A secondary data analysis of a cluster randomized controlled trial: improved cookstoves associated with reduction in incidence of low birthweight in rural Malawi

Predictors of low birthweight and comparisons of newborn birthweights among different groups of maternal factors at Rev. John Chilembwe Hospital in Phalombe district, Malawi: A retrospective record review

BACKGROUND

Birthweight has an impact on newborn's future health outcomes. Maternal factors, including age, delivery mode, HIV status, gestational age, parity and obstetric complications (preeclampsia or eclampsia [PE], antepartum hemorrhage [APH] and sepsis), however, have been shown as risk factors of low birthweight (LBW) elsewhere. For data-guided interventions, we aimed to identify predictors of LBW and compare newborn birthweights between different groups of maternal factors at Rev. John Chilembwe Hospital in Phalombe district, Malawi.

METHODS

Using a retrospective record review study design, we extracted data from maternity registers of 1244 women and their newborns from October, 2022 to March, 2023. Data were skewed. Median test was used to compare median birthweights. Chi-square or Fisher's exact tests were used to compare proportions of LBW among different groups of maternal factors. Multivariable logistic regression with stepwise, forward likelihood method was performed to identify predictors of LBW.

RESULTS

Median birthweight was 2900.00g (interquartile range [IQR]: 2600.00g to 3200.00g). Prevalence of LBW was 16.7% (n = 208). Proportions of LBW infants were higher in women with PE, APH, including women with sepsis than controls (10 [47.6%] of 21 vs 7 [58.3%] of 12 vs 191 [15.8%] of 1211, p < .001). Lower in term and postterm than preterm (46 [5.5%] of 835 vs 2 [3.7%] of 54 vs 160 [45.1%] of 355, p < .001). The odds of LBW infants were higher in preterm than term (AOR = 13.76, 95%CI: 9.54 to 19.84, p < .001), women with PE (AOR = 3.88, 95%CI: 1.35 to 11.18, p = .012), APH, including women with sepsis (AOR = 6.25, 95%CI: 1.50 to 26.11, p = .012) than controls.

CONCLUSION

Prevalence of LBW was high. Its predictors were prematurity, PE, APH and sepsis. Interventions aimed to prevent these risk factors should be prioritized to improve birthweight outcomes.

Effect of a chimney-fitted improved stove on pregnancy outcomes in Northwest Ethiopia: a randomized controlled trial

BACKGROUND

Exposure to household air pollution during pregnancy has been linked to adverse pregnancy outcomes. Improved stove was implemented in Ethiopia to reduce this exposure and related health problems. However, the effects of improved stove interventions on pregnancy outcomes remains uncertain.

METHOD

Individually randomized stove replacement trial was conducted among 422 households in six low-income rural kebeles of Northwestern Ethiopia. Pregnant women without known health conditions were recruited at ≤ 24 weeks gestation and randomized to an intervention or control group with a 1:1 ratio. A baseline survey was collected and a balance test was done. Two-sided independent samples t-test for continuous outcomes and chi-square for categorical variables were used to compare the effect of the intervention between the groups. Mean differences with 95% CIs were calculated and a p-value of < 0.05 was considered statistically significant.

RESULT

In this study, the mean birth weight was 3065 g (SD = 453) among the intervention group and not statistically different from 2995 g (SD = 541) of control group. After adjusting for covariates, infants born from intervention group weighed 55 g more [95% CI: - 43 to 170) than infants born from the control group, but the difference was not statistically significant (P = 0.274). The respective percentages for low birth weight were 8% and 10.3% for intervention and control groups respectively (P = 0.346). However, the average gestational age at delivery was higher among improved stove users (38 weeks (SD = 8.2) compared to control groups 36.5 weeks (SD = 9.6) with statistically significant difference at 0.91 weeks (95% CI: 0.52 to 1.30 weeks, p < 0.001). The corresponding difference in risk ratio for preterm birth is 0.94 (95% CI:0.92 to 0.97; p < 0.001). The percentages for maternal complications, stillbirth, and miscarriage in the intervention group were not statistically different from the control group.

CONCLUSIONS

While the increase in average birth weight among babies born to mothers using improved stoves was not statistically significant, babies had a longer gestational age on average, offering valuable health benefits. However, the study didn't find a significant impact on other pregnancy outcomes like stillbirth, miscarriage, or maternal complications.

TRIAL REGISTRATION

The study was registered at the Pan African Clinical Trial Registry website under the code PACTR202111534227089, ( https://pactr.samrc.ac.za/ (Identifier). The first trial registration date was (11/11/2021).