Postnatal home visits by community health workers are effective in reducing neonatal mortality if they occur within the first two days after birth, according to a prospective study conducted in Bangladesh.1 Among infants who survived their first day of life, the 28-day mortality rate was a third that among those who were visited that day by a health worker than among those who never received a visit (hazard ratio, 0.3). Similarly, among infants who survived for two or more days, the mortality rate among those visited by a health worker on day two was far lower than that among infants who were never visited (0.4). Infants whose initial visit from a health worker occurred after the second day had no survival advantage.

Researchers analyzed data from a 24-month study conducted in 2004–2005 to examine the effectiveness of three intervention approaches—community care, home care and usual care—for improving health outcomes among newborns and their mothers. Eight administrative districts in Sylhet, Bangladesh, were randomly chosen as sites for the home care intervention, in which female community health workers monitored their assigned catchment area (comprising approximately 4,000 people) for pregnancies and provided a range of counseling and care. Health workers were asked to visit the home of each pregnant woman twice before birth (at 12–16 and 30–34 weeks of gestation) and three times after delivery (on the first, third and seventh days of life) to counsel the family on preventive care, birth preparations, neonatal care, breastfeeding and recognizing when to seek medical care. In addition, the health workers assessed newborns for signs of serious illness and provided appropriate referrals (as well as antibiotics, if infection was suspected but the parents refused a referral). Given the challenges of providing care in this setting, health worker visits frequently did not coincide with the desired schedule—and in some cases did not occur at all. However, the workers, monitored by field supervisors, kept detailed records of their visits, and they made a final visit to each household during the fifth week after birth to collect information on neonatal outcomes, demographic characteristics, medical history and other relevant factors.

The investigators examined neonatal mortality (deaths in the first 28 days of life per 1,000 live births) among four groups of infants: those who received their first visit from a health worker on the first day of life, those whose initial visit occurred on day 2, those who weren't visited until days 3–6 and those who were first visited on the seventh day or later. For each group, the mortality rate was compared with the rate among infants who never received a visit but survived for an equal length of time. In addition to calculating mortality rates, the researchers estimated hazard ratios using models that controlled for economic status (based on the materials from which the family's home was constructed), parental education, gestational age at birth, multiple birth (whether the infant was a twin or triplet) and whether the pregnancy was the mother's first.

The analytic sample consisted of 9,211 newborns, of whom 31% were first visited by a health worker on day 1, 31% were first visited on day 2, 11% were first visited on one of the next four days and 13% on the seventh day or later. The remaining 14% never received a visit at all. The mothers of infants who were never visited tended to be wealthier and better educated than other mothers, and their births were more likely to be preterm.

The overall neonatal mortality rate was 29 per 1,000 live births. However, among infants who survived their first day of life, the mortality rate was substantially lower for those visited by a health care worker on day 1 than for those who never received a visit (21 vs. 65 per 1,000). Likewise, among infants who survived day 2, the mortality rate for those who were visited by a health worker on that day was a third the rate for infants who never received a visit (13 vs. 39 per 1,000). Mortality rates for infants whose first visit occurred on days 3–6 or on days 7–28 did not differ from rates for unvisited infants.

After adjustment for economic status, maternal education, gestational age, first pregnancy and multiple births, neonatal mortality was substantially lower among infants who were visited by a health care worker on their first day of life (hazard ratio, 0.3) or their second day (0.4) than among infants who never received a visit. Other factors positively or negatively associated with neonatal mortality among infants who survived their first or second day of life included first pregnancies (1.6–1.9), multiple births (3.6–6.7) and gestational age of at least 37 weeks (0.3). Mortality was not reduced if the first postnatal visit took place on day 3 or later.

Because infants whose parents refused referrals received extra home visits and treatment from health workers, the researchers were not able to assess whether a greater number of visits influenced infant survival. Nonetheless, the findings suggest that women in developing countries "where home delivery with unskilled attendants is the norm" should "receive a home visit and undergo assessment by a trained worker as soon as possible, preferably on the day of birth but no later than 48 hours after birth." The degree of benefits, the researchers add, likely depends on the content and quality of the intervention, and future research should examine strategies for facilitating prompt provision of postnatal care.—P. Doskoch


1. Baqui AH et al., Effect of timing of first postnatal care home visit on neonatal mortality in Bangladesh: a observational cohort study, BMJ, 2009, 339:b2826. doi: 10.1136/bmj.b2826.