INTRODUCTION

Career satisfaction among physicians across specialties has been found to be positively associated with self-reported income.1,2 However, studies examining a wide variety of non-health care occupations reveal that the perception of one’s pay as fair may be a better predictor of important work- and life-related outcomes than the actual amount of compensation.3,4 No studies, to our knowledge, have examined physicians’ perceptions of pay fairness and its relationship to relevant outcomes such as professional satisfaction and physical health.

A large body of occupational research has revealed the impact of perceived pay fairness on employees and their employers.5,6,7,–8 Perceptions of pay fairness are grounded in objective measures (e.g., extent to which one’s pay aligns with comparable rates in the external job market and among coworkers with similar skills and experience) and subjective considerations (e.g., feeling that one’s pay reflects the value of one's contribution in the workplace).9 Individuals who perceive their pay as fair are more likely to be satisfied at work and less likely to look for alternate employment options.10 Organizations with employees who believe that their pay is fair also generally outperform their peers.11,12,–13 However, it may be less apparent why individuals who feel they are inappropriately rewarded for their work effort would have worse health, which can lead to increased work absenteeism. Informed by the effort-reward imbalance model,14,15 the physiological impact of stressful psychosocial work environments on health is operative when individuals perceive their pay to be unfair.16 Perception of pay unfairness, for example, has been found to be associated with low heart rate variability, a stress-related indicator of functional impairment of the cardiovascular system, which increases the probability of coronary artery disease.17

For the first time, a majority of practicing physicians in the United States are employees rather than owners of their clinical practices.18,19 At the same time, more physicians, particularly those providing primary care, are considering or have moved into concierge practice in response to treatment time constraints and work effort-compensation concerns.20,21 In light of structural changes in the physician labor market, we report on a national survey of practicing physicians that assessed their perceptions of pay fairness. Specifically, we hypothesized that physicians who think their compensation is fair will be more satisfied at work, be less inclined to leave their medical practice, and report having better overall health.

METHODS

Study Participants and Data Collection

We selected a random sample of 4000 physicians, inclusive of all medical specialties, from the American Medical Association (AMA) Physician Masterfile, which includes data on all US physicians. Sampled physicians were listed as currently practicing, with resident physicians excluded. This sample was selected for a larger study evaluating physicians’ work motivation and their practice environments.

Between October 2014 and May 2015, we mailed three rounds of a self-administered survey to sampled physicians. A $10 bill was included in the second mailing to encourage participation. Subsequent to the survey mailings, 411 physicians were found to be study-ineligible (71 no longer practicing medicine, 54 no longer at the clinical practice on record, and 286 surveys returned as undeliverable), resulting in a final sample of 3589. The study protocol was reviewed and deemed exempt by the Office for the Protection of Research Subjects, University of Illinois at Chicago.

Survey Instrument

We reviewed the literature to preferentially include validated scales and questions. Cognitive interviews were conducted with 13 practicing physicians from different specialties and practice types to assess for overall survey coherence and individual item comprehension. The final instrument included an item that assessed physicians’ perceptions of pay fairness, as well as measures of work- and life-related outcomes such as work satisfaction, intent to leave current medical practice, and self-reported overall health.

Measures

Perceived pay fairness was evaluated by assessing agreement with the following statement: “My total compensation is fair.” The item had a five-point Likert response scale. Overall compensation was determined with a single item (“What was your approximate total compensation from the practice in the last year?”) that included an open-ended response.

Validated measures of physician work- and life-related well-being included single items for work satisfaction (“Overall, I am satisfied with my current work”),22 with a five-point “agree–disagree” response scale; intention to leave their current practice (“What is the likelihood that you will leave your current practice within two years?”),22 with a five-point response scale of definitely, likely, moderate, slight, and none; and overall health (“In general, how would you rate your health?”), with a five-point response scale of excellent, very good, good, fair, and poor.23

Data on respondents’ practice setting (solo, group, hospital, medical school, other), primary compensation (salary only, salary plus bonus, billing only, other), and employment type (employee, full or part owner, independent contractor) were also collected. Other physician characteristics including age, sex, race or ethnicity, and medical specialty were derived from the AMA Physician Masterfile.

Statistical Analyses

Responses were double-entered and imported into IBM SPSS Statistics version 21 software (IBM Corp., Armonk, NY). Differences between physician respondents and non-respondents by sex, race or ethnicity, and medical specialty were assessed using the Pearson χ2 test; differences in age were evaluated with the t test.

Multivariable logistic regression was conducted to assess the association between physician characteristics (e.g., internal medicine vs. non-primary care specialties) and perceived pay fairness. As the dependent variable, pay fairness perception was dichotomized as those who “agreed” or “strongly agreed” that their compensation was fair versus all other respondents. Next, a series of hierarchical logistic regressions were conducted in two steps to evaluate perceptions regarding pay fairness and its association with work satisfaction, intent to leave current practice, and health status, respectively. We dichotomized the responses (top 2 options and the bottom 3 options) for the three dependent variables assessing work satisfaction, intent to leave current practice, and overall health. The step 1 model had overall compensation, but not pay fairness perception, as an independent variable. Self-reported overall compensation data were categorized in $50,000 increments from $150,000 to $349,999, with a lower-end category of <$150,000 and upper-end category of ≥$350,000. Those who declined to provide data on their compensation were included in the model and categorized as “missing.” The step 2 model was specified with the addition of pay fairness perception as a dichotomized independent variable. To evaluate the potential for an interaction effect between pay fairness perception and overall compensation, step 2 models were specified with interaction terms. All models included physician age, sex, race or ethnicity, medical specialty, practice setting, primary compensation, and employment type as covariates.

RESULTS

Of the 3589 potential respondents, 2263 returned completed surveys, for a response rate of 63%. There was a higher proportion of white physicians in the respondent than non-respondent group (59% vs. 52%, P < 0.001), but no other significant differences between respondents and non-respondents were observed. Employed physicians constituted more than half (53%) of respondents, and 40% were owners of their practice. Nearly half of respondents (47%) worked in a group practice setting, while 17% were solo practitioners.

Overall Compensation and Pay Fairness Perception

One-fifth of respondents (21%) reported earning $350,000 or more in the previous year, while 15% earned less than $150,000. More than half (57%) agreed that their compensation was fair, while 30% disagreed or strongly disagreed (Table 1). Among respondents, 492 declined to answer the overall compensation survey item.

Table 1 Self-Reported Data from Physician Survey (N = 2263)*
Full size table

Work- and Life-Related Outcomes

Nearly three-quarters of physician respondents (73%) “agreed” or “strongly agreed” that they were satisfied with their current work. The likelihood of leaving one’s practice in the next 2 years was reported as “definitely” or “likely” by 18% of respondents, while 70% said “slight” or “none.” General overall health was reported as “very good” or “excellent” by 77% of respondents (Table 1).

Associations Between Physician Characteristics and Perceived Pay Fairness

Higher income levels were consistently associated with increased odds of perceived pay fairness, but there were no differences in pay fairness perceptions between physicians in primary care and non-primary care specialties (Table 2). Compared to physicians in a group practice setting, solo practitioners (OR, 0.61; 95% CI, 0.44–0.81; P < 0.001) had lower odds of agreeing that their pay was fair. Primary compensation as “salary plus bonus” was associated with higher odds (OR, 1.26; 95% CI, 1.00–1.60; P = 0.05) of perceiving one’s pay as fair, compared to those compensated primarily through “salary only.” No significant difference in perception of pay fairness was found between physicians who were employees and those who owned their medical practices. Neither age, sex, nor race influenced perceptions of pay fairness.

Table 2 Logistic Regression Showing Odds of Physicians Agreeing that Their Compensation Is Fair Based on Their Medical Specialty after Adjusting for Total Compensation, Practice, and Personal Characteristics*
Full size table

Associations Between Perceived Pay Fairness, Compensation, and Work- and Life-Related Outcomes

In step 1 logistic regression models, total compensation was found to be positively associated with work satisfaction, intention to remain in current practice, and self-reported health (Table 3). Compared to the reference group of physicians who earned less than $150,000 per year, compensation of ≥$350,000 was associated with a more than twofold increase in the odds of being satisfied with work (OR, 2.12; 95% CI, 1.49–3.01; P < 0.001), negligible likelihood of leaving their current practice (OR, 2.37; 95% CI, 1.67–3.35; P < 0.001), and more than 50% greater odds of having better self-reported health (OR, 1.60; 95%, 1.09–2.33; P = 0.03). After accounting for perceived pay fairness in the step 2 models, total compensation was not significantly associated with work satisfaction or overall health. Most of the associations between compensation and turnover intentions also became non-significant after adjusting for perceived pay fairness, with only those earning ≥$350,000 (OR, 1.75; 95% CI, 1.22–2.51; P < 0.01) retaining statistical significance. Physicians who thought their compensation was fair had increased odds of being satisfied with their current work (OR, 4.90; 95% CI, 3.94–6.08; P < 0.001), a diminished likelihood of leaving their current practice (OR, 2.46; 95% CI, 2.01–3.01; P < 0.001), and better overall health (OR, 1.33; 95% CI, 1.08–1.65; P < 0.01). Finally, the Wald coefficient in the step 2 logistic regression models did not reveal any significant interaction effect between overall compensation and pay fairness perception; thus the models displayed in Table 3 did not include interaction terms as independent variables.

Table 3 Hierarchical Logistic Regressions Showing Odds for Physician Work Satisfaction, Intent to Leave Current Practice, and Personal Health Based on Total Self-Reported Compensation and Perceived Pay Fairness*
Full size table

DISCUSSION

Perceptions create our experienced reality. Consistent with other studies examining the relationship between compensation and work-related outcomes,1,2 we found that physicians who earned more reported greater satisfaction at work and less inclination to leave their practice. However, these associations became largely non-significant after physicians’ perceptions of pay fairness were taken into account. Physicians who thought their compensation was fair experienced greater professional satisfaction and expressed less intention of turnover. In addition, pay fairness perception was positively associated with better self-reported health. Our study findings build on previous research of physician compensation,24,25,–26 and point to potential consequences for physicians and their patients. Studies have found that patients are more satisfied with their care when their physicians have higher professional satisfaction, 27 increased physician turnover is associated with worse patient experiences of care,28 patients whose physicians are compliant with health promotion practices such as screenings and vaccinations tend to undergo these preventive measures themselves,29 and physicians with a normal body mass index are more likely to engage patients with obesity in weight-loss discussions.30

We did not find differences in pay fairness perceptions between physicians who were employees and practice owners. That said, more physicians are entering employment relationships in which their compensation will be determined by an employer. Compensation has always been an essential element of any employment relationship. While compensation systems provide an economic and transactional mechanism for bringing labor inside an organization, they also serve an important social and relational function.31 It is well established that high-performing organizations create and nurture employment relationships that are built on mutual respect and trust.32,33 Compensation systems and employee perceptions of whether their pay is fair have “an important role in shaping whether people feel they are treated with dignity, trust, and respect, and whether they believe the values and culture of the organization are worthy of their fullest commitment and highest efforts.”34 Therefore, health care employers’ responses to physicians’ negative perceptions of pay fairness may be an increasing concern as more physicians are employed by medical groups and hospital systems.

According to organizational justice research, an essential determinant of perceived pay fairness is whether an individual understands and considers the procedures for determining compensation to be fair and just.35 This concept of procedural fairness is grounded in the belief that compensation determination procedures are applied consistently, are free from bias, and have mechanisms to correct inaccuracies.4,36,37 Procedural fairness research has also shown that compensation determination procedures that are perceived to be sound and appropriate may be as important as the specific amount of compensation.38,39 This notion of procedural fairness is particularly salient for health care organizations that employ physicians across various medical specialties. Given the long-standing pay differential between physicians in primary care and those in more procedure-intensive specialties,26 health care employers may need to be more transparent about how physician reimbursement is determined, and be prepared to explain the rationale for pay determination across different medical specialties. While we found no significant differences in pay fairness perceptions between primary care and non-primary care physicians, mechanisms that allow physicians to raise concerns about pay decisions and to have those concerns addressed could help minimize negative consequences related to perceived pay unfairness.

With respect to procedural fairness, physicians are probably more inclined to see their pay determination procedure as fair if they had input in the development of compensation methods. With implementation of the Medicare Access and CHIP Reauthorization Act of 2015, physician compensation methods are expected to be increasingly tied to explicit performance targets.40,41 If physicians think that pay-for-performance (P4P) measures used to determine their pay are appropriate (which is more likely if they help to identify such measures), there is a higher likelihood that they will perceive their pay as fair. On the other hand, if physicians believe that P4P does not promote high-quality patient care, they may be less inclined to think performance measures used in determining their pay are appropriate.42 One major concern among physicians is the determination of pay using performance measures that are beyond the control of individual physicians.43,44 Given this concern, adjusting quality measures for potentially health-related factors such as patient sociodemographic characteristics has garnered more attention, especially as P4P moves toward measuring outcomes, rather than processes, of care.45 As opposed to processes of care, which are largely under physician control, outcomes of care such as blood pressure and glycemic control are influenced by more than medical care. In July 2014, the National Quality Forum amended its blanket policy against the use of sociodemographic risk adjustment and has since been assessing the appropriateness of using such P4P risk adjustment.46

Several limitations of this study should be considered. First, a cross-sectional survey cannot assess causality between key study variables. Second, the respondent group had a higher proportion of white physicians than the non-respondent group, which may limit the generalizability of our results. Third, like other studies on physician compensation, we relied on self-reported income data, because it is neither feasible nor practical to acquire these data through other means. Given the sensitivity of this information, it was not surprising that one-fifth of our physician sample did not answer the total compensation survey item. While there were differences for physicians with missing income data in our hierarchical regression models, perceived pay fairness remained a significant independent variable in every specified model.

In our study, physicians who thought their overall compensation was fair had greater work satisfaction, lower likelihood of leaving their current practice, and better self-reported health. To promote a high-performing physician workforce in a rapidly changing health care system, medical practices and hospital systems may need to pay more attention to perceptions of pay fairness, which have implications for the well-being of physicians and their patients.