Job Satisfaction of Midwives in the Context of a Midwife-Led Project – A Repeat Measure Mixed Methods Study

Background: Job satisfaction of midwives is important to prevent skill shortage. Those working in midwife-led models of care were more satised than those working in standard care. Job satisfaction in the context of a midwife-led project was not researched previously. The aim of this study was to investigate job satisfaction before and after the implementation of a midwife-led intervention. Methods: Longitudinal observational study at three time points using quantitative and qualitative methods. A total of 43 midwives working in the labour ward participated in the online surveys and 5-7 in the focus group discussions. The surveys comprised questions from validated instruments. Descriptive and multivariable time series analysis were used for quantitative and content analysis for qualitative data. Results: Adjusted predicted scores decreased between t 0 and t 1 and subsequently increased at t 2 without reaching baseline values (e.g ‘professional support subscales’: between t 0 and t 1 : (0.65, 95% CI [0.45, 0.86] versus 0.26, 95% CI [0.08, 0.45], p=0.005) and t 0 and t 2 (0.65, 95% CI [0.45, 0.86] versus 0.29, 95% CI [0.12, 0.47], p=0.004). Focus group discussions revealed three themes: “Job satisfaction”, “Challenges with the implementation” and “Continuity of care”. Midwives perceived the additional tasks as stressors. Conclusion: The implementation of new interventions might increase work related stress and decrease job satisfaction in an early phase. This effect was stronger than the one of acquiring more autonomy and responsibility. Heads of institutions and policy makers should recognise the needs of support and additional resources for staff when planning new projects.


Data collection
Data was collected at three time points: before the implementation of the telephone debrie ng sessions as well as two and seven months afterwards. At each time point, an online questionnaire was sent to all eligible midwives and a focus group discussion was conducted.
The online surveys were based on different validated instruments to assess job satisfaction [18][19][20][21]. In particular, the German version of the midwifery speci c instrument to assess job satisfaction from Turnbull et al. [18,22], selected questions of the German version of the Copenhagen Psychosocial Questionnaire (COPSOQ) [23], self-translated questions of the domain 'decision authority' of the Leiden Quality of Work Life Questionnaire for Nurses (LQWLQ-N) [19] as well as socio-demographic questions of the STRAIN-project were included [24].
In the focus group discussions, the themes of the subscales of the instrument of Turnbull et. [18] ('professional satisfaction', 'professional support', 'client interaction' and 'professional development'), 'decision authority' of the Leiden Quality of Work Life Questionnaire for Nurses (LQWLQ-N) [19], continuity of care [25] as well as questions regarding the conduct and organisation of the telephone debrie ng sessions were incorporated in the semi-structured interview guide.
The study was approved by the Ethics Committee of the Canton of Zurich in Switzerland (BASEC-NR. Req-2017-00133).

Data preparation and analysis
Quantitative data of the three online questionnaires were merged using an anonymised ID-code. For categorial variable absolute and relative frequency and for metric variables mean, median, range and standard deviation were computed as appropriate. Scores of validated instruments were calculated as proposed by their developer. Repeatedly measured categorical variables with more than two categories were compared using Skillings-Mack tests. Generalised estimating equation (GEE) models of the Gaussian family with robust standard errors and with log link were used to assess adjusted temporal trajectories of instrument scores. Corresponding point estimates with 95% con dence intervals were reported. Statistical signi cance was established at p < 0.05. Stata Version 13 (StataCorp, College Station, TX, USA) was used for all statistical analyses.
Focus group discussion were transcribed verbatim and analysed using qualitative content analysis methods according to Kuckatz [26]. Deductive and inductive coding were applied and codes were grouped into themes. Code and theme names as well as citations were translated into English and translations were checked by a German and an English native speaker. Qualitative data was analysed using Atlas.ti 8.

Characteristics of participants
The median age of participants of the surveys was 33.5 years (range = 25.0-64.0 years, Table 1). Two third of the midwives (n = 28) were born in Switzerland and more than half had a Bachelor's degree (n = 24). The midwives had worked for a median of 9.2 years in the profession. Attitude toward telephone debrie ng More than half of the midwives had an open attitude towards the telephone debrie ng sessions before their implementation (54.3% absolutely or mostly agreed at t 0 , Table 2). The proportion of midwives recognising the telephone debrie ng sessions as an additional stress increased slightly without signi cant difference between t 0 , t 1 and t 2 . However, participants acknowledging the project being important for the women decreased signi cantly between t 0 and t 2 (p = 0.035).

Development of job satisfaction of midwives
The midwife-speci c instrument [18] showed a decrease and subsequently increases in the mean of the 'Professional satisfaction subscale', the 'Professional support subscales' and the Professional support' between t 0 and t 1 respectively t 2 ( Table 3). The mean scores of the 'Client interaction subscale' in contrast, increased slightly between t 0 and t 1 and more clearly between t 1 and t 2 . Other job and occupational related factors Signi cant differences were found for adjusted predictions for the 'Meaning of work' subscale of the COPSOQ, which were lower at t 1 compared to t 0 and for the 'Job satisfaction' subscale, which showed a decrease between t 0 and t 2 ( Table 4). The 'Intention to leave the profession' subscale increased signi cantly between t 0 and t 2 . interval; 3 COPSOQ = Copenhagen Psychosocial Questionnaire; a signi cant adjusted difference between t 0 and t 1 ; b signi cant adjusted difference between t 0 and t 3 ; *p < 0.05; **p < 0.001 Figure 1: Adjusted trajectory of sub scores over measurements

Focus group discussions
A total of eleven midwives participated in one or two of the three focus group discussions, seven in the rst, ve in the second and seven in the third one. The focus group discussions revealed three themes, which were important for the job situation: "Job satisfaction", "Challenges with the implementation" and "Continuity of care".
The theme "Job satisfaction" comprised general aspects. Most midwives mentioned being very satis ed with their job. Having a secure job and working in a good team were experienced as very important factors.
"I experience that we are a really good team and help each other…" (before the implementation) Factors negatively affecting job satisfaction were working shifts but also the request to complete an increasing number of tasks. A trend toward negative quotations was higher two and seven months after the implementation of the debrie ng session, when interviews were conducted during periods with high workloads.
"Working shifts makes it nearly impossible to work 100 percent. Thus, working several years full time and keep the social environment intact is not possible." (two months after the implementation) Regarding the "Challenges of the implementation", midwives highlighted the additional workload of conducting the telephone debrie ng sessions, which should be completed during their working time. They experienced it as stressful, but even so, some of them point out the good feeling it gives them afterwards.
"Once I've done it, I'm happy about it, but it is the same as for her (comment: name of the colleague), that it is always breathing down my neck." (two months after) The organisation of the telephone debrie ng sessions was especially challenging. Women were di cult to reach, and several calls or emails were needed to make contact. Midwives also pointed out their need to know if women pro ted from the telephone calls in order to nd a sense in the additional workload.
"This (comment: the debrie ng sessions) is really di cult to plan, this is a major challenge for me." (seven months after) "Continuity of care" was a controversial issue. Many participating midwives worked in a university hospital because they wanted demarcation from work.
However, seven months after the implementation of the telephone debrie ng sessions, some midwives came to appreciate the follow-up contact with the women which rounded off their care.
"And then, I'm coming back again after two, three months (…) This is nice for me, that work is not already nished but we return to the birth situation." (seven months after)

Discussion
To our knowledge, this was the rst study investigating job satisfaction of midwives in the context of a midwife-led project. Quantitative and qualitative data revealed that additional tasks increased work-related stress in the short term.
In contrast to studies investigating job satisfaction of midwives working in midwife-led models of care [8], our study did not show an increase in satisfaction but a decrease in the rst phase followed by an increase which did not reach the baselines values. Even though the implementation of the telephone debrie ng sessions comprised elements of midwife-led care such as fostering continuity of care as well as being initiated by the midwives and in their responsibility, the changes in the work situation were less substantial. However, the additional tasks lead to an increased workload and decreased satisfaction, and con rmed the results of other studies that workload is a potential work-related stressor for midwives [5,7,27]. The results of the current study were also consistent with those of one, which showed dissatisfaction in situations where additional tasks led to a very high workload [28].
The longitudinal character of our study might have added as a new aspect the development of job satisfaction in a very early phase of the implementation of an intervention. Two months after starting the telephone debrie ng sessions, the organisational aspects had priority and many midwives were not able to acknowledge the bene ts of the project. This might be one of the causes for the initial decrease and subsequent increase of job satisfaction during the study period. Midwives might have needed time to recognise the bene ts of the intervention. A Danish study showed that providing high-quality care led to an increased job satisfaction [4]. Interviews with the users of the telephone debrie ng sessions in our study emphasised the satisfaction of women and their bene ts of processing birth [17]. It might be that it was too early to estimate the long-term development of job satisfaction seven months after the implementation of the debrie ng sessions. Future study should plan longer follow-ups.
Strengths of our study were the use of questions from validated instruments combined with qualitative data from focus group discussions. Quotations of midwives provided a deeper insight leading to explanations for some quantitative ndings. However, the small sample size, which was due to the limited number of midwives working in the same hospital, was a limitation. Additionally, the single centre study might not have provided results which can be generalised for other settings. Midwives working in larger maternity units might not have the same interest in continuity of care as those working in smaller ones. During the study phase, which lasted nearly one year, staff turnover was observed causing incomplete follow-up data as well as new participants.
Additionally, the chances of changing external factors such as periods with higher and lower workloads or new regulations were also increased due to the duration of the study.

Conclusion
Our study showed a decreased job satisfaction in the early phase of a new project. This effect was stronger than the one of acquiring more autonomy and responsibility. This knowledge is important for heads of institutions and policy makers, because strategies to support midwives during implementation phases and additional resources might be necessary to prevent decreased job satisfaction. It remained unclear how satisfaction of midwives would develop during a longer time period. Future studies should investigate job satisfaction in the context of midwife-led projects in larger samples and different settings and consider a longer follow-up.
Adjusted prediction; CI:Con dence interval; COPSOQ; Copenhagen Psychosocial Questionnaire; LQWLQ-N:Leiden Quality of Work Life Questionnaire for Nurses; SD:Standard deviation Declarations Ethics approval and consent to participate All participants received written and oral information about the aim and the procedure of the study and gave written consent. The study was approved by the Ethics Committee of the Canton of Zurich in Switzerland (BASEC-NR. Req-2017-00133).

Consent for publication
Not applicable Availability of data and materials The datasets generated and analysed during the current study are not publicly available due to the sensitive nature of qualitative data from but are available from the corresponding author on reasonable request.

Competing interests
The authors declare that they have no competing interests.

Funding
The study was nanced by the State Secretariat for Education, Research and Innovation SERI, Switzerland. The funding body did not play any role in the design of the study and collection, analysis, and interpretation of data and in writing the manuscript.  Adjusted trajectory of sub scores over measurements