Perceptions of midwives on shortage and retention of staff at a public hospital in Tshwane District

BACKGROUND:  Midwifery is the backbone of women and child healthcare. The shortage of staff in maternity units is a crisis faced by many countries worldwide, including South Africa. OBJECTIVES:  This study aims to explore the perceptions of midwives on the shortage and retention of staff at a public institution. METHOD:  The study was conducted at one of the tertiary hospitals in Tshwane District, Gauteng Province. A total of 11 midwives were interviewed through face-to-face and focus group interviews. An explorative, descriptive generic qualitative design method was followed, and a non-probability, purposive sampling technique was used. Thematic coding analysis was followed for analysing data. RESULTS:  The impact of shortage of midwives was reported to be directly related to poor provision of quality care as a result of increased workload, leading to low morale and burnout. The compromised autonomy of midwives in the high obstetrics dependency units devalues the status of midwives. CONCLUSION:  Midwives are passionate about their job, despite the hurdles related to their day-to-day work environment. They are demoralised by chronic shortage of staff and feel overworked. Staff involvement in decision-making processes is a motivational factor for midwives to stay in the profession. The midwives need to be in the centre of the decision-making processes related to their profession. The revision of the scope of practice and classification of midwifery profession away from general nursing complex by the South African Nursing Council (SANC) could place midwifery in its rightful status.

In this qualitative study, a generic inductive approach with a descriptive design was followed to explore the perceptions on the shortage and retention of midwives. A qualitative design investigates people’s reports of their subjective opinions, attitudes, beliefs or reflections on their experiences of things in the outer world (Percy, Kostere & Kostere 2015:78). Focus groups and individual face-to-face semi-structured interviews were conducted among participants. The study was conducted at a particular academic hospital in Tshwane District, Gauteng Province. Among other services, the hospital provides for high-risk maternal and neonatal care. The antenatal and postnatal wards have 21 beds each, while the labour ward consists of 6 labour beds, 3 admission beds and 10 high care beds. The institution is a teaching hospital with gynaecology and obstetrics registrars, student doctors and midwifery students. The student midwives are placed to undergo periodic clinical training from the nursing college and the two universities which are affiliated to the hospital. The midwifery staffing of the units appears to be subminimum as there is evidence of consistent reliance on the agency staff to cover daily shifts. The population comprised midwives in the maternity wards. Non-probability purposive sampling method was used to select 11 midwives working in the maternity units at the time and who fitted the eligible criteria of two or more years’ experience practising as midwives. The researcher included six midwives in the focus group discussions. Individual face-to-face interviews were conducted with five midwives, two of whom also took part in the focus group discussions. Data collection instruments were designed and tested on a different group of midwives at a different setting to evaluate the effectiveness of the tool. Interview guides were used and upon satisfaction on responses received, the tool was then confirmed to be ideal to be used in the main interviews. Data were collected using an interview guide and a tape recorder. Midwives were interviewed during their working shift because of availability. A semi-structured interview with open-ended questions was used for the interviews. The interview guide comprised a wide range of open-ended questions, covering demographic factors, intentions of the midwives to either stay or leave the profession, reasons for their proposed intentions and perceptions on why other midwives opt for leaving. The same interview guide was used for focus group discussions and for individual interviews. The researcher requested permission from the operations manager about the interviews and made appointments with midwives. Data collection was started with one focus group discussion to get a broader perspective on the shortage and retention of staff, followed by individual face-to-face interviews to obtain in-depth data. Focus group discussion was performed to ensure that midwives can discuss the topic with other colleagues and for the researcher to obtain more information. Six midwives participated in the focus group discussion which was conducted by two interviewers. The first interviewer was there to ascertain that the discussions were conducted smoothly by supporting the participants in explaining the purposes and procedures of the interview and to obtain written consent from all participants. Because of the nature of business in the labour ward, the discussions were conducted in an empty labour room, so as to avoid moving the midwives away from their work stations. Individual face-to-face interviews were conducted on a different day from the focus group. The researcher decided to conduct face-to-face interviews with a few midwives with the view that focus group may limit freedom of expression, especially with individuals who are not confident to voice their personal views in a group setting. Five participants were interviewed in a quiet office in their unit of work. The tape recorder was used to record all the information discussed. The research questions which were asked were the same as the ones used in the focus group discussions. Data saturation was achieved during the face-to-face interview with participant number three. Two more participants were interviewed thus confirming data saturation. Thematic data analysis was followed to analyse the data (Percy et al. 2015:80). After familiarisation with collected data, themes were generated. Data were summarised by category from each transcript and arranged according to themes. The themes were supported by quotations from the raw data and then compared and contrasted with existing relevant literature. Names and other identifiers were changed to protect the privacy of participants (Green & Thorogood 2014:72). The participants in the focus group were identified as Participant#, while the individual interview participants were identified as IP#. Ethical approval was granted by the Research Ethics Committee of the University of South Africa (HSHDC/567/2016) and the Faculty of Health Sciences of the University of Pretoria for the Gauteng Province (252/2017). Written permission to conduct the study was obtained from the Chief Executive Officer of the hospital as a selected site. Verbal permission was also obtained from the operations manager before conducting interviews in the maternity ward. Confidentiality means not disclosing information gained from research in other settings such as through informal conversations (Green & Thorogood 2014:72). Protection of confidentiality was achieved by limiting the persons who had access to taped materials. Only the researcher had access to taped materials during the research study process. Harish, Kumar and Singh (2015:410) describe autonomy as the ability to decide for the self, free from control of others, and with sufficient level of understanding to arrive at a meaningful choice. The researcher explained the purpose and procedures of the research study, and participants were given an opportunity to read the information document before signing the consent forms. Consent forms were signed before the beginning of the interviews and group discussions. Trustworthiness of this study was assured by using Lincoln and Guba’s criteria as cited in Anney (2014:276–278) of credibility, transferability, dependability and confirmability. A pretest of the data collection tools was done on midwives from a different setting of research site which was not included in the main study. According to Colorafi and Evans (2016:23), dependability is a component of trustworthiness, which can be fostered by consistency in procedures across participants over time through various methods of data collection. The researcher used semi-structured interviews with the aid of an interview guide. All participants were asked the same questions in the same order to ensure dependability. Colorafi and Evans (2016:23) describe confirmability as reasonable freedom from researcher bias. Transparency in all the processes of the research study was portrayed through sharing the approach and methods of data collection with the ethics committees involved. The data collected were kept safe and would be available if needed by stakeholders. Grove, Burns and Gray (2013:199) state that internal validity (credibility) refers to the extent to which the effects detected in the study are a true reflection of reality, rather than the result of extraneous variables. The researcher did not alter the question, and the objectives were never altered throughout the research process. All participants were given the same opportunity to respond to the questions posed. External validity (transferability) is concerned with the extent to which the study findings can be generalised beyond the sample used in the study (Grove et al. 2013:202). For this study, generalisation was limited to only the population within the selected institution. According to Percy et al. (2015:79), external generalisation is not necessary, because the data are not quantifiable.

N/A

• Dima Health