key: cord-0030115-qar85zlr
authors: Offenberg, Luisa K.; Sommer, Samira T.; Schmidt, Manuela; Kasten, Stefanie; Bockheim, Florian; Gavrilov, Boris; Hunzelar, Carmen; Ikar, Nur; Oberholz, Maja P. S.; Paños-Willuhn, Joana L.; Weltermann, Birgitta
title: Higher chronic stress and less satisfaction with GP communication in migrants with unemployment experience: data from the representative German DEGS1 and the GPCare-1 study
date: 2022-04-20
journal: BMC Prim Care
DOI: 10.1186/s12875-022-01691-1
sha: 03b7a7db5039ca88d6e9814e723ab7506829c994
doc_id: 30115
cord_uid: qar85zlr

BACKGROUND: The impact of unemployment on health is well studied. However, information on associations of unemployment, migration background and general practitioner-patient communication is scarce. METHODS: Data from the representative German Health Interview and Examination Survey for Adults (DEGS1) of individuals in working age (n = 5938) were analysed stratified by unemployment and migration background. Using official weighting factors, the prevalence of chronic stress, having ≥1 chronic disease, having a GP and GP visits in the last 12 months was determined. Multivariate regression models were analysed for associations between unemployment, migration background, and other socio-demographic characteristics with GP visits and chronic stress. Data from the General Practice Care-1 (GPCare-1) study (n = 813 patients) were analysed for differences in patient-physician communication between unemployed with and without migration background. Reverse proportional odds models were estimated for associations of unemployment and migration background with physician-patient communication. RESULTS: In the DEGS1, 21.5% had experienced unemployment (n = 1170). Of these, 31.6% had a migration background (n = 248). Compared to unemployed natives, unemployed with migration background had higher chronic stress (mean: 14.32 vs. 13.13, p = 0.02), while the prevalence of chronic disease was lower (21.7% vs. 30.2%, p = 0.03). They were less likely to have a GP (83.6% vs. 90%, p = 0.02), while GP visits were similar (mean: 3.7 vs. 3.3, p = 0.26). Migration background and unemployment experience were not associated with GP visits, while both factors were significantly associated with higher chronic stress (both: p < 0.01). In GPCare-1, 28.8% had ever experienced unemployment (n = 215). Of these, 60 had a migration background (28.6%). The unemployed with migration background reported less frequently that the GP gives them enough space to describe personal strains (46.5% vs. 58.2%; p = 0.03), and that their problems are taken very seriously by their GP (50.8% vs. 73.8%; p = 0.04). In multivariate analyses, migration background showed a lower probability of having enough space to describe personal strains and feeling that problems were taken very seriously. CONCLUSION: Unemployment experience and migration background were associated with higher chronic stress. Only migration background was associated with less satisfaction regarding physician-patient communication.

In 2013, the International Labour Organization defined persons in unemployment as "all those of working age who were not in employment, carried out activities to seek employment during a specified recent period, and were currently available to take up employment given a job opportunity" [1] . In July 2020, 7.4% unemployed individuals were registered in Europe, and 4.4% in Germany respectively [2] . The Organisation for Economic Cooperation and Development (OECD) recently showed for Europe that foreign-born individuals were more frequently unemployed than natives [3] . These findings are a public health concern since unemployment was shown to be associated with a higher prevalence of chronic illnesses and mortality [4, 5] . Especially, higher rates of adverse mental outcomes such as depressive disorders, anxiety, insomnia and stress were shown [6, 7] . Based on 41 studies, a review by Norström et al. (2014) showed that most studies found a negative effect of unemployment on health but additional studies differentiating between certain subgroups are needed, e.g. on individuals with migration background [8] .

Scarce data on the impact of unemployment on the health of individuals with migration background showed for example that migrant men experienced the highest decrease compared to both males and females with employment in life satisfaction when becoming unemployed [9] . Furthermore, health status in unemployed migrants varies by country of origin, e.g., unemployed Turkish women, especially at high age, experiencing high levels of distress [10] . Given the current increase in migration populations, a better understanding of health outcomes in various migrant populations and the role of the health care system, especially primary care, is needed. In Germany alone the number of migrants increased between 2011 and 2016 from 6.3 to 9.2 million increasing the prevalence from 7.9 to 11.2% within the general population [11] .

General practitioners are relevant contact persons for individuals of all ages and backgrounds. Therefore, they have a strong position to address not only stress and chronic illnesses but also social issues such as unemployment. Although Zimmermann et al. (2018) showed that 43.3% of GPs were consulted at least three times a week by patients with work/unemployment problems [12] , data is limited regarding subpopulations with migration background. So far, research described differences between natives and individuals of different ethnicities in physician-patient communication in general [13, 14] , but did not specifically address physicianpatient communication regarding personal strains.

This study builds on data from the nationally representative German Health Interview and Examination Survey for Adults (DEGS1) and the General Practice Care-1 patient survey (GPCare-1). Using descriptive and multivariate analyses, it investigates associations between various socio-demographic parameters including unemployment experience and migration background with the number of GP visits in 12 months and the patientreported quality of GP-patient communication.

The study draws on data from two surveys: Conclusion: Unemployment experience and migration background were associated with higher chronic stress. Only migration background was associated with less satisfaction regarding physician-patient communication.

Keywords: Physician-patient communication, Primary health care, Chronic stress, Migrant, Unemployment, General practitioner 1. the representative German Health Interview and Examination Survey for Adults (DEGS1), 2. the General Practice Care 1 study (GPCare-1).

The DEGS1 study which is representative for the general German population was conducted by the Robert Koch Institute as part of the German health monitoring system (2008-2011). The DEGS1 study has a mixed design which allows for cross-sectional and longitudinal analyses. The study population was sampled by inviting participants according to the study protocol additionally to participants who had already participated in the German National Health Interview and Examination Survey 1998 (GNHIES98). The DEGS1 study included a total of 8152 participants. 3959 participants had already been part of the GNHIES98 while 4193 participants were newly recruited. In this paper participants were only included if data on experience of unemployment during the last 5 years was available; ergo, the study population in this paper consists of 5938 participants of working age (18 to 64 years old). More details regarding the study protocol of the DEGS1 can be found elsewhere [15] . The DEGS1 data used for this analysis were kindly provided by the Robert Koch Institute as public use file.

The following DEGS1 measurements of socio-demographic parameters were used for the analyses:

-Age (in years) -Sex (male, female) -Education level classified according to the International Standard Classification of Education [16] . -Socioeconomic status (SES) was calculated based on information regarding education, employment status and income, which was subsequently classified into low, middle and high socioeconomic status (for details see [17] ). -Number of underage individuals in household -Social support determined by the Oslo-3-Items-Social-Support Scale (Oslo-3) classified in three categories: low (3-8 points), middle (9-11 points), high (12-14 points) [18, 19] . -Financial dependency was determined by asking participants if they were the main breadwinner of the household.

Issues of migration and employment were addressed using the following items:

-Migration background was defined as the participant or at least one parent was born outside of Germany.

-Migration generation (first or second) -Nationality -Self-assessed knowledge of the German language -Current employment status (employed, unemployed) -Unemployment in the past 5 years (yes, no) -Length of unemployment during the past 5 years

In addition, measurements of chronic stress, illness and GP contacts were used for the analyses:

-Chronic stress was determined by the 12-item Screening Scale of the Trier Inventory for the Assessment of Chronic Stress (TICS-SSCS) which encompasses chronic worrying, work related und social overload, excessive demands and lack of social recognition in the past 3 months [20, 21] ; the sum score was classified into below average to average (0-11 points), above average (12-22 points) and high stress (23-48 points) [22] . -Having at least one chronic illness (e.g., diabetes, heart disease) -Having a GP (yes, no) -Self-reported number of GP visits in the last 12 months

Data collection for the General Practice Care 1 study (GPCare-1) was conducted from June until August 2020 in 12 primary care offices in the Greater Bonn region, Germany. The participating primary care offices belong to the academic teaching practice network associated with the Institute of General Practice and Family Medicine, University Hospital Bonn. In each participating practice, all adult patients who visited the practice during the time of recruitment were asked to participate. Patients were eligible if they had sufficient language skills and were mentally capable to fill the self-administered questionnaire in German, English, Turkish or Arabic. 813 participants between 18 and 91 years old filled out the questionnaire.

To allow for comparison, the GPCare-1 questionnaire used the DEGS1-questions on age, migrant status and gender, except that the third gender (diverse) was added.

Participants were asked for their highest level of education. Education level was then computed into three categories: low education (no school education/ secondary school up to 9th/up to 10th grade), middle education (high school (A-levels)/vocational school) and high education (university degree).

As no validated screening tool for patient-physician communication on social problems is available in German, eight questions were constructed based on existing questionnaires: the Patient Reactions Assessments Instruments (PRAD) [23] , the Medical Interview Satisfaction Scale (MISS) [24] , the patient requests form [25] and the patient-doctor relationship questionnaire (PDRQ-9) [26] . The first four questions addressed patients' experiences with their GP, the second four questions focussed on patients' preferences regarding their GP contact. Details on questionnaire items are provided in Table 5 . All items used a five option Likert-type answer (strongly disagree to strongly agree). The questionnaire was piloted by 40 individuals of the German general population with minor revisions thereafter.

In the DEGS1 population, all analyses were carried out using weighted complex samples procedures to allow for conclusions representative of the German general population. The standardized weighting factor was provided by the Robert Koch Institute and took age, gender, nationality, education, population for each federal state and BIK classification details into account. To allow weighting of participants who had already participated in 1998 the re-participation probability was estimated using logistic regression. For further details please see [27] . The number of cases is reported unweighted while the prevalence and the confidence interval is weighted. In Table 1 the experience of unemployment in the past 5 years was used as filter variable to stratify subpopulations. Weighted chi-squared tests and t-tests were applied. In Table 2 DEGS1 participants were filtered for unemployment experience in the past 5 years (n = 1170) and stratified by migration background. Weighted chi-squared tests for categorical data as well as t-tests for mean values were applied to determine differences between individuals with and without migration background. Considering all participants, one multiple linear regression was performed using migration background, unemployment experience, chronic stress and chronic illness as independent variables and GP visits as dependent variable. Age, gender and socioeconomic status were used as covariates. Outliers who had reported > 14 GP visits (n = 100; 1.7%) were excluded to meet the assumptions of the multiple linear regression. In 252 participants no data on the number GP visits was available so these cases were also excluded from the analysis (Table 3) . Additionally, a poisson analysis was performed using the same independent variables as in the multiple linear regression and chronic stress (TICS score) as dependent variable (Table 4 ). To account for missing values multiple imputation with chained equations with 25 iterations and repetitions was carried out [28] .

In the GPCare-1 dataset, descriptive analyses were performed. In the subpopulation of individuals with unemployment experience during their lifetime (n = 215), information on migration background was available in 210 cases. Chi-square tests and t-tests were used to determine differences between individuals with and without migration background (Table 5 ). For the multivariate models, two cases with diverse gender were excluded as the number of cases did not allow for subgroup analyses. The first multivariate model analysed for associations of the dependent variable "My doctor gives me enough space to describe personal strains" with gender, age, migration background, chronic disease, unemployment status, stress and education level ( Table 6 ). The second model estimated the relationship between the dependent variable "I get the feeling that my doctor takes my problems very seriously" and the same covariates (Table 7) . Both models were estimated as a reverse proportional odds model that takes the ordinality of the response scale into account: the probability of observing at least one category (e. g. partial or full agreement) was estimated [29] . Missing values were taken into account by applying multiple imputation with chained equations with 25 iterations and repetitions [28] .

In all analyses, statistical significance was set at p < 0.05 (two-tailed). Analyses were conducted using IBM Statistical Package for Social Sciences (SPSS 25.0) for Windows (IBM Corp., Armonk, NY, USA). Poisson and proportional odds regression models were calculated with R (Version 4.1.2).

The Charité-Universitätsmedizin Berlin Ethics' Committee had provided ethical approval for the DEGS1 study protocol in September 2008 (No. EA2/047/08) [15] . The Ethics' Committee of the Medical Faculty of the University of Bonn approved the study protocol for GPCare-1 in June 2020 (Ref. No. 215/20). All participating patients received verbal and written information on study procedures, anonymity, and confidentiality. Also, they were informed that participation was voluntary. No formal written consent was required as the return of the anonymous questionnaire indicated informed consent from the patient for their data to be used in the study. The GPCare-1 study was registered in the German Clinical Trials Register (DRKS00022330). Both studies were conducted in accordance with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. The sociodemographic and health characteristics of DEGS1 participants of working age (18-64 years old) were compared of individuals with and without experience of unemployment in the past 5 years a n or percent unless noted otherwise 

When comparing individuals with and without unemployment experience, the results showed that participants Fig. 1 .

The multiple linear regression analysis showed no significant association of migration background and unemployment experience with the number of GP visits when controlling for covariates. Higher chronic stress, female gender, lower SES, and having a chronic illness were significantly associated with a higher number of GP visits. For details, please see Table 3 . The sociodemographic characteristics and patients' communication experiences with GPs n = 813) were stratified by natives and individuals with migration background a denotes a p-value < 0.05 between results of original variables of natives and individuals with migration background with unemployment experience b n or percent unless noted otherwise Being female, younger, having a chronic illness, migration background, unemployment experience in the past 5 years and having a lower SES were all significantly associated with chronic stress (p < 0.01). For details, please see Table 4 .

A total of 813 patients participated in the data collection: their mean age was 51.6 years (SD ± 18.7 years), 59.3% were females, and the majority of participants had medium education (low 32.0%, medium 43.5%, 

Compared to natives, patients with migration background and unemployment experience had a significantly higher prevalence of strongly disagreeing with the statement that their GP takes their problems very seriously (migration background: 8.8% vs. natives: 3.5%, p = 0.04) and that the doctor gives them enough space to describe personal strains (migration background: 17.2% vs. natives: 4.3%; p = 0.03). Regarding the other six communication items, no significant differences between subgroups with and without migration background were found.

As outlined in Table 6 the multivariate results estimated that the probability of agreeing to the statement "My doctor gives me enough space to describe personal strains" was lower in individuals with migration background. However, having unemployment experience or not did not significantly change the probability of agreeing to the statement mentioned above.

Regarding the statement "I get the feeling that my doctor takes my problems very seriously" both migration background and having higher stress decreased the chances of agreeing to this statement. Having unemployment experience or not did not show significant results. For more details, please see Table 7 .

Based on the nationally representative DEGS1 data, our study showed a higher prevalence of chronic stress in individuals with migration background compared to natives with unemployment experience in the past 5 years. Higher chronic stress was significantly associated with unemployment experience and migration background. These findings are in line with results of Aichberger (2012), who observed higher distress levels in Turkish female migrants with unemployment experience in comparison to unemployed native women living in Berlin, Germany [30] . In contrast, researchers from Sweden reported no difference in the level of psychological distress between unemployed natives and migrants [31] . The contradicting results might be due to differences of the populations studied. In the current study population, most individuals with migration background and without a German nationality had a Turkish nationality. Study participants of the Swedish study included asylum seekers as well as immigrants from countries with guest worker' programs [31] . Differences could also be due to diverging employment policies in Sweden and Germany [32] .

Examining chronic stress in individuals with migration background is especially important since evidence from general practice offices in Israel has shown that general practitioners often missed to identify psychological distress in migrants [33] . These shortcomings are likely due to reasons on behalf of patients as well as physicians and the health care systems. For example, Schouten et al. (2006) showed that patients belonging to an ethnic minority tended to be less assertive and doctors seem to be interacting less with patients of ethnic minorities [14] .

In the GPCare-1 study, patients with a migration background were less likely to report enough space to talk about psychosocial problems with their GP and that their problems were taken seriously. To improve communication quality for migrant patients various approaches were shown to be effective, but are not necessarily available widespread, e.g., trainings for general practitioners in multicultural communication and social problems, resources for medical interpreters and social workers [34] [35] [36] . Also, migrants should be informed and encouraged to speak up in GP communication whom they might view as authority rather than personal resource [37] . This study did not show any difference in GP visits for unemployed individuals with and without migration background. This finding adds to the diverging picture of GP utilization by migrants as reported by Graetz et al. [38] . A study conducted by Glaesmer et al., which showed a higher number of GP visits by first generation migrants compared to natives in Germany, did not focus on the subgroup of individuals with unemployment experience which might explain the differences [39] .

The main topic of this study is novel in Germany, especially regarding chronic stress and physician-patient communication of migrant populations. Due to the cross-sectional design no causal relationships can be determined. Although nationally representative overall due to survey-specific weighting factors, it is recognized that the share of young men with migration background within the DEGS1 was higher than in the micro census [40] . For both subpopulations (unemployed with and without migration background) the DEGS1 question on unemployment in the last 5 years did not differentiate between job seekers directly after training and later in life. Data collection for the GPCare-1 study was conducted during the Covid-19 pandemic which might have led to higher chronic stress in patients. GPCare-1 data on GP-patient communication is used to complement the DEGS1 findings but does not allow for direct comparison as only the DEGS1 data are representative for the German population.

This study aims at increasing awareness on the association of unemployment and chronic stress in individuals with migration background. Migrants' perception of GP contacts are reassuring for the German health care system, but point towards a potential for improvement. Future research, also using qualitative approaches, will help to better understand the complex interactions between migrants and GPs.

Resolution concerning statistics of work, employment and labour underutilization

Unemployment statistics: Unemployment in the EU and the euro area

OECD. Foreign-born unemployment. 2020. Cited 9

Welfare state regimes, unemployment and health: a comparative study of the relationship between unemployment and self-reported health in 23 European countries

Subjektive gesundheit und Erwerbslosigkeit in Deutschland auf basis der EU-SILC-Daten von

Arbeitslosigkeit und gesundheit aus sozialmedizinischer Sicht

Regionale und individuelle Faktoren von Stresserleben in Deutschland: Ergebnisse einer repräsentativen Befragung mit dem perceived stress questionnaire (PSQ). Gesundheitswesen

How does unemployment affect self-assessed health? A systematic review focusing on subgroup effects

Do immigrants suffer more from job loss? Unemployment and subjective well-being in Germany. Demography

Community unemployment and immigrants' health in Montreal

Bundeszentrale für politische Bildung. Ausländische Bevölkerung | bpb: In absoluten Zahlen, Anteile der Altersgruppen in Prozent

Soziale Probleme in der hausärztlichen Versorgung -Häufigkeit, Reaktionen, Handlungsoptionen und erwünschter Unterstützungsbedarf aus der Sicht von Hausärztinnen und Hausärzten

Refugee experiences of general practice in countries of resettlement: a literature review

• fast, convenient online submission • thorough peer review by experienced researchers in your field • rapid publication on acceptance • support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year submit your research ?

German health interview and examination survey for adults (DEGS) -design, objectives and implementation of the first data collection wave

Die Umsetzung der Bildungsskala ISCED-1997 für die Volkszählung 1970, die Mikrozensus-Zusatzerhebung 1971 und die Mikrozensen 1976-2004 (Version 1). Mannheim: Zentrum für Umfragen

Messung des sozioökonomischen status in der Studie zur gesundheit Erwachsener in Deutschland (DEGS1)

Social support, negative life events and mental health

World Health Organization. EUROHIS developing common instruments for health surveys

Trierer Inventar zur Erfassung von chronischem Streß (TICS): Skalenkonstruktion, teststatistische Überprüfung und Validierung der Skala Arbeitsüberlastung

Factor structure and psychometric properties of the trier inventory for chronic stress (TICS) in a representative german sample

Chronischer stress bei Erwachsenen in Deutschland

Qualität der Arzt-Patienten-Beziehung -evaluation der deutschen version des patient reactions assessment instruments (PRA-D)

The medical interview satisfaction scale: development of a scale to measure patient perceptions of physician behavior

The patient requests form: a way of measuring what patients want from their general practitioner

A patient-doctor relationship questionnaire (PDRQ-9) in primary care: development and psychometric evaluation

Die erste Welle der Studie zur gesundheit Erwachsener in Deutschland (DEGS1): Stichprobendesign, response

Flexible imputation of missing data

Regression for categorical data

Socio-economic status and emotional distress of female Turkish immigrants and native German women living in Berlin

Employment status and psychological distress in a population-based crosssectional study in Sweden: the impact of migration

Germany and Sweden: Comparative Analysis. European Parliament Directorate General for Internal Policies

Identification of emotional distress among asylum seekers and migrant workers by primary care physicians: a brief report

Cultural diversity in patient participation: the influence of patients' characteristics and doctors' communicative behaviour

Implementing guidelines and training initiatives to improve cross-cultural communication in primary care consultations: a qualitative participatory European study

Challenges and facilitators for health professionals providing primary healthcare for refugees and asylum seekers in high-income countries: a systematic review and thematic synthesis of qualitative research

Die hausärztliche Betreuung von Migranten/−innen mit chronischem Schmerzsyndrom

Utilization of health care services by migrants in Europe -a systematic literature review

Health care utilization among first and second generation immigrants and nativeborn Germans: a population-based study in Germany

Beteiligung von menschen mit Migrationshintergrund an Gesundheitssurveys des Robert Koch-Instituts

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations

We owe special thanks to the Robert Koch Institute, Berlin, for kindly providing the data set and additional information on the DEGS1 survey. Also, we thank the participating practices of the research practice network of the Institute of General Practice and Family Medicine, University of Bonn, and their patients for their friendly participation. We would also like to thank Dr. Thomas Welchowski from the Institute of Medical Biometry, Informatics and Epidemiology (IMBIE) in Bonn for his valuable advice and support in conducting regression analyses. Additionally, we would like to thank Yelda Idik M. Sc. from the Institute of General Practice and Family Medicine, University of Bonn for coordinating the revision process. The study was performed within the framework of the dissertation program of the Institute of General Practice and Family Medicine which was kindly supported by the Medical Faculty of the University of Bonn.

Authors' contributions LO, BW, SS and SK conceptualized the study and its design. LO and MS developed the statistical approach and analysed the data. LO, BW, MS, SS and SK interpreted the data and results. LO wrote the first draft of the manuscript. LO, SS, SK, FB, BG, CH, NI, MO, JPW and BW developed the GPCare-1 questionnaire and collected the data of the GPCare-1 study. All authors discussed and approved the final manuscript.

Open Access funding enabled and organized by Projekt DEAL. The GPCare-1 study was performed within the framework of the dissertation program of the Institute of General Practice and Family Medicine which was kindly supported by the Medical Faculty of the University of Bonn.

The DEGS1 data set underlying this article were provided by the 'Health Monitoring' Research Data Centre at the Robert Koch Institute (RKI), which is accredited by the German Data Forum according to uniform and transparent standards (http:// www. ratswd. de/ en/ data-infra struc ture/ rdc). Data are accessible on application to interested scientists for anonymous scientific secondary analyses. Detailed information on access, application forms, and guidelines can be obtained from daten nutzu ng@ rki. de. The data set of the GPcare-1 patient study will be shared on reasonable request to the Institute of General Practice and Family Medicine of the University of Bonn, Germany.

The DEGS1 survey was consented with the Federal and State Commissioners for Data Protection and had been approved by the Charité-Universitätsmedizin Berlin Ethics Committee in September 2008 (No. EA2/047/08). Participants provided written informed consent before the interview and examination.The GPCare-1 study was approved by the ethical committee of the Medical Faculty of the University of Bonn in June 2020 (No. 215/20) . Patients received information on study procedures, confidentiality, anonymity, and were informed that participation was voluntary both verbally and in writing. Thus, no formal written consent was required as the return of the anonymous questionnaire indicated informed consent from the patient for their data to be used in the study. The GPcare-1 study is registered in the German Clinical Trial Register (DRKS00022330). Both studies were conducted in accordance with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.

Not applicable.

The authors declare no conflict of interest.