User login
METHODS: We surveyed a random sample of primary care physicians participating in Medicaid managed care (n = 670) or traditional Medicaid (n = 670). Primary outcomes measured were physicians’ satisfaction with Medicaid managed care, traditional Medicaid, and commercial managed care. Satisfaction was measured on a 5-point Likert-type scale.
RESULTS: The response rate was 52%. Physicians participating in Medicaid managed care were less likely to be satisfied or very satisfied with Medicaid managed care (28.6%) than with commercial managed care (40%) or their previous experience with traditional Medicaid (39.7%). Among physicians participating in traditional Medicaid, 29.8% were satisfied or very satisfied with traditional Medicaid. Physicians participating in Medicaid managed care were less satisfied with clinical autonomy under that system in comparison with their previous experience with traditional Medicaid (relative difference = 10.8%, P =.001). In multiple linear regression analyses, clinical autonomy (R2 = 0.40) was a strong predictor of overall satisfaction with Medicaid managed care.
CONCLUSIONS: Enhancing physicians’ clinical autonomy may result in improved satisfaction with Medicaid managed care. State Medicaid agencies should include physician satisfaction as a measure of Medicaid managed care plans’ quality.
Although national attempts at comprehensive health care reform have faltered, many state governments have developed and implemented reform initiatives. In particular, rising Medicaid expenditures and concerns about inadequate access to care have led 49 states and the District of Columbia to experiment with Medicaid managed care.1 The proportion of the United States Medicaid population enrolled in managed care plans increased from 10% in 1991 to 48% in 1997.1 In Missouri, rising Medicaid expenditures led to the state government’s decision to enroll Aid to Families and Dependent Children (AFDC) program recipients in the state’s eastern, central, and western regions in Medicaid managed care beginning in 1995.2,3 Traditional Medicaid continued to cover the AFDC populations in Missouri’s southern and northern regions.2
There are ongoing concerns about the quality of care provided to patients in Medicaid managed care plans.4-8 Low rates of physician participation have plagued traditional Medicaid programs and have adversely affected access and quality of care.9 Increasing primary care provider participation in Medicaid managed care programs is critical for improving the quality of care provided to Medicaid recipients.9 Thus, efforts to increase physician participation may be necessary to improve the quality in those programs.
Physician satisfaction has been associated with outcomes important to patients, insurers, and state governments. In one study, physician satisfaction was correlated with patient satisfaction, increased continuity of care, lower patient no-show rates, and more reasonable charges for follow-up visits.10 Less satisfied general practitioners in England were more likely to prescribe inappropriate drugs.11 Physician satisfaction was also found to predict physician turnover12 and was associated with participation in the Ontario physician strike.13 Therefore, physician satisfaction with Medicaid managed care may affect both quality of care and physician participation in the program.
A number of studies have examined physician satisfaction in traditional fee-for-service, managed care, and institutional settings. Clinical autonomy is strongly predictive of physician satisfaction.12,14-16 Organizational autonomy (the ability of a physician to control his or her workplace),14-16 satisfaction with income,13,15,17 and satisfaction with patient relationships10,12,13 have also been found to predict physician satisfaction.
From a theoretical perspective, autonomy is central to the role functioning of physicians in the physician-patient relationship.18 This relationship has therapeutic qualities beyond those of any prescribed or recommended intervention.19-21 Health care system changes that affect physician autonomy may have an impact on the nature of the physician-patient relationship, as well as affect physician satisfaction.22-24
The purposes of our study were to assess level of physician satisfaction with Medicaid managed care in comparison with traditional Medicaid and commercial managed care and to identify the determinants of physician satisfaction with Medicaid managed care.
Methods
Design
We performed a cross-sectional survey of primary care physicians participating in Medicaid managed care or in traditional Medicaid after the implementation of Medicaid managed care in eastern and central Missouri.
Questionnaire Development
We used a literature review to identify topics for the questionnaire.12,15-17,25-33 The questionnaire included items about physicians’ satisfaction with: their ability to care for patients; their communications with third-party payers; and their satisfaction with patient relationships and reimbursement. Physicians participating in Medicaid managed care were asked to answer these questions according to their current experience with Medicaid managed care and commercial managed care and their previous experiences with traditional Medicaid. Physicians participating in traditional Medicaid were asked to answer these questions according to their current experience. A single item was used to measure overall satisfaction with type of insurance. We measured global physician satisfaction with the practice of medicine using a previously validated scale.17 All responses were measured on a 5-point scale, where 1 = very dissatisfied and 5 = very satisfied. Other items measured the level of physicians’ input into their practices’ decision-making process, the number of Medicaid managed care plans in which they participated, and demographic information. The questionnaire was pilot-tested on groups of academic and community family physicians in Missouri. After it was revised, physician members of the Missouri Division of Medical Services quality improvement project reviewed it. The final instrument included 40 questions and took apporoximately 15 minutes to complete.*
Sample Selection
We sampled 2 groups of primary care physicians. The Medicaid managed care sample was made up of physicians in eastern and central Missouri participating in Medicaid managed care. Physicians in northern and southern Missouri participating in traditional Medicaid made up the traditional Medicaid sample. We defined primary care physicians as family physicians, general practitioners, primary care pediatricians, primary care internists, and obstetrician-gynecologists. The Missouri Division of Medical Services provided 2 databases of primary care physicians participating in Medicaid managed care or traditional Medicaid. A random sample of 670 physicians was drawn from each database. The sample size was determined by assuming an 80% power to detect a 10% difference of proportions between groups, with a 2-sided P value of .05 and a projected response rate of 50%. Subsequently, each sample was matched with the state licensure database. Physicians were not eligible to participate in the survey if they were identified as specialists either by the state licensure database or by self-report.
Survey Administration
We mailed the questionnaire 3 times in the fall and winter of 1996-1997. Each mailing was sent with a cover letter signed by the principal investigator and the head of the Missouri Division of Medical Services. Postcard reminders were mailed after the first and third mailings. Methods to increase the response rate included hand-written thank you notes, hand-signed signatures on the cover letter, and the inclusion of a token gift.34,35 One mailing was sent by Priority Mail.
Statistical Analysis
We performed a principal component analysis followed by an orthogonal rotation on the 15 items comprising the satisfaction portion of the survey instrument. That was followed by a Cronbach’s a internal reliability analysis on the resulting factors.
We used chi-square tests to compare respondents with nonrespondents. Comparisons within the Medicaid managed care sample were performed using the Wilcoxon signed-rank test, and comparisons between samples were done with the Wilcoxon rank sum test. Because of multiple comparisons, we set statistical significance at P = .01. We measured bivariate relationships between possible predictor variables and overall satisfaction with Medicaid managed care using Spearman rank correlation test coefficients. We performed multivariable analysis using forward stepwise multiple linear regression. In regression analyses, dependent variables met the .05 significance level for inclusion in the models. We deleted cases with missing data from our analysis.
We performed statistical tests with SAS software (version 6.1, SAS institute, Cary, NC) or SPSS software (version 7.5.2, SPSS, Chicago, Ill).
Results
Response Rate
We excluded the following from the study: 265 specialists; 48 physicians who were not participating in Medicaid programs, not in practice, or still in training; and 57 physicians whose surveys were not deliverable. There were 505 usable responses received from the remaining 970 eligible physicians, for a response rate of 52%. The response rates in the traditional Medicaid and Medicaid managed care samples were 57% and 46%, respectively. The characteristics of the respondents are shown in Table 1. When comparing respondents with nonrespondents in the Medicaid managed care sample, pediatricians and family physicians were more likely to respond than internists and obstetricians (P <.001), and rural physicians were more likely to respond than urban physicians (P = .022). In the traditional Medicaid sample, family physicians and pediatricians were also more likely to respond than were internists or obstetricians (P = .002), and allopathic physicians were more likely to respond than were osteopathic physicians (P = .036). There were no significant differences in measures of satisfaction between responses received after the third mailing and responses received before the third mailing (P = 0.15 by Kruskall-Wallis test for both comparisons).
Questionnaire Analysis
Factor analysis confirmed 3 measures of components of physician satisfaction: clinical autonomy, organizational autonomy, and patient relationships. The clinical autonomy scale was made up of 6 items. These items measured satisfaction with the physician’s ability to hospitalize patients, prescribe medicine, choose specialists, order tests, obtain prior authorization, and control number of patient visits. Four items comprised the organizational autonomy scale, which measured satisfaction with communications with insurers about billing and payment, enrollment, prior authorization, and utilization review. Three items comprised the patient relationship scale: satisfaction with continuity, the quality of patient relationships, and the intellectual challenge posed by patient problems. The internal consistency of these 3 scales measured by Cronbach’s a ranged from 0.76 to 0.90. Two reimbursement items were grouped together on factor analysis and were treated as a scale.
Overall Satisfaction by Insurance Type
Figure 1 The Figure shows that physicians were more likely to respond that they were satisfied or very satisfied with commercial managed care (40%) and their previous experience with traditional Medicaid (39.7%) than with Medicaid managed care (28.6%). There was no difference in satisfaction between Medicaid managed care and satisfaction with traditional Medicaid measured in the traditional Medicaid sample (29.8%).
Physician Satisfaction with Commercial Managed Care
Physicians expressed neutrality toward clinical autonomy (mean = 2.95, standard deviation [SD] = 0.84), neutrality or slight dissatisfaction with organizational autonomy (mean = 2.63, SD = 0.89) and reimbursement (mean = 2.49, SD = 0.98), and neutrality to satisfaction with patient relationships (mean = 3.22, SD = 0.84).
Comparison of Physician Satisfaction Components Among Insurance Types
We compared scores on satisfaction scales by determining the relative mean difference between various insurance types and Medicaid managed care. This was calculated by subtracting mean satisfaction scale scores for different insurance types from mean Medicaid managed care scores and dividing that result by the maximum possible range for that scale. For these comparisons, experienced researchers’ consensus opinion was that a 10% difference would be meaningful. In the Medicaid managed care sample, satisfaction with reimbursement was less with Medicaid managed care than with commercial managed care Table 2. Physicians expressed lower satisfaction with clinical autonomy in a Medicaid managed care system than with their previous traditional Medicaid experience. Satisfaction with Medicaid managed care reimbursement was lower than satisfaction with traditional Medicaid reimbursement measured in the traditional Medicaid sample.
Predictors of Satisfaction
Physicians’ satisfaction with components of satisfaction in a Medicaid managed care system was correlated with overall satisfaction with Medicaid managed care Table 3. Overall satisfaction was also correlated with comfort with workload, caring for patients less than 30 hours per week, being a woman, and being a family physician. Satisfaction was inversely correlated with hours worked, caring for patients more than 60 hours per week, and the number of Medicaid managed care plans in which a physician participated.
We developed 3 stepwise multiple linear regression models using overall satisfaction with Medicaid managed care as the dependent variable. We built 3 models to explore the effects of any instability in the regression models due to missing data. For model 1, satisfaction components were independent variables, and for model 2 all variables that were significant in bivariate analysis were independent variables. Independent variables entered into model 3 were global satisfaction with the practice of medicine, clinical autonomy, workload, number of Medicaid managed care plans in which physicians participated, and whether a physician worked more than 60 hours or less than 30 hours per week in patient care. The resulting models were similar: only model 3 is reported here.* In this regression model, satisfaction with clinical autonomy was the most important predictor of overall satisfaction with Medicaid managed care, accounting for 40% of the variance in overall satisfaction Table 4. Global satisfaction with the practice of medicine, satisfaction with reimbursement, and working less than 30 hours per week in direct patient care were less important predictors of overall satisfaction.
Discussion
This is the first study to measure physician satisfaction with Medicaid managed care in comparison with traditional Medicaid and commercial managed care. Although physicians were less satisfied with Medicaid managed care in comparison with commercial managed care and their previous traditional Medicaid experience, they were not less satisfied with Medicaid managed care in comparison to physician satisfaction with traditional Medicaid measured in the traditional Medicaid sample. This suggests that physicians’ satisfaction level with commercial managed care did not decline from their previous level of satisfaction with traditional Medicaid. The higher satisfaction reported for physicians’ previous experience with traditional Medicaid may be related to implementation of the new Medicaid managed care program.
Physicians’ satisfaction level with clinical autonomy in the Medicaid managed care and commercial managed care systems was similar, indicating that physicians view the methods used to affect their behavior in commercial and Medicaid managed care programs similarly. This finding is supported by the results of a study of Arizona’s Medicaid managed care program.36
Our finding that clinical autonomy was a strong predictor of overall satisfaction with Medicaid managed care is consistent with the findings of other recent studies of physician satisfaction37,38 and theoretical work that stresses the importance of autonomy in the development of professional identity and functioning.18 The strength of this association in our study (R2 = .40) highlights the importance physicians place on maintaining their authority to provide the care they believe will best serve their patients. In contrast, satisfaction with reimbursement was a weak predictor of overall satisfaction (R2 = .02). State Medicaid agencies attempting to improve physician satisfaction with Medicaid managed care should work first to improve clinical autonomy while providing reasonable reimbursement. In traditional Medicaid programs, both reimbursement and decreased interference with clinical decision making have been associated with increased rates of participation.39-41 Increasing physician satisfaction with clinical autonomy in Medicaid managed care plans may result in increased rates of physician participation.
Efforts to increase physician satisfaction with clinical autonomy may also increase quality of care.10,11 Low levels of satisfaction with clinical autonomy in managed care systems has been associated with a perception of lower quality of care.42 However, participation in managed care plans does not always result in lower satisfaction with clinical autonomy. Several studies have shown a strong association between participation in managed care plans and high levels of satisfaction with clinical autonomy.15,37,43 Physician involvement in decision-making processes that affect their daily work is related to increased satisfaction with autonomy.16,44 Thus, involving physicians and physician organizations in the formulation of guidelines that will affect physician behavior in Medicaid managed care plans could promote satisfaction with clinical autonomy while improving quality and controlling costs.
The strengths of this study include the development of 3 scales that have good face validity and content validity and high internal reliability. Our survey included physicians participating in traditional Medicaid, as well as Medicaid managed care. Most importantly, this is one of the first studies to address physician satisfaction early in the national experiment with Medicaid managed care plans.
Limitations
Our study may be limited by a response rate of 52%. Although we found no difference between late and early responders on measures of satisfaction and intention to continue participation, our results may be biased toward negative responders. This study was also limited to Missouri physicians. It may not be possible to generalize our findings to other states. Furthermore, we measured satisfaction early in the implementation of Medicaid managed care. Physician satisfaction may change as the Medicaid managed care and commercial markets evolve.
Conclusions
There may be a tendency for some to view an exploration of the effect of changes in the health care delivery system on physician autonomy and satisfaction as a case of professional self-interest, especially in the current climate where physicians are increasingly and publicly critical of the impact of managed care on quality of care. However, the larger sociological perspective—drawn from decades of empirical and theoretical research that links autonomy to physician satisfaction, the therapeutic relationship, and quality of care—shows the importance of autonomy and satisfaction.10,11,18,20,21 Finding ways to promote physician autonomy and satisfaction in Medicaid managed care settings is necessary to maintain the therapeutic relationship that forms the basis of high-quality care. The quality indicators of Medicaid managed care plan performance should include a measurement of physician satisfaction and autonomy.
Acknowledgments
Dr Gazewood was partially funded by the National Research Service Award 5-T32-P#17001-08 and by a Harrison Teaching Professorship of Generalist Medicine. The Missouri Division of Medical Services and the University of Missouri-Columbia Department of Family and Community Medicine small grant program also provided funding.
The authors acknowledge Paula McDonald for research assistance and secretarial support.
1. Iglehart JK. The American health care system: Medicaid. N Engl J Med 1999;340:403-8.
2. Jenkins ER. Managed care plus (MC+): the wave of the future for Missouri Medicaid? Mo Med 1995;92:222-3.
3. Coughlin TA, Ku L, Holahan J, Heslam D, Winterbottom C. State responses to the Medicaid spending crisis: 1988 to 1992. J Health Polit Policy Law 1994;19:837-64.
4. Mirvis DM, Chang CF, Hall CJ, Zaar GT, Applegate WB. TennCare: health system reform for Tennessee. JAMA 1995;274:1235-41.
5. Dallek G. A consumer advocate on Medicaid managed care. Health Aff 1996;15:174-7.
6. Halvorson GC. An HMO chief executive officer on Medicaid managed care. Health Aff 1996;15:170-1.
7. Gold M, Sparer M, Chu K. Medicaid managed care: lessons from five states. Health Aff 1996;15:153-66.
8. Chang CF, Kiser LJ, Bailey JE, et al. Tennessee’s failed managed care program for mental health and substance abuse services. JAMA 1998;279:864-9.
9. Freund DA, Hurley RE. Medicaid managed care: contribution to issues of health reform. Ann Rev Public Health 1995;16:473-95.
10. Linn LS, Brook RH, Clark VA, Davies AR, Fink A, Kosecoff J. Physician and patient satisfaction as factors related to the organization of internal medicine group practices. Med Care 1985;23:1171-8.
11. Melville A. Job satisfaction in general practice: implications for prescribing. Soc Sci Med 1980;14A:495-9.
12. Lichtenstein R. Measuring the job satisfaction of physicians in organized settings. Med Care 1984;22:56-68.
13. Kravitz RL, Linn LS, Shapiro MF. Physician satisfaction under the Ontario Health Insurance Plan. Med Care 1990;28:502-12.
14. Stamps PL, Piedmont EB, Slavitt DB, Haase AM. Measurement of work satisfaction among health professionals. Med Care 1978;16:337-52.
15. Schulz R, Girard C, Scheckler WE. Physician satisfaction in a managed care environment. J Fam Pract 1992;34:298-304.
16. Schulz R, Schulz C. Management practices, physician autonomy, and satisfaction: evidence from mental health institutions in the Federal Republic of Germany. Med Care 1988;26:750-63.
17. Stamps PL, Cruz NT. Issues in physician satisfaction: new perspectives. Annu Arbor, Mich: Health Administration Press; 1994;31.-
18. Hughes EC. Men and their work. New York, NY: The Free Press of Glencoe; 1958.
19. Liberman. Analysis of the placebo phenomenon. J Chron Dis 1963;15:761-83.
20. Frankel RM. The laying on of hands: aspects of the organization of gaze, touch, and talk in a medical encounter. In: Todd A, Fisher S, eds. The social organization of doctor-patient communication. Norwood, NJ: Ablex Publishing Corporation; 1993;71-106.
21. Bruhn JG. The doctor’s touch: tactile communication in the doctor-patient relationship. South Med J 1978;71:1469-73.
22. Shortell SM, Waters TM, Clarke KW, Budetti PP. Physicians as double agents: maintaining trust in an era of multiple accountabilities. JAMA 1998;280:1102-8.
23. Mechanic D, Schlesinger M. The impact of managed care on patients’ trust in medical care and their physicians. JAMA 1996;275:1693-7.
24. Balint J, Shelton W. Regaining the initiative: forging a new model of the patient-physician relationship. JAMA 1996;275:887-91.
25. Mawardi BH. Physician career satisfaction: another look. Annu Conf Res Med Educ 1980;19:52-6.
26. Linn LS, Yager J, Cope D, Leake B. Health status, job satisfaction, job stress, and life satisfaction among academic and clinical faculty. JAMA 1985;254:2775-82.
27. Baker LC, Cantor JC. Physician satisfaction under managed care. Health Aff 1993;12:258-70.
28. Murray JP. Physician satisfaction with capitation patients in an academic family medicine clinic. J Fam Pract 1988;27:108-13.
29. Mechanic D. The organization of medical practice and practice orientations among physicians in prepaid and nonprepaid primary care settings. Med Care 1975;13:189-204.
30. Skolnik NS, Smith DR, Diamond J. Professional satisfaction and dissatisfaction of family physicians. J Fam Pract 1993;37:257-63.
31. Suchman AL, Roter D, Green M, Lipkin M, Jr. Physician satisfaction with primary care office visits: Collaborative Study Group of the American Academy on Physician and Patient. Med Care 1993;31:1083-92.
32. Mainous AGd, Ramsbottom-Lucier M, Rich EC. The role of clinical workload and satisfaction with workload in rural primary care physician. Arch Fam Med 1994;3:787-92.
33. Kerstein J, Pauly MV, Hillman A. Primary care physician turnover in HMOs. Health Serv Res 1994;29:17-37.
34. Sallis JF, Fortmann SP, Solomon DS, Farquhar JW. Increasing returns of physician surveys. Am J Public Health 1984;74:1043.-
35. Maheux B, Legault C, Lambert J. Increasing response rates in physicians’ mail surveys: an experimental study. Am J Public Health 1989;79:638-9.
36. Silverstein G. Physicians’ perceptions of commercial and Medicaid managed care plans: a comparison. J Health Polit Policy Law 1997;22:5-21.
37. Schulz R, Scheckler WE, Moberg DP, Johnson PR. Changing nature of physician satisfaction with health maintenance organization and fee-for-service practices. J Fam Pract 1997;45:321-30.
38. Warren MG, Weitz R, Kulis S. Physician satisfaction in a changing health care environment: the impact of challenges to professional autonomy, authority, and dominance. J Health Soc Behav 1998;39:356-67.
39. Mitchell JB. Physician participation in Medicaid revisited. Med Care 1991;29:645-53.
40. Davidson SM, Perloff JD, Kletke PR, Schiff DW, Connelly JP. Full and limited medicaid participation among pediatricians. Pediatrics 1983;72:552-9.
41. Adams EK. Effect of increased Medicaid fees on physician participation and enrollee service utilization in Tennessee, 1985-1988. Inquiry 1994;31:173-87.
42. Kerr EA, Hays RD, Mittman BS, Siu AL, Leake B, Brook RH. Primary care physicians’ satisfaction with quality of care in California capitated medical groups. JAMA 1997;278:308-12.
43. Baker LC, Cantor JC, Miles EL, Sandy LG. What makes young HMO physicians satisfied? HMO Pract 1994;8:53-7.
44. Barr JK, Steinberg MK. Professional participation in organizational decision making: physicians in HMOs. J Community Health 1983;8:160-73.
METHODS: We surveyed a random sample of primary care physicians participating in Medicaid managed care (n = 670) or traditional Medicaid (n = 670). Primary outcomes measured were physicians’ satisfaction with Medicaid managed care, traditional Medicaid, and commercial managed care. Satisfaction was measured on a 5-point Likert-type scale.
RESULTS: The response rate was 52%. Physicians participating in Medicaid managed care were less likely to be satisfied or very satisfied with Medicaid managed care (28.6%) than with commercial managed care (40%) or their previous experience with traditional Medicaid (39.7%). Among physicians participating in traditional Medicaid, 29.8% were satisfied or very satisfied with traditional Medicaid. Physicians participating in Medicaid managed care were less satisfied with clinical autonomy under that system in comparison with their previous experience with traditional Medicaid (relative difference = 10.8%, P =.001). In multiple linear regression analyses, clinical autonomy (R2 = 0.40) was a strong predictor of overall satisfaction with Medicaid managed care.
CONCLUSIONS: Enhancing physicians’ clinical autonomy may result in improved satisfaction with Medicaid managed care. State Medicaid agencies should include physician satisfaction as a measure of Medicaid managed care plans’ quality.
Although national attempts at comprehensive health care reform have faltered, many state governments have developed and implemented reform initiatives. In particular, rising Medicaid expenditures and concerns about inadequate access to care have led 49 states and the District of Columbia to experiment with Medicaid managed care.1 The proportion of the United States Medicaid population enrolled in managed care plans increased from 10% in 1991 to 48% in 1997.1 In Missouri, rising Medicaid expenditures led to the state government’s decision to enroll Aid to Families and Dependent Children (AFDC) program recipients in the state’s eastern, central, and western regions in Medicaid managed care beginning in 1995.2,3 Traditional Medicaid continued to cover the AFDC populations in Missouri’s southern and northern regions.2
There are ongoing concerns about the quality of care provided to patients in Medicaid managed care plans.4-8 Low rates of physician participation have plagued traditional Medicaid programs and have adversely affected access and quality of care.9 Increasing primary care provider participation in Medicaid managed care programs is critical for improving the quality of care provided to Medicaid recipients.9 Thus, efforts to increase physician participation may be necessary to improve the quality in those programs.
Physician satisfaction has been associated with outcomes important to patients, insurers, and state governments. In one study, physician satisfaction was correlated with patient satisfaction, increased continuity of care, lower patient no-show rates, and more reasonable charges for follow-up visits.10 Less satisfied general practitioners in England were more likely to prescribe inappropriate drugs.11 Physician satisfaction was also found to predict physician turnover12 and was associated with participation in the Ontario physician strike.13 Therefore, physician satisfaction with Medicaid managed care may affect both quality of care and physician participation in the program.
A number of studies have examined physician satisfaction in traditional fee-for-service, managed care, and institutional settings. Clinical autonomy is strongly predictive of physician satisfaction.12,14-16 Organizational autonomy (the ability of a physician to control his or her workplace),14-16 satisfaction with income,13,15,17 and satisfaction with patient relationships10,12,13 have also been found to predict physician satisfaction.
From a theoretical perspective, autonomy is central to the role functioning of physicians in the physician-patient relationship.18 This relationship has therapeutic qualities beyond those of any prescribed or recommended intervention.19-21 Health care system changes that affect physician autonomy may have an impact on the nature of the physician-patient relationship, as well as affect physician satisfaction.22-24
The purposes of our study were to assess level of physician satisfaction with Medicaid managed care in comparison with traditional Medicaid and commercial managed care and to identify the determinants of physician satisfaction with Medicaid managed care.
Methods
Design
We performed a cross-sectional survey of primary care physicians participating in Medicaid managed care or in traditional Medicaid after the implementation of Medicaid managed care in eastern and central Missouri.
Questionnaire Development
We used a literature review to identify topics for the questionnaire.12,15-17,25-33 The questionnaire included items about physicians’ satisfaction with: their ability to care for patients; their communications with third-party payers; and their satisfaction with patient relationships and reimbursement. Physicians participating in Medicaid managed care were asked to answer these questions according to their current experience with Medicaid managed care and commercial managed care and their previous experiences with traditional Medicaid. Physicians participating in traditional Medicaid were asked to answer these questions according to their current experience. A single item was used to measure overall satisfaction with type of insurance. We measured global physician satisfaction with the practice of medicine using a previously validated scale.17 All responses were measured on a 5-point scale, where 1 = very dissatisfied and 5 = very satisfied. Other items measured the level of physicians’ input into their practices’ decision-making process, the number of Medicaid managed care plans in which they participated, and demographic information. The questionnaire was pilot-tested on groups of academic and community family physicians in Missouri. After it was revised, physician members of the Missouri Division of Medical Services quality improvement project reviewed it. The final instrument included 40 questions and took apporoximately 15 minutes to complete.*
Sample Selection
We sampled 2 groups of primary care physicians. The Medicaid managed care sample was made up of physicians in eastern and central Missouri participating in Medicaid managed care. Physicians in northern and southern Missouri participating in traditional Medicaid made up the traditional Medicaid sample. We defined primary care physicians as family physicians, general practitioners, primary care pediatricians, primary care internists, and obstetrician-gynecologists. The Missouri Division of Medical Services provided 2 databases of primary care physicians participating in Medicaid managed care or traditional Medicaid. A random sample of 670 physicians was drawn from each database. The sample size was determined by assuming an 80% power to detect a 10% difference of proportions between groups, with a 2-sided P value of .05 and a projected response rate of 50%. Subsequently, each sample was matched with the state licensure database. Physicians were not eligible to participate in the survey if they were identified as specialists either by the state licensure database or by self-report.
Survey Administration
We mailed the questionnaire 3 times in the fall and winter of 1996-1997. Each mailing was sent with a cover letter signed by the principal investigator and the head of the Missouri Division of Medical Services. Postcard reminders were mailed after the first and third mailings. Methods to increase the response rate included hand-written thank you notes, hand-signed signatures on the cover letter, and the inclusion of a token gift.34,35 One mailing was sent by Priority Mail.
Statistical Analysis
We performed a principal component analysis followed by an orthogonal rotation on the 15 items comprising the satisfaction portion of the survey instrument. That was followed by a Cronbach’s a internal reliability analysis on the resulting factors.
We used chi-square tests to compare respondents with nonrespondents. Comparisons within the Medicaid managed care sample were performed using the Wilcoxon signed-rank test, and comparisons between samples were done with the Wilcoxon rank sum test. Because of multiple comparisons, we set statistical significance at P = .01. We measured bivariate relationships between possible predictor variables and overall satisfaction with Medicaid managed care using Spearman rank correlation test coefficients. We performed multivariable analysis using forward stepwise multiple linear regression. In regression analyses, dependent variables met the .05 significance level for inclusion in the models. We deleted cases with missing data from our analysis.
We performed statistical tests with SAS software (version 6.1, SAS institute, Cary, NC) or SPSS software (version 7.5.2, SPSS, Chicago, Ill).
Results
Response Rate
We excluded the following from the study: 265 specialists; 48 physicians who were not participating in Medicaid programs, not in practice, or still in training; and 57 physicians whose surveys were not deliverable. There were 505 usable responses received from the remaining 970 eligible physicians, for a response rate of 52%. The response rates in the traditional Medicaid and Medicaid managed care samples were 57% and 46%, respectively. The characteristics of the respondents are shown in Table 1. When comparing respondents with nonrespondents in the Medicaid managed care sample, pediatricians and family physicians were more likely to respond than internists and obstetricians (P <.001), and rural physicians were more likely to respond than urban physicians (P = .022). In the traditional Medicaid sample, family physicians and pediatricians were also more likely to respond than were internists or obstetricians (P = .002), and allopathic physicians were more likely to respond than were osteopathic physicians (P = .036). There were no significant differences in measures of satisfaction between responses received after the third mailing and responses received before the third mailing (P = 0.15 by Kruskall-Wallis test for both comparisons).
Questionnaire Analysis
Factor analysis confirmed 3 measures of components of physician satisfaction: clinical autonomy, organizational autonomy, and patient relationships. The clinical autonomy scale was made up of 6 items. These items measured satisfaction with the physician’s ability to hospitalize patients, prescribe medicine, choose specialists, order tests, obtain prior authorization, and control number of patient visits. Four items comprised the organizational autonomy scale, which measured satisfaction with communications with insurers about billing and payment, enrollment, prior authorization, and utilization review. Three items comprised the patient relationship scale: satisfaction with continuity, the quality of patient relationships, and the intellectual challenge posed by patient problems. The internal consistency of these 3 scales measured by Cronbach’s a ranged from 0.76 to 0.90. Two reimbursement items were grouped together on factor analysis and were treated as a scale.
Overall Satisfaction by Insurance Type
Figure 1 The Figure shows that physicians were more likely to respond that they were satisfied or very satisfied with commercial managed care (40%) and their previous experience with traditional Medicaid (39.7%) than with Medicaid managed care (28.6%). There was no difference in satisfaction between Medicaid managed care and satisfaction with traditional Medicaid measured in the traditional Medicaid sample (29.8%).
Physician Satisfaction with Commercial Managed Care
Physicians expressed neutrality toward clinical autonomy (mean = 2.95, standard deviation [SD] = 0.84), neutrality or slight dissatisfaction with organizational autonomy (mean = 2.63, SD = 0.89) and reimbursement (mean = 2.49, SD = 0.98), and neutrality to satisfaction with patient relationships (mean = 3.22, SD = 0.84).
Comparison of Physician Satisfaction Components Among Insurance Types
We compared scores on satisfaction scales by determining the relative mean difference between various insurance types and Medicaid managed care. This was calculated by subtracting mean satisfaction scale scores for different insurance types from mean Medicaid managed care scores and dividing that result by the maximum possible range for that scale. For these comparisons, experienced researchers’ consensus opinion was that a 10% difference would be meaningful. In the Medicaid managed care sample, satisfaction with reimbursement was less with Medicaid managed care than with commercial managed care Table 2. Physicians expressed lower satisfaction with clinical autonomy in a Medicaid managed care system than with their previous traditional Medicaid experience. Satisfaction with Medicaid managed care reimbursement was lower than satisfaction with traditional Medicaid reimbursement measured in the traditional Medicaid sample.
Predictors of Satisfaction
Physicians’ satisfaction with components of satisfaction in a Medicaid managed care system was correlated with overall satisfaction with Medicaid managed care Table 3. Overall satisfaction was also correlated with comfort with workload, caring for patients less than 30 hours per week, being a woman, and being a family physician. Satisfaction was inversely correlated with hours worked, caring for patients more than 60 hours per week, and the number of Medicaid managed care plans in which a physician participated.
We developed 3 stepwise multiple linear regression models using overall satisfaction with Medicaid managed care as the dependent variable. We built 3 models to explore the effects of any instability in the regression models due to missing data. For model 1, satisfaction components were independent variables, and for model 2 all variables that were significant in bivariate analysis were independent variables. Independent variables entered into model 3 were global satisfaction with the practice of medicine, clinical autonomy, workload, number of Medicaid managed care plans in which physicians participated, and whether a physician worked more than 60 hours or less than 30 hours per week in patient care. The resulting models were similar: only model 3 is reported here.* In this regression model, satisfaction with clinical autonomy was the most important predictor of overall satisfaction with Medicaid managed care, accounting for 40% of the variance in overall satisfaction Table 4. Global satisfaction with the practice of medicine, satisfaction with reimbursement, and working less than 30 hours per week in direct patient care were less important predictors of overall satisfaction.
Discussion
This is the first study to measure physician satisfaction with Medicaid managed care in comparison with traditional Medicaid and commercial managed care. Although physicians were less satisfied with Medicaid managed care in comparison with commercial managed care and their previous traditional Medicaid experience, they were not less satisfied with Medicaid managed care in comparison to physician satisfaction with traditional Medicaid measured in the traditional Medicaid sample. This suggests that physicians’ satisfaction level with commercial managed care did not decline from their previous level of satisfaction with traditional Medicaid. The higher satisfaction reported for physicians’ previous experience with traditional Medicaid may be related to implementation of the new Medicaid managed care program.
Physicians’ satisfaction level with clinical autonomy in the Medicaid managed care and commercial managed care systems was similar, indicating that physicians view the methods used to affect their behavior in commercial and Medicaid managed care programs similarly. This finding is supported by the results of a study of Arizona’s Medicaid managed care program.36
Our finding that clinical autonomy was a strong predictor of overall satisfaction with Medicaid managed care is consistent with the findings of other recent studies of physician satisfaction37,38 and theoretical work that stresses the importance of autonomy in the development of professional identity and functioning.18 The strength of this association in our study (R2 = .40) highlights the importance physicians place on maintaining their authority to provide the care they believe will best serve their patients. In contrast, satisfaction with reimbursement was a weak predictor of overall satisfaction (R2 = .02). State Medicaid agencies attempting to improve physician satisfaction with Medicaid managed care should work first to improve clinical autonomy while providing reasonable reimbursement. In traditional Medicaid programs, both reimbursement and decreased interference with clinical decision making have been associated with increased rates of participation.39-41 Increasing physician satisfaction with clinical autonomy in Medicaid managed care plans may result in increased rates of physician participation.
Efforts to increase physician satisfaction with clinical autonomy may also increase quality of care.10,11 Low levels of satisfaction with clinical autonomy in managed care systems has been associated with a perception of lower quality of care.42 However, participation in managed care plans does not always result in lower satisfaction with clinical autonomy. Several studies have shown a strong association between participation in managed care plans and high levels of satisfaction with clinical autonomy.15,37,43 Physician involvement in decision-making processes that affect their daily work is related to increased satisfaction with autonomy.16,44 Thus, involving physicians and physician organizations in the formulation of guidelines that will affect physician behavior in Medicaid managed care plans could promote satisfaction with clinical autonomy while improving quality and controlling costs.
The strengths of this study include the development of 3 scales that have good face validity and content validity and high internal reliability. Our survey included physicians participating in traditional Medicaid, as well as Medicaid managed care. Most importantly, this is one of the first studies to address physician satisfaction early in the national experiment with Medicaid managed care plans.
Limitations
Our study may be limited by a response rate of 52%. Although we found no difference between late and early responders on measures of satisfaction and intention to continue participation, our results may be biased toward negative responders. This study was also limited to Missouri physicians. It may not be possible to generalize our findings to other states. Furthermore, we measured satisfaction early in the implementation of Medicaid managed care. Physician satisfaction may change as the Medicaid managed care and commercial markets evolve.
Conclusions
There may be a tendency for some to view an exploration of the effect of changes in the health care delivery system on physician autonomy and satisfaction as a case of professional self-interest, especially in the current climate where physicians are increasingly and publicly critical of the impact of managed care on quality of care. However, the larger sociological perspective—drawn from decades of empirical and theoretical research that links autonomy to physician satisfaction, the therapeutic relationship, and quality of care—shows the importance of autonomy and satisfaction.10,11,18,20,21 Finding ways to promote physician autonomy and satisfaction in Medicaid managed care settings is necessary to maintain the therapeutic relationship that forms the basis of high-quality care. The quality indicators of Medicaid managed care plan performance should include a measurement of physician satisfaction and autonomy.
Acknowledgments
Dr Gazewood was partially funded by the National Research Service Award 5-T32-P#17001-08 and by a Harrison Teaching Professorship of Generalist Medicine. The Missouri Division of Medical Services and the University of Missouri-Columbia Department of Family and Community Medicine small grant program also provided funding.
The authors acknowledge Paula McDonald for research assistance and secretarial support.
METHODS: We surveyed a random sample of primary care physicians participating in Medicaid managed care (n = 670) or traditional Medicaid (n = 670). Primary outcomes measured were physicians’ satisfaction with Medicaid managed care, traditional Medicaid, and commercial managed care. Satisfaction was measured on a 5-point Likert-type scale.
RESULTS: The response rate was 52%. Physicians participating in Medicaid managed care were less likely to be satisfied or very satisfied with Medicaid managed care (28.6%) than with commercial managed care (40%) or their previous experience with traditional Medicaid (39.7%). Among physicians participating in traditional Medicaid, 29.8% were satisfied or very satisfied with traditional Medicaid. Physicians participating in Medicaid managed care were less satisfied with clinical autonomy under that system in comparison with their previous experience with traditional Medicaid (relative difference = 10.8%, P =.001). In multiple linear regression analyses, clinical autonomy (R2 = 0.40) was a strong predictor of overall satisfaction with Medicaid managed care.
CONCLUSIONS: Enhancing physicians’ clinical autonomy may result in improved satisfaction with Medicaid managed care. State Medicaid agencies should include physician satisfaction as a measure of Medicaid managed care plans’ quality.
Although national attempts at comprehensive health care reform have faltered, many state governments have developed and implemented reform initiatives. In particular, rising Medicaid expenditures and concerns about inadequate access to care have led 49 states and the District of Columbia to experiment with Medicaid managed care.1 The proportion of the United States Medicaid population enrolled in managed care plans increased from 10% in 1991 to 48% in 1997.1 In Missouri, rising Medicaid expenditures led to the state government’s decision to enroll Aid to Families and Dependent Children (AFDC) program recipients in the state’s eastern, central, and western regions in Medicaid managed care beginning in 1995.2,3 Traditional Medicaid continued to cover the AFDC populations in Missouri’s southern and northern regions.2
There are ongoing concerns about the quality of care provided to patients in Medicaid managed care plans.4-8 Low rates of physician participation have plagued traditional Medicaid programs and have adversely affected access and quality of care.9 Increasing primary care provider participation in Medicaid managed care programs is critical for improving the quality of care provided to Medicaid recipients.9 Thus, efforts to increase physician participation may be necessary to improve the quality in those programs.
Physician satisfaction has been associated with outcomes important to patients, insurers, and state governments. In one study, physician satisfaction was correlated with patient satisfaction, increased continuity of care, lower patient no-show rates, and more reasonable charges for follow-up visits.10 Less satisfied general practitioners in England were more likely to prescribe inappropriate drugs.11 Physician satisfaction was also found to predict physician turnover12 and was associated with participation in the Ontario physician strike.13 Therefore, physician satisfaction with Medicaid managed care may affect both quality of care and physician participation in the program.
A number of studies have examined physician satisfaction in traditional fee-for-service, managed care, and institutional settings. Clinical autonomy is strongly predictive of physician satisfaction.12,14-16 Organizational autonomy (the ability of a physician to control his or her workplace),14-16 satisfaction with income,13,15,17 and satisfaction with patient relationships10,12,13 have also been found to predict physician satisfaction.
From a theoretical perspective, autonomy is central to the role functioning of physicians in the physician-patient relationship.18 This relationship has therapeutic qualities beyond those of any prescribed or recommended intervention.19-21 Health care system changes that affect physician autonomy may have an impact on the nature of the physician-patient relationship, as well as affect physician satisfaction.22-24
The purposes of our study were to assess level of physician satisfaction with Medicaid managed care in comparison with traditional Medicaid and commercial managed care and to identify the determinants of physician satisfaction with Medicaid managed care.
Methods
Design
We performed a cross-sectional survey of primary care physicians participating in Medicaid managed care or in traditional Medicaid after the implementation of Medicaid managed care in eastern and central Missouri.
Questionnaire Development
We used a literature review to identify topics for the questionnaire.12,15-17,25-33 The questionnaire included items about physicians’ satisfaction with: their ability to care for patients; their communications with third-party payers; and their satisfaction with patient relationships and reimbursement. Physicians participating in Medicaid managed care were asked to answer these questions according to their current experience with Medicaid managed care and commercial managed care and their previous experiences with traditional Medicaid. Physicians participating in traditional Medicaid were asked to answer these questions according to their current experience. A single item was used to measure overall satisfaction with type of insurance. We measured global physician satisfaction with the practice of medicine using a previously validated scale.17 All responses were measured on a 5-point scale, where 1 = very dissatisfied and 5 = very satisfied. Other items measured the level of physicians’ input into their practices’ decision-making process, the number of Medicaid managed care plans in which they participated, and demographic information. The questionnaire was pilot-tested on groups of academic and community family physicians in Missouri. After it was revised, physician members of the Missouri Division of Medical Services quality improvement project reviewed it. The final instrument included 40 questions and took apporoximately 15 minutes to complete.*
Sample Selection
We sampled 2 groups of primary care physicians. The Medicaid managed care sample was made up of physicians in eastern and central Missouri participating in Medicaid managed care. Physicians in northern and southern Missouri participating in traditional Medicaid made up the traditional Medicaid sample. We defined primary care physicians as family physicians, general practitioners, primary care pediatricians, primary care internists, and obstetrician-gynecologists. The Missouri Division of Medical Services provided 2 databases of primary care physicians participating in Medicaid managed care or traditional Medicaid. A random sample of 670 physicians was drawn from each database. The sample size was determined by assuming an 80% power to detect a 10% difference of proportions between groups, with a 2-sided P value of .05 and a projected response rate of 50%. Subsequently, each sample was matched with the state licensure database. Physicians were not eligible to participate in the survey if they were identified as specialists either by the state licensure database or by self-report.
Survey Administration
We mailed the questionnaire 3 times in the fall and winter of 1996-1997. Each mailing was sent with a cover letter signed by the principal investigator and the head of the Missouri Division of Medical Services. Postcard reminders were mailed after the first and third mailings. Methods to increase the response rate included hand-written thank you notes, hand-signed signatures on the cover letter, and the inclusion of a token gift.34,35 One mailing was sent by Priority Mail.
Statistical Analysis
We performed a principal component analysis followed by an orthogonal rotation on the 15 items comprising the satisfaction portion of the survey instrument. That was followed by a Cronbach’s a internal reliability analysis on the resulting factors.
We used chi-square tests to compare respondents with nonrespondents. Comparisons within the Medicaid managed care sample were performed using the Wilcoxon signed-rank test, and comparisons between samples were done with the Wilcoxon rank sum test. Because of multiple comparisons, we set statistical significance at P = .01. We measured bivariate relationships between possible predictor variables and overall satisfaction with Medicaid managed care using Spearman rank correlation test coefficients. We performed multivariable analysis using forward stepwise multiple linear regression. In regression analyses, dependent variables met the .05 significance level for inclusion in the models. We deleted cases with missing data from our analysis.
We performed statistical tests with SAS software (version 6.1, SAS institute, Cary, NC) or SPSS software (version 7.5.2, SPSS, Chicago, Ill).
Results
Response Rate
We excluded the following from the study: 265 specialists; 48 physicians who were not participating in Medicaid programs, not in practice, or still in training; and 57 physicians whose surveys were not deliverable. There were 505 usable responses received from the remaining 970 eligible physicians, for a response rate of 52%. The response rates in the traditional Medicaid and Medicaid managed care samples were 57% and 46%, respectively. The characteristics of the respondents are shown in Table 1. When comparing respondents with nonrespondents in the Medicaid managed care sample, pediatricians and family physicians were more likely to respond than internists and obstetricians (P <.001), and rural physicians were more likely to respond than urban physicians (P = .022). In the traditional Medicaid sample, family physicians and pediatricians were also more likely to respond than were internists or obstetricians (P = .002), and allopathic physicians were more likely to respond than were osteopathic physicians (P = .036). There were no significant differences in measures of satisfaction between responses received after the third mailing and responses received before the third mailing (P = 0.15 by Kruskall-Wallis test for both comparisons).
Questionnaire Analysis
Factor analysis confirmed 3 measures of components of physician satisfaction: clinical autonomy, organizational autonomy, and patient relationships. The clinical autonomy scale was made up of 6 items. These items measured satisfaction with the physician’s ability to hospitalize patients, prescribe medicine, choose specialists, order tests, obtain prior authorization, and control number of patient visits. Four items comprised the organizational autonomy scale, which measured satisfaction with communications with insurers about billing and payment, enrollment, prior authorization, and utilization review. Three items comprised the patient relationship scale: satisfaction with continuity, the quality of patient relationships, and the intellectual challenge posed by patient problems. The internal consistency of these 3 scales measured by Cronbach’s a ranged from 0.76 to 0.90. Two reimbursement items were grouped together on factor analysis and were treated as a scale.
Overall Satisfaction by Insurance Type
Figure 1 The Figure shows that physicians were more likely to respond that they were satisfied or very satisfied with commercial managed care (40%) and their previous experience with traditional Medicaid (39.7%) than with Medicaid managed care (28.6%). There was no difference in satisfaction between Medicaid managed care and satisfaction with traditional Medicaid measured in the traditional Medicaid sample (29.8%).
Physician Satisfaction with Commercial Managed Care
Physicians expressed neutrality toward clinical autonomy (mean = 2.95, standard deviation [SD] = 0.84), neutrality or slight dissatisfaction with organizational autonomy (mean = 2.63, SD = 0.89) and reimbursement (mean = 2.49, SD = 0.98), and neutrality to satisfaction with patient relationships (mean = 3.22, SD = 0.84).
Comparison of Physician Satisfaction Components Among Insurance Types
We compared scores on satisfaction scales by determining the relative mean difference between various insurance types and Medicaid managed care. This was calculated by subtracting mean satisfaction scale scores for different insurance types from mean Medicaid managed care scores and dividing that result by the maximum possible range for that scale. For these comparisons, experienced researchers’ consensus opinion was that a 10% difference would be meaningful. In the Medicaid managed care sample, satisfaction with reimbursement was less with Medicaid managed care than with commercial managed care Table 2. Physicians expressed lower satisfaction with clinical autonomy in a Medicaid managed care system than with their previous traditional Medicaid experience. Satisfaction with Medicaid managed care reimbursement was lower than satisfaction with traditional Medicaid reimbursement measured in the traditional Medicaid sample.
Predictors of Satisfaction
Physicians’ satisfaction with components of satisfaction in a Medicaid managed care system was correlated with overall satisfaction with Medicaid managed care Table 3. Overall satisfaction was also correlated with comfort with workload, caring for patients less than 30 hours per week, being a woman, and being a family physician. Satisfaction was inversely correlated with hours worked, caring for patients more than 60 hours per week, and the number of Medicaid managed care plans in which a physician participated.
We developed 3 stepwise multiple linear regression models using overall satisfaction with Medicaid managed care as the dependent variable. We built 3 models to explore the effects of any instability in the regression models due to missing data. For model 1, satisfaction components were independent variables, and for model 2 all variables that were significant in bivariate analysis were independent variables. Independent variables entered into model 3 were global satisfaction with the practice of medicine, clinical autonomy, workload, number of Medicaid managed care plans in which physicians participated, and whether a physician worked more than 60 hours or less than 30 hours per week in patient care. The resulting models were similar: only model 3 is reported here.* In this regression model, satisfaction with clinical autonomy was the most important predictor of overall satisfaction with Medicaid managed care, accounting for 40% of the variance in overall satisfaction Table 4. Global satisfaction with the practice of medicine, satisfaction with reimbursement, and working less than 30 hours per week in direct patient care were less important predictors of overall satisfaction.
Discussion
This is the first study to measure physician satisfaction with Medicaid managed care in comparison with traditional Medicaid and commercial managed care. Although physicians were less satisfied with Medicaid managed care in comparison with commercial managed care and their previous traditional Medicaid experience, they were not less satisfied with Medicaid managed care in comparison to physician satisfaction with traditional Medicaid measured in the traditional Medicaid sample. This suggests that physicians’ satisfaction level with commercial managed care did not decline from their previous level of satisfaction with traditional Medicaid. The higher satisfaction reported for physicians’ previous experience with traditional Medicaid may be related to implementation of the new Medicaid managed care program.
Physicians’ satisfaction level with clinical autonomy in the Medicaid managed care and commercial managed care systems was similar, indicating that physicians view the methods used to affect their behavior in commercial and Medicaid managed care programs similarly. This finding is supported by the results of a study of Arizona’s Medicaid managed care program.36
Our finding that clinical autonomy was a strong predictor of overall satisfaction with Medicaid managed care is consistent with the findings of other recent studies of physician satisfaction37,38 and theoretical work that stresses the importance of autonomy in the development of professional identity and functioning.18 The strength of this association in our study (R2 = .40) highlights the importance physicians place on maintaining their authority to provide the care they believe will best serve their patients. In contrast, satisfaction with reimbursement was a weak predictor of overall satisfaction (R2 = .02). State Medicaid agencies attempting to improve physician satisfaction with Medicaid managed care should work first to improve clinical autonomy while providing reasonable reimbursement. In traditional Medicaid programs, both reimbursement and decreased interference with clinical decision making have been associated with increased rates of participation.39-41 Increasing physician satisfaction with clinical autonomy in Medicaid managed care plans may result in increased rates of physician participation.
Efforts to increase physician satisfaction with clinical autonomy may also increase quality of care.10,11 Low levels of satisfaction with clinical autonomy in managed care systems has been associated with a perception of lower quality of care.42 However, participation in managed care plans does not always result in lower satisfaction with clinical autonomy. Several studies have shown a strong association between participation in managed care plans and high levels of satisfaction with clinical autonomy.15,37,43 Physician involvement in decision-making processes that affect their daily work is related to increased satisfaction with autonomy.16,44 Thus, involving physicians and physician organizations in the formulation of guidelines that will affect physician behavior in Medicaid managed care plans could promote satisfaction with clinical autonomy while improving quality and controlling costs.
The strengths of this study include the development of 3 scales that have good face validity and content validity and high internal reliability. Our survey included physicians participating in traditional Medicaid, as well as Medicaid managed care. Most importantly, this is one of the first studies to address physician satisfaction early in the national experiment with Medicaid managed care plans.
Limitations
Our study may be limited by a response rate of 52%. Although we found no difference between late and early responders on measures of satisfaction and intention to continue participation, our results may be biased toward negative responders. This study was also limited to Missouri physicians. It may not be possible to generalize our findings to other states. Furthermore, we measured satisfaction early in the implementation of Medicaid managed care. Physician satisfaction may change as the Medicaid managed care and commercial markets evolve.
Conclusions
There may be a tendency for some to view an exploration of the effect of changes in the health care delivery system on physician autonomy and satisfaction as a case of professional self-interest, especially in the current climate where physicians are increasingly and publicly critical of the impact of managed care on quality of care. However, the larger sociological perspective—drawn from decades of empirical and theoretical research that links autonomy to physician satisfaction, the therapeutic relationship, and quality of care—shows the importance of autonomy and satisfaction.10,11,18,20,21 Finding ways to promote physician autonomy and satisfaction in Medicaid managed care settings is necessary to maintain the therapeutic relationship that forms the basis of high-quality care. The quality indicators of Medicaid managed care plan performance should include a measurement of physician satisfaction and autonomy.
Acknowledgments
Dr Gazewood was partially funded by the National Research Service Award 5-T32-P#17001-08 and by a Harrison Teaching Professorship of Generalist Medicine. The Missouri Division of Medical Services and the University of Missouri-Columbia Department of Family and Community Medicine small grant program also provided funding.
The authors acknowledge Paula McDonald for research assistance and secretarial support.
1. Iglehart JK. The American health care system: Medicaid. N Engl J Med 1999;340:403-8.
2. Jenkins ER. Managed care plus (MC+): the wave of the future for Missouri Medicaid? Mo Med 1995;92:222-3.
3. Coughlin TA, Ku L, Holahan J, Heslam D, Winterbottom C. State responses to the Medicaid spending crisis: 1988 to 1992. J Health Polit Policy Law 1994;19:837-64.
4. Mirvis DM, Chang CF, Hall CJ, Zaar GT, Applegate WB. TennCare: health system reform for Tennessee. JAMA 1995;274:1235-41.
5. Dallek G. A consumer advocate on Medicaid managed care. Health Aff 1996;15:174-7.
6. Halvorson GC. An HMO chief executive officer on Medicaid managed care. Health Aff 1996;15:170-1.
7. Gold M, Sparer M, Chu K. Medicaid managed care: lessons from five states. Health Aff 1996;15:153-66.
8. Chang CF, Kiser LJ, Bailey JE, et al. Tennessee’s failed managed care program for mental health and substance abuse services. JAMA 1998;279:864-9.
9. Freund DA, Hurley RE. Medicaid managed care: contribution to issues of health reform. Ann Rev Public Health 1995;16:473-95.
10. Linn LS, Brook RH, Clark VA, Davies AR, Fink A, Kosecoff J. Physician and patient satisfaction as factors related to the organization of internal medicine group practices. Med Care 1985;23:1171-8.
11. Melville A. Job satisfaction in general practice: implications for prescribing. Soc Sci Med 1980;14A:495-9.
12. Lichtenstein R. Measuring the job satisfaction of physicians in organized settings. Med Care 1984;22:56-68.
13. Kravitz RL, Linn LS, Shapiro MF. Physician satisfaction under the Ontario Health Insurance Plan. Med Care 1990;28:502-12.
14. Stamps PL, Piedmont EB, Slavitt DB, Haase AM. Measurement of work satisfaction among health professionals. Med Care 1978;16:337-52.
15. Schulz R, Girard C, Scheckler WE. Physician satisfaction in a managed care environment. J Fam Pract 1992;34:298-304.
16. Schulz R, Schulz C. Management practices, physician autonomy, and satisfaction: evidence from mental health institutions in the Federal Republic of Germany. Med Care 1988;26:750-63.
17. Stamps PL, Cruz NT. Issues in physician satisfaction: new perspectives. Annu Arbor, Mich: Health Administration Press; 1994;31.-
18. Hughes EC. Men and their work. New York, NY: The Free Press of Glencoe; 1958.
19. Liberman. Analysis of the placebo phenomenon. J Chron Dis 1963;15:761-83.
20. Frankel RM. The laying on of hands: aspects of the organization of gaze, touch, and talk in a medical encounter. In: Todd A, Fisher S, eds. The social organization of doctor-patient communication. Norwood, NJ: Ablex Publishing Corporation; 1993;71-106.
21. Bruhn JG. The doctor’s touch: tactile communication in the doctor-patient relationship. South Med J 1978;71:1469-73.
22. Shortell SM, Waters TM, Clarke KW, Budetti PP. Physicians as double agents: maintaining trust in an era of multiple accountabilities. JAMA 1998;280:1102-8.
23. Mechanic D, Schlesinger M. The impact of managed care on patients’ trust in medical care and their physicians. JAMA 1996;275:1693-7.
24. Balint J, Shelton W. Regaining the initiative: forging a new model of the patient-physician relationship. JAMA 1996;275:887-91.
25. Mawardi BH. Physician career satisfaction: another look. Annu Conf Res Med Educ 1980;19:52-6.
26. Linn LS, Yager J, Cope D, Leake B. Health status, job satisfaction, job stress, and life satisfaction among academic and clinical faculty. JAMA 1985;254:2775-82.
27. Baker LC, Cantor JC. Physician satisfaction under managed care. Health Aff 1993;12:258-70.
28. Murray JP. Physician satisfaction with capitation patients in an academic family medicine clinic. J Fam Pract 1988;27:108-13.
29. Mechanic D. The organization of medical practice and practice orientations among physicians in prepaid and nonprepaid primary care settings. Med Care 1975;13:189-204.
30. Skolnik NS, Smith DR, Diamond J. Professional satisfaction and dissatisfaction of family physicians. J Fam Pract 1993;37:257-63.
31. Suchman AL, Roter D, Green M, Lipkin M, Jr. Physician satisfaction with primary care office visits: Collaborative Study Group of the American Academy on Physician and Patient. Med Care 1993;31:1083-92.
32. Mainous AGd, Ramsbottom-Lucier M, Rich EC. The role of clinical workload and satisfaction with workload in rural primary care physician. Arch Fam Med 1994;3:787-92.
33. Kerstein J, Pauly MV, Hillman A. Primary care physician turnover in HMOs. Health Serv Res 1994;29:17-37.
34. Sallis JF, Fortmann SP, Solomon DS, Farquhar JW. Increasing returns of physician surveys. Am J Public Health 1984;74:1043.-
35. Maheux B, Legault C, Lambert J. Increasing response rates in physicians’ mail surveys: an experimental study. Am J Public Health 1989;79:638-9.
36. Silverstein G. Physicians’ perceptions of commercial and Medicaid managed care plans: a comparison. J Health Polit Policy Law 1997;22:5-21.
37. Schulz R, Scheckler WE, Moberg DP, Johnson PR. Changing nature of physician satisfaction with health maintenance organization and fee-for-service practices. J Fam Pract 1997;45:321-30.
38. Warren MG, Weitz R, Kulis S. Physician satisfaction in a changing health care environment: the impact of challenges to professional autonomy, authority, and dominance. J Health Soc Behav 1998;39:356-67.
39. Mitchell JB. Physician participation in Medicaid revisited. Med Care 1991;29:645-53.
40. Davidson SM, Perloff JD, Kletke PR, Schiff DW, Connelly JP. Full and limited medicaid participation among pediatricians. Pediatrics 1983;72:552-9.
41. Adams EK. Effect of increased Medicaid fees on physician participation and enrollee service utilization in Tennessee, 1985-1988. Inquiry 1994;31:173-87.
42. Kerr EA, Hays RD, Mittman BS, Siu AL, Leake B, Brook RH. Primary care physicians’ satisfaction with quality of care in California capitated medical groups. JAMA 1997;278:308-12.
43. Baker LC, Cantor JC, Miles EL, Sandy LG. What makes young HMO physicians satisfied? HMO Pract 1994;8:53-7.
44. Barr JK, Steinberg MK. Professional participation in organizational decision making: physicians in HMOs. J Community Health 1983;8:160-73.
1. Iglehart JK. The American health care system: Medicaid. N Engl J Med 1999;340:403-8.
2. Jenkins ER. Managed care plus (MC+): the wave of the future for Missouri Medicaid? Mo Med 1995;92:222-3.
3. Coughlin TA, Ku L, Holahan J, Heslam D, Winterbottom C. State responses to the Medicaid spending crisis: 1988 to 1992. J Health Polit Policy Law 1994;19:837-64.
4. Mirvis DM, Chang CF, Hall CJ, Zaar GT, Applegate WB. TennCare: health system reform for Tennessee. JAMA 1995;274:1235-41.
5. Dallek G. A consumer advocate on Medicaid managed care. Health Aff 1996;15:174-7.
6. Halvorson GC. An HMO chief executive officer on Medicaid managed care. Health Aff 1996;15:170-1.
7. Gold M, Sparer M, Chu K. Medicaid managed care: lessons from five states. Health Aff 1996;15:153-66.
8. Chang CF, Kiser LJ, Bailey JE, et al. Tennessee’s failed managed care program for mental health and substance abuse services. JAMA 1998;279:864-9.
9. Freund DA, Hurley RE. Medicaid managed care: contribution to issues of health reform. Ann Rev Public Health 1995;16:473-95.
10. Linn LS, Brook RH, Clark VA, Davies AR, Fink A, Kosecoff J. Physician and patient satisfaction as factors related to the organization of internal medicine group practices. Med Care 1985;23:1171-8.
11. Melville A. Job satisfaction in general practice: implications for prescribing. Soc Sci Med 1980;14A:495-9.
12. Lichtenstein R. Measuring the job satisfaction of physicians in organized settings. Med Care 1984;22:56-68.
13. Kravitz RL, Linn LS, Shapiro MF. Physician satisfaction under the Ontario Health Insurance Plan. Med Care 1990;28:502-12.
14. Stamps PL, Piedmont EB, Slavitt DB, Haase AM. Measurement of work satisfaction among health professionals. Med Care 1978;16:337-52.
15. Schulz R, Girard C, Scheckler WE. Physician satisfaction in a managed care environment. J Fam Pract 1992;34:298-304.
16. Schulz R, Schulz C. Management practices, physician autonomy, and satisfaction: evidence from mental health institutions in the Federal Republic of Germany. Med Care 1988;26:750-63.
17. Stamps PL, Cruz NT. Issues in physician satisfaction: new perspectives. Annu Arbor, Mich: Health Administration Press; 1994;31.-
18. Hughes EC. Men and their work. New York, NY: The Free Press of Glencoe; 1958.
19. Liberman. Analysis of the placebo phenomenon. J Chron Dis 1963;15:761-83.
20. Frankel RM. The laying on of hands: aspects of the organization of gaze, touch, and talk in a medical encounter. In: Todd A, Fisher S, eds. The social organization of doctor-patient communication. Norwood, NJ: Ablex Publishing Corporation; 1993;71-106.
21. Bruhn JG. The doctor’s touch: tactile communication in the doctor-patient relationship. South Med J 1978;71:1469-73.
22. Shortell SM, Waters TM, Clarke KW, Budetti PP. Physicians as double agents: maintaining trust in an era of multiple accountabilities. JAMA 1998;280:1102-8.
23. Mechanic D, Schlesinger M. The impact of managed care on patients’ trust in medical care and their physicians. JAMA 1996;275:1693-7.
24. Balint J, Shelton W. Regaining the initiative: forging a new model of the patient-physician relationship. JAMA 1996;275:887-91.
25. Mawardi BH. Physician career satisfaction: another look. Annu Conf Res Med Educ 1980;19:52-6.
26. Linn LS, Yager J, Cope D, Leake B. Health status, job satisfaction, job stress, and life satisfaction among academic and clinical faculty. JAMA 1985;254:2775-82.
27. Baker LC, Cantor JC. Physician satisfaction under managed care. Health Aff 1993;12:258-70.
28. Murray JP. Physician satisfaction with capitation patients in an academic family medicine clinic. J Fam Pract 1988;27:108-13.
29. Mechanic D. The organization of medical practice and practice orientations among physicians in prepaid and nonprepaid primary care settings. Med Care 1975;13:189-204.
30. Skolnik NS, Smith DR, Diamond J. Professional satisfaction and dissatisfaction of family physicians. J Fam Pract 1993;37:257-63.
31. Suchman AL, Roter D, Green M, Lipkin M, Jr. Physician satisfaction with primary care office visits: Collaborative Study Group of the American Academy on Physician and Patient. Med Care 1993;31:1083-92.
32. Mainous AGd, Ramsbottom-Lucier M, Rich EC. The role of clinical workload and satisfaction with workload in rural primary care physician. Arch Fam Med 1994;3:787-92.
33. Kerstein J, Pauly MV, Hillman A. Primary care physician turnover in HMOs. Health Serv Res 1994;29:17-37.
34. Sallis JF, Fortmann SP, Solomon DS, Farquhar JW. Increasing returns of physician surveys. Am J Public Health 1984;74:1043.-
35. Maheux B, Legault C, Lambert J. Increasing response rates in physicians’ mail surveys: an experimental study. Am J Public Health 1989;79:638-9.
36. Silverstein G. Physicians’ perceptions of commercial and Medicaid managed care plans: a comparison. J Health Polit Policy Law 1997;22:5-21.
37. Schulz R, Scheckler WE, Moberg DP, Johnson PR. Changing nature of physician satisfaction with health maintenance organization and fee-for-service practices. J Fam Pract 1997;45:321-30.
38. Warren MG, Weitz R, Kulis S. Physician satisfaction in a changing health care environment: the impact of challenges to professional autonomy, authority, and dominance. J Health Soc Behav 1998;39:356-67.
39. Mitchell JB. Physician participation in Medicaid revisited. Med Care 1991;29:645-53.
40. Davidson SM, Perloff JD, Kletke PR, Schiff DW, Connelly JP. Full and limited medicaid participation among pediatricians. Pediatrics 1983;72:552-9.
41. Adams EK. Effect of increased Medicaid fees on physician participation and enrollee service utilization in Tennessee, 1985-1988. Inquiry 1994;31:173-87.
42. Kerr EA, Hays RD, Mittman BS, Siu AL, Leake B, Brook RH. Primary care physicians’ satisfaction with quality of care in California capitated medical groups. JAMA 1997;278:308-12.
43. Baker LC, Cantor JC, Miles EL, Sandy LG. What makes young HMO physicians satisfied? HMO Pract 1994;8:53-7.
44. Barr JK, Steinberg MK. Professional participation in organizational decision making: physicians in HMOs. J Community Health 1983;8:160-73.