Tosha B. Wetterneck, MD, MS

Hospitalists are a growing workforce numbering over 40,000 physicians, one‐third of whom are women.[1] Flexibility of work schedules and control over personal time have been the traditional selling points of the specialty.[2] Multiple studies of physician work life reveal growing physician dissatisfaction and a high prevalence of burnout.[3] To mitigate burnout risk, leaders in hospital medicine recognize the importance of creating a sustainable profession that offers both job and career satisfaction as well as work‐life balance and, importantly, fairness within the work environment.[4] Although success in some of these endeavors has been realized sporadically, sustaining work‐life balance and fairness in the specialty remains a work in progress, whereas evidence of high job attrition and pay inequities remain.[5, 6]

Pay inequity for women relative to men continues to be pervasive in medicine, including among early‐career physicians, researchers, and various specialists.[6, 7, 8, 9, 10, 11, 12, 13] The earnings gap seems to persist for physicians, even as federal efforts such as the Fair Pay Act of 2013 and the Paycheck Fairness Act of 2014 aim to end wage discrimination.[11, 14] Differences in specialty, part‐time status, and practice type do not mitigate the disparity.[8, 10, 15] Additional explanations have been proposed to explain the variability, including gender differences in negotiating skills, lack of opportunities to join networks of influence within organizations, and implicit or explicit bias and discrimination.[12, 16, 17, 18, 19, 20]

The earnings gap is also a consequence of what is commonly called the glass ceiling.[18, 19] Most agree that obstacles to fair advancement of women include absence of collaborative environments and role models who have successfully achieved work‐life balance.[17, 20, 21, 22] Somewhat surprisingly, women leaders in medicine seem to suffer greater income disparity than nonleaders; this income gap is prevalent among leaders in other elite professions as well.[7, 23] It is unknown whether women physicians' emphasis on work‐life balance, seen repeatedly in surveys, explains any of the pay disparity.[24] Little research to date has examined whether work‐life priorities of women in hospital medicine differ from men.

In this study, we sought to examine differences in job priorities between men and women hospitalists. In particular, we examine the relative prioritization of substantial pay to job satisfaction. We also examined gender differences in work patterns and earnings to explore potential sources of the persistent gender earnings gap.

METHODS

We analyzed data from the 20092010 Hospital Medicine Physician Worklife Survey, the design of which is detailed elsewhere.[4] Briefly, a 118‐item survey was administered by mail to a stratified sample of hospitalists from the Society of Hospital Medicine database and 3 large multisite hospitalist groups. A single survey item asked respondents to identify up to 4 out of 12 most important domains to their satisfaction with a hospitalist job. The domains were distilled from focus groups of nationally representative hospitalists as described previously,[4] and the survey item allowed up to one‐third of these domains to be identified as respondents' personal priorities. The list included: optimal variety of tasks, optimal workload, substantial pay, collegiality with other physicians, recognition by leaders, rewarding relationships with patients, satisfaction with nurses, optimal autonomy, control over personal time, fairness within organization, ample availability of resources to do job, and organizational climate of trust and belonging. We tabulated and ranked the frequency with which respondents selected each satisfaction domain by gender. Due to the nonstandard format of the survey item, we a priori decided to analyze only responses that were completed as instructed.

We also used demographic data including detailed work characteristics, clinical and nonclinical workload, total pretax earnings in 2009 as a hospitalist, and self‐identification as leader of their hospital medicine group. Respondent characteristics were tabulated and gender differences were tested using the t test, rank sum test, and the Fischer exact test as appropriate. We also listed the number of nonrespondents for each item. In estimating gender differences in earnings, we opted to use multiple‐imputation techniques to more conservatively account for greater variance inherent in the presence of missing data. Consistent with existing guidelines,[25] we demonstrated that item responses were not missing monotonically by visually inspecting patterns of nonresponse. We further demonstrated that data were missing at random by showing that response patterns of completed survey items did not predict whether or not a given variable response was missing using logistic regression models. We found no significant differences between respondents with complete and missing data. We verified that appropriate regression models for each variable on every other variable converged. We used Stata 13.1 (StataCorp, College Station, TX) to perform multiple imputations using chain equations (mi impute chain) to create 10 imputed tables for 7 normally distributed continuous variables using the ordinary least squares method, 3 non‐normally distributed variables using the predictive mean matching method, 2 nonordinal categorical variables using the multinomial logit method, and 1 binary variable using the logit method.[26, 27] Gender, pediatric specialty status, region of practice, and whether or not respondents prioritized substantial pay for job satisfaction were used as regular variables without missing data points.

Differences in earnings were assessed using a multivariate ordinary linear regression model applied to the imputed datasets fitted by forward selection of explanatory variables using P < 0.20 in bivariate analysis for inclusion and manual backward elimination of all statistically nonsignificant variables. We tested the significance of the women leader interaction term in the final parsimonious model. We used the usual significance threshold of P < 0.05 for inferences. Our analysis of publicly available anonymous data was exempt from IRB review.

RESULTS

Of the 816 survey respondents (response rate 25.6%), 40 either omitted the item soliciting work priorities or completed it incorrectly. Data from the remaining 776 respondents were used for the present analysis. Respondent characteristics are tabulated in Table 1. The characteristics of hospitalists by age, gender, specialty, practice model, and practice region were representative of US hospitalists from other surveys.[28]

Several gender differences were seen in the characteristics of hospitalists and their work (Table 1). Women compared to men hospitalists were less likely to be leaders, more likely to be pediatricians, work in university settings, and practice in Western states. Women compared to men, on average, were younger by 3 years, worked fewer full‐time equivalents (FTEs), worked a greater percentage of nights, and reported fewer billable encounters per shift. They were also more likely to be divorced or separated.

Job satisfaction priorities differed for women and men hospitalists. Table 2 lists job satisfaction domains in descending order of the frequency prioritized by men. The largest proportion of women and men prioritized optimal workload. However, although substantial pay was prioritized next most frequently by men, more women prioritized collegiality and control over personal time than substantial pay.

Key differences in individual characteristics, work patterns, and indicating substantial pay as a priority were associated with self‐reported total earnings in 2009 from respondents' work as a hospitalist. As shown in Table 3, the inclusion of detailed productivity measures such as FTE, days of monthly clinical work, and estimated number of daily billable encounters yielded a model that explained 33% of variance in earnings. After adjusting for significant covariates including pediatric specialty, practice model, geography, and amount and type of clinical work, the estimated underpayment of women compared to men was $14,581. Hospitalists who prioritized substantial pay earned $10,771 more than those who did not. The female x leader interaction term testing the hypothesis that gender disparity is greater among leaders than frontline hospitalists was not statistically significant ($16,720, P = 0.087) and excluded from the final model.

DISCUSSION

In a national stratified sample of US hospitalists, we found gender differences in job satisfaction priorities and hospitalist work characteristics. We also confirmed the persistence of a substantial gender earnings disparity. Lower earnings among women compared to men hospitalists were present in our data after controlling for age, pediatric specialty, practice model, geography, type of clinical work, and productivity measures. The gender earnings disparity noted in 1999[6] persists, although it appears to have decreased, possibly indicating progress toward equity. We showed that women hospitalists' relative tendency not to prioritize pay explains a significant portion of the residual income gap.

Hoff examined hospitalist earnings in a large national survey of hospitalists in 1999. Our cohorts differed in age and experience (both lower in the Hoff study). An estimated $24,000 ($124,266 vs. $148,132) earnings gap between women and men was greater than our estimate of $14,581 following an interval of 10 years. Although survey items differed, both studies found that women were less interested in pay than men when considering a hospitalist job. Hoff also found that work setting and attitudes about pay and lifestyle were significantly related to earnings. We extended the previous analyses of gender differences in job satisfaction priorities, work, and demographic characteristics to explain the earnings gap and understand how it may be remedied.

When considering job satisfaction, we found that more men than women prioritized substantial pay and that prioritization of substantial pay was directly related to higher earnings. Therefore, fewer women prioritizing pay partly explains women's lower earnings. Reasons for why fewer women prioritize pay were not assessed in this study but may include factors like being part of 2‐income households and competing commitments.[29] Priorities may even be influenced by empirically observed gender differences in discussions of financial matters, governed by cultural norms. Such norms may implicitly sanction employers to offer women less pay than men for the same or similar work. Women may disadvantage themselves by negotiating less or less well than men for higher starting and promotion salaries. They may be perceived more negatively than men when they do negotiate pay, leading to unintended negative consequences such as loss of social networks, decreased likability, and even loss of job offers.[29, 30, 31, 32]

More women prioritized optimal variety of tasks (6th most prevalent among women and 10th among men). Women who highly rate optimal variety of tasks as a job satisfier may choose positions in which they teach, perform research, and participate in hospital committees and quality‐improvement work, but offer lower pay. Yet hours per month doing nonclinical work was not significantly different between men and women, nor associated with earnings differences in our earnings models. Understanding whether women self‐select into hospitalist jobs with like‐minded colleagues to achieve complementary fit or end up supplementing their skills with hospitalists with different priorities may inform strategies to reduce the gender income disparity.[5] Unlike disparities between various hospital medicine groups, systematic disparities within practices risk generating low levels of organizational fairness and burnout among employees.

Not surprisingly, productivity was positively associated with earnings but did not fully account for the gender earnings gap. Our data demonstrated that women, on average, were associated with work characteristics that expectedly generate less compensation. For example, women were younger, more often part time, academic, pediatric, less often leaders, and reported fewer billable encounters compared to men. These differences account for some of the earnings gap between men and women, but these factors were controlled for in the earnings model. In addition, our analysis may have underestimated the gap by not incorporating loss of fringe benefits from part‐time status and not comprehensively counting incentive pay associated with high productivity. Other work patterns more commonly associated with women suggest an imbalance in reimbursement. More women than men work nights that are often compensated at higher rates than daytime work, yet their average pay was less, suggesting that compensation for night work may need to be adjusted to reflect its unique burdens and responsibilities.[33]

Although the gender pay gap was not more extreme among leaders compared to frontline hospitalists in our data, the trend, nonetheless, underscores an important consideration. Whereas clinical work is paid for in mostly measurable ways, pay for leadership may be influenced by intangible factors such as reputation, negotiation, and confidence that may disadvantage women relative to men.[7, 19, 23, 34, 35, 36] Efforts to overcome implicit gender bias should be most effective when we consciously couple fair promotion of women to leadership with fair compensation commensurate with their male peers.[21]

Our data are vulnerable to nonresponse bias.[1, 4, 5] Post hoc analyses demonstrated that distributions of age, gender, practice model and region of our respondents were similar to other nationally representative cohorts of hospitalists. Consequently, we believe our data can make valid estimates about a nationally representative sample of hospitalists. However, we acknowledge several additional weaknesses of self‐reported data, including recall bias and accuracy of productivity figures, which were rounded to variable significant digits by respondents. Earnings analysis using this data was intended to be exploratory, but the findings echoed analyses using more authoritative data sources.[11] Still, we made inferences conservatively by adopting multiple imputation techniques for dealing with nonresponse surveys in adherence to established reporting guidelines.[25] We also note several limitations relevant to multiple imputations. The greater prevalence of missing data for survey items soliciting earnings and the number of billable encounters suggest they were not truly missing at random as assumed. However, we showed that missingness is unrelated to the variables under study, justifying use of the technique. The wider measures of variance derived from multiple imputations make us vulnerable to not detecting associations that may exist.

The gender earnings gap found in hospital medicine echoes the gap found in multiple medical specialties, including but not limited to pediatrics, academic medicine, gastroenterology, and plastic surgery.[7, 8, 9, 11, 12, 13, 37] Hospital medicine employment models and practice patterns have important structural differences compared to previously studied populations that could mitigate factors contributing to women physicians' lower earnings. However, despite well‐defined working hours, lack of control over the number of patient encounters per day and high prevalence of hospital‐employed practice models, the gender earnings gap persists. We showed that lower prioritization for pay may reflect the self‐selection of women into lower paying jobs. Unmeasured factors, including implicit bias and differences in negotiations, social networks and mentoring opportunities[38, 39] may also contribute to pay differences between men and women hospitalists. As hospital medicine tackles gender inequities and other disparities, strategies to assess and address fair physician compensation must be on the table.

Hospitalists are a growing workforce numbering over 40,000 physicians, one‐third of whom are women.[1] Flexibility of work schedules and control over personal time have been the traditional selling points of the specialty.[2] Multiple studies of physician work life reveal growing physician dissatisfaction and a high prevalence of burnout.[3] To mitigate burnout risk, leaders in hospital medicine recognize the importance of creating a sustainable profession that offers both job and career satisfaction as well as work‐life balance and, importantly, fairness within the work environment.[4] Although success in some of these endeavors has been realized sporadically, sustaining work‐life balance and fairness in the specialty remains a work in progress, whereas evidence of high job attrition and pay inequities remain.[5, 6]

Pay inequity for women relative to men continues to be pervasive in medicine, including among early‐career physicians, researchers, and various specialists.[6, 7, 8, 9, 10, 11, 12, 13] The earnings gap seems to persist for physicians, even as federal efforts such as the Fair Pay Act of 2013 and the Paycheck Fairness Act of 2014 aim to end wage discrimination.[11, 14] Differences in specialty, part‐time status, and practice type do not mitigate the disparity.[8, 10, 15] Additional explanations have been proposed to explain the variability, including gender differences in negotiating skills, lack of opportunities to join networks of influence within organizations, and implicit or explicit bias and discrimination.[12, 16, 17, 18, 19, 20]

The earnings gap is also a consequence of what is commonly called the glass ceiling.[18, 19] Most agree that obstacles to fair advancement of women include absence of collaborative environments and role models who have successfully achieved work‐life balance.[17, 20, 21, 22] Somewhat surprisingly, women leaders in medicine seem to suffer greater income disparity than nonleaders; this income gap is prevalent among leaders in other elite professions as well.[7, 23] It is unknown whether women physicians' emphasis on work‐life balance, seen repeatedly in surveys, explains any of the pay disparity.[24] Little research to date has examined whether work‐life priorities of women in hospital medicine differ from men.

In this study, we sought to examine differences in job priorities between men and women hospitalists. In particular, we examine the relative prioritization of substantial pay to job satisfaction. We also examined gender differences in work patterns and earnings to explore potential sources of the persistent gender earnings gap.

METHODS

We analyzed data from the 20092010 Hospital Medicine Physician Worklife Survey, the design of which is detailed elsewhere.[4] Briefly, a 118‐item survey was administered by mail to a stratified sample of hospitalists from the Society of Hospital Medicine database and 3 large multisite hospitalist groups. A single survey item asked respondents to identify up to 4 out of 12 most important domains to their satisfaction with a hospitalist job. The domains were distilled from focus groups of nationally representative hospitalists as described previously,[4] and the survey item allowed up to one‐third of these domains to be identified as respondents' personal priorities. The list included: optimal variety of tasks, optimal workload, substantial pay, collegiality with other physicians, recognition by leaders, rewarding relationships with patients, satisfaction with nurses, optimal autonomy, control over personal time, fairness within organization, ample availability of resources to do job, and organizational climate of trust and belonging. We tabulated and ranked the frequency with which respondents selected each satisfaction domain by gender. Due to the nonstandard format of the survey item, we a priori decided to analyze only responses that were completed as instructed.

We also used demographic data including detailed work characteristics, clinical and nonclinical workload, total pretax earnings in 2009 as a hospitalist, and self‐identification as leader of their hospital medicine group. Respondent characteristics were tabulated and gender differences were tested using the t test, rank sum test, and the Fischer exact test as appropriate. We also listed the number of nonrespondents for each item. In estimating gender differences in earnings, we opted to use multiple‐imputation techniques to more conservatively account for greater variance inherent in the presence of missing data. Consistent with existing guidelines,[25] we demonstrated that item responses were not missing monotonically by visually inspecting patterns of nonresponse. We further demonstrated that data were missing at random by showing that response patterns of completed survey items did not predict whether or not a given variable response was missing using logistic regression models. We found no significant differences between respondents with complete and missing data. We verified that appropriate regression models for each variable on every other variable converged. We used Stata 13.1 (StataCorp, College Station, TX) to perform multiple imputations using chain equations (mi impute chain) to create 10 imputed tables for 7 normally distributed continuous variables using the ordinary least squares method, 3 non‐normally distributed variables using the predictive mean matching method, 2 nonordinal categorical variables using the multinomial logit method, and 1 binary variable using the logit method.[26, 27] Gender, pediatric specialty status, region of practice, and whether or not respondents prioritized substantial pay for job satisfaction were used as regular variables without missing data points.

Differences in earnings were assessed using a multivariate ordinary linear regression model applied to the imputed datasets fitted by forward selection of explanatory variables using P < 0.20 in bivariate analysis for inclusion and manual backward elimination of all statistically nonsignificant variables. We tested the significance of the women leader interaction term in the final parsimonious model. We used the usual significance threshold of P < 0.05 for inferences. Our analysis of publicly available anonymous data was exempt from IRB review.

RESULTS

Of the 816 survey respondents (response rate 25.6%), 40 either omitted the item soliciting work priorities or completed it incorrectly. Data from the remaining 776 respondents were used for the present analysis. Respondent characteristics are tabulated in Table 1. The characteristics of hospitalists by age, gender, specialty, practice model, and practice region were representative of US hospitalists from other surveys.[28]

Several gender differences were seen in the characteristics of hospitalists and their work (Table 1). Women compared to men hospitalists were less likely to be leaders, more likely to be pediatricians, work in university settings, and practice in Western states. Women compared to men, on average, were younger by 3 years, worked fewer full‐time equivalents (FTEs), worked a greater percentage of nights, and reported fewer billable encounters per shift. They were also more likely to be divorced or separated.

Job satisfaction priorities differed for women and men hospitalists. Table 2 lists job satisfaction domains in descending order of the frequency prioritized by men. The largest proportion of women and men prioritized optimal workload. However, although substantial pay was prioritized next most frequently by men, more women prioritized collegiality and control over personal time than substantial pay.

Key differences in individual characteristics, work patterns, and indicating substantial pay as a priority were associated with self‐reported total earnings in 2009 from respondents' work as a hospitalist. As shown in Table 3, the inclusion of detailed productivity measures such as FTE, days of monthly clinical work, and estimated number of daily billable encounters yielded a model that explained 33% of variance in earnings. After adjusting for significant covariates including pediatric specialty, practice model, geography, and amount and type of clinical work, the estimated underpayment of women compared to men was $14,581. Hospitalists who prioritized substantial pay earned $10,771 more than those who did not. The female x leader interaction term testing the hypothesis that gender disparity is greater among leaders than frontline hospitalists was not statistically significant ($16,720, P = 0.087) and excluded from the final model.

DISCUSSION

In a national stratified sample of US hospitalists, we found gender differences in job satisfaction priorities and hospitalist work characteristics. We also confirmed the persistence of a substantial gender earnings disparity. Lower earnings among women compared to men hospitalists were present in our data after controlling for age, pediatric specialty, practice model, geography, type of clinical work, and productivity measures. The gender earnings disparity noted in 1999[6] persists, although it appears to have decreased, possibly indicating progress toward equity. We showed that women hospitalists' relative tendency not to prioritize pay explains a significant portion of the residual income gap.

Hoff examined hospitalist earnings in a large national survey of hospitalists in 1999. Our cohorts differed in age and experience (both lower in the Hoff study). An estimated $24,000 ($124,266 vs. $148,132) earnings gap between women and men was greater than our estimate of $14,581 following an interval of 10 years. Although survey items differed, both studies found that women were less interested in pay than men when considering a hospitalist job. Hoff also found that work setting and attitudes about pay and lifestyle were significantly related to earnings. We extended the previous analyses of gender differences in job satisfaction priorities, work, and demographic characteristics to explain the earnings gap and understand how it may be remedied.

When considering job satisfaction, we found that more men than women prioritized substantial pay and that prioritization of substantial pay was directly related to higher earnings. Therefore, fewer women prioritizing pay partly explains women's lower earnings. Reasons for why fewer women prioritize pay were not assessed in this study but may include factors like being part of 2‐income households and competing commitments.[29] Priorities may even be influenced by empirically observed gender differences in discussions of financial matters, governed by cultural norms. Such norms may implicitly sanction employers to offer women less pay than men for the same or similar work. Women may disadvantage themselves by negotiating less or less well than men for higher starting and promotion salaries. They may be perceived more negatively than men when they do negotiate pay, leading to unintended negative consequences such as loss of social networks, decreased likability, and even loss of job offers.[29, 30, 31, 32]

More women prioritized optimal variety of tasks (6th most prevalent among women and 10th among men). Women who highly rate optimal variety of tasks as a job satisfier may choose positions in which they teach, perform research, and participate in hospital committees and quality‐improvement work, but offer lower pay. Yet hours per month doing nonclinical work was not significantly different between men and women, nor associated with earnings differences in our earnings models. Understanding whether women self‐select into hospitalist jobs with like‐minded colleagues to achieve complementary fit or end up supplementing their skills with hospitalists with different priorities may inform strategies to reduce the gender income disparity.[5] Unlike disparities between various hospital medicine groups, systematic disparities within practices risk generating low levels of organizational fairness and burnout among employees.

Not surprisingly, productivity was positively associated with earnings but did not fully account for the gender earnings gap. Our data demonstrated that women, on average, were associated with work characteristics that expectedly generate less compensation. For example, women were younger, more often part time, academic, pediatric, less often leaders, and reported fewer billable encounters compared to men. These differences account for some of the earnings gap between men and women, but these factors were controlled for in the earnings model. In addition, our analysis may have underestimated the gap by not incorporating loss of fringe benefits from part‐time status and not comprehensively counting incentive pay associated with high productivity. Other work patterns more commonly associated with women suggest an imbalance in reimbursement. More women than men work nights that are often compensated at higher rates than daytime work, yet their average pay was less, suggesting that compensation for night work may need to be adjusted to reflect its unique burdens and responsibilities.[33]

Although the gender pay gap was not more extreme among leaders compared to frontline hospitalists in our data, the trend, nonetheless, underscores an important consideration. Whereas clinical work is paid for in mostly measurable ways, pay for leadership may be influenced by intangible factors such as reputation, negotiation, and confidence that may disadvantage women relative to men.[7, 19, 23, 34, 35, 36] Efforts to overcome implicit gender bias should be most effective when we consciously couple fair promotion of women to leadership with fair compensation commensurate with their male peers.[21]

Our data are vulnerable to nonresponse bias.[1, 4, 5] Post hoc analyses demonstrated that distributions of age, gender, practice model and region of our respondents were similar to other nationally representative cohorts of hospitalists. Consequently, we believe our data can make valid estimates about a nationally representative sample of hospitalists. However, we acknowledge several additional weaknesses of self‐reported data, including recall bias and accuracy of productivity figures, which were rounded to variable significant digits by respondents. Earnings analysis using this data was intended to be exploratory, but the findings echoed analyses using more authoritative data sources.[11] Still, we made inferences conservatively by adopting multiple imputation techniques for dealing with nonresponse surveys in adherence to established reporting guidelines.[25] We also note several limitations relevant to multiple imputations. The greater prevalence of missing data for survey items soliciting earnings and the number of billable encounters suggest they were not truly missing at random as assumed. However, we showed that missingness is unrelated to the variables under study, justifying use of the technique. The wider measures of variance derived from multiple imputations make us vulnerable to not detecting associations that may exist.

The gender earnings gap found in hospital medicine echoes the gap found in multiple medical specialties, including but not limited to pediatrics, academic medicine, gastroenterology, and plastic surgery.[7, 8, 9, 11, 12, 13, 37] Hospital medicine employment models and practice patterns have important structural differences compared to previously studied populations that could mitigate factors contributing to women physicians' lower earnings. However, despite well‐defined working hours, lack of control over the number of patient encounters per day and high prevalence of hospital‐employed practice models, the gender earnings gap persists. We showed that lower prioritization for pay may reflect the self‐selection of women into lower paying jobs. Unmeasured factors, including implicit bias and differences in negotiations, social networks and mentoring opportunities[38, 39] may also contribute to pay differences between men and women hospitalists. As hospital medicine tackles gender inequities and other disparities, strategies to assess and address fair physician compensation must be on the table.

Hospitalists are a growing workforce numbering over 40,000 physicians, one‐third of whom are women.[1] Flexibility of work schedules and control over personal time have been the traditional selling points of the specialty.[2] Multiple studies of physician work life reveal growing physician dissatisfaction and a high prevalence of burnout.[3] To mitigate burnout risk, leaders in hospital medicine recognize the importance of creating a sustainable profession that offers both job and career satisfaction as well as work‐life balance and, importantly, fairness within the work environment.[4] Although success in some of these endeavors has been realized sporadically, sustaining work‐life balance and fairness in the specialty remains a work in progress, whereas evidence of high job attrition and pay inequities remain.[5, 6]

Pay inequity for women relative to men continues to be pervasive in medicine, including among early‐career physicians, researchers, and various specialists.[6, 7, 8, 9, 10, 11, 12, 13] The earnings gap seems to persist for physicians, even as federal efforts such as the Fair Pay Act of 2013 and the Paycheck Fairness Act of 2014 aim to end wage discrimination.[11, 14] Differences in specialty, part‐time status, and practice type do not mitigate the disparity.[8, 10, 15] Additional explanations have been proposed to explain the variability, including gender differences in negotiating skills, lack of opportunities to join networks of influence within organizations, and implicit or explicit bias and discrimination.[12, 16, 17, 18, 19, 20]

The earnings gap is also a consequence of what is commonly called the glass ceiling.[18, 19] Most agree that obstacles to fair advancement of women include absence of collaborative environments and role models who have successfully achieved work‐life balance.[17, 20, 21, 22] Somewhat surprisingly, women leaders in medicine seem to suffer greater income disparity than nonleaders; this income gap is prevalent among leaders in other elite professions as well.[7, 23] It is unknown whether women physicians' emphasis on work‐life balance, seen repeatedly in surveys, explains any of the pay disparity.[24] Little research to date has examined whether work‐life priorities of women in hospital medicine differ from men.

In this study, we sought to examine differences in job priorities between men and women hospitalists. In particular, we examine the relative prioritization of substantial pay to job satisfaction. We also examined gender differences in work patterns and earnings to explore potential sources of the persistent gender earnings gap.

METHODS

We analyzed data from the 20092010 Hospital Medicine Physician Worklife Survey, the design of which is detailed elsewhere.[4] Briefly, a 118‐item survey was administered by mail to a stratified sample of hospitalists from the Society of Hospital Medicine database and 3 large multisite hospitalist groups. A single survey item asked respondents to identify up to 4 out of 12 most important domains to their satisfaction with a hospitalist job. The domains were distilled from focus groups of nationally representative hospitalists as described previously,[4] and the survey item allowed up to one‐third of these domains to be identified as respondents' personal priorities. The list included: optimal variety of tasks, optimal workload, substantial pay, collegiality with other physicians, recognition by leaders, rewarding relationships with patients, satisfaction with nurses, optimal autonomy, control over personal time, fairness within organization, ample availability of resources to do job, and organizational climate of trust and belonging. We tabulated and ranked the frequency with which respondents selected each satisfaction domain by gender. Due to the nonstandard format of the survey item, we a priori decided to analyze only responses that were completed as instructed.

We also used demographic data including detailed work characteristics, clinical and nonclinical workload, total pretax earnings in 2009 as a hospitalist, and self‐identification as leader of their hospital medicine group. Respondent characteristics were tabulated and gender differences were tested using the t test, rank sum test, and the Fischer exact test as appropriate. We also listed the number of nonrespondents for each item. In estimating gender differences in earnings, we opted to use multiple‐imputation techniques to more conservatively account for greater variance inherent in the presence of missing data. Consistent with existing guidelines,[25] we demonstrated that item responses were not missing monotonically by visually inspecting patterns of nonresponse. We further demonstrated that data were missing at random by showing that response patterns of completed survey items did not predict whether or not a given variable response was missing using logistic regression models. We found no significant differences between respondents with complete and missing data. We verified that appropriate regression models for each variable on every other variable converged. We used Stata 13.1 (StataCorp, College Station, TX) to perform multiple imputations using chain equations (mi impute chain) to create 10 imputed tables for 7 normally distributed continuous variables using the ordinary least squares method, 3 non‐normally distributed variables using the predictive mean matching method, 2 nonordinal categorical variables using the multinomial logit method, and 1 binary variable using the logit method.[26, 27] Gender, pediatric specialty status, region of practice, and whether or not respondents prioritized substantial pay for job satisfaction were used as regular variables without missing data points.

Differences in earnings were assessed using a multivariate ordinary linear regression model applied to the imputed datasets fitted by forward selection of explanatory variables using P < 0.20 in bivariate analysis for inclusion and manual backward elimination of all statistically nonsignificant variables. We tested the significance of the women leader interaction term in the final parsimonious model. We used the usual significance threshold of P < 0.05 for inferences. Our analysis of publicly available anonymous data was exempt from IRB review.

RESULTS

Of the 816 survey respondents (response rate 25.6%), 40 either omitted the item soliciting work priorities or completed it incorrectly. Data from the remaining 776 respondents were used for the present analysis. Respondent characteristics are tabulated in Table 1. The characteristics of hospitalists by age, gender, specialty, practice model, and practice region were representative of US hospitalists from other surveys.[28]

Several gender differences were seen in the characteristics of hospitalists and their work (Table 1). Women compared to men hospitalists were less likely to be leaders, more likely to be pediatricians, work in university settings, and practice in Western states. Women compared to men, on average, were younger by 3 years, worked fewer full‐time equivalents (FTEs), worked a greater percentage of nights, and reported fewer billable encounters per shift. They were also more likely to be divorced or separated.

Job satisfaction priorities differed for women and men hospitalists. Table 2 lists job satisfaction domains in descending order of the frequency prioritized by men. The largest proportion of women and men prioritized optimal workload. However, although substantial pay was prioritized next most frequently by men, more women prioritized collegiality and control over personal time than substantial pay.

Key differences in individual characteristics, work patterns, and indicating substantial pay as a priority were associated with self‐reported total earnings in 2009 from respondents' work as a hospitalist. As shown in Table 3, the inclusion of detailed productivity measures such as FTE, days of monthly clinical work, and estimated number of daily billable encounters yielded a model that explained 33% of variance in earnings. After adjusting for significant covariates including pediatric specialty, practice model, geography, and amount and type of clinical work, the estimated underpayment of women compared to men was $14,581. Hospitalists who prioritized substantial pay earned $10,771 more than those who did not. The female x leader interaction term testing the hypothesis that gender disparity is greater among leaders than frontline hospitalists was not statistically significant ($16,720, P = 0.087) and excluded from the final model.

DISCUSSION

In a national stratified sample of US hospitalists, we found gender differences in job satisfaction priorities and hospitalist work characteristics. We also confirmed the persistence of a substantial gender earnings disparity. Lower earnings among women compared to men hospitalists were present in our data after controlling for age, pediatric specialty, practice model, geography, type of clinical work, and productivity measures. The gender earnings disparity noted in 1999[6] persists, although it appears to have decreased, possibly indicating progress toward equity. We showed that women hospitalists' relative tendency not to prioritize pay explains a significant portion of the residual income gap.

Hoff examined hospitalist earnings in a large national survey of hospitalists in 1999. Our cohorts differed in age and experience (both lower in the Hoff study). An estimated $24,000 ($124,266 vs. $148,132) earnings gap between women and men was greater than our estimate of $14,581 following an interval of 10 years. Although survey items differed, both studies found that women were less interested in pay than men when considering a hospitalist job. Hoff also found that work setting and attitudes about pay and lifestyle were significantly related to earnings. We extended the previous analyses of gender differences in job satisfaction priorities, work, and demographic characteristics to explain the earnings gap and understand how it may be remedied.

When considering job satisfaction, we found that more men than women prioritized substantial pay and that prioritization of substantial pay was directly related to higher earnings. Therefore, fewer women prioritizing pay partly explains women's lower earnings. Reasons for why fewer women prioritize pay were not assessed in this study but may include factors like being part of 2‐income households and competing commitments.[29] Priorities may even be influenced by empirically observed gender differences in discussions of financial matters, governed by cultural norms. Such norms may implicitly sanction employers to offer women less pay than men for the same or similar work. Women may disadvantage themselves by negotiating less or less well than men for higher starting and promotion salaries. They may be perceived more negatively than men when they do negotiate pay, leading to unintended negative consequences such as loss of social networks, decreased likability, and even loss of job offers.[29, 30, 31, 32]

More women prioritized optimal variety of tasks (6th most prevalent among women and 10th among men). Women who highly rate optimal variety of tasks as a job satisfier may choose positions in which they teach, perform research, and participate in hospital committees and quality‐improvement work, but offer lower pay. Yet hours per month doing nonclinical work was not significantly different between men and women, nor associated with earnings differences in our earnings models. Understanding whether women self‐select into hospitalist jobs with like‐minded colleagues to achieve complementary fit or end up supplementing their skills with hospitalists with different priorities may inform strategies to reduce the gender income disparity.[5] Unlike disparities between various hospital medicine groups, systematic disparities within practices risk generating low levels of organizational fairness and burnout among employees.

Not surprisingly, productivity was positively associated with earnings but did not fully account for the gender earnings gap. Our data demonstrated that women, on average, were associated with work characteristics that expectedly generate less compensation. For example, women were younger, more often part time, academic, pediatric, less often leaders, and reported fewer billable encounters compared to men. These differences account for some of the earnings gap between men and women, but these factors were controlled for in the earnings model. In addition, our analysis may have underestimated the gap by not incorporating loss of fringe benefits from part‐time status and not comprehensively counting incentive pay associated with high productivity. Other work patterns more commonly associated with women suggest an imbalance in reimbursement. More women than men work nights that are often compensated at higher rates than daytime work, yet their average pay was less, suggesting that compensation for night work may need to be adjusted to reflect its unique burdens and responsibilities.[33]

Although the gender pay gap was not more extreme among leaders compared to frontline hospitalists in our data, the trend, nonetheless, underscores an important consideration. Whereas clinical work is paid for in mostly measurable ways, pay for leadership may be influenced by intangible factors such as reputation, negotiation, and confidence that may disadvantage women relative to men.[7, 19, 23, 34, 35, 36] Efforts to overcome implicit gender bias should be most effective when we consciously couple fair promotion of women to leadership with fair compensation commensurate with their male peers.[21]

Our data are vulnerable to nonresponse bias.[1, 4, 5] Post hoc analyses demonstrated that distributions of age, gender, practice model and region of our respondents were similar to other nationally representative cohorts of hospitalists. Consequently, we believe our data can make valid estimates about a nationally representative sample of hospitalists. However, we acknowledge several additional weaknesses of self‐reported data, including recall bias and accuracy of productivity figures, which were rounded to variable significant digits by respondents. Earnings analysis using this data was intended to be exploratory, but the findings echoed analyses using more authoritative data sources.[11] Still, we made inferences conservatively by adopting multiple imputation techniques for dealing with nonresponse surveys in adherence to established reporting guidelines.[25] We also note several limitations relevant to multiple imputations. The greater prevalence of missing data for survey items soliciting earnings and the number of billable encounters suggest they were not truly missing at random as assumed. However, we showed that missingness is unrelated to the variables under study, justifying use of the technique. The wider measures of variance derived from multiple imputations make us vulnerable to not detecting associations that may exist.

The gender earnings gap found in hospital medicine echoes the gap found in multiple medical specialties, including but not limited to pediatrics, academic medicine, gastroenterology, and plastic surgery.[7, 8, 9, 11, 12, 13, 37] Hospital medicine employment models and practice patterns have important structural differences compared to previously studied populations that could mitigate factors contributing to women physicians' lower earnings. However, despite well‐defined working hours, lack of control over the number of patient encounters per day and high prevalence of hospital‐employed practice models, the gender earnings gap persists. We showed that lower prioritization for pay may reflect the self‐selection of women into lower paying jobs. Unmeasured factors, including implicit bias and differences in negotiations, social networks and mentoring opportunities[38, 39] may also contribute to pay differences between men and women hospitalists. As hospital medicine tackles gender inequities and other disparities, strategies to assess and address fair physician compensation must be on the table.

Person‐organization fit concerns the conditions and consequences of compatibility between people and the organizations for which they work.[1] Studies of other industries have demonstrated that person‐organization fit informs the way individuals join, perform in, and are retained by organizations.[2] Person‐job fit is a closely related subordinate concept that concerns the alignment of workers and their job in as much as workers have needs that their job supplies, or conversely, jobs have requirements that certain workers' abilities can help meet.[3] Explorations of job fit in physicians and their work have recently emerged in a few investigations published in medical journals.[4, 5, 6, 7, 8] Further expanding the understanding of fit between physicians and their employment is important, because the decline of solo practices and recent emphasis on team‐based care have led to a growing number of US physicians working in organizations.[9]

The movement of physicians into employed situations may continue if certain types of Accountable Care Organizations take root.[10] And physicians may be primed to join employer organizations based on current career priorities of individuals in American society. Surveys of medical residents entering the workforce reveal more physicians preferring the security of being employees than starting their own practices.[11] Given these trends, job fit will inform our understanding of how personal and job characteristics facilitate recruitment, performance, satisfaction, and longevity of physician employees.

BACKGROUND

Virtually all hospitalists work in organizationshospitalsand are employees of hospitals, medical schools, physician group practices, or management companies, and therefore invariably function within organizational structures and systems.[7] In spite of their rapid growth in numbers, many employers have faced difficulties recruiting and retaining enough hospitalists to fill their staffing needs. Consequently, the US hospitalist workforce today is characterized by high salaries, work load, and attrition rates.[12]

In this evolving unsaturated market, the attraction‐selection‐attrition framework[13] provides a theoretical construct that predicts that hospitalists and their employers would seek congruence of goals and values early in their relationship through a process of trial and error. This framework assumes that early interactions between workers and organizations serve as opportunities for them to understand if job fit is poor and dissociate or remain affiliated as long as job fit is mutually acceptable. Therefore, job switching on average is expected to increase job fit because workers and organizations gain a better understanding of their own goals and values and choose more wisely the next time.

Other theoretical frameworks, such as the job characteristic model,[14] suggest that over time as workers stay at the same job, they continue to maintain and improve job fit through various workplace‐ or self‐modification strategies. For example, seniority status may have privileges (eg, less undesirable call), or workers may create privileged niches through the acquisition of new skills and abilities over time. Hospitalists' tendency to diversify their work‐related activities by incorporating administrative and teaching responsibilities[15] may thus contribute to improving job fit. Additionally, as a measure of complementarity among people who work together, job fit may be influenced by the quality of relationships among hospitalists and their coworkers through their reorientation to the prevailing organizational climate[16, 17] and increasing socialization.[18] Finally, given that experiential learning is known to contribute to better hospitalist work performance,[19] job fit may affect productivity and clinical outcomes vis‐‐vis quality of work life.

To test the validity of these assumptions in a sample of hospitalists, we critically appraised the following 4 hypotheses:

• Hypothesis 1 (H1): Job attrition and reselection improves job fit among hospitalists entering the job market.
• Hypothesis 2 (H2): Better job fit is achieved through hospitalists engaging a variety of personal skills and abilities.
• Hypothesis 3 (H3): Job fit increases with hospitalists' job duration together with socialization and internalization of organizational values.
• Hypothesis 4 (H4): Job fit is correlated with hospitalists' quality of work life.

METHODS

Analysis was performed on data from the 2009 to 2010 Hospital Medicine Physician Worklife Survey. The sample frame included nonmembers and members of Society of Hospital Medicine (SHM). Details about sampling strategy, data collection, and data quality are available in previous publications.[7, 20] The 118‐item survey instrument, including 9 demographic items and 24 practice and job characteristic items, was administered by mail. Examples of information solicited through these items included respondents' practice model, the number of hospitalist jobs they have held, and the specific kinds of clinical and nonclinical activities they performed as part of their current job.

We used a reliable but broad and generic measure of self‐perceived person‐job fit.[21] The survey items of the 5‐point Likert‐type scale anchored between strongly disagree and strongly agree were: I feel that my work utilizes my full abilities, I feel competent and fully able to handle my job, my job gives me the chance to do the things I feel I do best, I feel that my job and I are well‐matched, I feel I have adequate preparation for the job I now hold. The quality of hospitalists' relationships with physician colleagues, staff, and patients as well as job satisfaction was measured using scales adapted from the Physician Worklife Study.[22] Organizational climate was measured using an adapted scale from the Minimizing Error, Maximizing Outcome study incorporating 3 items from the cohesiveness subscale, 4 items from the organizational trust subscale, and 1 item from the quality emphasis subscale that were most pertinent to hospitalists' relationship with their organizations.[23] Intent to leave practice or reduce work hours was measured using 5 items from the Multi‐Center Hospitalist Survey Project.[24] Frequency of participation in suboptimal patient care was measured by adapting 3 items from the suboptimal reported practice subscale and 2 items from the suboptimal patient care subscale developed by Shanafelt et al.[25] Stress and job burnout were assessed using validated measures.[26, 27] Detailed descriptions of the response rate calculation and imputation of missing item data are available in previous publications.[7, 20]

Mean, variance, range, and skew were used to characterize the responses to the job fit survey scale. A table of respondent characteristics was constructed. A visual representation of job fit by individual hospitalist year in current practice was created, first, by plotting a locally weighted scatterplot smoothing curve to examine the shape of the general relationship, and second, by fitting a similarly contoured functional polynomial curve with 95% confidence intervals (CI) to a plot of the mean and interquartile range of job fit for each year in current practice. Spearman partial correlations were calculated for job fit and each of the 5 items addressing likelihood of leaving practice or reducing workload adjusted for gender to control for the higher proportion of women who plan to work part time. Median (interquartile range) job fit was calculated for categories defined by the number of job changes and compared with the reference category (no job change) using the nonparametric rank sum test for comparing non‐normally distributed data. Multivariate logistic regression models were used to calculate the odds ratio (OR) of participating in each of several clinical and nonclinical hospitalist activities between respondents whose job fit score was optimal (5 on a 5‐point scale) and less than optimal controlling for covariates that influence the likelihood of participating in these activities (years in current practice, practice model, and specialty training). A Spearman correlation matrix was created to assess interscale correlations among organizational parameters (years in current practice, job fit, organizational climate, and relationship with colleagues, staff, and patients). Finally, a separate Spearman correlation matrix was created to assess the interscale correlations among individual worker parameters (job fit, suboptimal patient care, job burnout, stress, and job satisfaction). Statistical significance was defined as P value <0.05, and all analyses were performed on Stata 11.0 (StataCorp, College Station, TX). The Northwestern University institutional review board approved this study.

RESULTS

Respondents included 816 hospitalists belonging to around 700 unique organizations. The adjusted response rate from the stratified sample was 26%. Respondents and nonrespondents were similar with regard to geographic region and model of practice, but respondents were more likely to be members of the SHM than nonrespondents. Panel A of Table 1 shows the demographic characteristics of the respondents. The mean age was 44.3 years, and about one‐third were women. The average hospitalist had about 7 years of experience in the specialty and about 5 years with their current hospitalist job. The majority were trained in internal medicine or one of its subspecialties, whereas pediatricians, family physicians, and physicians with other training made up the remainder.

Job fit was highly skewed toward optimum fit, with a mean of 4.3 on a scale of 1 to 5, with a narrow standard deviation of 0.7. The poorest job fit was reported by 0.3%, whereas optimal fit was reported by 21% of respondents. Job fit plotted against years in current practice had a logarithmic appearance typical of learning curves (Figure 1). An inflection point was visualized at around 2 years. For the purposes of this article, we refer to hospitalists' experience in the first 2 years of a job as an assimilation period, which is marked by a steep increase in job fit early when rapid learning or attrition took place. The years beyond the inflection point are characterized as an advancement period, when a more attenuated rise in job fit was experienced with time. The Spearman correlation between job fit and years in practice during the advancement period was 0.145 (n = 678, P < 0.001). Panel B of Table 1 displays the characteristics of respondents separately for the assimilation and advancement cohorts. Assimilation hospitalists in our sample had a mean age of 41.9 years and mean on‐the‐job experience of 4.3 years, reflecting that many hospitalists in the first 2 years of a job have made at least 1 job change in the past.

Figure 1

To show the effects of attrition and reselection, we first evaluated the proposition that hospitalists experience attrition (ie, intend to leave their jobs) in response to poor fit. Table 2 shows the correlations between job fit and the self‐reported intent to leave practice or reduce workload separately for the assimilation and advancement periods. For hospitalists in the assimilation period, job fit was negatively correlated with intent to leave current practice within 2 years and to leave hospital medicine within 5 years (P = 0.010 and 0.043, respectively). Hospitalists with <2 years in their current job, therefore, tended to consider attrition but not workload reduction to deal with poor job fit. On the other hand, hospitalists in the advancement period considered both attrition and workload reduction strategies in response to poor fit (all P < 0.001).

In Table 3, we further compared the median job fit across categories for the number of job switches. The median job fit during the assimilation period of hospitalists who had made 1 job change was slightly but statistically higher than the job fit of their counterparts who never left their first job (4.4 vs 4.0, P = 0.046). This suggests that job switching by hospitalists early in their jobs is associated with improved job fit (H1). However, the fit during the assimilation period of hospitalists who switched jobs twice or more was statistically no different from the fit of those in their first jobs, suggesting that the effect of the attrition‐reselection strategy is weak or inconsistent. The job fit for advancement period hospitalists was also different across the job change and no‐change categories. However, in the case of hospitalists later in their jobs, the median job fit was slightly but statistically lower among those who made job changes, revealing the potential drop in job fit that occurs when a hospitalist already established in his or her job starts over again in a new setting.

We hypothesized that hospitalists who achieved high job fit within a particular job were more likely to have engaged in activities that utilize a wider spectrum of their abilities. As shown in Table 4, hospitalists in the highest quartile of job fit were associated with a general trend toward higher odds of participating in a variety of common clinical and nonclinical hospitalist activities, but only the odds ratio associated with teaching achieved statistical significance (OR: 1.53, 95% CI: 1.01‐2.31) (H2).

Socialization with peers and the gradual sharing of values within organizations are hypothesized mechanisms for increasing job fit with time. We found that the number of years in current practice was positively correlated with job fit (Spearman coefficient R = 0.149, P < 0.001), organizational climate (R = 0.128, P < 0.001), and relationship with nonphysician staff (R = 0.102, P < 0.01). The association between years in practice and relationship with physician colleagues were weaker (R = 0.079, P < 0.05). Consistent with the episodic nature of patients' encounters with hospitalists, the measure of patient relationships was not significantly associated with length of time in job. In addition, we found substantial correlations among job fit, organizational climate, and all the relational measures (all R > 0.280, P < 0.001), indicating that hospitalists increasingly share the values of their organizations over time (H3).

Finally, we also hypothesized that poor job fit is associated with poor performance and quality of work life. Strong correlations with job fit were noted for stress (R = 0.307, P < 0.001), job burnout (R = 0.360, P < 0.001), and job satisfaction (R = 0.570, P < 0.001). Job fit (R = 0.147, P < 0.001), job burnout (R = 0.236, P < 0.001), stress (R = 0.305, P < 0.001), and job satisfaction (R = 0.224, P < 0.001) were all significantly correlated with the frequency of participating in suboptimal care (H4).

DISCUSSION

In this exploratory analysis, we validated in the hospitalist workforce several assumptions about person‐job fit that have been observed in workers of other industries. We observed attrition‐reselection (ie, job switching) as a strategy used by physicians to achieve better fit early in their job tenure, whereas job modification appeared to be more effective than attrition‐reselection among physicians already established in their jobs. We provided weak but plausible evidence that physicians with optimal job fit had a tendency to participate in activities (eg, teaching) that engage a wider set of interests and abilities. We also demonstrated the growth in hospitalists sharing the values of their organization through the time‐dependent associations among organizational climate, relational measures, and job fit. Finally, we found that physicians with suboptimal job fit were more likely to report poor performance in their work compared to those indicating optimal fit.

Our previous analysis of data from the Hospital Medicine Physician Worklife Survey exposed the widely variable work characteristics of hospitalist jobs in the US market and the equally variable preferences and priorities of individual hospitalists in selecting their work setting.[7] The implication of our present study is that hospitalists achieve the high levels of observed job fit using various strategies that aid their alignment with their employment. One of these strategies involves time, but physician longevity in practice may be both a determinant and product of good job fit. Although early job attrition may be necessary for fitting the right hospitalists to the right jobs, employers may appreciate the importance of retaining experienced hospitalists not only for cost and performance considerations but also for the growth of social capital in organizations consisting of enduring individuals. As our data suggest that hospitalists grow with their jobs, physicians may experience better fit with jobs that flexibly couple their work demands with benefits that address their individual work‐life needs over time. Another implication of this study is that job fit is a useful and predictive measure of job selection, performance, and retention. In light of studies that expose the limitations of job satisfaction as a measure influenced more by workers' dispositional affect (ie, their temperament and outlook) than their compatibility with their jobs,[28] job fit may add a functional dimension to traditional employee feedback measures.

There are limitations to this analysis. The most notable is the low survey response rate. Two reasons contributed to the fairly low rate of return. First, the original sampling frame included many outdated addresses and names of individuals who did not meet inclusion criteria. Although all sampled individuals who would have been excluded from the study could not be identified, we calculated our response rate without accounting for the proportion of potential ineligibles in the denominator population [Response Rate 2 (RR2) according to standards of the American Association of Public Opinion Research].[29] Second, the response rates of physician surveys have seen a steady decline over the years.[30] Respondents to our survey may be older and more experienced than US hospitalists in general. Although concerns about bias from under‐reporting cannot be fully addressed, we believe that the study sample is adequate for this preliminary study intended to translate the evidence of observed phenomena from the nonphysician to the physician workforces. The suboptimal response characteristics (high skew and low variability) of the generic person‐job fit survey scale used in this study indicate that a reliable survey instrument specifically designed to measure physician‐job fit need to be constructed de novo and validated for any future study. Although we performed simple analyses to support our assertions, few of our subanalyses may be underpowered, contributing to overinterpretation of the data. Additional empirical work is also necessary to assess the generalizability of this study's claims in other medical and surgical specialties. Such studies would also allow measurement of the sensitivity and specificity of physicians' self‐identification of poor job fit. Finally, additional investigations of this time‐dependent construct are more appropriately performed using a longitudinal study design to overcome the limitations inherent in this cross‐sectional analysis. Our conclusions about the time‐dependent features of job fit may be explained by other characteristics such as generational and cultural differences among hospitalists with varying experience.

As the US healthcare system reorganizes to bolster accountability,[31] we anticipate increasing interdependence between physicians and their employer organizations. Ultimately, the desired accountability in healthcare is likely to be obtained if physicians function not only as passive and interchangeable employees but as active stakeholders in the achievement of each organization's goals. A methodology for assessing the alignment of physicians and their jobs will continue to be important along the way.

Person‐organization fit concerns the conditions and consequences of compatibility between people and the organizations for which they work.[1] Studies of other industries have demonstrated that person‐organization fit informs the way individuals join, perform in, and are retained by organizations.[2] Person‐job fit is a closely related subordinate concept that concerns the alignment of workers and their job in as much as workers have needs that their job supplies, or conversely, jobs have requirements that certain workers' abilities can help meet.[3] Explorations of job fit in physicians and their work have recently emerged in a few investigations published in medical journals.[4, 5, 6, 7, 8] Further expanding the understanding of fit between physicians and their employment is important, because the decline of solo practices and recent emphasis on team‐based care have led to a growing number of US physicians working in organizations.[9]

The movement of physicians into employed situations may continue if certain types of Accountable Care Organizations take root.[10] And physicians may be primed to join employer organizations based on current career priorities of individuals in American society. Surveys of medical residents entering the workforce reveal more physicians preferring the security of being employees than starting their own practices.[11] Given these trends, job fit will inform our understanding of how personal and job characteristics facilitate recruitment, performance, satisfaction, and longevity of physician employees.

BACKGROUND

Virtually all hospitalists work in organizationshospitalsand are employees of hospitals, medical schools, physician group practices, or management companies, and therefore invariably function within organizational structures and systems.[7] In spite of their rapid growth in numbers, many employers have faced difficulties recruiting and retaining enough hospitalists to fill their staffing needs. Consequently, the US hospitalist workforce today is characterized by high salaries, work load, and attrition rates.[12]

In this evolving unsaturated market, the attraction‐selection‐attrition framework[13] provides a theoretical construct that predicts that hospitalists and their employers would seek congruence of goals and values early in their relationship through a process of trial and error. This framework assumes that early interactions between workers and organizations serve as opportunities for them to understand if job fit is poor and dissociate or remain affiliated as long as job fit is mutually acceptable. Therefore, job switching on average is expected to increase job fit because workers and organizations gain a better understanding of their own goals and values and choose more wisely the next time.

Other theoretical frameworks, such as the job characteristic model,[14] suggest that over time as workers stay at the same job, they continue to maintain and improve job fit through various workplace‐ or self‐modification strategies. For example, seniority status may have privileges (eg, less undesirable call), or workers may create privileged niches through the acquisition of new skills and abilities over time. Hospitalists' tendency to diversify their work‐related activities by incorporating administrative and teaching responsibilities[15] may thus contribute to improving job fit. Additionally, as a measure of complementarity among people who work together, job fit may be influenced by the quality of relationships among hospitalists and their coworkers through their reorientation to the prevailing organizational climate[16, 17] and increasing socialization.[18] Finally, given that experiential learning is known to contribute to better hospitalist work performance,[19] job fit may affect productivity and clinical outcomes vis‐‐vis quality of work life.

To test the validity of these assumptions in a sample of hospitalists, we critically appraised the following 4 hypotheses:

• Hypothesis 1 (H1): Job attrition and reselection improves job fit among hospitalists entering the job market.
• Hypothesis 2 (H2): Better job fit is achieved through hospitalists engaging a variety of personal skills and abilities.
• Hypothesis 3 (H3): Job fit increases with hospitalists' job duration together with socialization and internalization of organizational values.
• Hypothesis 4 (H4): Job fit is correlated with hospitalists' quality of work life.

METHODS

Analysis was performed on data from the 2009 to 2010 Hospital Medicine Physician Worklife Survey. The sample frame included nonmembers and members of Society of Hospital Medicine (SHM). Details about sampling strategy, data collection, and data quality are available in previous publications.[7, 20] The 118‐item survey instrument, including 9 demographic items and 24 practice and job characteristic items, was administered by mail. Examples of information solicited through these items included respondents' practice model, the number of hospitalist jobs they have held, and the specific kinds of clinical and nonclinical activities they performed as part of their current job.

We used a reliable but broad and generic measure of self‐perceived person‐job fit.[21] The survey items of the 5‐point Likert‐type scale anchored between strongly disagree and strongly agree were: I feel that my work utilizes my full abilities, I feel competent and fully able to handle my job, my job gives me the chance to do the things I feel I do best, I feel that my job and I are well‐matched, I feel I have adequate preparation for the job I now hold. The quality of hospitalists' relationships with physician colleagues, staff, and patients as well as job satisfaction was measured using scales adapted from the Physician Worklife Study.[22] Organizational climate was measured using an adapted scale from the Minimizing Error, Maximizing Outcome study incorporating 3 items from the cohesiveness subscale, 4 items from the organizational trust subscale, and 1 item from the quality emphasis subscale that were most pertinent to hospitalists' relationship with their organizations.[23] Intent to leave practice or reduce work hours was measured using 5 items from the Multi‐Center Hospitalist Survey Project.[24] Frequency of participation in suboptimal patient care was measured by adapting 3 items from the suboptimal reported practice subscale and 2 items from the suboptimal patient care subscale developed by Shanafelt et al.[25] Stress and job burnout were assessed using validated measures.[26, 27] Detailed descriptions of the response rate calculation and imputation of missing item data are available in previous publications.[7, 20]

Mean, variance, range, and skew were used to characterize the responses to the job fit survey scale. A table of respondent characteristics was constructed. A visual representation of job fit by individual hospitalist year in current practice was created, first, by plotting a locally weighted scatterplot smoothing curve to examine the shape of the general relationship, and second, by fitting a similarly contoured functional polynomial curve with 95% confidence intervals (CI) to a plot of the mean and interquartile range of job fit for each year in current practice. Spearman partial correlations were calculated for job fit and each of the 5 items addressing likelihood of leaving practice or reducing workload adjusted for gender to control for the higher proportion of women who plan to work part time. Median (interquartile range) job fit was calculated for categories defined by the number of job changes and compared with the reference category (no job change) using the nonparametric rank sum test for comparing non‐normally distributed data. Multivariate logistic regression models were used to calculate the odds ratio (OR) of participating in each of several clinical and nonclinical hospitalist activities between respondents whose job fit score was optimal (5 on a 5‐point scale) and less than optimal controlling for covariates that influence the likelihood of participating in these activities (years in current practice, practice model, and specialty training). A Spearman correlation matrix was created to assess interscale correlations among organizational parameters (years in current practice, job fit, organizational climate, and relationship with colleagues, staff, and patients). Finally, a separate Spearman correlation matrix was created to assess the interscale correlations among individual worker parameters (job fit, suboptimal patient care, job burnout, stress, and job satisfaction). Statistical significance was defined as P value <0.05, and all analyses were performed on Stata 11.0 (StataCorp, College Station, TX). The Northwestern University institutional review board approved this study.

RESULTS

Respondents included 816 hospitalists belonging to around 700 unique organizations. The adjusted response rate from the stratified sample was 26%. Respondents and nonrespondents were similar with regard to geographic region and model of practice, but respondents were more likely to be members of the SHM than nonrespondents. Panel A of Table 1 shows the demographic characteristics of the respondents. The mean age was 44.3 years, and about one‐third were women. The average hospitalist had about 7 years of experience in the specialty and about 5 years with their current hospitalist job. The majority were trained in internal medicine or one of its subspecialties, whereas pediatricians, family physicians, and physicians with other training made up the remainder.

Job fit was highly skewed toward optimum fit, with a mean of 4.3 on a scale of 1 to 5, with a narrow standard deviation of 0.7. The poorest job fit was reported by 0.3%, whereas optimal fit was reported by 21% of respondents. Job fit plotted against years in current practice had a logarithmic appearance typical of learning curves (Figure 1). An inflection point was visualized at around 2 years. For the purposes of this article, we refer to hospitalists' experience in the first 2 years of a job as an assimilation period, which is marked by a steep increase in job fit early when rapid learning or attrition took place. The years beyond the inflection point are characterized as an advancement period, when a more attenuated rise in job fit was experienced with time. The Spearman correlation between job fit and years in practice during the advancement period was 0.145 (n = 678, P < 0.001). Panel B of Table 1 displays the characteristics of respondents separately for the assimilation and advancement cohorts. Assimilation hospitalists in our sample had a mean age of 41.9 years and mean on‐the‐job experience of 4.3 years, reflecting that many hospitalists in the first 2 years of a job have made at least 1 job change in the past.

Figure 1

To show the effects of attrition and reselection, we first evaluated the proposition that hospitalists experience attrition (ie, intend to leave their jobs) in response to poor fit. Table 2 shows the correlations between job fit and the self‐reported intent to leave practice or reduce workload separately for the assimilation and advancement periods. For hospitalists in the assimilation period, job fit was negatively correlated with intent to leave current practice within 2 years and to leave hospital medicine within 5 years (P = 0.010 and 0.043, respectively). Hospitalists with <2 years in their current job, therefore, tended to consider attrition but not workload reduction to deal with poor job fit. On the other hand, hospitalists in the advancement period considered both attrition and workload reduction strategies in response to poor fit (all P < 0.001).

In Table 3, we further compared the median job fit across categories for the number of job switches. The median job fit during the assimilation period of hospitalists who had made 1 job change was slightly but statistically higher than the job fit of their counterparts who never left their first job (4.4 vs 4.0, P = 0.046). This suggests that job switching by hospitalists early in their jobs is associated with improved job fit (H1). However, the fit during the assimilation period of hospitalists who switched jobs twice or more was statistically no different from the fit of those in their first jobs, suggesting that the effect of the attrition‐reselection strategy is weak or inconsistent. The job fit for advancement period hospitalists was also different across the job change and no‐change categories. However, in the case of hospitalists later in their jobs, the median job fit was slightly but statistically lower among those who made job changes, revealing the potential drop in job fit that occurs when a hospitalist already established in his or her job starts over again in a new setting.

We hypothesized that hospitalists who achieved high job fit within a particular job were more likely to have engaged in activities that utilize a wider spectrum of their abilities. As shown in Table 4, hospitalists in the highest quartile of job fit were associated with a general trend toward higher odds of participating in a variety of common clinical and nonclinical hospitalist activities, but only the odds ratio associated with teaching achieved statistical significance (OR: 1.53, 95% CI: 1.01‐2.31) (H2).

Socialization with peers and the gradual sharing of values within organizations are hypothesized mechanisms for increasing job fit with time. We found that the number of years in current practice was positively correlated with job fit (Spearman coefficient R = 0.149, P < 0.001), organizational climate (R = 0.128, P < 0.001), and relationship with nonphysician staff (R = 0.102, P < 0.01). The association between years in practice and relationship with physician colleagues were weaker (R = 0.079, P < 0.05). Consistent with the episodic nature of patients' encounters with hospitalists, the measure of patient relationships was not significantly associated with length of time in job. In addition, we found substantial correlations among job fit, organizational climate, and all the relational measures (all R > 0.280, P < 0.001), indicating that hospitalists increasingly share the values of their organizations over time (H3).

Finally, we also hypothesized that poor job fit is associated with poor performance and quality of work life. Strong correlations with job fit were noted for stress (R = 0.307, P < 0.001), job burnout (R = 0.360, P < 0.001), and job satisfaction (R = 0.570, P < 0.001). Job fit (R = 0.147, P < 0.001), job burnout (R = 0.236, P < 0.001), stress (R = 0.305, P < 0.001), and job satisfaction (R = 0.224, P < 0.001) were all significantly correlated with the frequency of participating in suboptimal care (H4).

DISCUSSION

In this exploratory analysis, we validated in the hospitalist workforce several assumptions about person‐job fit that have been observed in workers of other industries. We observed attrition‐reselection (ie, job switching) as a strategy used by physicians to achieve better fit early in their job tenure, whereas job modification appeared to be more effective than attrition‐reselection among physicians already established in their jobs. We provided weak but plausible evidence that physicians with optimal job fit had a tendency to participate in activities (eg, teaching) that engage a wider set of interests and abilities. We also demonstrated the growth in hospitalists sharing the values of their organization through the time‐dependent associations among organizational climate, relational measures, and job fit. Finally, we found that physicians with suboptimal job fit were more likely to report poor performance in their work compared to those indicating optimal fit.

Our previous analysis of data from the Hospital Medicine Physician Worklife Survey exposed the widely variable work characteristics of hospitalist jobs in the US market and the equally variable preferences and priorities of individual hospitalists in selecting their work setting.[7] The implication of our present study is that hospitalists achieve the high levels of observed job fit using various strategies that aid their alignment with their employment. One of these strategies involves time, but physician longevity in practice may be both a determinant and product of good job fit. Although early job attrition may be necessary for fitting the right hospitalists to the right jobs, employers may appreciate the importance of retaining experienced hospitalists not only for cost and performance considerations but also for the growth of social capital in organizations consisting of enduring individuals. As our data suggest that hospitalists grow with their jobs, physicians may experience better fit with jobs that flexibly couple their work demands with benefits that address their individual work‐life needs over time. Another implication of this study is that job fit is a useful and predictive measure of job selection, performance, and retention. In light of studies that expose the limitations of job satisfaction as a measure influenced more by workers' dispositional affect (ie, their temperament and outlook) than their compatibility with their jobs,[28] job fit may add a functional dimension to traditional employee feedback measures.

There are limitations to this analysis. The most notable is the low survey response rate. Two reasons contributed to the fairly low rate of return. First, the original sampling frame included many outdated addresses and names of individuals who did not meet inclusion criteria. Although all sampled individuals who would have been excluded from the study could not be identified, we calculated our response rate without accounting for the proportion of potential ineligibles in the denominator population [Response Rate 2 (RR2) according to standards of the American Association of Public Opinion Research].[29] Second, the response rates of physician surveys have seen a steady decline over the years.[30] Respondents to our survey may be older and more experienced than US hospitalists in general. Although concerns about bias from under‐reporting cannot be fully addressed, we believe that the study sample is adequate for this preliminary study intended to translate the evidence of observed phenomena from the nonphysician to the physician workforces. The suboptimal response characteristics (high skew and low variability) of the generic person‐job fit survey scale used in this study indicate that a reliable survey instrument specifically designed to measure physician‐job fit need to be constructed de novo and validated for any future study. Although we performed simple analyses to support our assertions, few of our subanalyses may be underpowered, contributing to overinterpretation of the data. Additional empirical work is also necessary to assess the generalizability of this study's claims in other medical and surgical specialties. Such studies would also allow measurement of the sensitivity and specificity of physicians' self‐identification of poor job fit. Finally, additional investigations of this time‐dependent construct are more appropriately performed using a longitudinal study design to overcome the limitations inherent in this cross‐sectional analysis. Our conclusions about the time‐dependent features of job fit may be explained by other characteristics such as generational and cultural differences among hospitalists with varying experience.

As the US healthcare system reorganizes to bolster accountability,[31] we anticipate increasing interdependence between physicians and their employer organizations. Ultimately, the desired accountability in healthcare is likely to be obtained if physicians function not only as passive and interchangeable employees but as active stakeholders in the achievement of each organization's goals. A methodology for assessing the alignment of physicians and their jobs will continue to be important along the way.

Person‐organization fit concerns the conditions and consequences of compatibility between people and the organizations for which they work.[1] Studies of other industries have demonstrated that person‐organization fit informs the way individuals join, perform in, and are retained by organizations.[2] Person‐job fit is a closely related subordinate concept that concerns the alignment of workers and their job in as much as workers have needs that their job supplies, or conversely, jobs have requirements that certain workers' abilities can help meet.[3] Explorations of job fit in physicians and their work have recently emerged in a few investigations published in medical journals.[4, 5, 6, 7, 8] Further expanding the understanding of fit between physicians and their employment is important, because the decline of solo practices and recent emphasis on team‐based care have led to a growing number of US physicians working in organizations.[9]

The movement of physicians into employed situations may continue if certain types of Accountable Care Organizations take root.[10] And physicians may be primed to join employer organizations based on current career priorities of individuals in American society. Surveys of medical residents entering the workforce reveal more physicians preferring the security of being employees than starting their own practices.[11] Given these trends, job fit will inform our understanding of how personal and job characteristics facilitate recruitment, performance, satisfaction, and longevity of physician employees.

BACKGROUND

Virtually all hospitalists work in organizationshospitalsand are employees of hospitals, medical schools, physician group practices, or management companies, and therefore invariably function within organizational structures and systems.[7] In spite of their rapid growth in numbers, many employers have faced difficulties recruiting and retaining enough hospitalists to fill their staffing needs. Consequently, the US hospitalist workforce today is characterized by high salaries, work load, and attrition rates.[12]

In this evolving unsaturated market, the attraction‐selection‐attrition framework[13] provides a theoretical construct that predicts that hospitalists and their employers would seek congruence of goals and values early in their relationship through a process of trial and error. This framework assumes that early interactions between workers and organizations serve as opportunities for them to understand if job fit is poor and dissociate or remain affiliated as long as job fit is mutually acceptable. Therefore, job switching on average is expected to increase job fit because workers and organizations gain a better understanding of their own goals and values and choose more wisely the next time.

Other theoretical frameworks, such as the job characteristic model,[14] suggest that over time as workers stay at the same job, they continue to maintain and improve job fit through various workplace‐ or self‐modification strategies. For example, seniority status may have privileges (eg, less undesirable call), or workers may create privileged niches through the acquisition of new skills and abilities over time. Hospitalists' tendency to diversify their work‐related activities by incorporating administrative and teaching responsibilities[15] may thus contribute to improving job fit. Additionally, as a measure of complementarity among people who work together, job fit may be influenced by the quality of relationships among hospitalists and their coworkers through their reorientation to the prevailing organizational climate[16, 17] and increasing socialization.[18] Finally, given that experiential learning is known to contribute to better hospitalist work performance,[19] job fit may affect productivity and clinical outcomes vis‐‐vis quality of work life.

To test the validity of these assumptions in a sample of hospitalists, we critically appraised the following 4 hypotheses:

• Hypothesis 1 (H1): Job attrition and reselection improves job fit among hospitalists entering the job market.
• Hypothesis 2 (H2): Better job fit is achieved through hospitalists engaging a variety of personal skills and abilities.
• Hypothesis 3 (H3): Job fit increases with hospitalists' job duration together with socialization and internalization of organizational values.
• Hypothesis 4 (H4): Job fit is correlated with hospitalists' quality of work life.

METHODS

Analysis was performed on data from the 2009 to 2010 Hospital Medicine Physician Worklife Survey. The sample frame included nonmembers and members of Society of Hospital Medicine (SHM). Details about sampling strategy, data collection, and data quality are available in previous publications.[7, 20] The 118‐item survey instrument, including 9 demographic items and 24 practice and job characteristic items, was administered by mail. Examples of information solicited through these items included respondents' practice model, the number of hospitalist jobs they have held, and the specific kinds of clinical and nonclinical activities they performed as part of their current job.

We used a reliable but broad and generic measure of self‐perceived person‐job fit.[21] The survey items of the 5‐point Likert‐type scale anchored between strongly disagree and strongly agree were: I feel that my work utilizes my full abilities, I feel competent and fully able to handle my job, my job gives me the chance to do the things I feel I do best, I feel that my job and I are well‐matched, I feel I have adequate preparation for the job I now hold. The quality of hospitalists' relationships with physician colleagues, staff, and patients as well as job satisfaction was measured using scales adapted from the Physician Worklife Study.[22] Organizational climate was measured using an adapted scale from the Minimizing Error, Maximizing Outcome study incorporating 3 items from the cohesiveness subscale, 4 items from the organizational trust subscale, and 1 item from the quality emphasis subscale that were most pertinent to hospitalists' relationship with their organizations.[23] Intent to leave practice or reduce work hours was measured using 5 items from the Multi‐Center Hospitalist Survey Project.[24] Frequency of participation in suboptimal patient care was measured by adapting 3 items from the suboptimal reported practice subscale and 2 items from the suboptimal patient care subscale developed by Shanafelt et al.[25] Stress and job burnout were assessed using validated measures.[26, 27] Detailed descriptions of the response rate calculation and imputation of missing item data are available in previous publications.[7, 20]

Mean, variance, range, and skew were used to characterize the responses to the job fit survey scale. A table of respondent characteristics was constructed. A visual representation of job fit by individual hospitalist year in current practice was created, first, by plotting a locally weighted scatterplot smoothing curve to examine the shape of the general relationship, and second, by fitting a similarly contoured functional polynomial curve with 95% confidence intervals (CI) to a plot of the mean and interquartile range of job fit for each year in current practice. Spearman partial correlations were calculated for job fit and each of the 5 items addressing likelihood of leaving practice or reducing workload adjusted for gender to control for the higher proportion of women who plan to work part time. Median (interquartile range) job fit was calculated for categories defined by the number of job changes and compared with the reference category (no job change) using the nonparametric rank sum test for comparing non‐normally distributed data. Multivariate logistic regression models were used to calculate the odds ratio (OR) of participating in each of several clinical and nonclinical hospitalist activities between respondents whose job fit score was optimal (5 on a 5‐point scale) and less than optimal controlling for covariates that influence the likelihood of participating in these activities (years in current practice, practice model, and specialty training). A Spearman correlation matrix was created to assess interscale correlations among organizational parameters (years in current practice, job fit, organizational climate, and relationship with colleagues, staff, and patients). Finally, a separate Spearman correlation matrix was created to assess the interscale correlations among individual worker parameters (job fit, suboptimal patient care, job burnout, stress, and job satisfaction). Statistical significance was defined as P value <0.05, and all analyses were performed on Stata 11.0 (StataCorp, College Station, TX). The Northwestern University institutional review board approved this study.

RESULTS

Respondents included 816 hospitalists belonging to around 700 unique organizations. The adjusted response rate from the stratified sample was 26%. Respondents and nonrespondents were similar with regard to geographic region and model of practice, but respondents were more likely to be members of the SHM than nonrespondents. Panel A of Table 1 shows the demographic characteristics of the respondents. The mean age was 44.3 years, and about one‐third were women. The average hospitalist had about 7 years of experience in the specialty and about 5 years with their current hospitalist job. The majority were trained in internal medicine or one of its subspecialties, whereas pediatricians, family physicians, and physicians with other training made up the remainder.

Job fit was highly skewed toward optimum fit, with a mean of 4.3 on a scale of 1 to 5, with a narrow standard deviation of 0.7. The poorest job fit was reported by 0.3%, whereas optimal fit was reported by 21% of respondents. Job fit plotted against years in current practice had a logarithmic appearance typical of learning curves (Figure 1). An inflection point was visualized at around 2 years. For the purposes of this article, we refer to hospitalists' experience in the first 2 years of a job as an assimilation period, which is marked by a steep increase in job fit early when rapid learning or attrition took place. The years beyond the inflection point are characterized as an advancement period, when a more attenuated rise in job fit was experienced with time. The Spearman correlation between job fit and years in practice during the advancement period was 0.145 (n = 678, P < 0.001). Panel B of Table 1 displays the characteristics of respondents separately for the assimilation and advancement cohorts. Assimilation hospitalists in our sample had a mean age of 41.9 years and mean on‐the‐job experience of 4.3 years, reflecting that many hospitalists in the first 2 years of a job have made at least 1 job change in the past.

Figure 1

To show the effects of attrition and reselection, we first evaluated the proposition that hospitalists experience attrition (ie, intend to leave their jobs) in response to poor fit. Table 2 shows the correlations between job fit and the self‐reported intent to leave practice or reduce workload separately for the assimilation and advancement periods. For hospitalists in the assimilation period, job fit was negatively correlated with intent to leave current practice within 2 years and to leave hospital medicine within 5 years (P = 0.010 and 0.043, respectively). Hospitalists with <2 years in their current job, therefore, tended to consider attrition but not workload reduction to deal with poor job fit. On the other hand, hospitalists in the advancement period considered both attrition and workload reduction strategies in response to poor fit (all P < 0.001).

In Table 3, we further compared the median job fit across categories for the number of job switches. The median job fit during the assimilation period of hospitalists who had made 1 job change was slightly but statistically higher than the job fit of their counterparts who never left their first job (4.4 vs 4.0, P = 0.046). This suggests that job switching by hospitalists early in their jobs is associated with improved job fit (H1). However, the fit during the assimilation period of hospitalists who switched jobs twice or more was statistically no different from the fit of those in their first jobs, suggesting that the effect of the attrition‐reselection strategy is weak or inconsistent. The job fit for advancement period hospitalists was also different across the job change and no‐change categories. However, in the case of hospitalists later in their jobs, the median job fit was slightly but statistically lower among those who made job changes, revealing the potential drop in job fit that occurs when a hospitalist already established in his or her job starts over again in a new setting.

We hypothesized that hospitalists who achieved high job fit within a particular job were more likely to have engaged in activities that utilize a wider spectrum of their abilities. As shown in Table 4, hospitalists in the highest quartile of job fit were associated with a general trend toward higher odds of participating in a variety of common clinical and nonclinical hospitalist activities, but only the odds ratio associated with teaching achieved statistical significance (OR: 1.53, 95% CI: 1.01‐2.31) (H2).

Socialization with peers and the gradual sharing of values within organizations are hypothesized mechanisms for increasing job fit with time. We found that the number of years in current practice was positively correlated with job fit (Spearman coefficient R = 0.149, P < 0.001), organizational climate (R = 0.128, P < 0.001), and relationship with nonphysician staff (R = 0.102, P < 0.01). The association between years in practice and relationship with physician colleagues were weaker (R = 0.079, P < 0.05). Consistent with the episodic nature of patients' encounters with hospitalists, the measure of patient relationships was not significantly associated with length of time in job. In addition, we found substantial correlations among job fit, organizational climate, and all the relational measures (all R > 0.280, P < 0.001), indicating that hospitalists increasingly share the values of their organizations over time (H3).

Finally, we also hypothesized that poor job fit is associated with poor performance and quality of work life. Strong correlations with job fit were noted for stress (R = 0.307, P < 0.001), job burnout (R = 0.360, P < 0.001), and job satisfaction (R = 0.570, P < 0.001). Job fit (R = 0.147, P < 0.001), job burnout (R = 0.236, P < 0.001), stress (R = 0.305, P < 0.001), and job satisfaction (R = 0.224, P < 0.001) were all significantly correlated with the frequency of participating in suboptimal care (H4).

DISCUSSION

In this exploratory analysis, we validated in the hospitalist workforce several assumptions about person‐job fit that have been observed in workers of other industries. We observed attrition‐reselection (ie, job switching) as a strategy used by physicians to achieve better fit early in their job tenure, whereas job modification appeared to be more effective than attrition‐reselection among physicians already established in their jobs. We provided weak but plausible evidence that physicians with optimal job fit had a tendency to participate in activities (eg, teaching) that engage a wider set of interests and abilities. We also demonstrated the growth in hospitalists sharing the values of their organization through the time‐dependent associations among organizational climate, relational measures, and job fit. Finally, we found that physicians with suboptimal job fit were more likely to report poor performance in their work compared to those indicating optimal fit.

Our previous analysis of data from the Hospital Medicine Physician Worklife Survey exposed the widely variable work characteristics of hospitalist jobs in the US market and the equally variable preferences and priorities of individual hospitalists in selecting their work setting.[7] The implication of our present study is that hospitalists achieve the high levels of observed job fit using various strategies that aid their alignment with their employment. One of these strategies involves time, but physician longevity in practice may be both a determinant and product of good job fit. Although early job attrition may be necessary for fitting the right hospitalists to the right jobs, employers may appreciate the importance of retaining experienced hospitalists not only for cost and performance considerations but also for the growth of social capital in organizations consisting of enduring individuals. As our data suggest that hospitalists grow with their jobs, physicians may experience better fit with jobs that flexibly couple their work demands with benefits that address their individual work‐life needs over time. Another implication of this study is that job fit is a useful and predictive measure of job selection, performance, and retention. In light of studies that expose the limitations of job satisfaction as a measure influenced more by workers' dispositional affect (ie, their temperament and outlook) than their compatibility with their jobs,[28] job fit may add a functional dimension to traditional employee feedback measures.

There are limitations to this analysis. The most notable is the low survey response rate. Two reasons contributed to the fairly low rate of return. First, the original sampling frame included many outdated addresses and names of individuals who did not meet inclusion criteria. Although all sampled individuals who would have been excluded from the study could not be identified, we calculated our response rate without accounting for the proportion of potential ineligibles in the denominator population [Response Rate 2 (RR2) according to standards of the American Association of Public Opinion Research].[29] Second, the response rates of physician surveys have seen a steady decline over the years.[30] Respondents to our survey may be older and more experienced than US hospitalists in general. Although concerns about bias from under‐reporting cannot be fully addressed, we believe that the study sample is adequate for this preliminary study intended to translate the evidence of observed phenomena from the nonphysician to the physician workforces. The suboptimal response characteristics (high skew and low variability) of the generic person‐job fit survey scale used in this study indicate that a reliable survey instrument specifically designed to measure physician‐job fit need to be constructed de novo and validated for any future study. Although we performed simple analyses to support our assertions, few of our subanalyses may be underpowered, contributing to overinterpretation of the data. Additional empirical work is also necessary to assess the generalizability of this study's claims in other medical and surgical specialties. Such studies would also allow measurement of the sensitivity and specificity of physicians' self‐identification of poor job fit. Finally, additional investigations of this time‐dependent construct are more appropriately performed using a longitudinal study design to overcome the limitations inherent in this cross‐sectional analysis. Our conclusions about the time‐dependent features of job fit may be explained by other characteristics such as generational and cultural differences among hospitalists with varying experience.

As the US healthcare system reorganizes to bolster accountability,[31] we anticipate increasing interdependence between physicians and their employer organizations. Ultimately, the desired accountability in healthcare is likely to be obtained if physicians function not only as passive and interchangeable employees but as active stakeholders in the achievement of each organization's goals. A methodology for assessing the alignment of physicians and their jobs will continue to be important along the way.

Over the past 15 years, there has been dramatic growth in the number of hospitalist physicians in the United States and in the number of hospitals served by them.13 Hospitals are motivated to hire experienced hospitalists to staff their inpatient services,4 with goals that include obtaining cost‐savings and higher quality.59 The rapid growth of Hospital Medicine saw multiple types of hospital practice models emerge with differing job characteristics, clinical duties, workload, and compensation schemes.10 The extent of the variability of hospitalist jobs across practice models is not known.

Intensifying recruitment efforts and the concomitant increase in compensation for hospitalists over the last decade suggest that demand for hospitalists is strong and sustained.11 As a result, today's cohort of hospitalists has a wide range of choices of types of jobs, practice models, and locations. The diversity of available hospitalist jobs is characterized, for example, by setting (community hospital vs academic hospital), employer (hospital vs private practice), job duties (the amount and type of clinical work, and other administrative, teaching, or research duties), and intensity (work hours and duties to maximize income or lifestyle). How these choices relate to job satisfaction and burnout are also unknown.

The Society of Hospital Medicine (SHM) has administered surveys to hospitalist group leaders biennially since 2003.1215 These surveys, however, do not address issues related to individual hospitalist worklife, recruitment, and retention. In 2005, SHM convened a Career Satisfaction Task Force that designed and executed a national survey of hospitalists in 2009‐2010. The objective of this study is to evaluate how job characteristics vary by practice model, and the association of these characteristics and practice models with job satisfaction and burnout.

METHODS

RESULTS

Response Rate

Of the 5389 originally sampled addresses, 1868 were undeliverable. Addresses were further excluded if they appeared in duplicate or were outdated. This yielded a total of 3105 eligible surveyees in the sample. As illustrated in Figure 1, 841 responded to the mailed survey and 5 responded to the Web‐based survey. After rejecting 67 non‐hospitalist respondents and 3 duplicate surveys, a total of 776 surveys were included in the final analysis. The adjusted response rate was 25.6% (776/3035). Members of SHM were more likely to return the survey than nonmembers. The adjusted response rate from hospitalists affiliated with the 3 sponsoring institutions was 6% (40/662). Because these respondents were more likely to be non‐members of SHM, we opted to analyze the responses from the sponsor hospitalists together with the sampled hospitalists. The demographics of the resulting pool of 816 respondents affiliated with over 650 unique hospitalist groups were representative of the original survey frame. We analyzed data from 794 of these who responded to the item indicating their hospitalist practice model. Demographic characteristics of responders and nonresponders to the practice model survey item were similar.

Figure 1

Components of Job Satisfaction

Hospitalists' rankings of the most important factors for job satisfaction revealed differences across models (Figure 2). Overall, hospitalists were most likely to consider optimal workload and compensation as important factors for job satisfaction from a list of 13 considerations. Local groups and academics were least likely to rank optimal workload as a top factor, and local group hospitalists were more likely to rank optimal autonomy than those of other models. Academic hospitalists had less concern for substantial pay, and more concern for the variety of tasks they perform and recognition by leaders, than other hospitalists.

Figure 2

Job Satisfaction and Burnout Risk

Differences in the ratings of 4 of the 11 satisfaction and job characteristic domains were found across the practice models (Figure 3). Multispecialty group hospitalists were less satisfied with autonomy and their relationship with patients than other practice models, and along with multistate groups, reported the highest perceived workload. Organizational fairness was rated much higher by local group hospitalists than other practice models. Despite these differences in work patterns and satisfaction, there were no differences found in level of global job satisfaction, specialty satisfaction, or burnout across the practice models. Overall, 62% of respondents reported high job satisfaction (4 on a 1 to 5 scale), and 30% indicated burnout symptoms.

Figure 3

DISCUSSION

In our sample of US hospitalists, we found major differences in work patterns and compensation across hospitalist practice models, but no differences in job satisfaction, specialty satisfaction, and burnout. In particular, differences across these models included variations in hospitalist workload, hours, pay, and distribution of work activities. We found that hospitalists perform a variety of clinical and nonclinical tasks, for many of which there are not standard reimbursement mechanisms. We also found that features of a job that individual hospitalists considered most important vary by practice model.

Previous analysis of this data explored the overall state of hospitalist satisfaction.16 The present analysis offers a glimpse into hospitalists' systems‐orientation through a deeper look at their work patterns. The growth in the number of hospitalists who participate in intensive care medicine, specialty comanagement, and other work that involves close working relationships with specialist physicians confirms collaborative care as one of the dominant drivers of the hospitalist movement. At the level of indirect patient care, nearly all hospitalists contributed to work that facilitates coordination, quality, patient safety, or information technology. Understanding the integrative value of hospitalists outside of their clinical productivity may be of interest to hospital administrators.

Global satisfaction measures were similar across practice models. This finding is particularly interesting given the major differences in job characteristics seen among the practice models. This similarity in global satisfaction despite real differences in the nature of the job suggests that individuals find settings that allow them to address their individual professional goals. Our study demonstrates that, in 2010, Hospital Medicine has evolved enough to accommodate a wide variety of goals and needs.

While global satisfaction did not differ among practice types, hospitalists from various models did report differences in factors considered important to global satisfaction. While workload and pay were rated as influential across most models, the degree of importance was significantly different. In academic settings, substantial pay was not a top consideration for overall job satisfaction, whereas in local and multistate hospitalist groups, pay was a very close second in importance to optimal workload. These results may prove helpful for individual hospitalists trying to find their optimal job. For example, someone who is less concerned about workload, but wants to be paid well and have a high degree of autonomy, may find satisfaction in local hospitalist groups. However, for someone who is willing to sacrifice a higher salary for variety of activities, academic Hospital Medicine may be a better fit.

There is a concerning aspect of hospitalist job satisfaction that different practice models do not seem to solve. Control over personal time is a top consideration for many hospitalists across practice models, yet their satisfaction with personal time is low. As control over personal time is seen as a draw to the Hospital Medicine specialty, group leaders may need to evaluate their programs to ensure that schedules and workload support efforts for hospitalists to balance work and homelife commitments.

There are additional findings that are important for Hospital Medicine group leaders. Regardless of practice model, compensation and workload are often used as tools to recruit and retain hospitalists. While these tools may be effective, leaders may find more nuanced approaches to improving their hospitalists' overall satisfaction. Leaders of local hospitalist groups may find their hospitalists tolerant of heavier workloads as long as they are adequately rewarded and are given real autonomy over their work. However, leaders of academic programs may be missing the primary factor that can improve their hospitalists' satisfaction. Rather than asking for higher salaries to remain competitive, it may be more effective to advocate for time and training for their hospitalists to pursue important other activities beyond direct clinical care. Given that resources will always be limited, group leaders need to understand all of the elements that can contribute to hospitalist job satisfaction.

We point out several limitations to this study. First, our adjusted response rate of 25.6% is low for survey research, in general. As mentioned above, hospitalists are not easily identified in any available national physician database. Therefore, we deliberately designed our sampling strategy to error on the side of including ineligible surveyees to reduce systematic exclusion of practicing hospitalists. Using simple post hoc methods, we identified many nonhospitalists and bad addresses from our sample, but because these methods were exclusionary as opposed to confirmatory, we believe that a significant proportion of remaining nonrespondents may also have been ineligible for the survey. Although this does not fully address concerns about potential response bias, we believe that our sample representing a large number of hospitalist groups is adequate to make estimations about a nationally representative sample of practicing hospitalists. Second, in spite of our inclusive approach, we may still have excluded categories of practicing hospitalists. We were careful not to allow SHM members to represent all US hospitalists and included non‐members in the sampling frame, but the possibility of systematic exclusion that may alter our results remains a concern. Additionally, one of our goals was to characterize pediatric hospitalists independently from their adult‐patient counterparts. Despite oversampling of pediatricians, their sample was too small for a more detailed comparison across practice models. Also, self‐reported data about workload and compensation are subject to inaccuracies related to recall and cognitive biases. Last, this is a cross‐sectional study of hospitalist satisfaction at one point in time. Consequently, our sample may not be representative of very dissatisfied hospitalists who have already left their jobs.

The diversity found across existing practice models and the characteristics of the practices provide physicians with the opportunity to bring their unique skills and motivations to the hospitalist movement. As hospitals and other organizations seek to create, maintain, or grow hospitalist programs, the data provided here may prove useful to understand the relationship between practice characteristics and individual job satisfaction. Additionally, hospitalists looking for a job can consider these results as additional information to guide their choice of practice model and work patterns.

Over the past 15 years, there has been dramatic growth in the number of hospitalist physicians in the United States and in the number of hospitals served by them.13 Hospitals are motivated to hire experienced hospitalists to staff their inpatient services,4 with goals that include obtaining cost‐savings and higher quality.59 The rapid growth of Hospital Medicine saw multiple types of hospital practice models emerge with differing job characteristics, clinical duties, workload, and compensation schemes.10 The extent of the variability of hospitalist jobs across practice models is not known.

Intensifying recruitment efforts and the concomitant increase in compensation for hospitalists over the last decade suggest that demand for hospitalists is strong and sustained.11 As a result, today's cohort of hospitalists has a wide range of choices of types of jobs, practice models, and locations. The diversity of available hospitalist jobs is characterized, for example, by setting (community hospital vs academic hospital), employer (hospital vs private practice), job duties (the amount and type of clinical work, and other administrative, teaching, or research duties), and intensity (work hours and duties to maximize income or lifestyle). How these choices relate to job satisfaction and burnout are also unknown.

The Society of Hospital Medicine (SHM) has administered surveys to hospitalist group leaders biennially since 2003.1215 These surveys, however, do not address issues related to individual hospitalist worklife, recruitment, and retention. In 2005, SHM convened a Career Satisfaction Task Force that designed and executed a national survey of hospitalists in 2009‐2010. The objective of this study is to evaluate how job characteristics vary by practice model, and the association of these characteristics and practice models with job satisfaction and burnout.

METHODS

RESULTS

Response Rate

Of the 5389 originally sampled addresses, 1868 were undeliverable. Addresses were further excluded if they appeared in duplicate or were outdated. This yielded a total of 3105 eligible surveyees in the sample. As illustrated in Figure 1, 841 responded to the mailed survey and 5 responded to the Web‐based survey. After rejecting 67 non‐hospitalist respondents and 3 duplicate surveys, a total of 776 surveys were included in the final analysis. The adjusted response rate was 25.6% (776/3035). Members of SHM were more likely to return the survey than nonmembers. The adjusted response rate from hospitalists affiliated with the 3 sponsoring institutions was 6% (40/662). Because these respondents were more likely to be non‐members of SHM, we opted to analyze the responses from the sponsor hospitalists together with the sampled hospitalists. The demographics of the resulting pool of 816 respondents affiliated with over 650 unique hospitalist groups were representative of the original survey frame. We analyzed data from 794 of these who responded to the item indicating their hospitalist practice model. Demographic characteristics of responders and nonresponders to the practice model survey item were similar.

Figure 1