US Physicians’ Reactions To ACA Implementation, 2012–17

1. Lindsay Riordan is a medical student in the Mayo Clinic Alix School of Medicine, in Rochester, Minnesota.

R Warsame

2. Rahma Warsame is a consultant in the Division of Hematology, Mayo Clinic Minnesota, in Rochester.

S Jenkins

3. Sarah Jenkins is a statistician in the Division of Biomedical Statistics and Informatics, Mayo Clinic Minnesota.

K Lackore

4. Kandace Lackore is a statistician in the Division of Biomedical Statistics and Informatics, Mayo Clinic Minnesota.

JE Pacyna

5. Joel E. Pacyna is a analyst in the Biomedical Ethics Research Program, Mayo Clinic Minnesota.

RM Antiel

6. Ryan M. Antiel is a fellow in the Division of General Surgery, Mayo Clinic Minnesota.

T Beebe

7. Timothy Beebe is a professor of health policy and management at the University of Minnesota, in Minneapolis.

M Liebow

8. Mark Liebow is a consultant in the Division of General Internal Medicine, Mayo Clinic Minnesota.

B Thorsteinsdottir

9. Bjorg Thorsteinsdottir is a consultant in the Division of Community Internal Medicine, Mayo Clinic Minnesota.

M Wynia

10. Matthew Wynia is director of the Center for Bioethics and Humanities, Anschutz Medical Campus, University of Colorado, in Aurora.

SD Goold

11. Susan Dorr Goold is a professor of internal medicine and a professor of health management and policy at the University of Michigan, in Ann Arbor.

M DeCamp

12. Matthew DeCamp is an associate professor in the Center for Bioethics and Humanities and Division of General Internal Medicine, Anschutz Medical Campus, University of Colorado.

M Danis

13. Marion Danis is head of the Section on Ethics and Health Policy, Department of Bioethics, Clinical Center, National Institutes of Health, in Bethesda, Maryland.

J Tilburt

14. Jon Tilburt is a consultant in the Division of General Internal Medicine and the Division of Health Care Policy and Research, Mayo Clinic Minnesota.

1. Lindsay Riordan is a medical student in the Mayo Clinic Alix School of Medicine, in Rochester, Minnesota.

3. Sarah Jenkins is a statistician in the Division of Biomedical Statistics and Informatics, Mayo Clinic Minnesota.

4. Kandace Lackore is a statistician in the Division of Biomedical Statistics and Informatics, Mayo Clinic Minnesota.

7. Timothy Beebe is a professor of health policy and management at the University of Minnesota, in Minneapolis.

9. Bjorg Thorsteinsdottir is a consultant in the Division of Community Internal Medicine, Mayo Clinic Minnesota.

10. Matthew Wynia is director of the Center for Bioethics and Humanities, Anschutz Medical Campus, University of Colorado, in Aurora.

11. Susan Dorr Goold is a professor of internal medicine and a professor of health management and policy at the University of Michigan, in Ann Arbor.

12. Matthew DeCamp is an associate professor in the Center for Bioethics and Humanities and Division of General Internal Medicine, Anschutz Medical Campus, University of Colorado.

13. Marion Danis is head of the Section on Ethics and Health Policy, Department of Bioethics, Clinical Center, National Institutes of Health, in Bethesda, Maryland.

14. Jon Tilburt is a consultant in the Division of General Internal Medicine and the Division of Health Care Policy and Research, Mayo Clinic Minnesota.

Corresponding Author: Jon Tilburt, MD, MPH, Mayo Clinic, 200 First Street SW, Rochester, MN 55905, ude.oyam@noj.trublit

Abstract

Physicians play a key role in implementing health policy, and U.S. physicians were split in their enthusiasm for the Patient Protection and Affordable Care Act (ACA) soon after its implementation began. We re-administered elements of a prior survey of U.S. physicians to a similar sample to understand how US physician opinions of the ACA may have changed over a crucial five-year implementation period (2012–2017), and compared responses across both surveys. Of the 1,200 physicians to whom we sent a survey, 489 responded in 2017 (RR: 41%). In the summer of 2017, a majority of respondents (60%) believed the ACA improved access to care yet many (43%) acknowledged it reduced affordability of health insurance coverage. More physicians (53%) agreed the ACA “would turn US health care in the right direction” than in 2012 (42%), despite reporting perceived worsening in several practice conditions over that same time period. After adjusting for specialty, political affiliation, practice setting, perceived social responsibility, age, and gender, only political affiliation (Democrat, Republican, Independent) was a significant predictor of support for the ACA in the 2017 results.

INTRODUCTION

Since the passage of the Patient Protection and Affordable Care Act (ACA) in 2010, more than 30 million Americans have gained access to insurance coverage (1). Passed along party lines, the ACA has been controversial since its inception, generating over 70 attempts at legislative repeal and multiple lawsuits, including a recent federal district court ruling in the Texas v. Azar case on December 14, 2018 deeming the legislation unconstitutional (2, 3). Overall, public opinion about the ACA remains split, with recent approval ratings hovering around 50% (4). Certain provisions of the Act are extremely popular across party lines, including provisions like dependent coverage until age 26 and insurability at usual prices for citizens with preexisting conditions. Public opinion on other provisions, including the mandate to buy insurance or pay a penalty, and the ACA’s impact on affordability of premiums, has been mixed (4).

Physicians’ unique views of the health care system are both shaped by, and have the potential to shape, patients’ perspectives on health reform. Since physicians are on the front lines of health care delivery, their perspectives are crucial to understanding what policy components work for them and their patients and what components do not. There have been limited analyses of physician responses to the ACA – all cross-sectional, none over time. A 2015 national survey of primary care physicians found 48% approving and 52% disapproving of the legislation (5). Another survey of primary care physicians in early 2017 indicated that only 15% of physicians supported complete ACA repeal, but the study did not report general approval ratings (6). We surveyed physicians across specialties in 2012 to gauge initial reactions to ACA implementation (7). In the summer of 2017, when House Republicans were trying to repeal large portions of the ACA, we repeated our 2012 survey to assess potential changes in perceptions and approval for the ACA in the interval five years.

Here, we present those results, including respondents’ impressions of its overall impact on U.S. healthcare, fairness of payment, and impressions of specific provisions. We compared those responses to our 2012 survey to determine how physician opinions toward the ACA have changed in the intervening five years and what demographic and attitudinal factors might be associated with their opinions. We also gauged whether other aspects of medical practice have changed over the same time period.

METHODS

Ethical Review

This study was approved by the Mayo Clinic Institutional Review Board.

Study Participants

In the summer of 2017, we mailed a nine-page survey entitled, “Physicians, Health Care Costs, and Society” to a random sample of 1,200 U.S. physicians drawn from the AMA Masterfile. Though compositionally similar to 2012, the sample was a new random sample not connected to the 2012 cohort. Survey samples in both years were representative of the US population of physicians. The initial wave of paper surveys included a $10 bill and postage-paid return envelope. Second and third surveys were mailed to non-responders at 30 and 60 days.

Survey Instrument

The 2017 sampling and item wording were identical to our 2012 survey, plus several additional items described below (7). Domains covered included physicians’ perspectives on health reform, societal responsibilities, medical decision-making, and cost of health care.

Dependent Measures

We assessed physicians’ overall support for the ACA based on their agreement with the following statement: “The Affordable Care Act, if fully implemented, would turn United States health care in the right direction.” (Response categories: “strongly disagree,” “moderately disagree,” “moderately agree,” and “strongly agree.”) We also assessed physician perceptions of reimbursement fairness in both surveys – “The ACA, if fully implemented, would make physician reimbursement…”(Response categories: “more fair,” “less fair,” “neither more nor less fair,” and “not sure”) (7).

Independent Measures

We collected several predictors for ACA approval including primary specialty, political affiliation, and practice setting type in both 2012 and 2017. At both time points we also measured respondents individual professional sense of social responsibility with the following statements: “I would favor limiting coverage for expensive drugs and procedures if that would help expand access to basic health care for those currently lacking such care” / “Every physician is professionally obligated to care for the uninsured and underinsured” / “Addressing societal health policy issues, as important as that may be, falls outside the scope of my professional obligations as a physician” (Response categories: “strongly disagree,” “moderately disagree,” “moderately agree,” and “strongly agree”)(8). To harmonize the 2017 items with other national metrics, we measured party affiliation (“A Republican,” “A Democrat,” “An Independent,” “Other party,” and “No preference/Not interested in politics”) instead of using the liberal/moderate/conservative measure from our 2012 survey (9).

In the 2017 survey only, we also assessed U.S. physicians’ experience in the past five years and their perspectives of the impact of ACA implementation on patient care using existing measures from other surveys (7, 9). We included several items gauging the overall impact (“positive,” “negative,” “none,” “not sure”) of the ACA on domains including, “quality of care,” “cost of care,” “access to health care and insurance overall,” “ability to receive care with pre-existing conditions,” “affordability of health insurance,” “the role of employers in providing health insurance,” “overall medical practice,” and “the ability of practice to meet patient demand.”

Finally, we also asked about general changes in practice conditions over the last five years, including “ability to provide high quality care to all patients,” “amount of time available to spend with each patient,” “amount of time spent on administrative tasks related to insurance,” “time managing patients’ opioids,” “ability to recruit and retain clinical staff,” and “patient satisfaction.” Response categories for those were “much improved,” “somewhat improved,” “about the same,” “somewhat worse,” “much worse,” and “not sure.”(10)

Analysis

Response rates were computed in accordance with the guidelines outlined by the American Association for Public Opinion Research (11). Survey responses were summarized with frequencies and percentages and were compared between physician characteristics using chi-square tests. For ease of analysis, we collapsed individual specialties into four specialty types – surgical, procedural sub-specialist, non-procedural sub-specialist, and primary care. Logistic regression was used to examine unadjusted and adjusted associations with agreement to the statement “The ACA, if fully implemented, would turn U.S. health care in the right direction”, including indicators of primary specialty, practice setting, political affiliation, social responsibility, age, gender, and year. Also, for ease of presentation for the dependent measures, we collapsed 4-point response variables (“Strongly Disagree,” “Moderately Disagree,” “Moderately Agree,” and “Strongly Agree”) into “Disagree/Agree”. Multi-variable regression analysis was selected to adjust for respondent demographics and enable comparison of statistically significant predictors in 2012 and 2017. P-values less than 0.05 were considered statistically significant. All analyses were performed using SAS version 9.4 (SAS Institute Inc., Cary, NC).

Limitations

Several factors limit our ability to draw conclusive inferences. First, the overall characteristics of practicing physicians (not just our respondents) may have shifted over this same period. Second, it is also possible that the 2012 sample was slightly less susceptible to response bias given its higher response rate (65%). Third, and more substantively, the meaning of the phrase, “if fully implemented,” may have changed over the last five years. Pieces of the ACA have been stripped away, watered down, or not enforced, changing in part what the ACA is. By the summer of 2017, the Trump administration had already weakened the individual mandate and begun eliminating cost sharing reduction payments through executive action. While we did not measure opinion about these specifics, it is possible that such efforts may have led to increased physician support for what remained of the ACA by mid-2017. Fourth, our questions about the ACA preceded questions about changes in practice environment in the survey instrument, leaving open the possibility of priming effects in our data regarding 5-year practice changes, though the overall favorability trend toward the ACA ran in the opposite direction to reports of worsening practice conditions. Additionally, asking respondents to recall attitudes over a 5-year period may introduce recall bias. Moreover, cross-sectional sampling at two time points may suggest, but does not establish, an association between implementation of the ACA and practice environment issues. Finally, our change from a measure of political ideology (liberal/moderate/conservative) to party identification (Democrat/Independent/Republican) between the 2012 and 2017 surveys prohibited us from comparing political associations across time (12).

RESULTS

Population

The combined, weighted response rate across the 2012 and 2017 surveys was 53%. The response rate for the 2012 survey previously reported was 65% (7). Of the 1,200 physicians surveyed in 2017, 489 returned surveys, for a response rate of 41%. Of the 489 returned surveys, 445 included at least 80% completed items for analysis.

Based on the information available in the AMA Masterfile sampling frame, 2017 respondents were slightly older than non-respondents [mean age 50.7 vs 48.9 years, p=0.02]. Significantly more respondents in 2017 were from small/solo, group/HMO, and government practice settings (p=0.003). There were no significant differences between respondents and non-respondents with respect to sex, region, or specialty. Respondent characteristics are reported in Exhibit 1 .

Exhibit 1:

Demographic data are from the AMA Masterfile, 2017, as well as investigator-developed demographic items.

ACA Improving U.S. Healthcare/Fairness of Physician Reimbursement

In 2017, 53% of physicians agreed (33% moderately, 20% strongly) that, “the Affordable Care Act, if fully implemented, would turn United States health care in the right direction.” This represented an 11 percentage-point increase in agreement since 2012 (42% agreement in 2012, p <0.001) (Exhibit 2 ).

The item reported in this figure is an investigator-developed survey item fielded in 2012 and repeated in 2017.

The proportion of physicians who believed ACA implementation would make physician reimbursement less fair decreased from 44% in 2012 to 34% in 2017 (p<0.0001).

Perceived Impact of ACA

Sizeable majorities of U.S. physicians perceived a positive impact of the ACA on access to health care and insurance overall (60%) and access to care for patients with pre-existing conditions (73%). Forty-three percent of physicians believed the ACA has had a negative impact on the affordability of health insurance coverage and 34 percent believed the ACA had a negative impact on the ability of their practices to meet patient demand ( Exhibit 3 ).

Exhibit 3:

Perceptions of the impact of ACA implementation and perceived practice changes among 445 U.S. physicians in 2017.

Questions related to ACA impact were developed by the investigative team. The practice-change item related to opioid management is also investigator-developed. The remaining practice-change items are from the Commonwealth Fund Survey (10).

Other Practice Conditions – Last Five Years

Physicians reported worsening practice conditions over the 2012–2017 period including the amount of time spent on administrative issues related to insurance (67%), amount of time available to spend with each patient (59%), as well as ability to recruit/retain clinical staff (42%) and time managing patients’ opioid use (34%) ( Exhibit 3 ).

Covariates of ACA Attitudes Over Time

We examined physician attitudes toward the ACA by political affiliation, practice type, primary specialty, and sense of social responsibility. Physicians’ overall positive attitude toward the ACA increased across all specialty categories in 2017 from 2012. In 2012, fewer physicians in surgical and procedural specialties reported overall support for the ACA compared to primary care physicians. Over time, surgical and procedural specialists, many of whom more readily identified as Republican, increased their overall support more than did primary care physicians. In 2012, 30 percent of surgeons reported overall support; in 2017, 47 percent of surgeons did. In 2012 36 percent of procedural specialists reported overall support; in 2017, 53 percent did (Data not shown in Exhibits). In fact, in 2017, both procedural and non-procedural specialists rated overall support for the ACA somewhat higher than primary care physicians (ORs 1.20 and 1.40, respectively), though the comparisons were not statistically significant ( Exhibit 4 ). In 2017, in both unadjusted and adjusted analyses, specialty category was no longer a significant predictor of ACA support.

Exhibit 4:

Multivariable model of associations between support for the ACA and characteristics of U.S. physician in 2012 and 2017.

Demographic data are from the AMA Masterfile sample, 2017. The following survey items in the table are from Antiel, 2009 (8): “I would favor limiting coverage for expensive drugs and procedures if that would help expand access to basic health care for those currently lacking such care.” / “Every physician is professionally obligated to care for the uninsured and underinsured.” / “Addressing societal health policy issues, as important as that may be, falls outside the scope of my professional obligations as a physician.”

Other characteristic groups included but not shown due to small N: non-clinical and other specialty, other party and no preference/not interested in politics, and medical school and other practice setting.

In 2017, self-identified Republicans were markedly less likely to support full implementation of the ACA in 2017 (18%) than Independents (51%) and Democrats (86%) (Data not shown in Exhibits).

Similar to what was found in 2012, in 2017 physicians in small/solo practices showed the least overall support for the ACA (32%) compared to physicians in Group/HMO practices (48%), in city/state/federal government roles (63%), non-government hospital roles (66%), or with medical school affiliations (100%)(Data not shown in Exhibits).

As in 2012, physicians’ beliefs about social responsibility were associated with their opinions about the ACA. In an unadjusted logistic regression model, those who agreed that “addressing societal health policy issues, as important as that may be, falls outside the scope of my professional obligations as a physician” were less likely to support the ACA (OR=0.64, p=0.04) than physicians who did not endorse that item. Those who agreed that “every physician is professionally obligated to care for the uninsured and underinsured (were more likely to support the ACA (OR=1.77, p=0.006) than physicians who did not agree. Physicians who endorsed the item, “I would favor limiting coverage for expensive drugs and procedures if that would help expand access to basic health care for those currently lacking such care” were also more likely to support the ACA than non-endorsers (OR=2.49, p <0.0001) (Data not shown in Exhibits). In the adjusted model, none of these items was statistically significant (Exhibit 4 ).

Multivariable Results

In a multivariable model of our 2017 data that included specialty, political affiliation, practice setting, social responsibility, age, and gender – factors found to be important in 2012 – only political affiliation was significantly associated with support for the ACA in 2017 (OR = 4.74 and 24.84 for Independents and Democrats respectively compared to Republicans, (p

DISCUSSION

Over the five-year period from 2012 to 2017, during which key provisions of the ACA were both celebrated and criticized as they were rolled out, U.S. physician support for ACA increased markedly. When asked about specific ACA provisions in 2017, physicians reported favorable impressions about improved insurance access and coverage of pre-existing conditions, while reporting unfavorable impressions about the law’s impact on affordability of insurance, patient demand, and the role of employers in providing insurance. Physicians’ experiences of the overall impact of the ACA on their own practices were split or neutral, and tended to run concurrent with concerns about broader undesireable practice trends not directly attributable to the ACA, including worsening time pressures, growing administrative burden, challenges managing patient opioid use, and worsening staffing concerns.

Agreement with the overall direction of the ACA increased across all medical specialties. The significantly lower approval by procedural and surgical specialties observed in 2012 was no longer significant in 2017, suggesting that some of the concern in 2012 may have abated by 2017 (13). Increased familiarity with the law and experience with its impact on professional practice may have partly assuaged earlier concerns. The most significant predictor for supporting full implementation of the ACA in 2017 was party affiliation, analogous to the association with political ideology we found in 2012 (14). In our multivariate model, specialty, practice type, social responsibility, and demographics were not significantly associated with support for or opposition to the ACA.

Changes we observed over the five years of the ACA’s implementation appear to reflect real changes in physician opinion during that period. This longitudinal analysis provides evidence of greater support for the ACA among U.S. physicians notwithstanding many perceived growing practice challenges in that five-year period.

Acknowledgments

To our knowledge, this study is the first analysis of how physicians’ opinions about the ACA have changed over the course of its implementation. Physicians identify elements of the ACA that are working and elements that challenge their practice and patients. A slight majority of U.S. physicians believe, after experiencing its implementation, the ACA is a net positive for U.S. healthcare. Their favorable impressions increased, despite reporting declining affordability of insurance, increased administrative burdens, and other challenges they and their patients faced.

References

1. Blumenthal D, Abrams M, Nuzum R. The Affordable Care Act at 5 Years . N Engl J Med . 2015; 372 ( 25 ):2451–8. [PubMed] [Google Scholar]

2. Dalen JE, Waterbrook K, Alpert JS. Why do so many Americans oppose the Affordable Care Act? Am J Med . 2015; 128 ( 8 ):807–10. [PubMed] [Google Scholar]

5. The Kaiser Family Foundation and The Commonwealth Fund. Experiences and Attitudes of Primary Care Providers Under the First Year of ACA Coverage Expansion . 2015. [PubMed]

6. Pollack CE, Armstrong K, Grande D. A View from the Front Line - Physicians’ Perspectives on ACA Repeal . N Engl J Med . 2017; 376 ( 6 ):e8. [PubMed] [Google Scholar]

7. Antiel RM, James KM, Egginton JS, Sheeler RD, Liebow M, Goold SD, et al. Specialty, political affiliation, and perceived social responsibility are associated with U.S. physician reactions to health care reform legislation . J Gen Intern Med . 2014; 29 ( 2 ):399–403. [PMC free article] [PubMed] [Google Scholar]

8. Antiel RM, Curlin FA, James KM, Tilburt JC. Physicians’ Beliefs and U.S. Health Care Reform — A National Survey . New England Journal of Medicine . 2009; 361 ( 14 ):e23. [PMC free article] [PubMed] [Google Scholar]

9. Ryan J, Doty MM, Hamel L, Norton M, Abrams MK, Brodie M. Priamry Care Providers’ Views of Recent Trends in Health Care Delivery and Payment . Commonwealth Fund ; 2015. [PubMed] [Google Scholar]

11. The American Association for Public Opinion Research. Standard Definitions: Final Dispositions of Case Codes and Outcome Rates for Surveys. 9th edition 2016. [Available from: https://www.aapor.org/AAPOR_Main/media/publications/Standard-Definitions20169theditionfinal.pdf.

12. Kiley J. In polarized era, fewer Americans hold a mix of conservative and liberal views . Pew Research Center October 23 , 2017. [Google Scholar]

14. Jacobs LR, Mettler S. Liking Health Reform But Turned Off By Toxic Politics . Health Affairs . 2016; 35 ( 5 ):915–22. [PubMed] [Google Scholar]