- Top of page
- CONFLICT OF INTEREST
Despite progress on Millennium Development Goals 4 and 5, maternal and newborn mortality continues to be a serious problem worldwide, particularly in low- and middle-income countries where 99% of these deaths occur. In Ethiopia, a country of nearly 85.9 million where the great majority of people 84% live in rural areas, maternal and newborn mortality remains a serious problem. In 2011, the maternal mortality ratio was estimated at 676 deaths per 100,000 live births; neonatal mortality was estimated at 37 deaths per 1000 live births.
Most experts agree that timely provision of a package of evidence-based practices can reduce maternal and newborn deaths. Delays in the receipt of this package have been associated with decision making by women and families, geographic and economic barriers, and perceived or actual poor quality of facility-based care.
Efforts to reduce delays in care provision have taken many forms. One prevalent strategy has been to extend the care supply chain by training not only the professional health care providers at peripheral health facilities but also by task-shifting through the training and deployment of more proximal frontline health workers, thereby better integrating health workers into the overall care system.[5, 6] Frontline health workers include volunteer community health workers, traditional birth attendants (TBAs), as well as a variety of other paid paraprofessionals (eg, health extension workers and lady health workers) who attend women during pregnancy or during the birth and/or provide health education. It has been shown that the extension of maternal and newborn health care by trained health workers yields improved care provision and self-care behaviors and often contributes to better birth outcomes[7-12] and reduced neonatal mortality.[13, 14] While most studies have shown positive outcomes, some have not. One study, for example, found that after the introduction of Essential Newborn Care training to community-based birth attendants, the rate of newborn death did not decrease in the week following implementation, although the stillbirth rate did decline.
Training has been the primary approach to prepare health workers to provide appropriate and timely maternal and newborn health care. Training efforts have focused on basic specific skills, such as recognizing postpartum hemorrhage and providing misoprostol (Cytotec), recognizing sick newborns and the need for referral, and home-based essential neonatal care. The American College of Nurse-Midwives’ Home-Based Life Saving Skills (HBLSS) program offers a comprehensive program of maternal and newborn health care training. The HBLSS program uses a cascade training approach, similar to other programs in which training extends across multiple levels of trainees and the numbers of people trained increases with each level (eg, health consultants, master trainers, and trainers). After receiving training, one level is responsible for training the next level down with support from higher levels, as needed. These programs differ in the extent to which they emphasize didactic or participatory methods such as hands-on training and practice and follow-up supervision after the initial training.
Once embedded in a system of use (the provision of community-based training and care provision) and ongoing coaching and observation, health workers retain maternal and newborn health skills.
Regional variations in retained skills are observed, reflecting regional differences in policy interpretation and implementation in the distribution and use of misoprostol (Cytotec).
Intercadre variations in retained skills are also observed, largely reflecting the extent to which members of each cadre are directly involved in the actual provision of maternal and newborn health care, especially during the critical birth-to-48-hours period.
Health worker training programs have used a number of strategies to evaluate and demonstrate change in health workers’ skills and knowledge. Evaluation efforts have entailed (solely or in combination) pre- and posttraining knowledge tests, observation of skills by trainers, objective simulated clinical examinations, health worker self-reports, and interviews with end users (community members who are the intended recipients of the care/education). Some have also conducted evaluations a while after the posttraining to assess retention of knowledge and skills over time.[8, 20] In all cases, the evaluations have demonstrated that training significantly enhances health worker skills and knowledge. Moreover, follow-up evaluations have indicated that, although scores may have slipped from immediate posttraining, health worker skills and knowledge remained substantially above baseline over time.[21, 22]
A major intervention component of the Maternal and Newborn Health in Ethiopia Partnership (MaNHEP) has entailed the provision of maternal and newborn health training to health workers, including health extension workers, community health development agents, and TBAs. Health extension workers are young women with 10 years of primary and secondary school education and one year of certificate-level health training that consists of 16 packages. Health extension workers provide select reproductive health services including antenatal and postnatal care, family planning, and birth attendance, although their level of hands-on experience with birth may be limited. Community health development agents have minimal, yet variable, levels of maternal and newborn health training. Community health development agents are members of the community who are chosen by the community or organization to work as volunteers in health-related activities such as community mobilization and health education. Traditional birth attendants do not have formal training, but may have learned skills passed on from older generations. They have long been trusted to provide labor and birth care to women in their communities, especially in rural areas of Ethiopia.
MaNHEP maternal and newborn health training—and the Community Maternal and Newborn Health (CMNH) Family Meetings that the health workers ultimately conducted with women and their families—are based on the HBLSS program. This program emphasizes antenatal care and birth preparedness practices, illness recognition, and referral, where possible. It also emphasizes the elimination of potentially harmful practices, such as prelacteal feeding, putting oil or butter on the cord stump, and bathing the newborn before 24 hours. CMNH training uses pictorial presentation, storytelling, discussion, experience sharing, and demonstrations followed by practice using pictorial checklists called Take Action Cards. Training seeks to create a safe and welcoming learning environment where everyone is respected for their unique contributions (eg, trainers sit on the floor with participants; sitting in a circle; use of the local language to encourage participation).
Health workers were trained in a cascade that extended across 4 levels of trainees in order to reach women and their family caregivers in the community. The training cascade (Figure 1) began with maternal and newborn health consultants training federal and regional staff (master training of trainers), followed by 3 levels of training of trainers (supervisors and coordinators, health extension workers, and guide teams). In total, 91 health extension workers and 638 guide team members were trained. Pre- and posttraining evaluation procedures and performance test results have been previously described.
Guide teams are composed of community health development agents, TBAs, wise women, and wise men who are consulted for problems related to mothers and newborns. Guide team members are selected because of their interest, experience, or current roles and responsibilities in the community, and their ability to teach and facilitate discussion. Guide team members commit to spend at least 4 to 6 hours weekly conducting CMNH meetings with pregnant women and their families. This work involves traveling within 3 to 4 villages.
CMNH training materials include a demonstration kit (newborn, placenta, uterus, and breast models; razor blade; 5% chlorine; cord tie; soap; waterproof placenta container; apron; cloths; and a cup, bowl, and spoon); a Take Action Card booklet; and a master list, guide team record, and trainers’ manual.
This article examined the degree to which health worker skills and knowledge were retained 18 months after initial training (January 2011-June 2012) and identified sociodemographic factors associated with 18-month skills assessment performance.
- Top of page
- CONFLICT OF INTEREST
Our results revealed that the majority of health workers retained the knowledge gained in maternal and newborn health training. In 18-month posttesting, substantial knowledge retention was noted in 12 of 14 care steps and 13 of 14 care steps of Prevent Problems Before Baby Is Born among health extension workers in the Amhara and Oromiya regions, respectively. Moreover, knowledge retention was also high for Prevent Problems After Baby Is Born among health extension workers in the Amhara (16 of 16 care steps) and Oromiya (14 of 16 care steps) regions. These findings complement prior work in other contexts such as Bangladesh, India, Liberia, and southern Ethiopia where the HBLSS training approach has been successful in transferring knowledge[21, 22, 27, 28] and behavior change among health workers, women, and family caregivers.
The results also highlighted areas for improvement across regions and cadres. Most notably, health workers scored lower on steps that used verbal cues. Select action steps, for example, were accomplished by reminding the family care team about what not to do during and after birth (eg, do not give oxytocin) rather than using a physical action to show the family what they should do (eg, rub the womb). These steps can be particularly difficult to remember to include in the CMNH meeting because they do not involve a physical action. This finding supports results from prior programmatic research that nonaction steps are the most challenging steps for health workers to remember in posttraining and 18-month posttests. Development of novel strategies to improve knowledge transfer and retention of these low-scoring areas are important points for future intervention.
The findings demonstrate that regardless of sociodemographic characteristics, maternal and newborn health service experience, and involvement in voluntary activities, all types of cadres (health extension worker, TBA, and community health development agent) have the ability to learn and practice lifesaving maternal and newborn health skills. The 3 health worker cadres varied significantly by key sociodemographic characteristics. For example, health workers had varied levels of experience, and significant differences were noted by age and marital status across the 3 health worker cadres. Furthermore, more than 90% of health extension workers and TBAs reported attending at least one birth since the posttraining posttest, and the majority of community health development agents and TBAs reported engagement in voluntary activities.
Regional variations were also observed related to birth attendance and referrals. For example, health extension workers and TBAs in Amhara attended a significantly higher number of births compared to health extension workers and TBAs in Oromiya. Moreover, TBAs in the Amhara region referred a significantly higher number of cases than did TBAs in the Oromiya region. These results complement findings that significantly more women in Amhara received birth care from a skilled provider or health extension worker compared to women in Oromiya.
Regional and cadre differences were also observed in the retention of knowledge and skills for the prevention of problems before and after birth. For example, the average 18-month posttest score for misoprostol storage at home was significantly lower among all cadres in the Amhara as compared to cadres in Oromiya. This finding may be accounted for by regional differences in misoprostol distribution and utilization. Specifically, Oromiya permits health extension workers, as well as well-trained TBAs and community health development agents, to keep and provide misoprostol. Amhara permits only health extension workers to keep and provide it.
Overall, the demonstrated knowledge and skills of the care steps was less consistent among community health development agents and TBAs in Amhara compared to community health development agents and TBAs in Oromiya. This finding is surprising given the level of engagement in CMNH family meetings among community health development agents and TBAs in the Amhara region (93% and 82%, respectively) compared to community health development agents and TBAs in the Oromiya region (43% and 49%, respectively). Far fewer community health development agents and TBAs in Amhara, however, were members of MaNHEP quality improvement teams than were community health development agents and TBAs in Oromiya (Community Health Development Agent 28% vs 46%, TBA 10% vs 51%, respectively). Quality improvement teams composed of community leaders and health workers meet regularly to discuss practical community-based solutions to maternal and newborn health problems. It is possible that engagement in these teams reinforces CMNH content. Future work is needed to better discern the possible reasons for these conflicting patterns (eg, disparities in level of support following training, differences in performance scoring practices between regions).
Regression analyses demonstrated that having more years of education is associated with significantly higher 18-month posttest scores for Prevent Problems Before Baby Is Born. Other variables such as age, cadre type, marital status, birth experience, and number of referrals had no effect on 18-month posttest scores. This finding suggests that, regardless of prior maternal and newborn health service experience and sociodemographic characteristics, community-based health workers from all cadres are able to learn and provide maternal and newborn health services.
For the topic of Prevent Problems After Baby Is Born, we noted that health workers who attended more births and who were older scored lower compared to younger health workers and those who attended fewer births. While these findings were surprising, they may simply reflect the influence of Traditional birth attendants in the analysis. Traditional birth attendents are older, on average, than other cadres and report the highest overall number of attended births, yet TBAs scored as low or lower on the topic than did community health development agents and health extension workers.
The maternal and newborn health training program has important implications for countries such as Ethiopia where the majority of the population lives in rural areas and has low literacy. The training program emphasizes building the capacity of health workers, who are closest to pregnant women and their families, and demonstrates an effective way to deliver maternal and newborn health education and services at the community level. In addition, the CMNH program that the health workers provide utilizes pictorial presentation, storytelling, discussion, experience sharing and demonstrations, thereby eliminating the barrier that low literacy can at times pose. Furthermore, the training cascade approach enables health workers to continuously practice the steps involved in maternal and newborn health care and increase their confidence. The training also provides an opportunity to build team spirit among diverse health worker groups who may not otherwise interact.
Some limitations are inherent to this type of performance assessment study. First, it is possible that different maternal and newborn health specialists scored the posttraining and 18-month posttests for individual trainees, making a comparability of scores less reliable. We minimized this potential bias by providing training to maternal and newborn health specialists, using co-scoring with consultants until scoring was consistent, and by using a similar checklist at pretesting before training, posttraining posttesting, and 18-month posttesting. Another limitation is that 3 care steps in Prevention Problems Before Baby Is Born and 5 steps in Prevention Problems After Baby Is Born involved more than a single action. For example, “wipe baby's face and dry baby as soon as baby is born” involves 2 separate steps, although it was worth only one point in testing. Trainees were only given credit for the step if they completed all substeps, which may have resulted in lower performance scores overall. In the future, care steps with more than one action should be divided into multiple steps so that respondents can receive credit for the skills they have retained.