Neonatal mortality risks among preterm births in a rural Bangladeshi cohort.

BACKGROUND: Preterm birth leads to an estimated 35% of neonatal deaths worldwide. Our study analyses neonatal mortality risks among preterm births in rural Bangladesh. METHODS: Trained community health workers (CHW) prospectively collected data between June 2007 and September 2009. RESULTS: Among 32 126 livebirths, 22.3% were preterm (delivered at <37 weeks gestation) and almost half (46.4%) of all neonatal deaths occurred among preterm babies. Preterm babies who were born as the first child {[risk ratio (RR) 1.4; 95% confidence interval (CI) 1.1, 1.8]} and in the poorest households [RR 1.7; 95% CI 1.2, 2.4] were at higher mortality risk. Birth and newborn care preparedness was associated with lower risk of mortality [RR 0.3; 95% CI 0.2, 0.4] while preterm infants who had symptoms of infection [RR 5.6; 95% CI 4.3, 7.1] or whose mother suffered antenatal complications [RR 1.4; 95% CI 1.1, 1.8] were at higher mortality risk. Elimination of excess neonatal deaths caused by preterm would decrease population-level neonatal mortality rate by 31.0% [95% CI 27.60%, 34.5%]. CONCLUSION: Given that 87% of preterm births and 60% of preterm deaths were in late or moderate preterm infants, and that 87% preterm babies received a visit from CHW within third day of life, a home-based essential care package delivered by CHWs for sick preterm infants, specifically focused on birth preparedness, skin-to-skin care, immediate breast feeding, early recognition of danger signs, and linked through referral to intensive and quality care in health facilities, could be an effective approach in low resource settings.

Determinants and pattern of care seeking for preterm newborns in a rural Bangladeshi cohort.

BACKGROUND: Despite the increased burden of preterm birth and its complications, the dearth of care seeking data for preterm newborns remains a significant knowledge gap. Among preterm babies in rural Bangladesh, we examined: 1) determinants and patterns of care seeking, and 2) risk analysis for care-seeking from qualified and unqualified providers. METHOD: Trained community health workers collected data prospectively from 27,460 mother-liveborn baby pairs, including 6,090 preterm babies, between June 2007 and September 2009. Statistical analyses included binomial and multinomial logistic regressions. RESULTS: Only one-fifth (19.7%) of preterm newborns were taken to seek either preventive or curative health care. Among care-seeker preterm newborns, preferred providers included homeopathic practitioners (50.0%), and less than a third (30.9%) sought care from qualified providers. Care-seeking from either unqualified or qualified providers was significantly lower for female preterm babies, compared to male babies [Relative Risk Ratio (RRR) for unqualified care: 0.68; 95% Confidence Interval (CI): 0.58, 0.80; RRR for qualified care: 0.52; 95% CI: 0.41, 0.66]. Among preterm babies, care-seeking was significantly higher among caregivers who recognized symptoms of illness [RR: 2.14; 95% CI: 1.93, 2.38] or signs of local infection (RR: 2.53; 95% CI: 2.23, 2.87), had a history of child death [RR: 1.21; 95% CI: 1.07, 1.37], any antenatal care (ANC) visit [RR: 1.41; 95% CI: 1.25, 1.59]. Birth preparedness (RRR: 1.24; 95% CI: 1.09, 1.68) and any ANC visit (RRR: 1.73; 95% CI: 1.50, 2.49) were also associated with increased likelihood of care seeking for preterm babies from qualified providers. CONCLUSION: To improve care seeking practices for preterm babies and referral of sick newborns to qualified providers/facilities, we recommend: 1) involving community-preferred health care providers in community-based health education and awareness raising programs; 2) integrating postnatal care seeking messages into antenatal counselling; and 3) further research on care seeking practices for preterm babies.

Incidence and risk factors of preterm birth in a rural Bangladeshi cohort.

BACKGROUND: Globally, about 15 million neonates are born preterm and about 85% of global preterm birth occurs in Asia and Africa regions. We aimed to estimate the incidence and risk factors for preterm birth in a rural Bangladeshi cohort. METHODS: Between June 2007 and September 2009, community health workers prospectively collected data from 32,126 mother-live-born baby pairs on household socio-demographic status, pregnancy history, antenatal care seeking and newborn gestational age determined by recall of date of last menstrual period. RESULTS: Among all live births, 22.3% were delivered prior to 37 weeks of gestation (i.e. preterm); of which 12.3% were born at 35-36 weeks of gestation (late preterm), 7.1% were born at 32-34 weeks (moderate preterm), and 2.9% were born at 28-31 weeks of gestation (very preterm). Overall, the majority of preterm births (55.1%) were late preterm. Risk of preterm birth was lower among women with primary or higher level of education (RR: 0.92; 95% CI: 0.88, 0.97), women who sought antenatal care at least once during the index pregnancy (RR: 0.86; 95% CI: 0.83, 0.90), and women who had completed all birth preparedness steps (RR: 0.32; 95% CI: 0.30, 0.34). In contrast, risk of preterm birth was higher among women with a history of child death (RR: 1.05; 95% CI: 1.01, 1.10), who had mid-upper arm circumference (MUAC) ≤250 mm, indicative of under nutrition (for women having MUAC <214 mm the risk was higher; RR: 1.26; 95% CI: 1.17, 1.35), who reported an antenatal complication (RR: 1.32; 95% CI: 1.14, 1.53), and who received iron-folic acid supplementation for 2-6 months during the index pregnancy (RR: 1.33; 95% CI: 1.24, 1.44). CONCLUSIONS: In resource poor settings with high burden of preterm birth, alike Bangladesh, preterm birth risk could be reduced by close monitoring and/or frequent follow-up of women with history of child death and antenatal complications, by encouraging women to seek antenatal care from qualified providers, to adopt birth preparedness planning and to maintain good nutritional status. Additional research is needed to further explore the associations of antenatal iron supplementation and maternal nutritional status on preterm birth.

Impact of 4.0% chlorhexidine cord cleansing on the bacteriologic profile of the newborn umbilical stump in rural Sylhet District, Bangladesh: a community-based, cluster-randomized trial.

BACKGROUND: Randomized trials from South Asia indicate umbilical cord chlorhexidine cleansing reduces mortality and omphalitis. No community-based data are available on bacteriological profile of the cord, early neonatal colonization dynamics, or impact of cord cleansing on colonizing organisms. Such data could clarify the design of scaled chlorhexidine interventions. METHODS: Umbilical swabs were collected at home (days 1, 3, 6) after birth from infants participating in a trial of 3 cord-care regimens (no chlorhexidine, single cleansing, multiple cleansing) in Sylhet, Bangladesh. Overall and organism-specific positivity rates were estimated by cord-care regimen and by day of collection. RESULTS: Between September 2008 and October 2009, 1923 infants contributed 5234 umbilical swabs. Positivity rate was high (4057 of 5234, 77.5%) and varied substantially across groups. Immediate (day 1) reductions in cord colonization were observed in single- (prevalence rate ratio = 0.75, 95% confidence interval: 0.70-0.81) and multiple- (prevalence rate ratio = 0.71, 95% confidence interval: 0.66-0.77) cleansing groups. Reductions persisted and increased in magnitude through day 6 only if babies received multiple applications. On days 1, 3, and 6, respectively, multiple cleansing consistently reduced invasive organisms such as Escherichia coli (49%, 64%, and 42% lower), Klebsiella pneumoniae (46%, 53%, and 33% lower), and Staphylococcus aureus (34%, 84%, and 85% lower). CONCLUSIONS: Cord cleansing with 4.0% chlorhexidine immediately after birth reduces overall and organism-specific colonization of the stump. Reductions are greater and sustained longer with daily cleansing through the first week of life, suggesting that programs promoting chlorhexidine cleansing should favor multiple over single applications.