World-wide, each year, some 60 million births occur at home, and the pregnancy outcomes appear considerably worse than births that occur in a medical facility . Much is still unknown about the attendants at these home births, particularly in regards to their training, delivery practices, access to medical equipment and testing, and their interaction with the formal health care system [4–9]. Prior investigations have generally been small pilot studies, conducted in limited areas [7–9]. Because these HBAs appear to play such a crucial role in pregnancy outcomes in many developing countries, we felt it important to obtain more detailed information about HBA knowledge, training, and delivery practices. The Global Network for Women’s and Children’s Health Research is a multi-site international consortium of research programs and birth registries which operates in regions of the world where maternal-newborn mortality rates are the highest. Thus, its members were well suited to carry-out this study.
This survey was conducted in geographic areas in which the NICHD Global Network has been conducting research for about 10 years [10–12]. This history suggests that in all likelihood, the results of the HBAs in this study demonstrated a higher level of baseline training and practices than if the survey had been conducted in adjacent, non-Global Network areas where HBA training and research studies had not occurred. For example, one clear difference noted between the DRC with almost no clean birth kit use and other participating sites with active use of clean birthing kits was likely explained by the fact that, because there were no current, ongoing Global Network studies in the area of the DRC where the HBA survey was conducted, clean birthing kits were not generally available. The high use of uterotonics in the Belgaum site was likely explained by a recent misoprostol research study conducted at that site.
The Indian ANMs, who conduct some home births, are different from the traditional HBAs in many ways. They generally have had a number of years of elementary and high school education before entering a Ministry of Health training program lasting a minimum of 18 months . The ultimate goal is for them to perform deliveries in clinics throughout India. Therefore, even though they perform home births, we elected to summarize their data separately from the more traditional home birth attendants.
Focusing on the traditional HBAs, a number of conclusions can be drawn. The HBAs are generally older women with little schooling and are generally poor. Most do not have an indoor toilet, only 54% have electricity, most lack a gas or electric stove for cooking and less than half have access to a cell phone or some sort of transportation. Only about 30% can read or write and many cannot read numbers, tell time, or use a calendar. Very few have had more than a month of professional training. Compared to the required level of training for birth attendants in most developed countries, which generally ranges from three to more than 10 years post high school, this is a very low level of training.
In nearly all sites, the HBAs performed 1 to 4 deliveries per month. This relatively low range is likely explained in part by the often wide dispersion of communities geographically, and the lack of transportation available to many home birth attendants, restricting their potential clients to those within walking distance. This low number of deliveries means that most HBAs will experience a life threatening emergency only occasionally. The skills necessary to manage those emergencies, even if previously taught and mastered, often are forgotten because of disuse . For example, it is estimated that only 3% of newborns require bag and mask resuscitation . For a HBA delivering 40 babies a year, that skill will need to be applied perhaps once a year, and when taught, will likely soon be forgotten. In our experience from a previous study, many HBAs who had been provided a bag and mask and taught how to use it, no longer had this equipment in their possession a year after the study was over. (Personal observation RLG, EMM, CB) .
With few exceptions, HBAs had limited access to medical testing and equipment, and additionally, limited training to carry these out adequately. Testing that is standard practice in most high and middle income countries, such as measuring blood pressure or testing for anemia were available to less than half of them. Tests for infectious diseases, with the exception of HIV at some sites, were only rarely available. Few of the HBAs had a stethoscope or could repair a vaginal laceration. There were also practices employed, such as shaking or spanking the baby, that are generally viewed as ineffective and potentially harmful to newborn health. In the face of hemorrhage, less than half the HBAs reported the use of uterine massage, an important birth management practice that should be within the scope of home birth attendants’ delivery skills . There are, however, some generally recommended practices such as drying the baby, clearing mucus from the baby’s mouth, and teaching exclusive breast feeding that are routinely used by the home birth attendants.
Most HBAs claimed to refer women with medical problems, particularly hemorrhage. Nevertheless, in several of the sites, nearly half of the home birth attendants also reported having never visited their referral hospital. Also the percentage of the women referred by home birth attendants who actually arrived at the hospital is unknown. From other unpublished data collected in these sites (RLG, EMM), we know that the actual rate of referral in these areas from home to any health facility is very low.
This study had several strengths. Our sample included a large number of HBAs from sites in six countries. The questionnaire covered a wide range of variables, including: demographics, practice capabilities, diagnostic testing, equipment and reporting/referral practices. Potential weaknesses include the fact that the study was carried out in areas where a number of research projects involving HBA training have been performed, and thus the results may not be representative of all HBAs in those countries, and may overestimate the HBA knowledge and/or skills compared to other regions. Also, the data presented are derived from self-reported surveys, and did not involve actual observation. Nevertheless, we believe this is one of the most comprehensive multi-country studies to evaluate skills and practices of a large sample of practicing HBAs.
We believe the information derived from this study will be useful to health care providers and policy makers as they try to improve pregnancy outcomes in their regions. It reinforces the arguments of those who believe that moving more births into facilities with better trained birth attendants represents the best chance to reduce the very high maternal and perinatal mortality rates in many low income countries. However, as we await this transition, the survey also highlights some areas where training might improve outcomes for HBA-conducted deliveries—which still comprise most of the births that occur today in many resource-limited settings. For example, teaching uterine massage in the immediate post-partum period might reduce post-partum hemorrhage and lead to a concomitant decrease in maternal mortality, while training not to spank or shake newborns might prevent some neonatal injuries.
Until about 10 years ago, when there was little hope that large numbers of developing country home deliveries could be relocated to hospitals, many efforts were made to upgrade the skills of HBAs. In most studies, there appeared to be little improvement in pregnancy outcomes associated with this training. However, more recently, a study in Tibet found that community health workers could, in fact, be effectively trained to perform uterine massage, as well as in appropriate neonatal resuscitation techniques . Results from a recent Global Network trial suggest that training HBAs in newborn resuscitation might reduce the number of births formerly characterized as stillbirths . Examining the results of the controlled trials of traditional birth attendant training, the authors of a Cochrane Review summarized the limited data available by stating that there is potential to reduce perinatal mortality through HBA training, especially when done in conjunction with building a stronger linkage to the health system [14, 17]. It is clear, however, that these reductions do not typically reach levels of perinatal mortality achievable with high quality hospital care for the mother and infant. Treating many of the conditions that result in the deaths of mothers, fetuses and many newborns (e.g. obstructed labor, placental abruption, preeclampsia/eclampsia, intrapartum asphyxia) usually requires high levels of diagnostic skills, and treatments typically available in hospitals including cesarean section and blood transfusion . These interventions, now collectively known as emergency obstetric and newborn care or EmONC, when made available in hospitals to populations who formerly delivered at home, have been associated with substantial improvements in pregnancy outcomes. The low levels of literacy and formal schooling found in most HBAs in our study, suggest that even with additional training, many HBAs will not be able to acquire the skills to perform high level obstetric and newborn care.
Another approach might be to train skilled birth attendants such as nurse midwives with the intent that they practice in the communities performing home births. If strongly linked to the health system with rapid transport of patients with obstetric complications to a facility able to deal the complications, decreases in maternal and perinatal mortality will likely occur. However, in many low income countries, transportation even across short distances is a problem, and without timely access to a facility for delivery, important decreases in mortality with this approach are not likely to occur.