Growth of ambulatory surgery centers more likely in higher-resourced counties with lower deprivation, 2014-2021

Introduction

The United States’ healthcare system is increasingly seeking to optimize the site of care for surgery by shifting procedures to ambulatory surgical centers (ASCs). Empirical data are needed to understand the impact of these trends on historically underserved communities. Financial incentives may lead to ASC openings in more affluent areas as hospitals seek to increase their surgical volume and profitability.^1,2 The objectives of this study were to characterize trends in the ASC industry and to identify market-level characteristics associated with ASC openings.

Methods

We performed a cross-sectional study of county-level data derived from the Area Health Resources File (AHRF) in 2019. County-level characteristics obtained from AHRF are derived from different years of census surveys and the American Community Survey.³ The most recent year available was chosen for each variable in the AHRF. We also estimated county-level Area Deprivation Index (ADI) using population-weighted means for ADI block group data from the Neighborhood Atlas in 2019.⁴

The primary outcome of interest was opening an ASC within a given county between 2014 and 2021. Using the 20% sample of Medicare fee-for-service (FFS) beneficiaries’ Carrier Files from 2011 to 2021, we identified ASCs that received Medicare reimbursement each year. We defined an ASC opening as one that occurred when an ASC did not submit charges in the three prior years. We then created a flag for whether an ASC opened in a given county during the study period. This conservative definition of ASC opening limited the study period of openings to 2014 to 2021. The flag for ASC opening was merged with the analytic sample that included county-level characteristics for 3230 counties in 2019.

We then constructed a multivariable logistic regression model to assess the county-level predictors of an ASC opening. Our final model included a flag for the presence of an ASC in the county for the year prior to an opening. As a sensitivity analysis, we compared different thresholds for the definition of an ASC opening and our findings were unchanged. All P-values were two-sided, and 0.05 was set as the threshold for significance. Statistical analyses were performed using SAS version 9.4 (SAS Institute Inc, Cary, NC, USA). This study was reviewed by the Harvard T.H. Chan School of Public Health Institutional Review Board and deemed exempt due to the use of de-identified data.

Results

The total number of annual ASCs identified in our sample increased from 5012 in 2011 to 5293 in 2021 (P < 0.001 for trend). In the multivariable model, the adjusted rate of ASC openings was higher among counties with an existing ASC (44.7% vs 13.8%, P < 0.001) (Table 1) (Full model results in Appendix Table). Counties from the lowest ADI were more likely to open an ASC when compared with counties from the highest quartile (31.9% vs 19.9%, P < 0.001). Counties with the highest quartile of black beneficiaries were more likely to open an ASC than counties from the lowest quartile (28.7% vs 22.0%, P =0.003). Counties from the highest quartile of physicians per capita were more likely to open an ASC than counties from the lowest quartile (31.4% vs 21.8%, P < 0.001). Counties from the highest quartile for average number of hospital beds were more likely to open an ASC than counties from the lowest quartile (32.6% vs 20.4%, P < 0.001).

Table 1.

Adjusted rates of new ASC opening by market characteristics at the county level, 2014-2021.

		Adjusted rate of opening new ASC
	Odds ratio	%	Delta	P-value	Global P-value
Existing ASC
No	REF	13.8%	REF	—	<0.001
Yes	5.95	44.7%	30.9%	<0.001	—
Area Deprivation Index
Lowest Quartile	REF	31.9%	REF	—	<0.001
2nd Quartile	0.68	25.1%	−6.8%	0.002
3rd Quartile	0.55	22.7%	−9.2%	<0.001	—
Highest Quartile	0.42	19.9%	−12.0%	<0.001	—
Rurality
Urban	REF	25.4%	REF	—	0.351
Rural	0.69	23.8%	−1.6%	0.351	—
% Black
Lowest Quartile	REF	22.0%	REF	—	0.010
2nd Quartile	1.15	22.8%	0.8%	0.689	—
3rd Quartile	1.11	26.2%	4.2%	0.045	—
Highest Quartile	1.24	28.7%	6.7%	0.003	—
Average Number Operating Rooms
Lowest Quartile	REF	24.8%	REF	—	0.521
2nd Quartile	1.07	26.7%	2.0%	0.323	—
3rd Quartile	1.28	24.0%	−0.8%	0.724	—
Highest Quartile	2.06	24.1%	−0.7%	0.838	—
Number MDs per capita
Lowest Quartile	REF	21.8%	REF	—	<0.007
2nd Quartile	1.26	21.4%	−0.3%	0.917	—
3rd Quartile	1.02	24.2%	2.5%	0.214	—
Highest Quartile	1.00	31.4%	9.6%	<0.001	—
Average Number Hospital Beds
Lowest Quartile	REF	20.4%	REF	—	0.001
2nd Quartile	0.93	21.6%	1.2%	0.580	—
3rd Quartile	1.21	25.1%	4.7%	0.051	—
Highest Quartile	1.73	32.6%	12.3%	<0.001	—

Models adjusted for variables listed above and % persons below poverty level, median household income, Persons Under Age 25 With 4 or More Years College, Average Number Medicare/Medicaid Dually Eligible, Medicare Advantage Penetration, Population Over Age 65, % White, % Hispanic, Cardiovascular disease per 10 000, Pulmonary disease per 10 000, Total Number Gastroenterology Surgeons per 10 000, Total Number General Surgeons per 10 000, Total Number Neurological Surgeons per 10 000, Total Number Ophthalmology, Primary Care, Hosp Full-Time Staff per 10 000, Total Number Orthopedic Surgeons per 10 000, Total Number Skilled Nursing Facility Beds per 10 000, Number General Practice, Primary Care, Hospital Full-Time Staff per 10 000.

Source: Authors’ analysis of data from a 20% sample of Medicare fee-for-service (FFS) beneficiaries’ Carrier Files from 2011 to 2021, the Area Health Resources File (AHRF), and Area Deprivation Index (ADI) data. Complete model results included in Appendix.

Discussion

ASC openings were more likely to occur in counties with greater existing healthcare resources and higher socioeconomic advantage. These findings suggest that the benefits of ASC growth may not be shared equitably.

ASC growth has historically been less regulated than acute care hospitals, and there are strong financial incentives to prioritize ambulatory surgery for low-risk and more profitable case-mix.⁵ These trends may be further compounded as the number of surgical procedures deemed appropriate by CMS for outpatient surgery and “ASC-eligible” continues to grow annually and CMS recently proposed eliminating the inpatient-only list for surgical procedures.^6,7

Approximately half of the US states do not require a “certificate of need” for ASC openings.⁸ Whether this regulatory landscape contributed to patterns in ASC growth is an area of future inquiry with a potential policy-focused solution that may control healthcare costs while ensuring that new healthcare facilities meet community needs.

Our study was limited to Medicare FFS beneficiaries and may not capture the full scope of care delivered at ASCs. Additionally, we were not able to verify ASC openings through further investigation. Our definition of ASC opening is likely an overestimation, though the extent of this is unclear. Finally, our county-level analysis does not capture within-county heterogeneity and dynamics that may affect decisions to open ASCs or access to ambulatory surgical care.

Supplementary material

Supplementary material is available at Health Affairs Scholar online.

Funding

Dr. T.C.T. reported receiving grants from Arnold Ventures, the National Center for Advancing Translational Sciences, National Institutes of Health to Harvard Catalyst, the Harvard Clinical and Translational Science Center, and financial contributions from Harvard University and its affiliated academic healthcare centers.