gms | German Medical Science

In Germany, almost 20 percent of all births are attended by self-employed midwives [1]. In the German federal state of Bavaria, a total of 107,000 babies were born in 2013, of which around 2,200 (two percent) were brought into the world at home or in a birthing centre [2]. These are not the only babies cared for by freelance midwives, however. In Bavaria, there is a particularly large number of freelance midwives attending births in hospitals: 70 percent of the 116 labour wards (n= 81) are staffed by self-employed midwives working shifts. In just 30 percent (n= 33) of Bavarian labour wards are the midwives actually employed by the hospital [2]. Thus, across Germany, but particularly in Bavaria, freelance midwives make a crucial contribution to ensuring comprehensive provision of intrapartum care (both within and outside the hospital setting). Recent coverage in the Bavarian media has reported labour ward closures, personnel shortages, midwives being stretched and leaving mothers unattended whilst labouring in the hospital [2]. In light of the fact that freelance midwives are leaving the profession, health policy intervention is necessary. Yet, the field of health policy-making suffers from a lack of detailed data on the diverse composition of the profession, its pressures and its resources. To date, scientific studies on the working conditions, mandate to provide care and departure from the profession of self-employed midwives in Germany and Bavaria have been inadequate. The current study aims to help close this research gap by examining the situation regarding freelance midwives in Bavaria leaving the profession entirely and/or no longer providing intrapartum care.

Here, a differentiated view of independent midwives’ work structures is essential, both for this study and for the general public debate as well as the public debate on health and the midwifery profession. Bearing this in mind, as part of this study, I developed the following definitions for the work structures of freelance midwives providing intrapartum care in Germany:

Midwifery work is traditionally classified according to occupational setting. For example, in Section 134a of Book 5 of the German Social Code (SGB V) on the payment of midwives by the health insurance companies, a fundamental distinction is made between midwives providing intrapartum care outside the hospital setting and those providing such care within a hospital. In Germany, it is also customary to use another criterion to classify midwives: since 1985, midwives have been able to choose whether they would like to practice as freelancers or as salaried members of staff [12]. Another option that is popular among midwives today is to combine a position of employment with freelance work. Figure 1 [Fig. 1] illustrates the classification described above.

The work model of independent or self-employed midwives providing intrapartum care can essentially be divided into two different systems [12]:

1.

Although independent midwives working within the “shift work” model generally continuously cover shifts in hospital delivery rooms, they are not actually employed by the hospital but rather charge the health insurance providers or insurance companies of the women they care for directly. They attend to all women who present themselves at the hospital to give birth during their shift and, as a team, these freelance midwives ensure the hospital delivery rooms are staffed around the clock.

2.

Independent midwives working in a caseload model enter into a direct treatment contract with the women on their books and, from around 37 weeks gestation are on call 24 hours a day ready to attend their clients’ births. What clearly distinguishes this second model in the context of this study is, first, the continuity of antenatal, intrapartum and postpartum care (including breastfeeding counselling), and, second, the 24-hour on-call service provided by the midwife for her clients.

In her dissertation, Birgit Reime [9] analysed the work stress of midwives from a sociological perspective. The primary occupational pressures in the midwifery profession that Reime identified were: working hours, night, weekend and shift work and on-call duty. She also reveals midwives’ low pay as a stress factor as well as practitioners from other areas questioning their professional competences.

In Switzerland, the low pay and irregular working hours of freelance midwives were highlighted as early as 1993 [4]. Independent working, high levels of responsibility, the extensive use of their own skills and the diverse range of tasks were identified as positive features of midwifery. The close relationship midwives develop with the women they care for and the feeling that they are doing something meaningful were also identified as positive aspects of the job. Society’s recognition of their occupation was particularly important for freelance midwives. However, the fact that it is so difficult to find cover was perceived as a work stress. Tensions with members of other professions and low pay were added to the list of negative aspects. Responsibility and professional autonomy was generally seen as positive factors, although, when midwives are faced with difficult decisions, this could also become a stress factor.

In terms of structural stress factors, the Accident Prevention and Insurance Association for the Health and Welfare Services (Berufsgenossenschaft für Gesundheitsdienst und Wohlfahrtspflege, BGW) is aware of the excessively long and irregular working hours and night shifts [8]. The BGW also highlights the fact that occupational stress factors are not systematically recorded and, since 2011 has been supporting the doctoral student, Nina Reitis, to gather information on freelance midwives’ working conditions. As part of her research project, in January 2012, midwife and scholar Reitis conducted a validated postal survey of 1,000 self-employed midwives in Germany [10]. The findings of the study have not yet been published but initial evaluations of the demographic data and midwives’ workload were presented at the German Society for Midwifery Science’s 2^nd International Conference in February 2014. These initial findings indicated above-average levels of stress among the sample surveyed compared to other healthcare professions studied by the BGW.

In 2011, the German Federal Ministry of Health commissioned the IGES Institute to conduct a study on the state of pay and supply in midwifery care [1]. The study forecast that the majority of midwives working in Bavaria (both within and outside the hospital setting) would experience increasing financial insecurity and that a growing number would leave the profession. However, the misleading choice of study title (which referred exclusively to midwifery care outside the hospital setting) gave the impression that only independent midwives attending homebirths or births in birthing centres were affected. The relevance of the study’s findings for freelance midwives working in hospitals was thus obscured, which, among other things, resulted in the press receiving false information [5].

The IGES study identified seven factors that could result in self-employed midwives considering a reduction in their working hours or giving up their jobs entirely [1]: excessive workload, dissatisfaction with working hours, incompatibility of the job with private and family life, low pay, high cost of professional liability insurance, high overall running costs and other professional interests. According to the study, many midwives are ambivalent about their jobs. The majority of midwives responded that they felt they were doing something meaningful in their daily work they were satisfied at the end of their working day. Professional autonomy and the freedom to make their own decisions were also seen as very positive aspects of working as a freelancer. Nevertheless, one in four self-employed midwives often or very often considered giving up their jobs.

The phenomenon of healthcare professionals giving up their careers was also examined by Josephson et al. [6] in a study of Swedish nurses. According to the study, work-related pressures elicited stress responses which, in turn, caused physical or psychiatric disorders. In order to combat this, 18 percent of the study population changed to a new field of work, gave up work or took long-term sick leave. The researchers involved in the study interpreted a change in or abandonment of career as a strategy to prevent the illnesses that develop as a result of unhealthy psychosocial working conditions.

In summary, the literature shows that freelance midwives primarily value the social and emotional aspects of their work, such as social recognition, the sense of purpose they get from their profession and the close relationships they develop in their working life [1], [9], [10]. They missed job security and opportunities for promotion, however [1], [9]. The stresses they experience can largely be attributed to occupational and organisational structure: low income, long working hours and an unfavourable work-life balance are mentioned especially frequently [1], [9], [10]. Added to this are the increasing costs of professional liability insurance for midwives providing intrapartum care, which represents an additional stress for self-employed midwives [1], [7], [10].

An analysis of the literature produced an interesting finding: the existence of intense pressure but at the same time also a high level of job satisfaction among freelance midwives. This phenomenon was frequently described in the available literature [1], [4], [8], [9] but has not yet been interpreted further. Moreover, it became apparent that it is difficult to estimate the future development of the provision of intrapartum care by self-employed midwives. The present study seeks to address this research need.

The effort-reward imbalance (ERI) model [13], [15], [16], [14] is an established model used in psychology to measure stress at work. In the context of working conditions in the healthcare sector, the model has proven to be a good predictor of imminent changes in and the intent to leave a given profession [3], [6].

Siegrist [13] uses the term effort to describe elements of investment in occupational contexts which are equalised by rewards in the form of job characteristics such as remuneration, esteem and/or career opportunities. If there is a disparity between these two weights, Siegrist [14] then refers to an occupational effort-reward imbalance. This is illustrated in Figure 2 [Fig. 2].

Prolonged effort-reward imbalances elicit strong negative emotions and sustained stress responses in the individual affected. If the cost of the effort invested in the occupational context outweighs the benefits of the reward received for that effort, the risk of damaging effects on health caused by stress increases. The third component of the ERI model is overcommitment – excessive dedication to a job, which can modify the individual balance between effort and reward.

A review of studies on the model [15] demonstrates the empirical robustness of the ERI model. The report summarises the results of six European cross-sectional and longitudinal studies that use the ERI model as a theoretical frame of reference. A significant association could be identified between elevated relative risk of coronary heart disease and/or cardiovascular risk factors and mental illness, particularly depression and alcohol dependency, and effort-reward imbalances in the occupational context.

The validity and reliability (internal consistency) of the individual questions used in the effort-reward imbalance measurement tool were evaluated by a wide range of international researchers [11], [17]. A research group at the University Hospital of Düsseldorf (Universitätsklinikum Düsseldorf), led by Johannes Siegrist, is constantly working to improve the questions. One of the group’s publications on the quality of the instrument [17] emphasises the high scale reliability. Using a confirmatory factor analysis based on data from a representative sample of employed individuals in Germany, the authors confirm that the scales have satisfactory psychometric properties. According to the authors, the theoretical structure of the model used was empirically substantiated, which indicates high validity. The correlation between the individual scales was calculated using Cronbach’s alpha. The values for internal consistency were higher than alpha=0.7 for all scales. Additionally, using coefficient calculations (r_max=–0.25), the scales for effort and reward were identified as having a high degree of selectivity.

The study aims to establish the extent to which stress at work and the rewards received by self-employed midwives for their efforts are evenly balanced. To do so, we measure the prevalence of imbalanced effort and reward situations among freelance midwives providing intrapartum care in Bavaria. In addition, we examine the extent to which the prevalence of effort-reward imbalances differs in the different work models (midwives working outside the hospital setting, freelance attending midwives who form the core staff covering shifts in hospital labour wards and freelance attending midwives with on-call duty).

The Ethics Commission of the Institutional Review Board at Ludwig-Maximilians-Universität (LMU) in Munich (Ethikkommission der Medizinischen Fakultät der Ludwig-Maximilians-Universität München) gave the study a positive evaluation (Nr. 204-14). We conducted an anonymised online cross-sectional survey of self-employed midwives in Bavaria providing intrapartum care (from all the work models mentioned). Midwives were invited to participate via the email distribution list of the Bavarian Midwifery Association (Bayerischer Hebammenverband) and via a “domino effect” in midwifery-related forums and Internet platforms. The following inclusion criteria were determined: participants had to be working as a self-employed midwife, had to have provided intrapartum care as one of their services in 2013 and had to be working principally in Bavaria.

A total of 107 invitations were sent for the data survey; 70 midwives completed the questionnaire (65 percent). Of these, 25 midwives had to be excluded from the data analysis, primarily because they had already stopped providing birth assistance in the last few years and therefore no longer fulfilled the inclusion criteria. Ultimately, we were able to evaluate data from 45 freelance midwives who had attended births in Bavaria in 2013 (42 percent).

To capture the ERI model, the most recent standardised and validated short version of the effort-reward imbalance questionnaire [16] was used. This comprises a total of 16 questions. The first three questions focus on the effort items, the next seven questions collect information on the reward items. Calculating the effort-reward imbalance ratio scores enables us to draw conclusions on the prevalence of an imbalanced effort and reward situation (ERI ratio scores ≥1) in the study population.

In addition, sociodemographic information on the participant was collected (professional experience; number of births/year; town/federal state) as well as information on their job situation (on-call duty and/or shift work; birth attendance outside the hospital setting and/or within a hospital).

The statistical evaluation was conducted using SPSS© 17.0 (IBM Deutschland GmbH, Ehingen) software. Descriptive information on the composition of the sample was provided and the sums of the effort and reward variables were calculated. On their own, the totals are not yet meaningful, however. Therefore, in a next step, the ERI ratio is calculated based on the following formula:

The numerator in the ratio comprises the values from the effort variables and the denominator contains the values from the reward variables. The k variable adjusts the formula and is composed of a quotient of the total number of reward variables and effort variables. If the ERI ratio is greater than 1, an effort-reward imbalance is identified [13].

We conducted an analysis of differences in terms of the prevalence of effort-reward imbalances between self-employed midwives covering hospital shifts as core midwives and midwives providing intrapartum care outside the hospital setting (MANOVA). Unfortunately, it was not possible to draw comparisons with midwives working within the on-call model as the number of available cases was too small. The mean value of the overcommitment variables and their correlation with the ERI ratio were calculated (linear regression).

The efforts and rewards of self-employed midwives derived from the literature reviewed were matched with the effort-reward imbalance model and presented in Figure 3 [Fig. 3].

In a first step, the individual items from the ERI questionnaire were analysed: under efforts, respondents particularly highlighted inadequate pay (mean 1.71; SD=0.73), as well as the anticipated deterioration of their job situation (mean 3.16; SD=0.71). In terms of the most important reward, on the other hand, midwives referred to the deserved recognition from people important to them in their professional lives (colleagues/superiors; mean 3.13, SD=0.7).

A total of 33 of the 45 midwives surveyed had an ERI score of ≥1 (mean 1.2, SD=0.3). The prevalence of effort-reward imbalances in work was thus equivalent to 73 percent.

Significant differences were only anticipated between independent attending midwives covering hospital shifts and those working outside the hospital setting as the number of freelance midwifes providing an on-call service participating in the survey was very low (n=2). On average, the ERI scores for the midwives working outside the hospital setting was lowest with an arithmetic mean of 1.03 (SD: 0.3 / 95% CI 0.78–1.25). This group was followed by attending midwives combining hospital work with out-of-hospital intrapartum care; their mean value was 1.19 (SD: 0.3 / 95% CI 1.01–1.26). The midwives exclusively covering shift work in hospitals had the highest ERI scores with a mean value of 1.33 (SD: 0.3 / 95% CI 1.1–1.58), which means they were most affected by effort-reward imbalances.

The prevalence of effort-reward imbalances (≥1) was 50 percent among midwives working outside the hospital setting and 86 percent among self-employed midwives working as core staff covering shifts in hospitals. This variance was confirmed as significant using a Chi² test (p=0.26).

The reported overcommitment was mid-table with a mean value of 15.51 (SD=3.02). In addition, there was also a positive linear correlation between overcommitment and the effort-reward imbalance ratio in the sample examined (F=4.997; p=0.03; R²=0.18).

Using Siegrist’s model of effort-reward imbalances at work [13], [16], [14] as part of this study we conducted a cross-sectional survey of self-employed midwives attending births in Bavaria in all the aforementioned settings (working outside the hospital, covering shifts as core midwives in a hospital labour ward or similar, and freelance midwives providing on-call intrapartum care). The results show that for 73 percent (n=33) of all midwives participating in the survey, there was an imbalance between professional demands and the rewards they receive in exchange for their investment. The prevalence of these effort-reward imbalances in the total sample examined means that for almost three-quarters (n=33) of the midwives surveyed, the stresses of attending births in a freelance capacity is not equalised by the rewards received. This study thus provides empirical evidence that, for many midwives (both those working freelance within a hospital and those attending births outside the hospital setting) the ratio between the effort put into their jobs and the rewards they receive is unbalanced. The results of the study at item level substantiate the findings from the available literature: self-employed midwives are under financial pressure and are worried about their work situation deteriorating in future. This is mostly equalised or rewarded on the social and emotional level.

According to the standard interpretation of the effort-reward imbalance, the midwives affected can expect negative health implications due to work stress. However, the model can also be used as a predictor of future job leavers in the midwifery profession. The study findings can therefore act as a stimulus to improve the job situation and working conditions of midwives: if the working conditions remain the same or even deteriorate, we can also anticipate that increasing numbers of midwives will leave the profession due to the high prevalence of effort-reward imbalances. The study forecasts this to be the case especially for independent midwives covering shifts in hospital labour wards or similar.

As a result, it will become increasingly difficult for hospitals to find enough self-employed midwives to maintain continuous cover for all shifts on their labour wards. This, in turn, could result in labour wards staffed by freelance midwives being forced to close. The comprehensive provision of care to expectant mothers or women in labour is thus jeopardised.

Apart from completely abandoning the profession, however, midwives also have the option of shifting the range of services they offer to other areas of midwifery (prenatal and postnatal care and breastfeeding counselling, for instance). Birth attendance, which is really at the heart of midwife care, is thus sidelined. Consequently, it becomes harder and harder to find midwives to provide continuity in antenatal, intrapartum and postpartum care.

It was unclear why only two freelance midwives offering an on-call service participated in the study. It is possible that they could not be reached with the study design used or that there are, in fact, only a small number of midwives working within this model in Bavaria. We can therefore not emphasise the importance of a system of professional registration and the resultant statistics on midwives’ different spheres of work enough.

The study has various limitations: the ERI model can highlight situations of psychological stress and can also be used as a predictor of the intent to leave a profession on the population level. However, the decision to abandon the profession is a personal one for each midwife and, as such, cannot be fully represented using quantitative, aggregate data. The sampling error, the restricted access to the field and the challenges of calculating the target population all reduce the generalisability of the study results. The online survey method also creates a selection bias, as midwives must have access to the necessary technical equipment, motivation and skills to participate

The present work illustrates that the systematic capture and investigation of the structures and conditions of the midwifery profession in Germany is a relevant objective for healthcare research. Due to the limited available data on work-related stress experienced by self-employed midwives, the findings of this study are therefore still salient, despite the aforementioned limitations.

Almost three quarters of the freelance midwives participating in this study experience a high level of work-related stress based on an effort-reward imbalance. In light of these findings, we can anticipate that self-employed midwives in Bavaria will make the decision to leave the profession in the future.

For midwives working outside the hospital setting, this means less freedom of choice in terms of where the women on their books give birth.

Since 70 percent (n=81) of labour wards in Bavaria are staffed by self-employed midwives working shifts, the anticipated number of freelance attending midwives abandoning midwifery will be especially problematic.

If self-employed midwives limit their range of services at the expense of birth attendance, Bavaria would therefore face the threat of a universal shortage of supply in the provision of intrapartum care.

This article is based on a master’s thesis in the field of Public Health at Ludwig-Maximilians-Universität (LMU) in Munich.

The authors declare that they have no competing interests.

The research funding program Lehre@LMU gave compen-sation for the material costs of Clara Mössinger’s master’s thesis. The company Weleda provided a package of its products as an incentive for midwives to participate in the survey.