew to a greater extent than for any other clinician category (Figures 3 and 4). Specifically, the normalized primary care physician-utilization gap decreased across urbanicity-based, income-based, and ADI-based groups (Figure 4). Specifically, a $1 decline in the inflation-adjusted conversion factor was associated with a 0.65 (95% CI: 0.56, 0.74), 0.50 (95% CI: 0.42, 0.58), and 0.63 (95% CI: 0.55, 0.71) percentage-point increase in the urbanicity, income, and ADI care-utilization gaps, respectively (Supplemental Figure 1). Hence, the $30.17 decline in the inflation-adjusted conversion factor was associated with 19.60 (95% CI: 16.83, 22.36), 15.11 (95% CI: 12.77, 17.44), and 18.91 (95% CI: 16.48, 21.34) percentage-point increases in the urbanicity, community-income, and ADI care utilization gaps, respectively (Figure 3). For family practice physicians specifically, the $30.17 decline in the inflation-adjusted conversion factor was associated with 21.31 (95% CI: 13.37, 29.26), 14.02 (95% CI: 7.12, 20.92), and 19.55 (95% CI: 11.86, 27.24) percentage-point increases in the urbanicity, community-income, and ADI care utilization gaps, respectively.

Figure 4.

Trends in use of primary care physician services by year, and conversion-factor-associated gaps in primary care. Legend: Panels A, C, and E show the percentage of Medicare beneficiaries with any primary care physician services by urbanicity, community-income-per-capita, and ADI. Panels B, D, and F show the actual care-utilization gaps and the care-utilization gap as predicted based on the inflation-adjusted conversion factor controlling for the population-to-primary care physician ratio. The gaps (solid blue lines) are normalized to reflect the same distribution of beneficiaries for all years. The modeled gaps (dash blue lines) are the linear prediction of the gap based on the inflation-adjusted conversion factor. ADI, Area Deprivation Index

The results are similar for other provider specialties. For example, the $30.17 inflation-adjusted conversion factor decline was associated with 5.05 (95% CI: 2.16, 7.93) and 5.05 (95% CI: 3.29, 6.82) percentage-point increases in the community-income and ADI care-utilization gaps, respectively, for medical specialties (Figure 3). For surgical specialties, these percentage-point increases were 3.72 (95% CI: 1.99, 5.45), 6.51 (95% CI: 4.00, 9.02), and 7.54 (95% CI: 5.65, 9.42) in the urbanicity, community-income, and ADI care-utilization gaps, respectively. With few exceptions, the statistical associations between the care-utilization gaps and the inflation-adjusted conversion factor are such that a decline in the inflation-adjusted conversion factor is associated with an increase in care-utilization gap (Supplemental Figure 1). These exceptions are no significant association for the medical specialties group (urbanicity gap), cardiology (urbanicity gap), hematology/oncology (urbanicity and community-income gaps), obstetrics/gynecology (urbanicity and ADI gaps), and physician/NPP (urbanicity gap). Only for obstetrics/gynecology (community-income gap; -0.08 percentage points) and psychiatry (urbanicity gap; -0.01 percentage points) was there a significant negative association between the inflation-adjusted conversion factor and the gaps, but the magnitudes of these associations were near zero.

Beneficiaries with care from NPPs rose substantially between 2005 and 2023, and care-utilization gaps grew as well. For example, normalized NPP care utilization rose from 17.9% to 57.5% for metropolitan areas compared with 15.1% to 51.5% for small/rural areas, reflecting a gap increase from 2.75 to 6.00 percentage points over this period (Supplemental Table 3). There was also an increase in the community-income care-utilization gap over the study period (5.52 to 7.90 percentage points). Conversely, by ADI, more deprived communities had higher NPP utilization than those less deprived, and this gap shrank by -4.25 percentage points. Normalized projections of changes in the care-utilization gaps for NPP care associated with the $30.17 decline in the inflation-adjusted conversion factor were 17.36 (95% CI: 9.42, 25.30), 16.06 (95% CI: 9.42, 22.70), and (16.82; 95% CI: 12.48, 21.17) percentage-point increases in the urbanicity, community-income, and ADI gaps, respectively (Figure 3).

Discussion

This study of nearly 33 million beneficiary-years (2005-2023) found that the care-utilization rate for any physician care decreased over the study period. Decreases were greater for beneficiaries in small/rural versus metropolitan areas, lower-income versus higher-income communities, and worst-ADI (most-deprived) versus best-ADI (least-deprived) communities. Hence, each of these gaps widened over the study period. We found that these gaps grew as the inflation-adjusted conversion factor declined. Between 2005 and 2025, the inflation-adjusted conversion factor declined $30.17 or 48.3% and this decline was associated with a 2.46 to 7.00 percentage-point increase in the care-utilization gap across the three community comparisons for physician care.

Various socioeconomic factors have been associated with differences in utilization of care,^32-35 which is a commonly used proxy for access to care.^41,42 Specifically, a review study found that rural populations have less care utilization than urban populations. ³² This is evident in our results as the urbanicity gap for any physician care increased 1.76 percentage points; there were larger increases for primary care physician specialties (13.32) and surgical specialties (5.27). The Agency for Healthcare Research and Quality found that poor and low-income households had worse access for 79% and 71% of access measures than high-income households, respectively, with no instances of better access for any measure. ³³ Our results quantified significant income care-utilization gaps and found that these gaps increased as the conversion factor decreased for all clinicians groups except obstetrics/gynecology physicians, but we note that this is a statistical association not causation. As for ADI, a review study found more disadvantaged areas had reduced primary care access. ³⁵ This is evident in our results with increasing ADI care-utilization gaps associated with conversion factor declines.

The Medicare Payment Advisory Commission (MedPAC) found that, based on its surveys, Medicare beneficiaries reported adequate access to care (with some beneficiaries experiencing difficulty finding a new primary care physician) and that this access is comparable to privately insured patients. ⁴³ MedPAC’s conclusion is a general statement, which on the surface, appears to conflict with our results, but it does not for several reasons. First, healthcare utilization surveys generally report higher utilization that actually occurs. ⁴⁴ Second, MedPAC reports utilization for Medicare beneficiaries in aggregate and compares them externally to privately insured patients. Our results are internal comparisons among groups of Medicare beneficiaries. Third, we show changes in access over time, while MedPAC does not.

Various studies have shown the impact of inflation-adjusted conversion factor declines on physician reimbursement.^1-11 In addition to these physician impacts, our study observed that conversion factor declines are associated with reduced access for patients. Various studies have found associations between physician reimbursement and access.^21-24,45-48 Other studies have found indirect associations between physician reimbursement and access based on the reimbursement differences across insurance types,^{17,25-28,30,31} particularly with the ability to get an appointment.^16-19 While our results show care-utilization declines were widespread across physician specialties consistent with the literature on reimbursement and access, of particular interest was that these declines were greater among more disadvantaged groups yielding a widening care-utilization gap, which was associated with declines in the inflation-adjusted conversion factor. Hence, by inference, there is a statistical association between declining Medicare reimbursement and worsening health equity among Medicare fee-for-service beneficiaries.

Surprisingly, the increases in care-utilization gaps were particularly acute for primary care physician services with 10.34 to 13.32 percentage-point increases in the gaps. Given primary care’s importance, in 2009 and 2021, CMS increased RVUs for evaluation and management with the intent to bolster primary care.^1,36 These evaluation-and-management increases resulted in conversion factor declines to maintain budget neutrality within the MFPS, which affected all specialties including primary care. As many NPPs are primary care providers, it is possible that some of the growing care-utilization gaps for primary care physicians is offset by growth in the number of NPPs, which grew 96% between 2013 and 2022 and are expected to continue grow rapidly (40% for nurse practitioners and 28% for physician assistants between 2023 and 2033). ⁴⁹ However, our results show that despite large increases in the number of Medicare beneficiaries receiving care from NPPs, community-income and ADI care-utilization gaps increased for any physicians or NPP care as well as for NPP-specific urbanicity and community-income care-utilization gaps.

The associations between reimbursement and care-utilization found in this study suggest that Medicare reimbursement reductions have consequences for patients, particularly given the ubiquitous nature of economic tradeoffs in a resource constrained environment. The results of this study show that the costs of these tradeoffs have been disproportionately borne by beneficiaries of lower socioeconomic status as declines in the inflation-adjusted conversion factor are associated with increased care-utilization gaps, which imply reduced health equity. Indeed, a study of reimbursement and patient retention in human immunodeficiency viruses found that an increase in Medicaid reimbursement was associated with increased retention of non-Hispanic Black patients with the authors concluding that increased reimbursement may foster health equity. ⁵⁰ Another study of Medicare reimbursement declines for surgical care concluded that this approach to controlling rising expenditures may threaten surgical-care access. ⁵¹ Our finding of conversion factor-associated increases in care-utilization gaps for surgical specialties suggests that access to surgical care has been affected by conversion factor declines.

Given the association between worsening access disparities among patients and the declining conversion factor, policies for mitigating these disparities need to be considered. Likely, eliminating budget neutrality is unrealistic. Short of that, one option is to begin indexing the conversion factor for inflation as well as adjusting total Medicare Part B funding for changes in the Medicare population such that there is a per beneficiary per year amount that is statutorily indexed for inflation. The Omnibus Budget Reconciliation Act of 1989 (enacted in 1992), which included budget neutrality, required that any change to the MPFS that increased expenditures more than a $20 million would trigger a compensating change in the conversion factor. ⁵² This trigger is not adjusted for inflation, but if it were, it would be $47.3 million in 2026 dollars. Other options include paying for disposable supplies and advanced equipment from the Hospital Outpatient Prospective Payment System and the Ambulatory Surgical Center System rather than the MPFS and redirecting savings from the Merit-based Incentive Payment System back into the MFPS.

This study has limitations. First, while the study established a significant correlation between the declining inflation-adjusted conversion factor and normalized population care-utilization differences, correlation is not causation. Second, other factors, such as COVID, the implementation of the Affordable Care Act and Medicare Access and CHIP Reauthorization Act, population growth, growth in the number of physicians/NPPs, and the Geographic Practice Cost Index, may also impact utilization. These were accounted for using year effects. Third, utilization is not the same as access, but it is a common proxy for access.^41,42 Fourth, this study focused on Medicare fee-for-service beneficiaries whose care is reimbursed through the MPFS. Hence, the results may differ for Medicare Advantage patients. Fifth, Medicare data designates NPPs but not the specific physician specialty that they support. Hence, while many NPPs work in primary care, we could not compute care-utilization gaps specific to primary care that included primary care physicians and NPPs working in primary care. Finally, the community-income and ADI measures are based on beneficiary ZIP code. We could not control for income and deprivation measures on a more granular basis.

Conclusion

We found increasing care-utilization gaps by urbanicity, community-income, and ADI. The size of these gaps was temporally associated with declines in the inflation-adjusted conversion factor such that decreases in it were statistically associated with a widening gap. The widening of these gaps was substantial given large decrease in the inflation-adjusted conversion factor over the last two decades. Hence, decreased physician payments resulting from the Medicare budget neutrality mandate may exacerbate disparities in access to care for traditionally underserved Americans. Policy makers should change the mechanisms that have led to the substantial inflation-adjusted conversion factor declines to mitigate these disparities.

ORCID iD

Eric W. Christensen https://orcid.org/0000-0003-1377-0787

Ethical Considerations

NA. Study was based on CMS data for deidentified beneficiaries.

Consent to Participate

NA. Study was based on CMS data for deidentified beneficiaries.

Data Availability Statement

The data using in this study are under a data use agreement with CMS and cannot be shared.*