In Pakistan, a cadre of community midwives was introduced in 2007 to increase the proportion of skilled birth attendants in the country (USAID, 2012). The community midwives participate in an 18-month training programme, approved by the Pakistan Nursing Council (PNC). PNC accreditation is based on a minimum qualification plus work experience in the community setting, and that the applicant should be a permanent resident of that community area. Evidence suggests that the role of skilled birth attendant, or community midwife, is significant for the prevention and instant treatment of maternal and child complications (Bhutta et al, 2008).
The World Health Organization (WHO)'s State of the World's Midwifery (United Nations Population Fund and WHO, 2014) report scrutinises the worldwide midwifery landscape across 73 low- and middle-income countries and emphasised the investment in high-quality midwifery as imperative for the prevention of maternal and newborn deaths. Since 2011, governments, non-governmental institutions, and reporting countries have taken crucial steps to support midwifery. However, there is still a lack of adequately educated, confident, and experienced midwives to support the health of women and infants.
Internationally, midwives moving into self-employed practice, although registered and competent to practice, identified that they needed support to develop confidence and to enhance professional expertise (Kensington, 2006; Lennox et al, 2012; Cummins et al, 2015). Studies have revealed a lack of support in two important areas. The first is the absence of mechanisms such as mentorship, and the second is lack of administrative and reporting structures (Booth et al, 2006; McDonald et al, 2010). This lack of mentorship support for midwives has also been reported in the Pakistani context. Mohammad et al (2015) carried out a qualitative study on the experiences of community midwives, in which participants reported that once they completed their training, they lacked support during their transition process, which hindered their ability to practise as independent community midwives. A similar study conducted by Sarfaraz and Hamid (2014) also identified the participant's concerns about lack of support in the field. The challenges show that midwives in Pakistan are struggling to gain their professional identity as independent midwives (Saleem et al, 2015). These studies show that the same issues that have been reported internationally.
‘For organisations and communities, the benefits of mentorship are related to a motivated and satisfied workforce, as well as staff retention. A well supported professional workforce can offer higher quality care to the mother and child’
Benefits of mentorship
Turkmani et al (2014) highlighted that the implementation of a mentorship programme is a very powerful professionalising strategy in midwifery. It is considered a long-term partnership, which is mutually beneficial and voluntary in nature, whereby a knowledgeable person (mentor) supports the growth and development of a less-experienced person (mentee) with his or her leadership potential (Canadian Nurses Association, 2004). It helps to develop competency (Henderson and Malko-Nyhan, 2006), decrease work-related stresses in a practical environment (Beecroft et al, 2006; Chen and Lou, 2014), and apply theoretical knowledge to clinical practice (McKenna, 2003). Formal mentorship training may help midwives to develop leadership skills, enhance their personal and professional relationships in a community setting, and increase their confidence (Theobald and Mitchell, 2002; Lennox et al, 2008). To a mentor, this may provide personal satisfaction and motivation, enthusiasm for her career, and an enhanced sense of professional development (McKenna, 2003; Lennox et al, 2008). For organisations and communities, its benefits are related to a motivated and satisfied workforce, as well as staff retention (Myall and Lathlean, 2008). Moreover, a robust, well supported professional workforce has the ability to offer a higher quality of care to the mother and child (Steele, 2009).
Mentorship around the world
Mentoring is also associated with graduate programmes, such as in New Zealand, which has had a national Midwifery First Year in Practice (MFYP) programme for new graduate midwives since 2007. A retrospective study of the graduates' experience of MFYP between 2007 and 2010 showed a high regard for the mentorship programme (Dixon et al, 2015). The programme was seen as an important source of support for all graduate midwives, both employed and self-employed (Kensington et al, 2016). In a recent Canadian study, mentorship training initiatives aimed to increase registered nurses' and midwives' competencies and confidence (Lau et al, 2016). In Pakistan, however, no such initiative has been taken up.
Preparation for mentorship
Before running an effective mentorship programme, mentors need to be adequately trained and updated to prepare for this challenging role (Finnerty et al, 2006). Mentors need to be made aware that mentoring is a demanding task and in order for them to continue to mentor, they must understand their role and their responsibilities (Finnerty et al, 2006).
The aim of the study, therefore, was to pilot an education programme about mentorship for community midwives. The mentoring model used was a developmental one, which aimed to equip community midwife mentors with increased role competency, self-confidence and self-determination in their practice and in that of their mentees in the future.
Method
Study design and setting
This study used a quasi-experimental pre- and post-design. A total of 50 community midwives were recruited from 19 districts of Sindh, Pakistan. The study took place from November 2015 to September 2016.
Ethical approval
Approval for the study was obtained from the Midwifery Association of Pakistan and Ethical Review Committee of Aga Khan University. The participants were informed of their research rights, such as withdrawing from the study at any stage. Written consent was obtained and confidentiality was assured.
Inclusion criteria
The total study population consisted of community midwives who had successfully completed the 18-month education programme from an accredited midwifery school and had 3 or more years of experience in community setting. They were licensed by the PNC to practise midwifery independently.
Intervention
As an intervention, participants were provided with a 2-day workshop on mentorship, with 4 hours of teaching each day. The primary investigator developed mentorship sessions based on the literature review of published articles and the input of experts. The topics of the sessions were: defining mentorship, steps of mentorship process, confidentiality in mentoring, problem solving in mentoring, and accountability. The primary investigator who delivered the sessions had experience of teaching and experience of working with midwives.
Data collection
In May 2016, community midwives enrolled in the 2-day mentorship workshop. Pre-test and post-test questionnaires were administered before and after the workshop. Altogether, three training workshops were conducted with different participants in groups of 16, 20 and 14 participants, making a total of 50 participants. The study participants were selected from a list of community midwives, according to eligibility criteria. This exercise was carried out with the help of the educational committee of the Midwifery Association of Pakistan. Eligible participants were approached by telephone. Follow-up was carried out after 3 months, also via telephone, in August 2016 to assess whether they had initiated working as a mentor.
Data collection tool
Mentoring Knowledge Scale
A Mentoring Knowledge Scale was developed, based on the literature review and the key concepts in the mentoring training. It consisted of a total of 14 multiple choice questions that measured participants' knowledge regarding mentorship. The questions included in the knowledge scale were related to the significance of mentoring, communication skills, problem solving skills, and establishing boundaries in mentoring. Higher total scores reflect greater knowledge regarding mentorship. An a coefficient of 0.92 was identified for the mentoring knowledge scale. A coefficient of 0.7 is considered acceptable for a newly developed tool (Burns and Grove, 2005). The mean difficulty index was 0.61 and the discrimination index was 0.29.
Mentoring perception scale
The ‘Expected Costs and Benefits of being a Mentor’ scale, created by Ragins and Scandura (1999), consisted of 41 items that measured an individual's perceptions about the costs and benefits of mentorship. The items related to the benefits are measured on a five-point Likert-type scale, ranging from ‘strongly disagree, to ‘strongly agree’. The items related to cost are reverse-scored items. Higher scores reflect more positive perceptions about mentorship. For the costs and benefits scale, a coefficient of 0.83 and 0.89 was noted, respectively. This is comparable to previous studies, such as those by Rohatinsky (unpublished).
Willingness to Mentor Scale
The Willingness to Mentor scale (Ragins and Scandura, 1999) is an eight-item scale that measures participants' readiness for mentoring. The items were measured on a five-point Likert-type scale ranging from ‘strongly disagree’ to ‘strongly agree’. Higher scores reflect greater willingness for mentorship. The coefficient for the tool was 0.82.
Data analysis
Data was entered onto a Microsoft Excel spreadsheet and was tested on SPSS version 19. All tests were two-tailed with a significance level of P<0.05. The demographic variables were analysed through descriptive statistics: mean and standard deviation were used for continuous variables and frequency and proportion were used for categorical variables. For comparison of the pre-test and post-test scores, a paired, t-test was used. Effect size for knowledge was also calculated. Moreover, as the measure of training effectiveness, the participant's average normalised gain (g) was calculated along with percent absolute gain and percent relative gain.
Results
The demographic characteristics of the community midwives are shown in Table 1.
| Categories | Classification | n | % | Mean (±SD) |
|---|---|---|---|---|
| Age (years) | 20–30 | 35 | 70 | |
| 31–40 | 12 | 24 | ||
| 41–50 | 3 | 6 | ||
| 50 or more | 0 | 0 | ||
| Mother tongue | Urdu | 6 | 12 | |
| Sindhi | 35 | 70 | ||
| Punjabi | 4 | 8 | ||
| Pushto | 5 | 10 | ||
| Balochi | 0 | 0 | ||
| Others | 0 | 0 | ||
| Experience as a community midwife | Birthing centre | 13 | 26 | |
| Hospital | 13 | 26 | ||
| Both | 24 | 48 | ||
| Others | 0 | 0 | ||
| Years of work experience as a community midwife | 3 | 25 | 50 | 4.0 Years (±1.27) |
| 4 | 10 | 20 | ||
| 5 | 7 | 14 | ||
| 6 | 5 | 10 | ||
| 7 | 3 | 6 | ||
| Marital status | Single | 27 | 54 | |
| Married | 21 | 42 | ||
| Widowed | 2 | 4 | ||
| Divorced | 0 | 0 | ||
| Separated | 0 | 0 | ||
| Others | 0 | 0 | ||
| Awareness of the concept of mentorship | Yes | 7 | 14 | |
| No | 43 | 86 | ||
| Attended mentorship training | Yes | 0 | 0 | |
| No | 50 | 100 | ||
| Previous experience of mentorship | Yes | 2 | 4 | |
| No | 48 | 96 | ||
| Role | Mentor | 0 | 0 | |
| Mentee | 2 | 4 | ||
| Other | 0 | 0 | ||
| Type of mentoring | Formal | 0 | 0 | |
| Informal | 2 | 4 |
Paired comparison of knowledge scores
The mean test scores of the mentorship knowledge improved from 45% (6.3 ± 2.38) at baseline to 62% (8.6 ± 2.23) post-intervention. The knowledge questionnaire showed significant increase in the mean scores (P=<0.001). The knowledge score equated to a high effect size of 1.03, which was calculated based on the parameter given by Becker (2000). Absolute gain of knowledge among participants was 17% and relative gain was 37%. Also, the participants' average normalised gain was 31% (Table 2).
| Knowledge | Mean (±SD) | P-value | Effect size | Absolute gain | Relative gain | Average normalised gain (g) |
|---|---|---|---|---|---|---|
| Pre-test | 6.3 (±2.38) | <0.001* | 1.03 | 17% | 37% | 31%a |
| Post-test | 8.6 (±2.23) |
Paired comparison of perceptions of mentoring
The average pre- and post-scores for perceptions about cost (P<0.001) were significant, since the P-value was less than 0.05 at 95% confidence level (-10.88, -5.52). The average pre- and post-scores for benefit scale in mentoring perception (P=0.061) were not found to be significant before and after the training since the value was not less than 0.05 at 95% confidence level (-6.96, -0.20) (Tables 3 and 4).
| Cost | Mean (±SD) | 95% CI of mean | P-value | |
|---|---|---|---|---|
| Upper | Lower | |||
| Pre-test | 48.86 (±8.62) | -10.88 | -5.52 | <0.001* |
| Post-test | 57.06 (±7.61) | |||
| Benefit | Mean (±SD) | 95% CI of mean | P-value | |
|---|---|---|---|---|
| Upper | Lower | |||
| Pre-test | 71.26 (±11.92) | -6.96 | -0.20 | 0.061 |
| Post-test | 74.84 (±7.95) | |||
Paired comparison of willingness to mentor
The average pre- and post-scores for willingness to mentor (P<0.001) were significant in the pre- and post-test, since the P-value was less than 0.05 at 95% confidence level (-8.73, -5.39) (Table 5).
| Willingness | Mean (±SD) | 95% CI of mean | P-value | |
|---|---|---|---|---|
| Upper | Lower | |||
| Pre-test | 27.36 (±5.84) | -8.73 | -5.39 | <0.001* |
| Post-test | 34.42 (±3.82) | |||
Follow-up
Follow-up telephone calls were conducted with participants 3 months after the initial study. When the participants were contacted through telephone calls for the follow-up, 43 out of the 50 participants responded. Among these 43 community midwives, 31 (72%) participants reported that they had begun to work as a mentor in their respective fields (Table 6). The reported reasons for not initiating the mentoring process were: busy work schedules, attending other training sessions, no one seeking help, and personal commitments. However, all those who had not started to work as a mentor reported their intention to initiate mentorship in the near future. During the telephone conversation, most of the participants reported that they appreciated the mentoring process.
| Categories | Classification | n | % |
|---|---|---|---|
| How many participants responded to the call? | Yes | 43 | 86 |
| No | 7 | 14 | |
| Have you initiated working as a mentor? | Yes | 31 | 72 |
| No | 12 | 28 |
Discussion
The findings of the present study indicated that the mentorship training had a positive impact on the knowledge of the participants. These findings affirmed the findings from previous studies conducted by Komaratat and Oumtanee (2009), Wallen et al (2010), and Pfund et al (2014), who demonstrated positive outcomes of mentorship training on the knowledge of the participants with training sessions similar to those in the present study. In contrast, a study by Lau (2016) showed no significant improvement in the participants' knowledge after a mentorship workshop; however, the participants indicated that after attending the workshop, they did gain awareness and insight about the mentoring relationship. This discrepancy could be due to the difference in the duration of the workshop and the particular tool used to assess the knowledge of the participants. The duration of the workshop was 1 half day (3 hours) compared to the 2 days (8 hours) in the present study, and the mentorship knowledge scale consisted of true or false questions, rather than multiple choice questions in the present study. There was a significant and meaningful change in overall knowledge from a single mentoring workshop, which was shown by the large effect size in overall knowledge scores and which is consistent with other literature (Lau et al, 2016).
Results also showed that mentorship training had a positive impact on the perceptions of the community midwives. The findings of this study build on the previous findings (Lee et al, 2009; Wallen et al, 2010) and show that perceptions, particularly about the cost versus benefit of mentoring, significantly improved from the pre- to post-test. However, the findings did not show any significant improvement regarding the benefits of mentoring. This may be due to the already positive perceptions of community midwives about mentoring, which is in line with the study conducted by Lee et al (2009).
The results of the study indicated that participating in a 2-day mentoring workshop had positive effects on the midwives' willingness to mentor. These findings concur with the studies by Ann et al (1999), Waters et al (2003), Finnerty et al (2006), McDonald et al (2010), and Wallen et al (2010). In contrast, the study by Jones (2004) found that mentoring was seen as an added responsibility in the midwives' role and, therefore, many midwives did not show willingness to mentor. This may be due to a lack of mentorship training, leading to confidence problem (Jones, 2004). Surprisingly, little research is available to further understand the relationship of training programmes on willingness to mentor.
This study also identified that the majority of the participants initiated work as a mentor in their respective community fields. These findings are congruent with the study conducted by Komaratat and Oumtanee (2009) and Turkmani et al (2014). The reasons for not beginning to work as a mentor were cited as busy work schedule, attending other training sessions, and personal commitments. Similar types of challenges were reported by midwives in the study conducted by Ali et al (2015). However, long-term longitudinal follow-up evaluations should be conducted to determine whether changes are maintained over time.
Recommendations
In this study, the normalised learning gain was more than 30%, which shows that the educational intervention was effective. According to Thomson et al (2001), the pre-defined target of normalised learning gain was 30%, which represents the value on which the educational intervention is considered effective. Based on the findings, the authors recommend that continuous educational trainings should be planned to reinforce the importance of mentoring in a midwifery setting. The study also suggests that the significance of the concept of mentoring needs to be embedded in the midwifery curriculum in order to acquaint midwifery students with the mentorship concept from the beginning. The authors also propose the need to have governmental support for the sustainability of the mentorship programme; a need that was identified from participants' feedback.
Limitations
The study design may have limitations that had an impact on to internal validity, such as maturation, history, and testing. The short duration of the training, however, reduced the impact associated with history and maturation. The testing effect could be resolved by including a control group that did not go through the training; however, the inclusion of a control group in this study was in many ways impractical. There was no ‘usual’ or ‘standardised’ training following the mentoring programme; therefore it would have been difficult to maintain the participants' motivation during the study. Moreover, due to time constraints, the study did not examine the long-term impact of the mentorship training.
Conclusion
The study has provided evidence that mentoring training was successful in improving the baseline knowledge, perceptions, and willingness of community midwives to mentor. Overall, the training was well received by the community midwives and this helped them initiate work as mentors in their respective areas. Administrative support by the government is recommended for the smooth running of mentoring workshop in the present environment and perhaps a programme for on-going support for mentors and mentees into the future. It is also recommended that a larger study need to be conducted in implementing a whole mentorship programme and that the education about mentoring needs to be offered on an on-going basis.
‘The significance of the concept of mentoring needs to be embedded in the midwifery curriculum in order to acquaint midwifery students with the mentorship concept from the beginning’