gms | German Medical Science

GMS Zeitschrift für Hebammenwissenschaft

Deutsche Gesellschaft für Hebammenwissenschaft e.V. (DGHWi)

ISSN 2366-5076

Job demands and job resources: determinants of early career exit of midwives

Research Article

Search Medline for

  • corresponding author Anja Alexandra Schulz - University of Education Freiburg, Institute for Everyday Culture, Sports and Health, Department Research Methods, Freiburg, Germany
  • Theresa Laschewski - University of Education Freiburg, Institute for Everyday Culture, Sports and Health, Department Research Methods, Freiburg, Germany
  • Markus Antonius Wirtz - University of Education Freiburg, Institute for Everyday Culture, Sports and Health, Department Research Methods, Freiburg, Germany

GMS Z Hebammenwiss 2021;8:Doc03

doi: 10.3205/zhwi000022, urn:nbn:de:0183-zhwi0000228

This is the English version of the article.
The German version can be found at: http://www.egms.de/de/journals/zhwi/2021-8/zhwi000022.shtml

Received: August 11, 2020
Accepted: April 12, 2021
Published: September 13, 2021

© 2021 Schulz et al.
This is an Open Access article distributed under the terms of the Creative Commons Attribution 4.0 License. See license information at http://creativecommons.org/licenses/by/4.0/.


Abstract

Background: Adverse work factors are assumed to be important determinants of premature departure from the midwifery profession. Occupational psychological models can provide a basis for theory-based modelling and operationalisation.

Aims: The aim of the study was a) to identify job demands and resources associated with early career exit from the perspective of midwives and b) the specification of the components of the job demands-resources (JD-R) model for the field of midwifery, taking job-specific work factors and conditions into account.

Methods: Five guideline-based qualitative interviews were conducted with active and former midwives from various areas of practice. Guideline development was based on the JD-R model. Data analysis was performed using Kuckartz’s qualitative content analysis.

Results: Central demands (lack of time and staff, tasks not related to the professional role, inadequate staffing ratios) contrast with positive job-related and personal resources (support from colleagues, optimism, coping strategies). The work factors reported align with the structure of the JD-R model. The centralisation of care structures proves to be an important structural framework condition that needs to be further clarified regarding its implications for the model structure in the context of midwives' work.

Conclusion: The study results expand the theoretical basis for the operationalisation of determinants of premature career exit of midwives and create an integrated view of varying activity and context characteristics.

Keywords: job demands, job resources, career exit, midwifery, qualitative content analysis


Background

The shortage of trained midwives is a growing problem for ensuring comprehensive, local, high-quality care for pregnant women, women in childbirth and new mothers. The increase in early career departure among midwives is seen as one of the key reasons for this shortage [18], [30], [36]. Occupational psychological models, such as the demand-control-(support) model (DCS model) [19], [20] or the effort-reward imbalance model (ERI model) [33], can provide a theory-based approach. They can explain (premature) departure from the midwifery profession by identifying cause-effect relationships between (psychosocial) job demands and perceived workload with a simultaneous lack of occupational resources (e.g. autonomy, appreciation). Adverse aspects of the job situation, which can have a negative impact on midwives’ work and thus also the quality of care they provide to the women they support, are assumed to be key predictors of career exit. The DCS model considers job demands and job latitude or autonomy in performing professional tasks to be important determinants [19], [20]. By applying the DCS model in the care sector, a correlation could be proved between the construct of job strain and intention to leave the profession [41]. According to the ERI model, a gratification crisis emerges when the occupational commitment invested is met with too low appreciation or recognition (e.g. through salary, security, opportunities for promotion) [33]. Using the ERI model in a cross-sectional survey, Mössinger et al. [26] provided evidence of existing gratification crises among freelance midwives. Because of methodological limitations, it was not possible to ascertain whether or not the ERI score is also a suitable predictor of premature career exit.

Based on the assumptions of the DCS and ERS models, Demerouti et al. [8] developed the job demands-resources model (JD-R model) [2], [3], [6], [7], [8] to predict burnout and its positive antipode, work commitment. Contrary to the understanding at the time that burnout symptoms were to be located primarily in social occupational fields (e.g. education, health care, social work) [24], the JD-R model was intended to be suitable not only for the (human) service sector, but also for different types of activity (e.g. industrial or transportation sectors) [8]. Therefore, the JD-R model also takes into account the variety of different combinations of work factors that may be associated with stress in different ways depending on the employment relationship and professional field. Figure 1 [Fig. 1] shows two key basic theoretical assumptions of the JD-R model. The focus of this article is primarily on the first of these two basic assumptions. The second, slightly more complex basic assumption is of importance if the model is intended as a basis for practical action. Only the main theoretical principles of the latter will be described.

Basic assumption 1: Despite the wide range of work factors, they can be assigned to the central constructs of job demands and job resources [3]: Job demands concern physical, psychological and social aspects of work. They function primarily as a stressor when people are subjected to high or poorly structured job demands, which require a sustained high level of physical and/or psychological effort to tackle [8]. Job resources facilitate the achievement of work-related objectives and reduce job demands and the accompanying physical and psychological consequences. In addition, they stimulate personal development [2]. Resources are subdivided into psychological, physical, organisational and social.

Basic assumption 2: In the professional context, burnout and job commitment are the product of two divergent processes. In the impairment process a sustained accumulation of job demands (e.g. time pressure) with repeatedly unsuccessful attempts to tackle these demands (exhausted resources) leads to reduced health and wellbeing, which manifests itself as health problems and exhaustion [7]. Job demands should not be considered negative per se. Adverse consequences arise mainly when, having successfully addressed these demands, the employee is unable to adequately recover (distress) [25].

In the motivational process it is assumed that job resources function as motivational stimuli, positively influencing both job commitment and perceived capability [3]. Resources can have an intrinsically motivating effect by promoting the growth, learning and self-development of employees. Job resources contribute to the achievement of work objectives and constitute extrinsic motivation [3]. As a result of an interaction effect, available resources can mitigate the negative consequences of the impairment process [8]. This, in turn, means that a deficit of job resources makes it difficult to achieve work goals and, due to the negative feelings (e.g. frustration, failure) that arise as a result, motivation decreases. As a result, employees may emotionally distance themselves from their own work in the long run and, as a result, sometimes develop cynicism about their own work [9]. In general, a wide range of job resources along with increased motivation of staff are correlated with a stronger work-role attachment and thus a higher probability of commitment to the organisation or the profession [9].

Initially, the role of the individual was only of secondary importance in the JD-R model. An important extension of the model involved modelling the construct of personal resources as an additional predictor [9] (see Figure 1 [Fig. 1]). Self-efficacy, optimism and organisation-based self-esteem (OBSE) are considered relevant personal resources in the work context [17]. The OBSE construct represents the extent to which individuals see themselves as important, competent and capable within the organisation [28]. The mediator variable personal resources postulated in the model could be partially verified in a psychometric structural analysis with respect to the prediction of burnout and job commitment [44]. While the assumed effect of personal resources on the impairment process could not be confirmed, it did have a mediating effect with respect to the causal relation of job resources and job commitment within the framework of the motivational process [44]. Moreover, it was also possible to identify a mediation effect between job resources and emotional exhaustion. Further, Xanthopoulou et al. [44] identified personal resources as a significant predictor of job resources, and, at the same time, job resources can promote the development of personal resources.

The JD-R model appears to provide a suitable theoretical framework for the midwifery profession. It can be assumed that, in midwifery, work factors will be systematically differently perceived and weighted because of the differences between midwives’ areas of practice (antenatal care, attendance at (non-) hospital births, postnatal care, or a combination) as well as various different types of employment relationship (employed, freelance, hospital-affiliated midwife, or a mixture of these). A study by Cramer et al. [5] on the relationship between working conditions and emotional wellbeing in midwives confirms this hypothesis. The authors were able to provide evidence of a correlation between emotional wellbeing as well as burnout and the work setting of midwives (hospital, outpatient setting). This correlation was ascribed, inter alia, to different work factors in the respective professional settings [5]. The JD-R model provides an appropriate basic theoretical assumption for this aspect.

The validity of the generic JD-R model has been empirically substantiated for different work settings, populations and countries [9]. However, the model has rarely been investigated in midwifery care. A study conducted in the Netherlands, for example, used elements of the model as a basis for a focus group discussion with new midwives [21], without explicitly exploring the model’s suitability as a theoretical basis for this area of research. Although the JD-R model is predominantly used to predict burnout and job commitment, studies suggest that burnout is highly associated with the intention to leave the profession [34].


Aim

The research project aims to provide an enhanced understanding of the processes that determine the probability of premature departure from the midwifery profession. The long-term objective is to be able to use this as a basis to positively influence this process through interventions with the aim of preventing early career exit. According to the model outlined by Craig et al. [4], qualitative research studies are important to be able to adequately identify structural characteristics of daily working practices. In line with the features of qualitative research, an analysis of specific case types is considered appropriate to acquire a more in-depth understanding (inductive conception phase).

Thus, the aim of the study was to identify and describe job demands and resources, which are, from the midwife’s perspective, associated with early career exit. The data basis served to substantiate work factors postulated in the JD-R model specifically in relation to the midwifery profession, from the midwife’s perspective.


Methods

Study design

To answer the research question, a qualitative study design was selected. Between November and December 2019, five standardised, qualitative, semi-structured interviews were conducted in German with midwives from southern Germany. The qualitative approach is particularly well suited in terms of openness, processuality, context and subject orientation aspects, as it makes it possible to acquire information on individual reasons and backgrounds from the perspective of the midwives being interviewed [12].

During the interviews, the main focus was on the first basic assumption of the JD-R model: the subdivision of relevant work factors into two generic constructs. Accordingly, the guideline used for the interviews was conceived in relation to the constructs of job demands and job resources. It includes six main guiding questions (see Table 1 [Tab. 1]). The comprehensibility of the guiding questions was tested in two cognitive interviews [29].

Sampling

For the sample selection, we chose a deductive recruitment strategy combined with gatekeepers [22]. Three multipliers from the German Midwives’ Association and 14 midwives were contacted by email. When composing the interview sample, care was taken to ensure that specific case types of midwifery practice (main work context: employed, freelance, hospital-associated) were represented so that an adequate structural range of variation was covered [22].

To ensure that inclusion criteria were fulfilled, a number of ad-hoc questions were emailed in advance (addressing, e.g. employment relationship, professional experience). The result was a total of five firm interview acceptances; two individuals had already left the midwifery profession at the time of the interview and three were still practicing midwifery. The two women who had left the profession had worked as midwives for 17 and 20 years, respectively, and had worked both as employed and free-lance midwives. They left the profession two and six years ago, respectively. The average professional experience of the interviewees was 11.5 years (standard deviation: 9.05 years) and ranged from 1 year to 21 years.

Conduct of the study

All participants received a letter in advance providing them with information about the study aim and interview process as well as data protection aspects in accordance with GDPR. The latter included information about the handling and processing of personal data, their individual rights regarding the processing of personal data, the process of pseudonymised transcription as well as encryption, the type and duration of storage of the data. All participants submitted a written declaration of informed consent. The face-to-face interviews were open and narrative-generating with a focus on job demands and job resources [22]. The location of the interviews varied between the professional and private premises of the interviewees as well as a room at a university. The interviews were digitally recorded (recording device: ROLAND R-05) and lasted 59 minutes, on average (standard deviation: 18.2 minutes). The audio recordings were secured on an externally encrypted storage medium.

Data preparation and data analysis

The audio recordings were transcribed and then pseudonymised according to Kuckartz’s five moderate basic rules [23]. Data analysis was deductive based on the structuring qualitative content analysis developed by Kuckartz ([23] p.97):

1.
Model-based, deductive creation of overarching categories (job demands, job resources) and main categories (for demands: emotional/physical strain, shift work, conflicts; for resources: support, autonomy, reward, feedback), based on the construct level of the JD-R model. Personal resources were included in the category system as separate mediator variables.
2.
Content analysis, based on the deductive system category, conducted for 50 percent of the data material in two independent runs. The positive/negative work factors reported by the midwives interviewed were subdivided into subcategories. This served to structure the data material and to substantiate the model-based work factors postulated specifically for the midwifery profession. The categorisation followed the main categories in terms of content. Thus, for instance, adverse work factors, which the midwives associated with emotional strain, were assigned to the main category of the same name. To ensure the separability and distinctness, all subcategories were linked to the respective definition.
3.
Coding of remaining data material, based on the category system from steps one and two, in two independent runs. For validation purposes, steps two and three were conducted independently by two members of the project team. Inconsistencies in the coding were discussed and resolved in a consensual process.

Results

The structuring of the results was oriented towards the general constructs of the JD-R model (see Table 2 [Tab. 2]). Due to the wealth of material collected, we focus on distinguishing between the work factors that comply with the theory and those perceived by the midwives themselves, as well as the identification of specific (structural) features.

Job demands

The job demands set out in the JD-R model emotional strain, physical strain, shift work and conflicts are consistent with the midwives’ perceived job demands. Some of the derived subcategories include job-specific demands as well as causal attributions for emotional job strains that are particularly relevant for the midwifery profession. The subcategory staff shortage was assigned to the main category emotional strain. In the hospital setting, the midwives often perceive the volume of tasks that have to be performed as disproportionately high compared to the number of midwives on duty. This would increase the number of clients who have to be cared for at the same time during the birth process, which, in turn would result in emotional strain for the midwives. In addition, the midwives mentioned aspects such as activities that were not related to their professional role and tasks requiring a high level or organisational effort as causes of emotional strain.

High level of responsibility and risk of potential damage claims, which midwives would have to bear the cost of retrospectively are reported to be further causes of emotional strain. The high level of responsibility plays a particularly important role for freelance midwives, resulting in a need for high professional liability insurance premiums and a high overall investment in practicing the profession. In addition, conflicts between the midwives and the medical staff would lead to emotional strain. This could be caused by differences in birth-specific views as well as unclear areas of responsibility and tasks.

Job resources

The job resources that are set out in the JD-R model support, reward, autonomy and feedback are largely consistent with the main categories representing midwives’ perceived job resources. The subcategories identified by the midwives did not, in general, exhibit any specific aspects that were not included in the theoretical model. Only for the main category support was support from colleagues emphasised as a key job resource. With regard to freelance midwifery, the aspects of self-responsibility and independent decision-making were highlighted as valuable resources.

Personal resources

Personal resources play a special role in the JD-R model as a mediator variable in the interaction between job demands and burnout or job commitment. From the data it was possible to identify the personal resources of optimism and passion for the midwifery profession. One midwife described the birth event as “(…) a new miracle every time (…)” (I-1: 26:16). She was grateful to have the opportunity to support families during such a special time. The idea that optimism at work served as a personal resource was reflected in another statement, where the midwife interviewed reported that she “(…) could not quite detach herself from the job, I would wish for better working conditions but actually don’t want to leave the profession altogether” (I-3: 48:12). Another personal resource is the way individuals cope with job demands. As effective coping strategies, second jobs with low levels of responsibility or supervision as well as conversations with work colleagues or family members were emphasised. One midwife reported that:

“[there were] some colleagues, (…) who did not [cope] very well and they left with real burnout (…) and needed a long time to find their way back” (I-5: 45:40).

Centralisation of care structures

One external condition identified by the midwives interviewed was centralisation of care structures. This means that centres are established as central medical care structures. From the midwives’ point of view, centralisation resulted in a higher workload in large maternity hospitals, as

“(…) [the] catchment area [for each hospital] gets larger and larger. This [leads to] increasing numbers of births, too few staff and a large number of absences due to illness, which has resulted in a lot of burnout [and] exhaustion in recent months (…)” (I-1: 15:29).

One midwife reported that the centralisation aspect was associated with dissatisfaction arising among staff in major hospitals. Large hospitals employ many different actors at the same time, which may have a negative impact on the birth process. In addition, the longer journey to the nearest hospital resulting from centralisation presents a major challenge for both midwives and birthing mothers living in rural areas. Moreover, births in small, rural maternity wards are “less complex” and “faster” (I-4:11:07) as, because of the shorter distance to hospital, birthing mothers do not have to go into hospital at the first signs of labour (e.g. latency phase).

According to the midwives, closures of small hospitals can directly influence premature departure from the profession because

“[after the closure of a maternity ward], many midwives stop working as midwives. It is the same for many [of my] colleagues, they don’t want to pay the price to work in an environment where they don’t even like working very much” (I-4: 31:10).

Indirectly, centralisation may act as a mediator variable, as closures of small, non-centrally located maternity departments result in a longer distance to travel to the nearest hospital. In turn, this may increase the probability of intention, to leave the profession. One of the midwives interviewed explains this relationship as follows.

“[The midwifery profession is] a woman’s job and the [midwives] mostly have roots and they have their houses and their children and their husband where they live, and they are no longer willing to accept these long journeys to the hospitals” (I-4: 31:41).

Moreover, because of the closures of regional maternity departments and considering the invariable shortage of skilled staff, the workload in the maternity departments adjoining the centrally located hospitals may increase. One consequence could be an increase in the stress level (e.g. due to having to care for many women simultaneously) as well as the perceived emotional strain, which could lead to an overall increase in work dissatisfaction. The centralisation aspect is a relevant condition for midwives in terms of job stress in the midwifery profession. The midwives’ statements indicate that the external factor of centralisation of care structures could influence or even cause different phenomena postulated in the JD-R model.

Specific features observed during the coding process

During the coding process the theoretical constructs job demands and job resources were, to a certain extent, difficult to distinguish from one another in midwives’ reported experience. Responsibility can be understood negatively in the sense of the necessity to take on responsibilities in the event of an emergency or damage claim. Accordingly, it is perceived as emotionally stressful:

“The idea that it is fate or nature or that it had been someone else’s fault, everyone always looks to the midwife first, to see whether she has made a mistake” (I-1: 13:56).

In contrast, responsibility can also be understood posi-tively in the sense of the desirable latitude for the individual to shape how they do their job. This is especially the case in care situations where midwives are free to make their own decisions and respond to the individual needs of the mother and baby:

“Women can simply be left to give birth; without constantly being interrupted with something (…) I mean, the doctors actually only come if there is something important, the rest can be done on the telephone, and actually, the birth process should be undisturbed from the start right up until the baby is born” (I-4: 12:34)

In the model, the relationship between job demands and job resources is regarded as an interaction. In the midwives’ reports, barely any distinction was drawn between the contribution made by the individual components job demands and job resources. Instead, they were addressed as integrated components. Thus, although the postulated interaction is shown to be relevant, distinguishing characteristics of interaction remain opaque. A similar difficulty arose concerning in differentiating the work factors related to emotional or physical strain. In the case of the subcategory staff shortages and activities not related to their professional role, the interview participants’ classification was ambiguous classification. Therefore, the final classification of one of the forms of strain was made by the researcher during the analysis process. This issue could not be completely resolved, despite a coding guideline with anchor quotes and defined subcategories.


Discussion

The aim of this study was to assess perceived job demands and job resources, from the midwives’ perspective, which are associated with the intention to leave the profession. The participating midwives reported a range of job demands and job resources, some of which could occur outside the midwifery context (e.g. lack of time, (dis)satisfaction with financial remuneration) and others which were specific to the midwifery profession (e.g. having multiple clients to care for simultaneously). The identification of work factors specific to the midwifery context based on the data is supported by a comparison with existing empirical research findings. The job demands identified are, for the majority, consistent with the findings of other surveys of midwives in Germany. For example, midwives from the federal states of Bavaria [31] and Saxony [32] associated job-related aspects such as too low income, lack of recognition, an overly high share of tasks not related to their professional role, and insufficient time to be able to adequately care for their clients with a high level of job stress in the midwifery profession. A nationwide expert report on the current situation regarding in-patient care also highlights the problem of staffing bottlenecks in midwifery regarding job satisfaction and job stress [1]. Furthermore the additional effort required as a result of having to care for several women simultaneously was emphasised.

The subjectively reported work factors are in line with the first basic assumption of the JD-R model, that job demands and job resources are the two most important work factors. This will be illustrated using the example of the construct of personal resources. Overall, the respondents mentioned optimism, passion for the job and self-efficacy resulting from successful coping as relevant personal resources. Together with OBSE, optimism and self-efficacy were considered fundamental components of individual adaptive capacity [17] and are regarded as relevant elements of the JD-R model. Within the framework of associated job resources and job commitment, personal resources act as mediators which help explain the variance in the outcome variables job exhaustion and job commitment [44].

The partially unclear demarcation between the reported work factors reflects the initial assumption of high diversity of perceived work factors in the midwifery profession. Depending on the midwife’s area of practice, different job-related factors are weighted differently and perceived differently in terms of whether they are adverse or beneficial. The need for work factors that can be adjusted depending on the area of practice supports the use of the JD-R model. Due to the generic approach it enables a high level of flexibility if information about diverging work factors is to be assessed. It thus seems particularly appropriate for future surveys in the field of job stress in the midwifery profession [8].

The external factor centralisation of care structures needs further clarification. Over the course of a decade (2007–2017), the percentage of maternity departments in Germany decreased by 28 percent [38], [39]. It was primarily small and less centrally located maternity departments that were affected by closures. Economic reasons, staff shortages and quality aspects (e.g. better quality results due to a higher number of cases) play an important role in this decision [13], [37]. Although there are efforts to centralise in all medical care structures, this trend is particularly relevant for inpatient and outpatient midwifery care. Already in 2014, the National Association of Statutory Health Insurance Funds (GKV-Spitzenverband) predicted a care bottleneck in midwifery, which could be attributed to an increase in resignations from contract and hospital-affiliated midwives as a result of centralisation measures [14]. Stahl et al. [37] address this issue, arguing that this trend coincides with an increase in professional liability insurance premiums for midwives. Consequently, freelance midwives opted to no longer supervise hospital births, deciding instead to switch to ante and postnatal services. This is important, because in 2017 a total of 20 percent of hospital births were supported by freelance hospital-affiliated midwives [10]. Between 2008 and 2010, a quarter of freelance midwives reported having stopped working in hospitals [11]. The liability insurance subsidy introduced in 2015 was not enough to mitigate the consequences of midwives leaving the profession [37]. These direct effects of centralisation measures on inpatient and outpatient midwifery are reflected in our findings.

Generally, the current study demonstrated uncertainty regarding how to categorise the centralisation aspect in the structure of the JD-R model and the phenomena it describes. To ascertain the extent to which this external factor is a specific characteristic of the midwifery context and has a tangible impact on the postulated interaction processes within the JD-R model more detailed analyses in future studies are required.


Limitation

Due to the small sample size, the generalisability of this study’s findings is limited. It cannot be ruled out that the midwives who had left the profession, in particular, may in retrospect, report more negatively about their job situation at the time than the respondents who are still practicing. The self-serving bias may strengthen this tendency by attributing the reasons for departure from the profession primarily to external factors and ascribing them less strongly to personal decisions, deficits or difficulties [35]. Furthermore, it is possible for cognitive distortions to have arisen due to recall bias [40] or a halo effect [42]. Moreover, the social perception of the role of midwives in society at large as well as its social desirability could result in distorted judgement and perception processes in the interview situation [27]. Above all, studies with larger and representative samples are needed to acquire a more comprehensive and differentiated picture of the job resources and demands of the different areas of midwifery practice. Furthermore, it is important to allow for testing of the relevance of the centralisation of care structures as a mediator variable within the interaction framework. This provides the basis for further verification of the application of the JD-R model in the midwifery context. In addition to the three forms of work models in midwifery practice investigated, hybrid forms have to be considered. It can be assumed that midwives conducting a mixture of different forms of practice will be exposed to different types of job strain than midwives who work exclusively in the hospital or non-hospital context.

Heuristic models like the JD-R model can provide a meaningful explanatory approach to address occupational psychology questions. However, when interpreting the data, it is important take into account the limited context-specific measurement quality of the model [9]. This implies that surveys in the work context rely primarily on data from self-assessments of the employed individuals. The resulting measurements have a subjective reference standard and must regarded as limited in terms of their reliability, validity and sensitivity. If a study’s aim is to provide an objective characterisation of working conditions, bias components due to subjective sources of variances have to be considered. Ideally external observations should be integrated.

This study contributes to expanding the theoretical basis of the research field of midwifery care. By this means, measures shall be identified and evaluated which to ensure that midwives remain in the profession in the long term and, on this basis, to counteract supply bottlenecks in obstetric care [43]. The longitudinal NEXT study is a pioneering example. Based on a theoretical model, care staff from ten EU countries were surveyed about their job situation and the reasons for their premature departure from the profession [15]. The findings served as a basis for domain-specific analyses and the development of targeted measures to address the situation with a view to shaping the future care sector and preventing premature career exit [16]. To be able to transfer this target-oriented approach to the midwifery context requires a) an empirical validation of the postulated structure of the JD-R model for the midwifery context and b) the analysis of the causal relationship between job demands and job resources individually, as well as their interplay in respect of occupational psychological experiences and especially premature departure from the midwifery profession [8].


Conclusion

The midwives reported a variety of work factors, which were weighted differently depending on the area of practice. The fact that the subjectively perceived work factors were closely aligned with the deductive, model-based category system shows that the construct level of the JD-R model appears to be a suitable tool for assessing midwife-specific differences in work factors. The aspect of centralisation of care structures and its possible impact on adverse and beneficial conditions and the characteristics of the different areas of midwifery practice requires further clarification, however. The JD-R model provides a well-founded theoretical basis for the varied areas of midwifery practice, which should be considered when designing a study to analyse determinants of premature departure from the midwifery profession.


Notes

The present study is associated with the overarching research project “Structural analysis of midwifery care in the rural Ortenau district” and was implemented at a separate point in time as an extension of that research project. The main study was reviewed by the Ethics Commission of the Germany Society for Psychology and no issues of ethical concern were identified (Az: MAW 022019). Given that experts were interviewed as part of this project and the focus was on their expert assessments and evaluations, no personal patient data was collected and we are therefore not required to seek approval from the local ethics committee for this part of the study. All interviews were conducted in German, the quotations are translations.

Competing interests

The authors declare that they have no competing interests.


References

1.
Albrecht M, Loos S, an der Heiden I, Temizdemir E, Ochmann R, Sander M, et al. Stationaere Hebammenversorgung [Inpatient midwifery care]. 2019 [Access: 31 Mar 2021]. Available from: https://www.iges.com/e6/e1621/e10211/e24893/e24894/e24895/e24897/attr_objs24976/IGES_stationaere_Hebammenversorgung_092019_ger.pdf External link
2.
Bakker AB, Demerouti E. Job demands-resources theory: Taking stock and looking forward. J Occup Health Psychol. 2017;22(3):273-85. DOI: 10.1037/ocp0000056 External link
3.
Bakker AB, Demerouti E. The Job Demands-Resources model: state of the art. J Manag Psychol. 2007;22(3):309-28. DOI: 10.1108/02683940710733115 External link
4.
Craig P, Dieppe P, Macintyre S, et al. Developing and evaluating complex interventions: The new Medical Research Council guidance. Brit Med J. 2008;337:979-83. DOI: 10.1055/a-1101-6960 External link
5.
Cramer E, Hunter B. Relationships between working conditions and emotional wellbeing in midwives. Women Birth. 2019;32(6):521-32. DOI: 10.1016/j.wombi.2018.11.010 External link
6.
Demerouti E, Bakker AB. The Job Demands-Resources model – Challenges for future research. SA J Industr Psychol. 2011;37(2):a974. DOI: 10.4102/sajip.v37i2.974 External link
7.
Demerouti E, Bakker AB, Nachreiner F, Schaufeli WB. A model of burnout and life satisfaction amongst nurses. J Adv Nurs. 2000;32(2):454-64. DOI: 10.1046/j.1365-2648.2000.01496.x External link
8.
Demerouti E, Bakker AB, Nachreiner F, Schaufeli WB. The job demands-resources model of burnout. J Appl Psychol. 2001;86(3):499-512. DOI: 10.1037/0021-9010.86.3.499 External link
9.
Demerouti E, Nachreiner F. Zum Arbeitsanforderungen-Arbeitsressourcen-Modell von Burnout und Arbeitsengagement – Stand der Forschung. Z Arb Wiss. 2019;73(2):119-30. DOI: 10.1007/s41449-018-0100-4 External link
10.
Deutscher Hebammenverband. Zahlenspiegel zur Situation der Hebamme 6/2017 [Numerical mirror on the situation of the midwife in June 2017]. 2017 [Access: 31 Mar 2021]. Available from: https://www.unsere-hebammen.de/w/files/tour-1/dhv-zahlenspiegel_web.pdf External link
11.
Deutscher Hebammenverband. Zahlenspiegel zur Situation der Hebammen 11/2019 [Numerical mirror on the situation of the midwives in November 2019]. 2019 [Access: 31 Mar 2021]. Available from: https://www.hebammenverband.de/index.php?eID=tx_securedownloads&p=3805&u=0&g=0&t=1597087504&hash=0aff59b9f1bbe9092676499ae5995b3f8d90c6f7&file=/fileadmin/user_upload/pdf/Presse/2019_11_Zahlenspiegel_zur_Situation_der_Hebammen.pdf External link
12.
Flick U. Qualitative Sozialforschung: Eine Einfuehrung [Qualitative Social Research: An Introduction]. 8th ed. Reinbek bei Hamburg: Rowohlt Taschenbuch Verlag; 2017.
13.
George W. Regionale Gesundheitsversorgung als Chance zukunftsfaehiger Ziel- und ressourcensteuerung im Gesundheitswesen. In: George W, Bonow M, editors. Gesundheitsversorgung. Lengerich, Westf.: Pabst Science Publishers; 2007. p.21-37.
14.
GKV-Spitzenverband. Zusammenfassung des GKV-Spitzenverbandes als Eraenzung zum Abschlussbericht der interministeriellen Arbeitsgruppe „Versorgung mit Hebammenhilfe“ [Summary of the GKV-Spitzenverband as a supplement to the final report of the interministerial working group “Provision of Midwifery Assistance”]. 2014 [Access: 31 Mar 2021]. Available from: https://www.gkv-spitzenverband.de/media/dokumente/presse/presse_themen/hebammen_1/14-01-07_Zusammenfassende_Stellungnahme_GKV-SV_Leseversion.pdf External link
15.
Hasselhorn HM, Mueller BH, Tackenberg P. Die Untersuchung des vorzeitigen Ausstiegs aus dem Pflegeberuf in Europa - die europaeische NEXT-Studie. In: Bundesanstalt für Arbeitsschutz und Arbeitsmedizin, editor. Berufsausstieg bei Pflegepersonal: Arbeitsbedingungen und beabsichtigter Berufsausstieg bei Pflegepersonal in Deutschland und Europa. Dortmund/Berlin/Dresden: Wirtschaftsverl. NW, Verl. fuer Neue Wiss.; 2005. (Schriftenreihe der Bundesanstalt für Arbeitsschutz und Arbeitsmedizin). p.11-20.
16.
Hasselhorn HM, Tackenberg P, Mueller BH. Vorzeitiger Berufsausstieg aus der Pflege in Deutschland als zunehmendes Problem fuer den Gesundheitsdienst - eine Ubersichtsarbeit. Gesundheitswesen. 2003;65(1):40-6. DOI: 10.1055/s-2003-36918 External link
17.
Hobfoll SE. Social and psychological resources and adaptation. Review Gen Psychol. 2002;6(4):307-24. DOI:10.1037//1089-2680.6.4.307 External link
18.
Jarosova D, Gurkova E, Palese A, Godeas G, Ziakova K, Song MS, et al. Job satisfaction and leaving intentions of midwives: analysis of a multinational cross-sectional survey. J Nurs Manag. 2016;24(1):70-9. DOI: 10.1111/jonm.12273 External link
19.
Karasek RA. Job Demands, Job Decision Latitude, and Mental Strain: Implications for Job Redesign. Admin Sci Quart. 1979;24(2):285-308. DOI: 10.2307/2392498 External link
20.
Karasek RA, Theorell T. Healthy work: Stress, productivity, and the reconstruction of working life. New York, NY: Basic Books; 1999.
21.
Kool L, Feijen-de Jong EI, Schellevis FG, Jaarsma DADC. Perceived job demands and resources of newly qualified midwives working in primary care settings in The Netherlands. Midwifery. 2019;69:52-8. DOI:10.1016/j.midw.2018.10.012 External link
22.
Kruse J, Schmieder C, Weber KM, Dresing T, Pehl T. Qualitative Interviewforschung: Ein integrativer Ansatz. 2nd ed. Weinheim, Basel: Beltz Juventa; 2014.
23.
Kuckartz U. Qualitative Inhaltsanalyse: Methoden, Praxis, Computerunterstuetzung. 3rd ed. Weinheim, Basel: Beltz Juventa; 2016.
24.
Maslach C. Understanding Burnout: Definitional issues in analyzing a complex phenomenon. In: Paine WS, editor. Job stress and burnout. Beverly Hills, CA: Sage; 1982. p.29-40.
25.
Meijman TF, Mulder G. Psychological Aspects of Workload. In: Drenth PJD, Thierry H, editors. Handbook of Work and Organizational Psychology. Hove: Psychology Press; 1998. p.5-33.
26.
Moessinger C, Weigl M, zu Sayn-Wittgenstein F zu. Stress bei der Arbeit – Warum freiberufliche Hebammen in Bayern die Geburtshilfe aufgeben: Eine Querschnittsstudie [Effort-reward imbalance of independent midwives in one federal state of Germany (Bavaria): A cross-sectional study]. GMS Z Hebammenwiss. 2019;(6):20-5. DOI: 10.3205/ZHWI000013 External link
27.
Mummendey H. Methoden und Probleme der Kontrolle sozialer Erwuenschtheit (Social Desirability). Zeitschrift für differentielle und diagnostische Psychologie. 1981;2(3):199-218.
28.
Pierce JL, Gardner DG. Self-Esteem Within the Work and Organizational Context: A Review of the Organization-Based Self-Esteem Literature. J Manag. 2004;30(5):591-622. DOI: 10.1016/j.jm.2003.10.001 External link
29.
Pohontsch N, Meyer T. Das kognitive Interview - Ein Instrument zur Entwicklung und Validierung von Erhebungsinstrumenten. Rehabilitation. 2015;54(1):53-9. DOI: 10.1055/s-0034-1394443 External link
30.
Pugh JD, Twigg DE, Martin TL, Rai T. Western Australia facing critical losses in its midwifery workforce: a survey of midwives' intentions. Midwifery. 2013;29(5):497-505. DOI: 10.1016/j.midw.2012.04.006 External link
31.
Sander M, Albrecht M, Loos S, Stengel V, Kleinschmidt L. Studie zur Hebammenversorgung im Freistaat Bayern [Study on midwifery care in a German federal state (Bavaria)]. Berlin; 2018 [Access: 31 Mar 2021]. Available from: https://www.stmgp.bayern.de/wp-content/uploads/2018/08/hebammenstudie_vollfassung.pdf External link
32.
Sander M, Albrecht M, Temizdemir E. Hebammenstudie Sachsen: Studie zur Erfassung der Versorgungssituation mit Hebammenleistungen in Sachsen sowie zur Moeglichkeit der kontinuierlichen landesweiten Erfassung von Daten über Hebammenleistungen. 2019 [Access: 31 Mar 2021]. Available from: https://publikationen.sachsen.de/bdb/artikel/33820 External link
33.
Siegrist J, Starke D, Chandola T, Godin I, Marmot M, Niedhammer I, et al. The measurement of effort-reward imbalance at work: European comparisons. Soc Sci Med. 2004;58(8):1483-99. DOI: 10.1016/S0277-9536(03)00351-4 External link
34.
Simon D, Tackenberg P, Hasselhorn HM, Kuemmerling A, Buescher A, Mueller BH. Auswertung der ersten Befragung der NEXT-Studie in Deutschland [Evaluation of the first survey of the NEXT study in Germany]. 2005 [Access: 31 Mar 2021]. Available from: http://www.altenpflege-online.net/content/download/146776/2922740/file/ap_09_10_pflegepraxis_next.pdf External link
35.
Six B, Caspar F. Attributionsfehler. In: Wirtz MA, editor. Dorsch - Lexikon der Psychologie. 19th ed. Bern: Hogrefe AG; 2020. p.217.
36.
Stahl K. Arbeitssituation von angestellten Hebammen in deutschen Kreisssaelen – Implikationen für die Qualitaet und Sicherheit der Versorgung [Working situation of employed midwives in German delivery rooms – implications for quality and safety of care]. Z Evid Fortbild Qual Gesundhwes. 2016. DOI: 10.1016/j.zefq.2016.07.005 External link
37.
Stahl K, Hildebrandt H, y Lehen C, Doering R, Siegmund-Schultze E. Verbesserung der Situation der Geburtshilfe in Baden-Wuerttemberg: Massnahmenplan [Improving the situation of obstetrics in a German federal state (Baden Wuerttemberg): plan of action]. 2019 [Access: 31 Mar 2021]. Available from: https://sozialministerium.baden-wuerttemberg.de/fileadmin/redaktion/m-sm/intern/downloads/Downloads_Runder-Tisch-Geburtshilfe/RTG_Massnahmenplan-OptiMedis_April-2019.pdf External link
38.
Statistisches Bundesamt. Grunddaten der Krankenhaeuser 2007 [Basic data of hospitals 2007]. 2008.
39.
Statistisches Bundesamt. Grunddaten der Krankenhaeuser 2017 [Basic data of hospitals 2017]. 2018.
40.
Sterne JAC, Hernán MS, McAleenan A, Reeves BC, Higgins JPT. Assessing risk of bias in a non-randomized study. In: Higgins JPT, Thomas J, Chandler J, Cumpston M, Li T, Page, MJ, Welch VA, editors. Cochrane handbook for systematic reviews of interventions. 2nd ed. Hoboken, NJ: Wiley-Blackwell; 2020. p.621-41.
41.
Widerszal-Bazyl M, Radkiewicz P, Hasselhorn HM, Conway P. Die Analyse von Pflegearbeit unter Anwendung des Demand-Control-Support-Modells in zehn europaeischen Laendern. In: Bundesanstalt für Arbeitsschutz und Arbeitsmedizin, editor. Berufsausstieg bei Pflegepersonal: Arbeitsbedingungen und beabsichtigter Berufsausstieg bei Pflegepersonal in Deutschland und Europa. Dortmund/Berlin/Dresden: Wirtschaftsverl. NW, Verl. fuer Neue Wiss.; 2005. (Schriftenreihe der Bundesanstalt für Arbeitsschutz und Arbeitsmedizin). p.109-15.
42.
Wirtz MA. Hof-Effekt. In: Wirtz MA, editor. Dorsch – Lexikon der Psychologie. 19th ed. Bern: Hogrefe AG; 2020. p.772.
43.
Wirtz MA, Schulz A. Evidenzbasierung in der Rehabilitation – Studiendesigns und konzeptuelle Grundlagen der empirischen Analyse der Wirksamkeit von Rehabilitationsmassnahmen. Rehabilitation. 2020;59(5):303-14. DOI: 10.1055/a-1064-6587 External link
44.
Xanthopoulou D, Bakker AB, Demerouti E, Schaufeli WB. The role of personal resources in the job demands-resources model. Int J Stress Manag. 2007;14(2):121-41. DOI: 10.1037/1072-5245.14.2.121 External link