Developing and implementing a model of equitable distribution of mentorship in districts with spatial inequities and maldistribution of human resources for maternal and newborn care in Rwanda

Background: The shortage of health care providers (HCPs) and inequity in their distribution along with the lack of sufficient and equal professional development opportunities in low-income countries contribute to the high mortality and morbidity of women and newborns. Strengthening skills and building the capacity of all HCPs involved in Maternal and Newborn Health (MNH) is essential to ensuring that mothers and newborns receive the required care in the period around birth. The Training, Support, and Access Model (TSAM) project identified onsite mentorship at primary care Health Centers (HCs) as an approach that could help reduce mortality and morbidity through capacity building of HCPs in Rwanda. This paper presents the results and lessons learnt through the design and implementation of a mentorship model and highlights some implications for future research. Methods: The design phase started with an assessment of the status of training in HCs to inform the selection of Hospital-Based Mentors (HBMs). These HBMs took different courses to become mentors. A clear process was established for engaging all stakeholders and to ensure ownership of the model. Then the HBMs conducted monthly visits to all 68 TSAM assigned HCs for 18 months and were extended later in 43 HCs of South. Upon completion of 6 visits, mentees were requested to assist their peers who are not participating in the mentoring programme through a process of peer mentoring to ensure sustainability after the project ends. Results: The onsite mentorship in HCs by the HBMs led to equal training of HCPs across all HCs regardless of the location of the HC. Research on this mentorship showed that the training improved the knowledge and self-efficacy of HCPs in managing postpartum haemorrhage (PPH) and newborn resuscitation. The lessons learned include that well trained midwives can conduct successful mentorships at lower levels in the healthcare system. The key challenge was the inconsistency of mentees due to a shortage of HCPs at the HC level. Conclusions: The initiation of onsite mentorship in HCs by HBMs with the support of the district health leaders resulted in consistent and equal mentoring at all HCs including those located in remote areas.

This paper aims to describe the development, implementation as well as results of the mentorship model for HCPs providing maternal and neonatal care in 68 health centers of the 3 districts of the Northern Province of Rwanda. The indicators that were assessed for the results included the distribution and consistency of mentoring to the catchment areas of all of the DHs included in this project. The data on the initial assessment of the study locations were also analysed. The 68 HCs involved in this study were located in Rulindo, Gicumbi and Gakenke districts in the Northern Province of Rwanda. The onsite mentorship programme was implemented by TSAM project and its stakeholders in the HCs in the three districts. These 3 districts were assigned to the TSAM project as per the Memorandum of Understanding between the project and the Ministry of Health. This research describes the mentorship programme. The design phase of this programme started with an initial assessment of the status of training in HCs in these districts. The results of this assessment informed the selection of midwives to serve as Hospital-Based Mentors (HBMs) for health care providers at the HCs. The research further describes the other steps of the design phase, namely the refresher courses on Emergency, Obstetric Neonatal Care (EmONC) for selected Hospital Based Mentors and training on the mentoring approach to be adopted. These HBMs also benefited from other courses with cross-cutting themes such as ethics, inter-professional collaboration, gender, maternal mental health and Gender-Based Violence (GBV). The initial training makes this mentorship programme unique, as the additional training was designed to allow HBMs to manage the mothers and newborns in an integrated manner. In addition, HBMs were drawn from the mentees who had followed the mentoring programme at the respective hospitals under the same project framework. Upgrading of some mentees to HBMs through cascades of special training presents the second unique aspect of our mentorship programme. The fact that mentors were drawn from a cadre of DH mentees allowed them to conduct mentorship more effectively. The design phase involved information meetings to establish a clear process for engaging all stakeholders and to ensure ownership of the model. Following the initial assessment and training, 23 HBMs conducted monthly visits to the 68 HCs located in catchment areas of five TSAM assigned hospitals in North for 18 months, from October 2018 to March 2020. This model was extended later in other HCs located in the catchment area of 5 hospitals in the Southern Province. Participants to the mentoring programme described in this manuscript were nurses or midwives providing maternal and/or neonatal care in 68 HCs of 3 concerned districts in the Northern province. Data on the number of mentees who attended the mentorship programme during its lifetime as well as the data on the initial assessment were analysed for this study. The data were entered in a database designed for the project using Excel. Analysis was done in Microsoft Excel to generate tables and descriptive statistics. Locations of HCs were collected using Global Positioning System (GPS) devices and the coordinates were used to generate maps in ArcGIS 10.7. The geographic data were analysed to identify spatial disparity of midwives using data on the number of midwives and population density for each hospital catchment area. Population data were obtained from the Health Information System of the hospital. In addition, data were analysed according to the number of nurses and midwives in the health centers that had received training on EmONC. Mentoring data was collected on the number of mentoring visits in the DH catchment areas segregated by gender for the nurse and midwife mentees. Also, data on the number of mentoring visits were collected based on the professional qualification of the midwives and nurses. Prior to designing the mentoring programme at the HC level, a rapid assessment in health facilities located in TSAM-assigned hospitals was conducted by researchers working on the project. The assessment aimed to determine the availability of staff who provide MNH care and the status of training on MNH for those staff. The results of the assessment allowed the project and its stakeholders not only to know the number of mentees that would be available but also it informed the initial selection of competent hospital-based mentors (HBMs). This assessment was conducted cognizant of the fact that over the past 2 decades, HCPs have received some off-site training related to maternal and newborn care [14, 19, 21]. The findings of the assessment are presented below: The results revealed that efforts are being made to equip maternity departments of HCs with midwives. Of the 68 HCs located in the catchment area of TSAM-assigned hospitals, 43 (63 %) had at least one midwife by the time of the assessment. However, as seen in Fig. 1 below, the density of midwives per 10000 population varied widely both within and across the districts as well as across the hospital catchment areas. HCs that were easily accessible had more midwives than those in more remote areas. For example, it is easy to realize that the density of midwives in Gicumbi district is higher compared to Gakenke district while even in Gakenke district itself, the density of midwives is higher in the catchment area of Nemba hospital than that of Ruli hospital. This disparity points to the need for the development of the TSAM mentorship programme to build the capacity of HCPs providing MNH in all HCs. Spatial distribution of the midwives per HCs in Rulindo, Gicumbi and Gakenke districts. This Figure shows the spatial distribution of the number of midwives per HCs in Rulindo, Gicumbi and Gakenke districts. Produced using ESRI 2019. ArcGIS Desktop: Release 10.7.1. Redlands, CA: Environmental Systems Apart from the availability of midwives in the HCs, the assessment revealed that there is an inequity in terms of densities of health care providers who benefited from the training on the EmONC across hospital catchment areas and districts. As shown on Fig. 2 below, the density of health care providers who received training on EmONC per 10000 population is higher in Gicumbi district and Gakenke district than in Rulindo district. However, even in the Gakenke district, the density is higher in the catchment area of Nemba hospital than in Ruli hospital catchment area. Likewise, even in Rulindo district which has a poor density, there is a disparity in the catchment area between Kinihira and Rutongo, further highlighting the need for the TSAM programme to implement a mentorship programme that focuses on EmONC to reach all HCs to provide consistent mentoring despite the disparities. Spatial distribution of midwives trained on EmONC per HCs in Rulindo, Gicumbi and Gakenke districts. This Figure represents the spatial distribution of midwives who benefited from the training on EmONC in Rulindo, Gicumbi and Gakenke districts. Produced using ESRI 2019. ArcGIS Desktop: Release 10.7.1. Redlands, CA: Environmental Systems The initial assessment also allowed the project and its stakeholders to examine the spatial distribution of HCPs with additional training on Essential Newborn Care (ENC). As with the EmONC, there were geographical disparities with HCPs trained in ENC which is thought to be helpful for staff providing care to mothers and newborns (Fig. 3). Spatial distribution of HCP trained on ENC per HC in Rulindo, Gicumbi and Gakenke districts. This Figure shows the spatial distribution of health care providers who benefited from the training on ENC in Rulindo, Gicumbi and Gakenke districts. Produced using ESRI 2019. ArcGIS Desktop: Release 10.7.1. Redlands, CA: Environmental Systems Mentorship is a flexible teaching and learning process that serves specific objectives of the HCPs and health care services [13–15, 21]. Given the disparities in maternal health services [22] and previous training programs that were provided to midwives, the TSAM project targeted staff providing care to women and newborns. This approach can reach many people at the same time without disturbing the routine work of service providers. More importantly, some maternity departments of HCs are staffed with midwives who have only recently graduated. Apart from the fact that they have not received any additional training to strengthen their capacity in the area of EmONC, they have limited opportunities for CPD beyond pre-service training and have limited access to experienced clinicians from whom they could learn key skills. For these HCPs, pre-service training and refresher courses alone may not translate into improved maternal health service delivery. Hence, the TSAM mentorship programme aimed to provide the bridge between traditional didactic training to hands-on practical training approaches. This was based on the underlying assumption that skills acquired during pre-service training are usually lost in the absence of CPD. Through TSAM’s mentorship approach, key essential MNH skills were imparted to enable nurses and midwives to effectively perform tasks that they did not feel confident in performing, either due to lack of knowledge, hands-on skills or both. The following section describes the development, implementation, and preliminary outputs of an onsite mentorship programme in the TSAM-assigned HCs in Rwanda. The mentorship model was implemented by HBMs who were midwives practicing in the DHs responsible for oversight of the HCs. The majority of these midwives had been mentored in a separate programme for HCPs at the DH level by national mentors with the support of the same project [22]. Potential mentors were selected with that input from the administration of DHs and the selection considered the following key criteria as per the mentorship guidelines of the Ministry of Health [23]. The mentorship model development consisted of several phases. The first was providing a refresher course on Emergency Obstetric and Newborn Care (EmONC) for 25 potential mentors. The candidate HBMs was proportional to the number of HCs within the catchment area of each hospital. The second phase consisted of providing training in mentoring and Cross-Cutting Themes (CCTs) including Gender, Ethics and Inter-Professional Collaboration to successful candidates to EmONC refresher training. Thirdly, there were induction meetings held for each DH to introduce the programme and ensure it is owned by beneficiaries. The final phases were the implementation of the onsite mentorship visits followed by monitoring and evaluation activities. Once again, it is worthy to mention that HBMs were drawn from former mentees. After a successful design phase, each HBM was assigned to between 2 and 3 HCs. The mentorship field visits were organized for all 68 HCs in the Northern Province. One-day monthly visits were conducted by HBMs from October 2018 to March 2020 (18 months). During the mentoring period, key services areas mentored include the labor ward, post-natal care and antenatal care services. Activities carried out by HBMs include management of cases with the mentees, bedside teaching, presentation on key selected topics in morning staff meetings based on the needs, and training of mentees using simulation. Logbooks were used to track the participation of mentees in different mentorship activities. The topics covered by HBMs were the components of Essential Newborn Care (ENC) and those of EmONC. HBMs had benefited from refresher courses on these topics to be covered. During the mentorship period, different activities were completed by mentors and mentees including ward round, assisted delivery, newborn resuscitation, and post natal care. In addition, mannequins were used to teach skills for different components. Upon completion of each of the visits, reports were written and submitted to the director of nursing at each hospital for compiling the report at the hospital level. The overall coordination of the mentoring programme was done by a CPD programme manager within the TSAM project. The formats of the reports by the mentors were developed by the TSAM project CPD team and presented during both the training on mentoring and induction meetings. The completed reports were then submitted to the CPD manager for the TSAM project to be compiled and analyzed and a report prepared. To ensure that the mentorship model was implemented as designed and experiences were shared, bi-annual evaluation meetings were organized for each hospital. These meetings brought together the same participants as those who attended the induction meetings before initiating the mentorship in HCs. The monitoring meetings aimed to share the key messages from the report of the mentorship visits, discuss the successes and challenges of the mentorship visits as well as develop the strategies to overcome the challenges. The key points that emerged from the coordination meetings are summarized below:

N/A

• Dima Health