Clean fuel-based interventions for reducing non- communicable disease burden at community level

SUMMARY

The is article describes the results of an extension of the Ghana Randomized Air Pollution and Health Study (GRAPHS) in which 700 mother–child pairs were followed up after the children reached 1 year of age to assess the effect of use of clean household fuels on blood pressure (BP) of young children. GRAPHS was a cluster randomized controlled trial with two intervention arms: (i) An improved biomass stove and (ii) a liquified petroleum gas (LPG) stove with continuous LPG supply till the child’s first birthday to assess the effect of household air pollution (HAP) on birth weights and childhood pneumonia. The study also assessed exposure to carbon monoxide (CO) at four time points during the prenatal period (at the time of enrollment and 3 times spaced at 3-week intervals) and 3 times during the postnatal period (when the child was 1 month, 4 months and 1 year old) in the whole cohort as well as prenatal and postnatal exposure to PM2.5 in a conveniently selected subset.

Intention-to-treat analysis was used to examine the effect of the cookstove intervention on BP Z-scores. Reverse distributed lag models were used to examine time-varying associations between prenatal and 1st-year-of-life CO exposures and BP Z-scores and generalized linear regression was used to examine associations between PM2.5 exposure and BP Z-scores. Multivariable model 1 accompany the lack of such access.³ It is estimated that HAP was adjusted for child sex and body mass index (BMI), maternal ethnicity and enrollment BP, second-hand tobacco smoke exposure and household asset index. Multivariable model 2 additionally adjusted for child age 4-y PM_2.5 exposure.

Key findings from the study include lower diastolic BP Z-scores in children whose mothers were enrolled in the LPG arm as compared to the control group, with significantly lower systolic BP and diastolic BP in females. The reduction was equivalent on average to 1.64 mmHg (95% confidence interval: 0.25, 2.95) or 3% reduction in diastolic BP at 4 years of age. Multivariable model 2 identified a sensitive window in late gestation where high CO exposure was associated with elevated systolic BP (p=0.02) and diastolic BP Z-scores (p<0.01) at 4 years. Similarly, higher CO exposure in the latter part of the 1st year of life was linked to higher diastolic BP Z-scores (p<0.01). No significant association was found between prenatal exposure to PM2.5 and BP Z-scores at 4 years of age. The study concludes that exposure–response analyses support an association between early-life HAP exposure and BP at 4 years of age.

COMMENTS

The study’s title implies examination of the relationship between prenatal exposure to HAP and BP in children; however, it also looked at the relationship between postnatal exposure to HAP and BP in children and an intention-to-treat analysis to look at differences in BP in children across the various study arms of the GRAPHS cluster randomized controlled trial. Comments on the adequacy of the sample size to detect significant differences in childhood BP, method of selection of sub-sample and justification for convenient sampling of participants to examine association between PM2.5 exposure and childhood BP are lacking. Differential distribution of covariates such as gestational hypertension, paternal BP, time and season of BP measurement among study arms can also lead to erroneous estimation of the association between HAP and childhood BP.

It has been mentioned that fieldworkers visited the households on weekly basis to assess and facilitate stove repairs along with assessing maternal health, but there is a lack of information on compliance with the interventions and steps taken to improve compliance. The effectiveness of even the most advanced cookstoves is contingent upon their consistent use. Factors such as ease of operation, performance in cooking staple foods, user familiarity and fuel costs also influence the use of cookstoves.¹ Stove use monitoring systems provide an objective measure of stove usage patterns by recording stove temperature at regular intervals eliminating the biases associated with recall and self- reporting, offering a reliable assessment of stove usage.²

The total number of people worldwide lacking access to clean fuels for cooking has dropped from 2.9 billion in 2010 to 2.1 billion in 2022. However, people living in low- and middle- income countries, especially in Sub-Saharan Africa and South Asia, bear a disproportionate brunt of the health hazards that responsible for 3.2 million deaths including over 237 000 deaths in children in 2020.⁴

Barker hypothesizedthatadverseconditionsduring foetal development are associated with an increased risk of developing chronic diseases in adulthood.⁵ The exposures in his hypothesis extend beyond just low birth weight and include stress, either nutritional or non-nutritional, during critical periods of development.⁶ The study also highlights the impact of environmental stresses such as intrauterine HAP exposure on increased childhood BP, which in turn increases the risk of adult hypertension and adverse cardiovascular outcomes in later life.^7,8

In India, 41% of all households and 55% of rural households do not use clean fuels for cooking.⁹ Further, non-communicable diseases show a rising trend and accounted for 63% of all deaths in 2018, and cardiovascular diseases contributed to 27% of the overall mortality.¹⁰ It is thus imperative that Indian households shift to biomass stoves and cleaners fuels for cooking not only for health reasons but also from the women empowerment perspective. The Pradhan Mantri Ujjwala Yojana launched by the Ministry of Petroleum and Natural Gas in 2016 aims to support this by providing free LPG connections to households that are not using clean fuels. This study, by showing that HAP adversely affects the health not only of adults but also of infants and young children, further supports steps to promote the use of cleaner fuels.

The topic of delayed effects and long-term impact of HAP demands further research and evidence generation as studies have given conflicting evidence.^11,12 Existing long-term large cohorts such as PURE,¹³ CAARS,¹⁴ and HDSS¹⁵ can be leveraged for this purpose and large-scale surveys such as the National Family Health Survey can incorporate data collection on HAP and ambient air pollution.