Abstract
Objective
To examine the relationship of physician versus hospital ownership of small‐ and medium‐sized practices with spending and utilization of care.
Data Source/Study Setting/Data Collection
Survey data for 1,045 primary care‐based practices of 1–19 physicians linked to Medicare claims data for 2008 for 282,372 beneficiaries attributed to the 3,010 physicians in these practices.
Study Design
We used generalized linear models to estimate the associations between practice characteristics and outcomes (emergency department visits, index admissions, readmissions, and spending).
Principal Findings
Beneficiaries linked to hospital‐owned practices had 7.3 percent more emergency department visits and 6.4 percent higher total spending compared to beneficiaries linked to physician‐owned practices.
Conclusions
Physician practices are increasingly being purchased by hospitals. This may result in higher total spending on care.
Keywords: Ambulatory/outpatient care, health care organizations and systems, ownership/governance
Health care reform appears to be accelerating two decades‐old trends: physician practices are increasingly owned by hospitals, and physicians in small practices are moving into medium‐sized and large practices (Burns, Goldsmith, and Sen 2013; Casalino 2014). Between 2013 and 2015, the percentage of physicians in practices with 1–24 physicians (i.e., small‐ and medium‐sized practices) decreased from approximately 48 percent to approximately 42 percent; the percentages were similar for primary care physicians (Muhlestein and Smith 2016). Estimates of the percentage of physicians employed by hospitals vary (Burns, Goldsmith, and Sen 2013; Kane 2015); a recent study suggested that the percentage increased from 29 to 33 percent between 2012 and 2014 (Physicians Advocacy Institute 2016). Recent policy developments—such as the Medicare Access and CHIP Reauthorization Act (MACRA)—are likely to further accelerate these trends (Squires and Blumenthal 2016). Little is known about the performance of hospital‐owned practices or about the performance of small‐ and medium‐sized practices (Casalino 2006; Casalino et al. 2014).
To address these questions, we linked Medicare claims data to a large, unique dataset—the National Study of Small‐ and Medium‐Sized Physician Practices (NSSMPP)—one of the largest surveys of medical practices ever conducted in the United States. In a previous article using this data, we found that practices with 1–2 physicians had 33 percent fewer preventable admissions than practices with 10–19 physicians, that practices with 3–9 physicians had 27 percent fewer preventable admissions than 10–19 physician practices, that preventable admission rates for 1–2 and 3–9 physician practices were as good or better than the national average for practices of all sizes, and that hospital‐owned small‐ and medium‐sized practices had higher rates of preventable admissions than physician‐owned practices (Casalino et al. 2014). In this article, we analyze the relationship between practice size and ownership and measures of utilization and Medicare spending.
Data and Methods
We linked survey data from the NSSMPP to Medicare claims data. NSSMPP has been previously described (Rittenhouse et al. 2011). Briefly, NSSMPP involved 40‐minute telephone surveys, conducted between July 2007 and March 2009, with the physician leader or administrator of a national sample of practices with 1–19 physicians. One thousand, seven hundred forty‐five practices responded; the adjusted response rate was 63.2 percent (Appendix S1). For this article, we analyzed data for the 3,010 physicians in the 1,045 practices that included at least 33 percent primary care physicians, were not community health centers, had patients attributed to them in 2008, and were not academic faculty practices (Appendix S1). We attributed 282,379 Medicare beneficiaries to these physicians, using claims data for 2008 from a national sample of 999,990 Medicare beneficiaries who during that year had seen a physician operating in a NSSMPP practice at least once, were enrolled in Part A and B coverage continuously, were not in the End Stage Renal Disease Program or Medicare Advantage, and survived throughout the year. We attributed each beneficiary to the physician with whom the beneficiary had a plurality of outpatient evaluation and management visits, with tie‐breakers as described in Appendix S2.
Our objective in this study was to examine the influence of outpatient ownership and size on beneficiary spending and utilization information. As such, our outcome variables included three utilization measures and five spending measures. The utilization measures were emergency room visits, hospital index admissions (all hospital admissions not including readmissions), and the readmission rate for beneficiaries with at least one hospitalization. The spending measures included hospital inpatient spending, hospital outpatient spending, postacute care spending, physician service spending, other spending (e.g., ambulatory surgery centers, outpatient imaging, and durable medical equipment), and total spending (the sum of all spending categories). These utilization and spending measures were constructed using beneficiary‐level data that were aggregated by CMS and provided in the 2008 Master Beneficiary Summary File. We geographically adjusted the spending measures as described in Appendix S3.
In our regression models, our primary independent variables of interest are group size (1–2 physicians, 3–9 physicians, and 10–19 physicians) and ownership (hospital‐owned or physician‐owned). Of our 1,045 practices, 871 were physician‐owned, and of these, 37.3 percent were 3–9 physician practices and 4.4 percent were 10–19 physician practices. The 174 hospital‐owned practices were larger on average (p < .001), with 55.8 percent having 3–9 physicians and 8.6 percent having 10–19 physicians.
Additional control variables are described in table footnotes and in Appendix S4.
Given the count nature of our dependent variables, we estimated our regressions using a generalized linear model with a log‐link and a Poisson count nature as chosen (separately for total spending and each utilization measure) by a modified Park test (Buntin and Zaslavsky 2004). Associations are shown as the percent change in the outcome. Survey weights are used in all analyses, and standard errors are clustered at the level of the primary sampling unit.
The study was approved by the Institutional Review Board at Weill Cornell Medical College. Analyses were performed using Stata 14.1.
Results
Total spending per beneficiary/year was $7,514, of which 38 percent was spent on hospital inpatient care, 14 percent on hospital outpatient care, 24 percent on physician services, 14 percent on postacute care, and 14 percent on other spending (Table 1). In part because the beneficiaries in our study survived all of 2008, average beneficiary/year spending of $7,514 is lower than average standardized spending of $8,359 in 2008 (CMS, 2017). Each beneficiary had, on average, 0.48 emergency room visits per year and 0.24 index admissions per year. The 30‐day readmission rate among beneficiaries with an index admission was 11.2 percent, which is lower than the national average and may also reflect our healthier population on average.
Table 1.
Descriptive Statistics, N = 282,372
| Mean | Standard Deviation | |
|---|---|---|
| Outcomes | ||
| Total spending | $7,514 | $13,271 |
| Hospital inpatient spending, % of Total Spending | $2,532, 33.7% | $7,675 |
| Hospital outpatient spending, % of Total Spending | $1,017, 13.5% | $2,467 |
| Postacute care spending, % of Total Spending | $1,064, 14.2% | $4,553 |
| Other spending, % of Total Spending | $1,077, 14.3% | $1,410 |
| Physician services spending, % of Total Spending | $1,823, 24.3% | $2,517 |
| Emergency room visits | 0.480 | 1.006 |
| Hospital index admissions | 0.242 | 0.544 |
| Readmission rate* | 11.2% | 0.245 |
| Practice characteristics | ||
| Physician‐owned | 85.0% | — |
| Hospital‐owned | 15.0% | — |
| 1–2 physician practice | 24.9% | — |
| 3–9 physician practice | 66.3% | — |
| 10–19 physician practice | 8.8% | — |
| Percent of primary care physicians | 94.9% | 0.121 |
| Patient‐centered medical home score Quartile 1 (0–11.76) | 28.2% | — |
| Patient‐centered medical home score Quartile 2 (>11.76–23.53) | 30.8% | — |
| Patient‐centered medical home score Quartile 3 (>23.53–37.5) | 21.2% | — |
| Patient‐centered medical home score Quartile 4 (>37.5–100) | 19.9% | — |
| Beneficiary characteristics | ||
| Number of chronic conditions (0–25) | 5.976 | 2.991 |
| Beneficiary age | 76.645 | 6.854 |
| Beneficiary sex, female | 61.2% | — |
| Dual eligible | 7.9% | — |
| White (non‐Hispanic) | 90.4% | — |
| Black (of African American) | 5.4% | — |
| Hispanic | 2.0% | — |
| Asian/Pacific Islander | 1.5% | — |
| American Indian/Alaska Native/Other/Unknown | 0.7% | — |
| Original reason for entitlement—old age and survivors insurance | 93.3% | — |
| Original reason for entitlement—disability insurance benefits (DIB) | 6.7% | — |
| Original reason for entitlement—end‐stage renal disease (ESRD) | 0.02% | — |
| Original reason for entitlement—both DIB and ESRD | 0.01% | — |
| Environment characteristics | ||
| Large central metro county | 14.7% | — |
| Large fringe metro county | 24.8% | — |
| Medium metro county | 24.9% | — |
| Small metro county | 13.0% | — |
| Micropolitan county | 14.2% | — |
| Noncore county | 8.3% | — |
| Zip code income <$20k | 0.3% | — |
| Zip code income [$20k to $30k) | 9.7% | — |
| Zip code income [$30k to $40k) | 31.5% | — |
| Zip code income [$40k to $50k) | 24.0% | — |
| Zip code income [$50k to $60k) | 14.0% | — |
| Zip code income [$60k to $70k) | 7.8% | — |
| Zip code income [$70k to $80k) | 5.1% | — |
| Zip code income >$90k | 4.4% | — |
| Zip code income missing | 3.0% | — |
| Survey period | ||
| July 1–December 31, 2007 | 3.5% | — |
| January 1–June 30, 2008 | 2.2% | — |
| July 1–December 31, 2008 | 66.8% | — |
| January 1–March 31, 2009 | 27.5% | — |
Authors’ analysis of NSPO survey data linked to Medicare claims data.
*The readmission rate is calculated for only individuals with an index hospitalization (N = 53,256).
Our sample was also older than the mean FFS beneficiary in year 2008 (76.6 compared to 72.0), was more likely to be female (61.2 percent compared to 55.9 percent), and was more likely to be white, non‐Hispanic (90.4 percent compared to 81.7 percent). Among other things, these differences may reflect our exclusion of end‐stage renal disease beneficiaries and requirement that the individual received outpatient care (Centers for Medicare & Medicaid Services 2017). Given that important differences exist between our sample and the national FFS sample, our results should not be construed as being applicable for all Medicare FFS beneficiaries.
Eighty‐five percent of practices were physician owned (Table 1). Nearly 25 percent of practices were 1–2 physician practices; 66.3 percent had 3–9 physicians; 8.8 percent had 10–19 physicians. Ninety‐five percent of physicians in the practices were primary care physicians (although practices were included in our analyses if they included at least 33 percent primary care physicians, most practices turned out to be all primary care).
Table 2 provides the associations between practice ownership and size and utilization measures. Beneficiaries attributed to hospital‐owned practices had 7.3 percent more emergency department visits per beneficiary/year than beneficiaries attributed to physician‐owned practices (p = .055). Index admissions and readmission rates did not differ significantly between hospital‐owned and physician‐owned practices. Larger practices had statistically significantly higher rates of ED visits than 1–2 physician practices: 6.9 percent higher in 3–9 physician practices and 13.5 percent higher in 10–19 physician practices. However, hospital readmission rates were lower in larger practices: Beneficiaries attributed to 3–9 physician practices had approximately 15 percent lower readmissions (p < .01), and 10–19 physician practices each had approximately 10 percent lower readmissions per beneficiary/year compared to 1–2 physician practices, although this latter association was imprecisely estimated and is not statistically significant.
Table 2.
Percent Change in Hospital Utilization Associated with Physician Practice Characteristics
| (1) | (2) | (3) | |
|---|---|---|---|
| Emergency Room Visits | Hospital Index Admissions | Readmission Rate | |
| Owned by a hospital | 7.3 | −0.8 | 4.1 |
| [3.8] | [2.7] | [4.0] | |
| 3–9 physicians | 6.9* | 6.2* | −14.9** |
| [3.1] | [2.8] | [5.4] | |
| 10–19 physicians | 13.5* | 19.7* | −10.2 |
| [3.6] | [1.5] | [8.8] | |
| Subpopulation | 282,379 | 282,379 | 53,256 |
| Mean | 0.480 | 0.242 | 0.112 |
| SD | 1.005 | 0.542 | 0.243 |
Authors’ analysis of NSPO survey data linked to Medicare claims data. Each column presents the percent change for the category compared to the reference. Each column is estimated using a separate regression. All equations adjust for the percentage of primary care physicians in each practice, PCMH quartile, HRR, urbanicity, ZIP code‐level income, survey period, 26 comorbidities diagnosed prior to 2008, beneficiary age, race/ethnicity, gender, dual‐eligibility, and original reason for entitlement. Results can be interpreted as the difference per beneficiary/year compared to reference categories of physician‐owned, 1–2 physicians. Standard errors are in brackets.
*p < .05, **p < .01, ***p < .001.
Table 3 provides the associations between practice ownership and size with spending. Total spending per Medicare beneficiary per year for hospital‐owned practices was 6.4 percent higher (p < .05). Hospital‐owned practices had significantly higher spending on hospital outpatient services (35.7 percent) and postacute care services (17.1 percent), but lower spending on physician services (−8.3 percent). Three to nine physician practices had 7.7 percent higher total spending compared to 1–2 physician practices, mainly because of higher hospital outpatient spending, but 10–19 physician practices had only 2.0 percent higher spending compared to 1–2 physician practices (not statistically significant). Within these large practices, there was substantially more spending on hospital inpatient and outpatient care and less spending on all other categories.
Table 3.
Percent Change in Spending Associated with Physician Practice Characteristics
| (1) | (2) | (3) | (4) | (5) | (6) | |
|---|---|---|---|---|---|---|
| Total Spending | Hospital Inpatient Spending | Hospital Outpatient Spending | Postacute Care Spending | Other Spending | Physician Services Spending | |
| Owned by a hospital | 6.4* | −2.1 | 35.7* | 17.1* | 1.2 | −8.3* |
| [1.4] | [4.0] | [7.5] | [7.0] | [3.8] | [2.0] | |
| 3–9 physicians | 7.7* | 6.3* | 23.4* | 1.5 | 6.2 | 3.0 |
| [1.6] | [2.7] | [3.9] | [3.4] | [3.3] | [2.4] | |
| 10–19 physicians | 2.0 | 12.7** | 27.3* | −11.8** | −13.4** | −12.0* |
| [2.5] | [4.1] | [6.6] | [4.4] | [5.1] | [3.3] | |
| Mean | 7,514 | 2,532 | 1,017 | 1,064 | 1,077 | 1,823 |
| SD | 13,240 | 7,656 | 2,461 | 4,542 | 1,407 | 2,511 |
Authors’ analysis of NSPO survey data linked to Medicare claims data. Each column presents the percent change for the category compared to the reference. Each column is estimated using a separate regression. All equations adjust for the percentage of primary care physicians in each practice, PCMH quartile, HRR, urbanicity, ZIP code‐level income, survey period, 26 comorbidities diagnosed prior to 2008, beneficiary age, race/ethnicity, gender, dual‐eligibility, and original reason for entitlement. Results can be interpreted as the difference per beneficiary/year compared to reference categories of physician‐owned, 1–2 physicians. Standard errors are in brackets.
*p < .05, **p < .01, ***p < .001.
Full results for Tables 2 and 3 including coefficients for control variables are available online (Online Exhibit S1 and S2).
Discussion
Our study, which includes a very large national sample of practices, is the first to compare utilization and total spending for hospital‐owned versus physician‐owned small‐ and medium‐sized primary care‐based practices. Approximately 42 percent of physicians work in these practices. Research to date suggests that physicians in small‐ and medium‐sized practices provide care that equals or exceeds the quality and spending performance of larger practices (Landon et al. 2008). We find that total spending per beneficiary per year was 6.4 percent ($481) higher in hospital‐owned practices compared to physician‐owned practices. This higher spending resulted primarily from higher spending on hospital services and on postacute care; spending for physician services was significantly lower in hospital‐owned practices. Rates of hospital admissions and readmissions were not significantly higher for hospital‐owned practices, so our results suggest that higher hospital spending is driven by greater intensity of hospital outpatient services (e.g., diagnostic imaging and outpatient surgery), rather than from an increased frequency of admissions of beneficiaries to the hospital.
Total spending did not differ consistently by practice size, except that 3–9 physician practices had 7.7 percent higher total spending ($579) per beneficiary per year compared to 1–2 physician practices, mainly because of higher spending on hospital services. The reason for this finding is not clear. Utilization of services did not vary consistently by group size: 10–19 physician practices had much higher rates of ED visits and hospital admissions compared to 1–2 physician practices, but much lower readmission rates.
Our findings add to—and are generally consistent with—the small body of literature on the comparative performance of hospital‐owned versus physician‐owned practices. Other studies, in a variety of settings, suggest that hospital ownership of practices is associated with higher total spending per patient (McWilliams et al. 2013; Baker, Bundorf, and Kessler 2014; Robinson and Miller 2014; Capps, Dranove, and Ody 2015; Neprash et al. 2015).
Given the rapid movement of physicians into hospital employment and into larger practices and the accelerating impact that key health reform policies are likely having on these two trends, it is important for researchers and policy makers to note that hospital‐owned practices and larger practices (at least in the range up to 19 physicians) do not necessarily perform better, or even as well, as physician‐owned practices and smaller practices (Landon et al. 2008; Weeks et al. 2010; McWilliams et al. 2013; Carlin, Dowd, and Feldman 2014; Kralewski et al. 2015). Hospital‐owned practices and larger practices have more resources—capital, information technology, and leaders whose time can be dedicated to finding ways to improve quality—to systematically improve the care they provide (Casalino et al. 2003, 2013, 2016; Friedberg et al. 2007). It is possible that these potential advantages do not outweigh—at least not yet—the potential advantages of small practices, for example, the strong mutual knowledge and trust that can develop among patients, staff, and physicians in the small practice setting.
Our study is subject to at least five limitations. First, although it is based on a very large, randomly selected sample of small‐ and medium‐sized practices weighted to be representative of the United States, we cannot claim that it is a precisely representative sample, because no “gold standard” dataset exists that contains the population of U.S. physician practices. Second, by design, our study does not include practices that are primarily composed of specialist physicians or practices that include 20 or more physicians. However, approximately 42 percent of physicians still work in practices with fewer than 20 physicians (Muhlestein and Smith 2016). Third, our data on practice characteristics and incentives rely on the accuracy of self‐report from a single leader of each practice. Fourth, our study is observational and demonstrates only an association, not necessarily a causal relationship, between practice characteristics and utilization and spending. Fifth, our analysis was performed using data from 2008, which may reduce the applicability of our results to today's rapidly changing health care delivery environment.
Conclusion
Our findings suggest four implications:
First, it may be worthwhile for researchers and funders of research to give more attention to comparing the performance of physician practices that differ from each other in size and ownership.
Second, it would be worthwhile for policy makers to give additional consideration to the likely effects—intended and unintended—of policies (such as Meaningful Use, MACRA, and paying higher rates for services in hospital‐owned facilities than for the same services in physician‐owned practices [GAO 2015]) on the types of practices that survive. MACRA, for example, is very complex, will eventually require reporting from practices to CMS on numerous measures, and offers a guaranteed bonus to physicians in “advanced alternative payment models” while threatening physicians not in these models with large penalties (up to 9 percent of Medicare revenues). Faced with this complexity and risk, many physicians in small and medium‐sized practices may decide it is safer and easier to seek hospital employment (Casalino 2017; Schneider and Hall 2017).
From a policy point of view, this might be a good thing (even if it is not what some physicians and their patients prefer), if hospital‐owned or large practices provide better care, but this is exactly where more research is needed. This does not mean that such policies should not be created, but it does mean that they should be structured, if possible, in such a way that they do not, practically speaking, mean that certain types of practice are likely to prevail and others to disappear regardless of the cost or quality of care they provide. Many initiatives of the Center for Medicare and Medicaid Innovation (CMMI) show awareness of this issue; in addition, the recently released final MACRA rules show that policy makers are aware of this issue and try to issue rules that can accommodate small practices. Nevertheless, the overall effect of MACRA may well be to drive physicians into larger organizations (Squires and Blumenthal 2016).
Third, policy makers might consider increasingly vigorous enforcement of antitrust laws, including more intensive scrutiny of the effects of vertical integration (hospital acquisition of physician practices) on the quality and cost of care (Gaynor, Ho, and Town 2015; Greaney and Ross 2016). Hospitals and medical groups large enough to negotiate much higher payment rates from health insurers (Vladeck and Rice 2009; Berenson et al. 2012; Reschovsky and Rich 2015) can drive smaller hospitals and groups out of business regardless of their relative performance on quality and cost. This can occur because higher payment rates can provide much more revenue than the comparatively small amounts of revenue that can be gained by scoring well on performance measures (Berenson, Ginsburg, and Kemper 2010).
Finally, hospital and physician leaders might try to find ways to combine the advantages of large organizational size with the advantages that can be provided by the human scale environment of small‐ and medium‐sized practices (Mostashari 2016).
Supporting information
Appendix SA1: Author Matrix.
Appendix S1: The Practice Sample.
Appendix S2: Attributing Beneficiaries to Physicians and Practices.
Appendix S3: Spending Categories and Standardization.
Appendix S4: Control Variables, including Measurement of Medical Home Processes.
Exhibit S1: Association between Hospital Utilization and Outpatient Practice Characteristics, Full Regression Results.
Exhibit S2: Association between Spending and Outpatient Practice Characteristics, Full Regression Results.
Acknowledgments
Joint Acknowledgment/Disclosure Statement: The Commonwealth Fund provided financial support for obtaining Medicare data and for conducting the analyses and writing the article. The Robert Wood Johnson Foundation provided support for the National Survey of Small and Medium‐Sized Physician Practices.
Disclosures: None.
Disclaimers: The statements, findings, conclusions, views, and opinions contained and expressed in this article are based in part on data obtained under license from the following IMS Health information services: Healthcare Organizational Services, (2007) IMS Health Inc. All rights reserved. The statements, findings, conclusions, views, and opinions contained and expressed herein are not necessarily those of IMS Health Inc. or any of its affiliated or subsidiary entities.
References
- Baker, L. C. , Bundorf M. K., and Kessler D. P.. 2014. “Vertical Integration: Hospital Ownership of Physician Practices Is Associated with Higher Prices and Spending.” Health Affairs (Millwood) 33 (5): 756–63. [DOI] [PubMed] [Google Scholar]
- Berenson, R. A. , Ginsburg P. B., and Kemper N.. 2010. “Unchecked Provider Clout in California Foreshadows Challenges to Health Reform.” Health Affairs 29 (4): 699–705. [DOI] [PubMed] [Google Scholar]
- Berenson, R. A. , Ginsburg P. B., Christianson J. B., and Yee T.. 2012. “The Growing Power of Some Providers to Win Steep Payment Increases from Insurers Suggests Policy Remedies May Be Needed.” Health Affairs (Millwood) 31 (5): 973–81. [DOI] [PubMed] [Google Scholar]
- Buntin, M. B. , and Zaslavsky A. M.. 2004. “Too Much Ado about Two‐Part Models and Transformation? Comparing Methods of Modeling Medicare Expenditures.” Journal of Health Economics 23 (3): 525–42. [DOI] [PubMed] [Google Scholar]
- Burns, L. R. , Goldsmith J. C., and Sen A.. 2013. “Horizontal and Vertical Integration of Physicians: A Tale of Two Tails” In Annual Review of Health Care Managmenet: Revisiting the Evolution of Health Systems Organization, edited by Friedman L., Goes J., and Savage G. T., pp. 39–117. Bingley, UK: Emerald Group Publishing Limited. [DOI] [PubMed] [Google Scholar]
- Capps, C. , Dranove D., and Ody C.. 2015. The Effect of Hospital Acquisitions of Physician Practices on Prices and Spending. Working Paper Series. Chicago, IL: Institute for Policy Research, Northwestern University. [DOI] [PubMed]
- Carlin, C. S. , Dowd B., and Feldman R.. 2014. “Changes in Quality of Health Care Delivery after Vertical Integration.” Health Services Research 50 (4): 1043–68. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Casalino, L. P. 2006. “Which Type of Medical Group Provides Higher‐Quality Care?” Annals of Internal Medicine 145 (11): 860–1. [DOI] [PubMed] [Google Scholar]
- Casalino, L. P. . 2014. “Accountable Care Organizations–The Risk of Failure and the Risks of Success.” New England Journal of Medicine 371 (18): 1750–1. [DOI] [PubMed] [Google Scholar]
- Casalino, L. P. . 2017. “The Medicare Access And CHIP Reauthorization Act and the Corporate Transformation of American Medicine.” Health Affairs (Millwood) 36 (5): 865–9. [DOI] [PubMed] [Google Scholar]
- Casalino, L. , Gillies R. R., Shortell S. M., Schmittdiel J. A., Bodenheimer T., Robinson J. C., Rundall T., Oswald N., Schauffler H., and Wang M. C.. 2003. “External Incentives, Information Technology, and Organized Processes to Improve Health Care Quality for Patients with Chronic Diseases.” Journal of the American Medical Association 289 (4): 434–41. [DOI] [PubMed] [Google Scholar]
- Casalino, L. P. , Wu F. M., Ryan A. M., Copeland K., Rittenhouse D. R., Ramsay P. P., and Shortell S. M.. 2013. “Independent Practice Associations and Physician‐Hospital Organizations Can Improve Care Management for Smaller Practices.” Health Affairs (Millwood) 32 (8): 1376–82. [DOI] [PubMed] [Google Scholar]
- Casalino, L. P. , Pesko M. F., Ryan A. M., Mendelsohn J. L., Copeland K. R., Ramsay P. P., Sun X., Rittenhouse D. R., and Shortell S. M.. 2014. “Small Primary Care Physician Practices Have Low Rates of Preventable Hospital Admissions.” Health Affairs (Millwood) 33 (9): 1680–8. [DOI] [PubMed] [Google Scholar]
- Casalino, L. P. , Chen M. A., Staub C. T., Press M. J., Mendelsohn J. L., Lynch J. T., and Miranda Y.. 2016. “Large Independent Primary Care Medical Groups.” Annals of Family Medicine 14 (1): 16–25. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Centers for Medicare & Medicaid Services . 2017. “State and County Level Demographic, Cost, Utilization, and Quality Data” [accessed on June 1, 2017]. Available at http://download.cms.gov/Research-Statistics-Data-and-Systems/Statistics-Trends-and-Reports/Medicare-Geographic-Variation/Downloads/State_County_Table_All.zip
- Friedberg, M. W. , Coltin K. L., Pearson S. D., Kleinman K. P., Zheng J., Singer J. A., and Schneider E. C.. 2007. “Does Affiliation of Physician Groups with One Another Produce Higher Quality Primary Care?” Journal of General Internal Medicine 22 (10): 1385–92. [DOI] [PMC free article] [PubMed] [Google Scholar]
- GAO . 2015. Increasing Hospital‐ Physician Consolidation Highlights Need for Payment Reform. Washington, DC.
- Gaynor, M. , Ho K., and Town R. J.. 2015. “The Industrial Organization of Health‐Care Markets.” Journal of Economic Literature 53 (2): 235–84. [Google Scholar]
- Greaney, T. L. , and Ross D.. 2016. “Navigating through the Fog of Vertical Merger Law: A Guide to Counselling Hospital‐Physician Consolidation under the Clayton Act.” Washington Law Review 91: 199–252. [Google Scholar]
- Kane, C. K. 2015. Updated Data on Physician Practice Arrangements: Inching Toward Hospital Ownership. Chicago, IL.
- Kralewski, J. , Dowd B., Knutson D., Tong J., and Savage M.. 2015. “The Relationships of Physician Practice Characteristics to Quality of Care and Costs.” Health Services Research 50(3): 710–29. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Landon, B. E. , Normand S. L., Meara E., Zhou Q., Simon S. R., Frank R., and McNeil B. J.. 2008. “The Relationship between Medical Practice Characteristics and Quality of Care for Cardiovascular Disease.” Medical Care Research and Review 65 (2): 167–86. [DOI] [PubMed] [Google Scholar]
- McWilliams, J. , Chernew M. E., Zaslavsky A. M., Hamed P., and Landon B. E.. 2013. “Delivery System Integration and Health Care Spending and Quality for Medicare Beneficiaries.” Journal of the American Medical Association Internal Medicine 173 (15): 1447–56. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Mostashari, F. 2016. “The Paradox of Size: How Small, Independent Practices can Thrive in Value‐Based Care.” Annals of Family Medicine 14 (1): 5–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Muhlestein, D. B. , and Smith N. J.. 2016. “Physician Consolidation: Rapid Movement from Small to Large Group Practices, 2013–15.” Health Affairs (Millwood) 35 (9): 1638–42. [DOI] [PubMed] [Google Scholar]
- Neprash, H. T. , Chernew M. E., Hicks A. L., Gibson T., and McWilliams J. M.. 2015. “Association of Financial Integration between Physicians and Hospitals with Commercial Health Care Prices.” Journal of the American Medical Association Internal Medicine 175 (12): 1932–9. [DOI] [PubMed] [Google Scholar]
- Physicians Advocacy Institute . 2016. Physician Practice Acquisition Study.
- Reschovsky, J. D. , and Rich E.. 2015. “Hospital Acquisition of Physician Groups: On the Road to Value‐Based or Higher‐Priced Care?” Journal of the American Medical Association Internal Medicine 175 (12): 1939–41. [DOI] [PubMed] [Google Scholar]
- Rittenhouse, D. R. , Casalino L. P., Shortell S. M., McClellan S. R., Gillies R. R., Alexander J. A., and Drum M. L.. 2011. “Small and Medium‐Size Physician Practices Use Few Patient‐Centered Medical Home Processes.” Health Affairs (Millwood) 30 (8): 1575–84. [DOI] [PubMed] [Google Scholar]
- Robinson, J. C. , and Miller K.. 2014. “Total Expenditures per Patient in Hospital‐Owned and Physician‐Owned Physician Organizations in California.” Journal of the American Medical Association 312 (16): 1663–9. [DOI] [PubMed] [Google Scholar]
- Schneider, E. C. , and Hall C. J.. 2017. “Improve Quality, Control Spending, Maintain Access—Can the Merit‐Based Incentive Payment System Deliver?” New England Journal of Medicine 376 (8): 708–10. [DOI] [PubMed] [Google Scholar]
- Squires, D. , and Blumenthal D.. 2016. Do Small Physician Practices Have a Future? New York: The Commonwealth Fund.
- Vladeck, B. C. , and Rice T.. 2009. “Market Failure and the Failure of Discourse: Facing up to the Power of Sellers.” Health Affairs (Millwood) 28 (5): 1305–15. [DOI] [PubMed] [Google Scholar]
- Weeks, W. B. , Gottlieb D. J., Nyweide D. E., Sutherland J. M., Bynum J., Casalino L. P., Gillies R. R., Shortell S. M., and Fisher E. S.. 2010. “Higher Health Care Quality and Bigger Savings Found at Large Multispecialty Medical Groups.” Health Affairs (Millwood) 29 (5): 991–7. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Appendix SA1: Author Matrix.
Appendix S1: The Practice Sample.
Appendix S2: Attributing Beneficiaries to Physicians and Practices.
Appendix S3: Spending Categories and Standardization.
Appendix S4: Control Variables, including Measurement of Medical Home Processes.
Exhibit S1: Association between Hospital Utilization and Outpatient Practice Characteristics, Full Regression Results.
Exhibit S2: Association between Spending and Outpatient Practice Characteristics, Full Regression Results.