Advancing Sexual and Reproductive Health and Rights
 
International Perspectives on Sexual and Reproductive Health
Volume 37, Number 1, March 2011
DIGEST

Training Program for Birth Attendants Reduces Neonatal and Perinatal Mortality in Zambian Clinics

A birth attendant training program that has been shown in a multinational trial to reduce stillbirth and perinatal mortality rates among neonates weighing at least 1.5 kilograms—most of them born outside of hospitals—also reduced mortality among infants of all weights born in Zambian clinics, a related study revealed.1 Following implementation of the program, the rates of seven-day neonatal mortality (deaths during the first week of life) and perinatal mortality (seven-day neonatal deaths plus stillbirths) declined by about a third. However, a new analysis from the original multinational trial revealed that the intervention did not yield the same benefits for the small subgroup of very low birth weight infants.2

The major causes of death during the first seven days of life are asphyxia, low birth weight, prematurity and infections. To help prevent such deaths, the World Health Organization created the Essential Newborn Care course, which improves birth attendants' ability to provide evidence-based newborn care. Researchers tested the program, in conjunction with the American Academy of Pediatrics' Neonatal Resuscitation Program, in a "train-the-trainer" trial involving a total of 96 rural communities in Argentina, Democratic Republic of the Congo, Guatemala, India, Pakistan and Zambia.

The two programs were introduced sequentially. In each country, after baseline data on birth outcomes had been collected, the Essential Newborn Care program was implemented and outcome data were collected for 4–9 months. Next, the Neonatal Resuscitation Program was initiated (except in Argentina), and outcome data were collected for 12 months. The year of implementation varied by country; data were collected for the first program in 2005–2007, and the second in 2006–2008.

At each study site, two "master trainers" completed the two programs and were trained in all study procedures. These trainers then trained a community coordinator, who in turn trained the community's birth attendants. The courses included material on cleanliness, neonatal care, initiation of breathing and resuscitation, prevention of hypo-thermia, breast-feeding, infant care and recognition of complications. The birth attendants taught relevant material to the mothers after delivery and collected data on neonatal outcomes at delivery and at seven days. To reduce bias, baseline data were collected after the birth attendants had been trained but before the intervention was implemented. The investigators hypothesized that the intervention would reduce rates of all-cause seven-day neonatal mortality; secondary outcomes included rates of stillbirth, perinatal mortality and death due to birth asphyxia (failure to begin or continue breathing at birth). The study used an intent-to-treat design in which all births in a community were included, regardless of whether a trained attendant was present.

According to a prior analysis, rates of stillbirth declined among infants weighing at least 1.5 kilograms following introduction of the Essential Newborn Care program. Rates of perinatal mortality also declined, although only among infants delivered by an attendant; no change occurred in rates of all-cause neonatal death during the first seven days, and no further benefits were detected after introduction of the Neonatal Resuscitation Program.

In current analysis, the investigators examined outcomes among the 1,096 very low birth weight infants (<1.5 kilograms) in the study. At each stage of the study, most births occurred at home (52–64%); in 10–19% of all births, no attendant, midwife or physician was present. Logistic regression analyses revealed no reduction in rates of stillbirth, perinatal mortality or all-cause seven-day neonatal mortality following introduction of the Essential Newborn Care program. One benefit did emerge during the Neonatal Resuscitation Program portion of the trial: Rates of death from birth asphyxiation declined from 17 to 3 deaths per 1,000 live births. However, no other outcomes showed improvement.

These findings, the researchers note, are not surprising: Very low birth weight infants "are likely to require advanced care," and the necessary interventions (e.g., prenatal steroid treatment, caesarean section) are not typically available for home births or births that take place in primary-care facilities. However, because the program has shown benefits among infants weighing more than 1.5 kilograms, who constitute the vast majority of neonates, Essential Newborn Care training "should continue to be advocated for all births, because it can reduce markedly the rates of stillbirths, neonatal deaths and perinatal deaths."

Moreover, the results were more encouraging in a related trial that assessed the intervention's effectiveness for institutional deliveries in Zambia, one of the few countries where the mortality rate among children aged five or younger has increased since 1990. The study was conducted in 18 urban health centers in Lusaka and Ndola (the country's two largest cities) in 2004–2006; these clinics accounted for 98% of institutional low-risk deliveries in these cities. Other than the focus on institutional (rather than all) deliveries, the study protocol, including selection of outcomes, was identical to that of the six-country study, and all 123 midwives who performed deliveries at the institutions were trained. The researchers again used logistic regression models to examine outcomes.

In total, 71,689 infants were born during the study period. After implementation of the Essential Newborn Care program, all-cause seven-day neonatal mortality decreased from 12 per 1,000 births to seven per 1,000 births (relative risk, 0.6), in part because of declines in levels of birth asphyxia and infection; no change was observed in rates of death due to low birth weight or infant malformation. In addition, perinatal mortality fell from 18 to 13 deaths per 1,000 births (0.7), although the stillbirth rate did not change.

After the Neonatal Resuscitation Program training was implemented, the rate of seven-day neonatal mortality increased from six per 1,000 to 10 per 1,000 (relative risk, 1.5), though no change in stillbirth occurred. The increase in seven-day mortality, the researchers believe, was an artifact of suboptimal follow-up rates, especially during the first part of the study; for example, seven-day outcomes were available for only 75% of infants during the Essential Newborn Care portion of the study. When the investigators used a generalized estimating equation model to impute the missing death data, the estimated rate of seven-day mortality decreased from 36 to 25 per 1,000 live births following implementation of the Essential Newborn Care program, and declined further to 16 deaths per 1,000 after the Neonatal Resuscitation Program was started. This trend makes more sense, the investigators note, as "it is unlikely that additional resuscitation training could have increased 7-day neonatal mortality rates."

In addition to the low seven-day follow-up rate early in the project, the study's main limitation was its before/after design; the researchers point out that although a cluster-randomized design would have been more desirable, it would have required a far greater sample size to detect statistically significant differences in outcomes. While additional research is needed "to confirm the effectiveness and to assess the sustainability" of Essential Newborn Care training "in other institutional settings," the investigators believe that the findings to date suggest that "neonatal care packages seem to be an effective way to improve neonatal outcomes in the developing world."

—P. Doskoch

REFERENCES

1. Carlo WA et al., Newborn care training of midwives and neonatal and perinatal mortality rates in a developing country, Pediatrics, 2010, 126(5):e1064–e1071.

2. Carlo WA et al., High mortality rates for very low birth weight infants in developing countries despite training, Pediatrics, 2010, 126(5):e1072–e1080.