Nitrogen dioxide exposures from LPG stoves in a cleaner-cooking intervention trial.

Background: Liquefied petroleum gas (LPG) stoves have been promoted in low- and middle-income countries (LMICs) as a clean energy alternative to biomass burning cookstoves.

Objective: We sought to characterize kitchen area concentrations and personal exposures to nitrogen dioxide (NO) within a randomized controlled trial in the Peruvian Andes. The intervention included the provision of an LPG stove and continuous fuel distribution with behavioral messaging to maximize compliance.

Methods: We measured 48-hour kitchen area NO concentrations at high temporal resolution in homes of 50 intervention participants and 50 control participants longitudinally within a biomass-to-LPG intervention trial. We also collected 48-hour mean personal exposures to NO among a subsample of 16 intervention and 9 control participants. We monitored LPG and biomass stove use continuously throughout the trial.

Results: In 367 post-intervention 24-hour kitchen area samples of 96 participants' homes, geometric mean (GM) highest hourly NO concentration was 138 ppb (geometric standard deviation [GSD] 2.1) in the LPG intervention group and 450 ppb (GSD 3.1) in the biomass control group. Post-intervention 24-hour mean NO concentrations were a GM of 43 ppb (GSD 1.7) in the intervention group and 77 ppb (GSD 2.0) in the control group. Kitchen area NO concentrations exceeded the WHO indoor hourly guideline an average of 1.3 h per day among LPG intervention participants. GM 48-hour personal exposure to NO was 5 ppb (GSD 2.4) among 35 48-hour samples of 16 participants in the intervention group and 16 ppb (GSD 2.3) among 21 samples of 9 participants in the control group.

Discussion: In a biomass-to-LPG intervention trial in Peru, kitchen area NO concentrations were substantially lower within the LPG intervention group compared to the biomass-using control group. However, within the LPG intervention group, 69% of 24-hour kitchen area samples exceeded WHO indoor annual guidelines and 47% of samples exceeded WHO indoor hourly guidelines. Forty-eight-hour NO personal exposure was below WHO indoor annual guidelines for most participants in the LPG intervention group, and we did not measure personal exposure at high temporal resolution to assess exposure to cooking-related indoor concentration peaks. Further research is warranted to understand the potential health risks of LPG-related NO emissions and inform current campaigns which promote LPG as a clean-cooking option.