Background

There are three crucial aspects concerning working conditions in the context of theirsustainability and their influence on medical students: First, the physicians, whoset the students an example of current working conditions. Second, the studentscorresponding perception of the working conditions. Third, the resulting needs andexpectations of medical students about their future working conditions. The firstand the latter aspect has been subject of many research studies [1–5]. However, the second one as a link between the current and to-beanalysis, hasn’t been a subject of scientific research in Germany so far. TheiCept-Study wants to examine the students’ perception of working conditions,answering the following pivotal questions:

  1. 1.

    How do medical students perceive stress-related working conditions of their supervising physicians?

  2. 2.

    Is the perception realistic?

  3. 3.

    Are there differences in the perception regarding age, specialty or state?

Generally speaking, the perception of working conditions depends on the directobservation of physicians at work and on the informal information from the peergroup or the media who is influencing the perception of medical students [6–9].

Whether the perception is congruent with reality cannot be answered with currentdata. Knowing that the perception has a major impact on the specialty choice ofmedical students makes the importance even more evident [6]. Furthermore there is an upcoming shortage of qualified employees inGerman hospitals: The “Deutsches Krankenhausinstitut” (DKI) assumes afurther requirement of 37.370 physicians until 2019 [10]. In cooperation with PricewaterhouseCoopers the institute for economicresearch (WifOR) estimated an additional need of 56.000 physicians until 2020 and106.000 until 2030 [11, 12]. The “Kassenärztliche Bundesvereinigung” and the“Bundesärztekammer” quantified the need to an extend to 71.625missing physician until 2020 [13]. These data suggest a threatening shortage of qualified medical employeesand thus an urgent need for motivated medical students, willing to work inhospitals.

In this study, two stress models were used for determining stress-related workingconditions.

  1. 1.

    The job-demand-control (JDC) model of Karasek et al.: In this theoretic model two parameters are confronted. On the one side the “job demand” and on the other side the “control” in terms of scope of action respectively scope for decision-making. Karasek et al. postulate that an imbalance between too high “job demand” and too little “control” (JDC-ratio > 1) results in “mental strain” [14, 15]. A current survey from 2012 interviewed medical employees in hospitals and proved the importance of the JDC model regarding the development of stress-related symptoms [16].

  2. 2.

    The effort-reward-imbalance (ERI) model of Siegrist et al.: This model postulates an imbalance between the “effort” at work and the corresponding “reward” as an intrinsic stress factor with all its negative psychological and physical manifestations. The negative consequences develop from domination of the “effort” in relation to the “reward” (ER-ratio > 1). There are three different types of “reward”: money, respect/acknowledgment and career advancement [17]. A 2006 published meta-analysis showed that high job demands, lack of social support, job insecurity and low appreciation raised the incidence rate of mental illnesses [18].

In the iCept-Study both models are combined, since thereby both extrinsic(JDC) and intrinsic (ERI) stress factors are taken into account. The importance ofboth models on the well being of employees was shown in a study that examined theirinfluence on the incident rate of myocardial infarction [19].

Methods

The iCept-Study is designed as an anonymized online-survey. Therefore the study isorientated towards the “international codex of market and socialresearch” and, because it will be administered in Germany, towards therespective declaration for the federal republic of Germany [20, 21]. Furthermore the “standards for quality assurance ofonline-surveys” will be taken into account [22]. The necessary scientific standards of quality can be found in the“Norm DIN ISO 20252:2006; Markt-, Meinungs- und Sozialforschung –Vokabular und Anforderungen”.

Sample

In this Study two samples will be recruited: physicians and medical students. Thesample of physicians will be used as the control-group, the medical students asthe experimental-group. Both samples will be chosen randomly from the members ofthe Marburger Bund, a professional organization and labor union of employedphysicians. On the cut-off date, the July 1st 2012, the MarburgerBund had 83.123 physicians and 19.223 medical students as members. The memberswill be contacted through e-mail in a standardized form, which will bedistributed with the kind support of the Marburger Bund. The e-mail will containa personalized link to the iCept-Study. In addition, Marburger Bund internalmedia like the “MB-Newsletter” or the “Marburger Bund Zeitung(MBZ)” will be used. This inclusion criterion is out of date and will notbe used in this study, since medical students DO have relevant clinicalinternships in the younger semesters.

ICept questionnaire

The iCept Questionnaire is built to assess mainly the above-mentioned twotheoretic stress models (JDC and ERI). For that purpose the questionnaire isbased upon the short questionnaire for work place analysis (KFZA) ofPrümper et al. [23] and the questionnaire for the effort-reward-imbalance (ERI) ofSiegrist et al. [24].

The KFZA is an established and validated questionnaire since 1995 and has beenused in many studies especially in hospitals [25–27]. Moreover it is listed by the federal institute of work safety andoccupational medicine (BAuA) as a universal screening method with satisfyingquality criteria [28]. The KFZA is also the basis for the “IMPULS-Test” ofMolnar et al. and other questionnaires [29–31]. It consists of 26 items and 11 scales.

The effort-reward-imbalance questionnaire (ERI-questionnaire) by Siegrist et al.has been developed in 2004 to assess the identical stressor (ERI). The qualitycriteria are satisfying (Crohnbach’s α > 0,7). The answer formatis a 5-point Likert scale, whereas current data suggest a 4-point Likert scaleto be more suitable [24, 32]. The questionnaire exists in a long (26 items) and a short (16 items)version [32, 33]. The short version has been used in many different studies [34, 35] and is also listed by the federal institute of work safety andoccupational medicine (BAuA) as a screening method with satisfying qualitycriteria [36].

Developing the iCept questionnaire, the items of KFZA and ERI-questionnaire havebeen reviewed for practicability at the clinical workplace. Also the items mustbe answerable for medical students from their point of view. Only items andscales fulfilling these criteria were used. The kind approval from Prof. JochenSiegrist (ERI-questionnaire) and Prof. Andrea Abele-Brehm (adapted KFZA) to usetheir questionnaires in this study has been obtained.

The stressors defined by the JDC are covered by the KFZA: the “jobdemand” is measured through the scales “suitably demandingwork” (QL1/QL2) and “suitable volume of work” (QN1/QN2); the“control” is assessed through the scale “scope foraction” (HS4-HS6) [37]. In addition, the scales “cooperation” (ZU2/ZU3) and“social support” (SR1-SR3) are taken from the KFZA. The itemsHS4-HS6 and an additional scale, “social climate” (SK1, SK2), havebeen taken from the adapted version of KFZA by Abele [31].

There is a slight correlation of both, ERI and KFZA questionnaires, regarding thescales “job demands” and “effort”. Different studiesshowed a correlation between the scales from r=0,3 to r=0,6 [38, 39]. Thus the scale “effort” is measured by items of bothquestionnaires (ERI2/ERI5 and QN1/QN2). Also the scale “reward” ismeasured by both questionnaires (ERI7/ERI8/ERI10 and SR1/SR2).

The overall job satisfaction is measured by a single item (JS1) from the“Job Diagnostic Survey” (JDS) of Schmidt et al. [40] That a single item can be used to measure the job satisfaction hasbeen shown in a meta-analysis, postulating a correlation of r=0,67 between“single-item measures” and “scale measures” regardingjob satisfaction [41].

The sociodemographic data are assessed according to the “demographicstandards” of the federal institution of statistics [42]. The following data will be collected:

  • EM1: Gender

  • EM2: Age

  • EM3: Specialty

  • EM4: State

  • EM5: Semester (only students)

  • EM6: Position (only physician)

Figure 1 shows all items with their target parameter.

Figure 1
figure 1

Allocation of iCept-items to target parameter.

Full size image

There will be two slightly different questionnaires administered: one for medicalstudents and one for physicians. The items of both versions only differgrammatically but not content wise or semantically: The items for thephysicians’ version will be written in the first-person singular, whereasthe version for students will be in the third-person singular. So there is nochange to any substantial degree. The items will be answered on a 4-point Likertscale (strongly disagree, disagree, agree, strongly agree). The complete iCeptquestionnaire contains 20 items and 5 more sociodemographic items (seeTable  1) and will take about 5–10 min ofthe participants’ time.

Table 1 iCept-questionnaire for medical students
Full size table

In order to keep the influence of the peer group or the media on thestudents’ perception as low as possible, the students will be asked onlyto rate the latest clinical internship.

The survey will be generated with the web based online survey tool“2ask” from the amundis communications GmbH. The Leibniz institutefor social science recommends this tool [43].

Statistical data analysis

The statistical data analysis will be performed with SPSS Statistics. As Figure1 shows, the scales “effort” and“job demand” are measured by 4 items, the scale “reward”by 5 items and the scale “control” by 3 items. Considering the4-point Likert scale, the scale sum scores varies:

  • Scale sum score “effort” (xeff):4≤ xeff ≥ 16

  • Scale sum score “job demand” (xjob):4≤ xjob ≥ 16

  • Scale sum score “reward” (xrew):5≤ xrew ≥ 20

  • Scale sum score “control” (xcon):3≤ xcon ≥ 12

In order to draw first conclusions about the stressors ERI and JDC the ratiobetween the respective scale sum scores are calculated (ER-ratio and JDC-ratio).To adjust the unequal number of items a correction factor, based on the numberof items, is used (ceri=1,25 for the scale “effort” andcjdc= 0,75 for the scale “job demand”).

ER − Ratio = x eff x rew × c eri JDC − Ratio = x job x con × c jdc

Values > 1 of the ER/JDC-ratio indicate stress with possible adverse healtheffects [24, 44].

Besides this relative component, indicating an imbalance between the scales, theabsolute component will also be calculated, indicating possible eustress. Forthis purpose the sum scale scores of “effort” and“reward” respectively “job demand” and“control” will be summed up (ER-Sum, JDC-Sum).

ER − Sum = x eff 4 + x rew 5 JDC − Sum = x job 4 + x con 3 .

For the analysis values > 5 (as a sufficient condition) and an ER/JDC-ratio =1 (as a necessary condition) will be taken as an indicator for healthy stress(eustress).

Seven items (ZU1/ZU2, SK1/SK2, SR1-SR3) reflect the psychosocial aspect of stressand will be analyzed separately as well as the “overall jobsatisfaction” item (JS1).

Discussion

The iCept-Study is thought to be the basis of ongoing further investigationsregarding the perception of working conditions in hospitals. The results shall servethe facilitation of improving working conditions. Especially the rough transitionfrom medical school to the first job in a hospital makes it very important to knowthe students’ perception in order to smooth that transition. Anotherimplication the perception of working conditions has, is the influence on thestudents’ specialty choice. This hasn’t been a subject of research inGermany so far and therefor could be considered as a future subject in theiCept-Study.