Burnout and job satisfaction among physicians working in urologic departments who immigrated to Germany,taking into account the income structure of their country of origin

2.1 Development and structure of the questionnaire

With the aim of conducting a study on the topic of “Assessment of their urological working environment by foreign clinicians in Germany” (EUTAKD is the German acronym of this study title), an interdisciplinary working group was constituted in January 2020. A survey comprising 101 items was developed based on the questions of a Saxon cross-sectional study, which was subsequently adapted and validated in 12 structured separate interviews with urologists (born in Germany and abroad) to verify the comprehensibility of the questions [2, 25]. The development, structuring and evaluation of the questionnaire was carried out in close consideration of the CHERRIES (Checklist for Reporting Results of Internet E-Surveys) guideline [26].

The EUTAKD questionnaire contains 12 items on the general characteristics of the participants (questions 1-10,39,40), 25 items on medical training and activity (questions 11-31,33-36), three items on immigration to Germany (questions 32,37,38), the complete MBI with a total of 22 items (questions 41.1-41.22) as well as 39 items on private and job satisfaction (42.1-42.39).

The MBI is a robust-validated instrument for quantifying PBO and contains three dimensions: A) emotional exhaustion (EE), B) depersonalization/loss of empathy (DP) and C) reduction in personal accomplishment (RPA) [4–11]. The sum score of the dimensions EE and DP correlates positively with PBO, while the sum score of the dimension RPA correlates negatively with PBO. The three burnout dimensions can be categorized into low, moderate and high risk of burnout: for EE 0–16, 17–26, and≥27, respectively; for DP: 0–6, 7–12, and≥13, respectively; and for RPA: ≥39, 32–38, and 0–31, respectively [4–10, 12–20]. The three burnout dimensions can also be dichotomized into high (critical) vs. low-moderate burnout risk. A high risk of burnout therefore exists at EE≥27, DP≥13 and/or RPA≤31 [4–10, 12–20]. A dual-combined end point was then still formed from the aggregated assessment of the risk constellations in the PBO dimensions EE and DP (EE≥27 and/or DP≥13).

Response options regarding satisfaction with private-life and job comprised a 5-point Likert scale (1-extremely dissatisfied, 2-rather dissatisfied, 3-partly dissatisfied/partly satisfied, 4-rather satisfied, 5-extremely satisfied).

2.2 Study implementation

The EUTAKD study was directed at all physicians currently working at a urology clinic in Germany who were born in a country other than Germany and had obtained non-German citizenship there (even if German citizenship had been applied for or obtained in the meantime). All physicians who did not meet these criteria were excluded from the study. An ethics vote has been given by the ethics committee of the University of Regensburg on 22.4.2020 (ZE: 20-1808-101).

The survey was conducted via Survey-Monkey^® (SurveyMonkey Inc., San Monte, USA) from August 1, 2020 until October 31, 2020. To raise public awareness for the EUTAKD study, access to the Survey-Monkey^® was advertised by means of a QR code in corresponding articles in the journals Uro-Forum and Uro-News (which are freely and widely available for every urologist in Germany), as well as by an information campaign via the Society of Urological Residents (GeSRU) [27]. Furthermore, all presidents of the regional urological societies were contacted twice and asked for their support by forwarding the study invitation to the various clinics of their regional society. Funding was provided by the Hans-Neuffer-Foundation.

2.3 Research question and statistical methods

The present analyses address the question of the extent to which IS-COO influences overall job satisfaction, satisfaction with the current income (received in Germany) and the risk of critical PBO during urological professional practice in Germany.

Endpoints of the study were the individual dimensions of burnout, i.e. EE, DP and RPA as well as the dual-combined endpoint of EE and DP, whereby a critical risk of PBO was defined as EE≥27 and/or DP≥13 [28]. In addition, overall job satisfaction and income satisfaction were further study endpoints. All endpoints were evaluated as sum-scores and dichotomized: satisfaction items were categorized as rather/extremely satisfied (Likert-scale 4-5) vs. partly satisfied/partly dissatisfied to extremely dissatisfied (Likert-scale 4-5 vs. 1-3). To evaluate the influence of time already spent in Germany by the study participants on the various endpoints, time was dichotomized in the median (≥7 vs. <7 years).

IS-COO based on income per capita reflecting the entire country population was classified into the three categories low-, middle- and high-income according to the World Bank criteria: Countries with an income of less than 1,045 USD/year are referred to as low-income countries (LIC) and those with a per capita income of more than 12,745 USD/year as high-income countries (HIC) [29]. The range of values in between (>1,045 to < 12,745 USD/year) represents the income per capita of a middle-income country (MIC).

All categorical group comparisons were performed exclusively with the Chi² test, whereas the comparisons of the sum scores in the individual PBO categories were always performed with the Mann-Whitney test (moreover, it was always specified in the tables exactly which statistical tests were used in the individual group comparison).

Multivariate-logistic regression models (MLRM) were used to test the influence of the following criteria on the various binary study endpoints: (1) IS-COO (categorical), (2) time in Germany (dichotomized in median), (3) gender (male vs. female) and (4) satisfaction with income (dichotomized in rather/extremely satisfied vs. partly/partly to extremely dissatisfied). The influence of the independent variables on the defined endpoints is presented with odds ratio (OR) including the 95% -confidence interval.

Data analysis was performed with SPSS 26.0 (IBM Corp., Armonk, NY, USA). The p-values given are always two-sided, the significance level was considered statistically significant at p≤.05 for all tests.

4.1 Summary of findings

To the best of the authors’ knowledge, these results of the EUTAKD study are absolutely novel and add data to the international literature on migrant physicians: ad 1) IS-COO has no significant influence on job satisfaction or the probability of burnout after immigration to a high-income country (such as Germany), ad 2) the lack of influence of IS-COO on these two endpoints remains if the duration of stay in Germany is taken into account (in other words, both early and late after immigration to Germany, the subdivision according to income structure of the country of origin has no influence on job satisfaction and burnout probability), ad 3) except for the burnout dimension RPA (which it improves), satisfaction with income earned in Germany shows no significant influence on other endpoints (including overall job satisfaction), ad 4) the time already spent in Germany by the urological colleagues is significantly associated with a lower risk for burnout (in other words, the longer the colleagues with migration background live in Germany, the lower the probability of critical burnout), and ad 5) the achievement of a critical PBO category in the dimensions DP and RPA are lowered by the dichotomized “time in Germany” (the median was 7 years in our study) by 72% and 91%, respectively.

4.2 Rationale of the present study and possible explanations for its findings, as well as considerations for future research approaches

With a per capita income averaged over all occupational groups of 57,810 USD/year, Germany is one of the high-income countries [29]. In 2019, 52,522 foreign doctors were employed in Germany [1]. How many foreign physicians work specifically in urology is currently not recorded. Urology is one of the specialties with a higher risk of PBO compared to other specialties. Previous studies described burnout rates of 52% for urologists in Ireland and the United Kingdom and up to 64% for US urologists [4, 12]. In the present study, a combined endpoint of the burnout dimensions EE and DP revealed a burnout high-risk category in 40% of the participants. The fact that the proportion of participants with burnout is comparatively low can possibly be explained by methodological differences, and differences between the characteristics of the study populations. In addition, Dyrbye et al. did not use the complete MBI in the US study to survey PBO [13]. In various studies, an inverse relationship between job satisfaction and burnout was observed [2–5].

Interestingly, in the present study participants with a length of stay in Germany of at least 7 years had a significantly lower risk of burnout as compared to doctors with a shorter length of stay. Length of stay and burnout may be mutually dependent – those who are dissatisfied and suffer from burnout leave the host country more quickly (and vice versa). This assumption is also supported by the observation that a smaller proportion of those who have been in Germany for at least 7 years intends to leave Germany and/or urology (41.3% vs. 60.0%, p = .119). It is likely that a longer period of residence leads to people acquiring resources that can reduce the risk of burnout: such as familiarity with the German healthcare system, recognition of professional qualifications acquired abroad, improved language skills, adaptation to the culture in Germany and the development of social networks. However, the observation that even among colleagues with a long-term stay in Germany, more than 40% would like to leave either the host country and/or the urological specialty raises concerns for the functionality of urology clinics in Germany.

Contrary to the initial working hypothesis, in the present analysis IS-COO had neither a significant influence on job satisfaction nor on satisfaction with the current income. Thus, the initial hypotheses of our study could not be confirmed and subsequent hypotheses presupposing this connection had to be rejected. An association between IS-COO and PBO was also not proven. A significant impact of satisfaction with income on PBO could be shown in the dimension RPA, but not on EE and DP.

In a recent study among intensive care physicians in China, Wang et al. demonstrated that higher income satisfaction was an independent protective factor against severe burnout [30]. A possible explanation for the lack of correlation in our study could be that job satisfaction is possibly multifactorial. Other determinants than IS-COO could have a major influence on job satisfaction. This would be good news for the development and implementation of interventions to improve job satisfaction, since determinants of job satisfaction identified in the EUTAKD and previous studies, such as working hours, stress reduction, small group discussions, mindfulness exercises and structured technical training, can be influenced and modified by the employer and also by the employee [8–10]. A sense of calling and a feeling of being significant is also described as a protection against dissatisfaction [31].

Interestingly, a study of Polish doctors by Dubas-Jakóbczy et al. showed that one of the main reasons for the intention to emigrate is the higher income in the immigration country [22]. Bazoukis et al. came to almost identical conclusions based on a study of Greek doctors [23]. So while the prospect of a higher income in the immigration country may be a driver of emigration, IS-COO is not necessarily associated with job satisfaction in the immigration country. In this comparison, it should be taken into account that the groups of LICs, MICs and HICs studies may include countries that are not comparable to Poland and Greece. Furthermore, the specific reasons for emigration were not considered in the present analysis.

Nevertheless, in physician migration it certainly plays a role which expectations are actually fulfilled in the immigration country or which aspects actually prove to be relevant. A possible discrepancy between expectation and reality can possibly also have an influence on job satisfaction and burnout risk among doctors.

The relationship between (expected and actual) income and job satisfaction or burnout should be investigated in further studies with prospective study designs. At the very least, further studies should explicitly consider the reasons for migration and the extent to which expectations have been fulfilled. In addition, future studies should address intervention options that can help reduce the relatively high proportion of physician colleagues with a migration background who want to leave Germany and/or the urological specialty again.

4.3 Strengths and limitations

To the best of the authors’ knowledge, this study is the first to provide data on the influence of IS-COO on burnout and job satisfaction of migrant physicians. Thus, the present analysis functions as a pioneer study for the urological specialty and probably also beyond.

When interpreting the study results, some limitations of the EUTAKD study should be taken into account: 1.) The response rate cannot be specified, since the number of migrant urologists working in German hospitals is unknown. It is not possible to characterize foreign urologists in Germany who decided not to participate in the study or who were not reached by the extensive and high-profile recruitment measures (see Study Implementation in the Material and Methods section). Thus, statements on the representativeness of the study results cannot be made. 2.) Not all participants who started the survey completed it (Fig. 1). Thus, only 60 complete data sets were available for the 39 satisfaction items, which had a strict implication for the formation of the multivariate models (to avoid overfitting). 3.) Due to the cross-sectional design, no causal or predictive statements can be made. 4.) IS-COO does not necessarily reflect the individual earnings and asset situation of physicians in these countries. 5.) This analysis only considered income structure of the country of origin but not working conditions and employment rights in the country of origin.

Therefore, the EUTAKD study is considered a descriptive and hypothesis generating pilot study [2, 32]. More comprehensive studies on this topic can certainly build on these initial results. Subsequent studies may consider a case-control study: immigrants-nonimmigrants in order to elucidate the influence of migration on job satisfaction and burnout.