Reconsidering gas as clean energy: Switching to electricity for household cooking to reduce NO2-attributed disease burden

Nitrogen dioxide (NO2) is a prevalent air pollutant in urban areas, originating from outdoor sources, household gas consumption, and secondhand smoke. The limited evaluation of the disease burden attributable to NO2, encompassing different health effects and contributions from various sources, impedes our understanding from a public health perspective. Based on modeled NO2 exposure concentrations, their exposure–response relationships with lung cancer, chronic obstructive pulmonary disease, and diabetes mellitus, and baseline disability-adjusted life years (DALYs), we estimated that 1,675 (655–2,624) thousand DALYs were attributable to NO2 in urban China in 2019 [138 (54–216) billion Chinese yuan (CNY) economic losses]. The transition from gas to electricity for household cooking was estimated to reduce the attributable economic losses by 35%. This reduction falls within the range of reductions achieved when outdoor air meets the World Health Organization interim target 3 and air quality guidelines for annual NO2, highlighting the significance of raising awareness of gas as a polluting household energy for cooking. These findings align with global sustainable development initiatives, providing a sustainable solution to promote public health while potentially mitigating climate change.


Source-specific exposure model
We have developed a source-specific exposure model [1] to simulate NO2 exposure concentration from various sources (Figure S1).The model first established an equation to calculate the indoor concentration of NO2 based on the conservation of mass combined with indoor dynamics of NO2.It then separated the indoor concentration of NO2 from outdoor sources ("aCout") and indoor sources ("S/V"), denoted as Cin,o and Cin,i in Figure S1, respectively.Human activities both indoors and outdoors were taken into account to calculate the NO2 exposure concentration from outdoor ("Cexp,o") and indoor ("Cexp,i") sources.A twostage Monte Carlo approach was applied in the model to obtain the variability distribution of exposure.
We validated the model by comparing the modelled concentrations with available field studies, ensuring the credibility of the modelled exposure (Table S1).
The input parameters used in this model to estimate NO2 exposure concentration in urban China in 2019 included the outdoor concentration of NO2 from 1,497 monitoring stations in 333 cities in China, the emission rate of NO2 from gas cooking and smoking, the surface removal rate of NO2 indoors, air change rates when windows were open, closed, and when operating range hoods, the volume of air in residences, and habits of cooking, smoking, ventilation, and outdoor activities.These parameters were detailed in Table 1 of our previous study [1], and here we only describe one of the key parametersestimation of NO2 emissions from gas cooking.This estimation relies on a comprehensive dataset of both the rate of NO2 generation from gas combustion and the cooking habits of residences in urban areas in China.The rate of NO2 generation from gas combustion was from a study conducted in the United Kingdom [2].To adapt this data to the Chinese context, we adjusted based on household gas consumption and total cooking duration specific to China [3].Regarding cooking habits, we conducted an extensive survey spanning all 31 provinces in China, which yielded a substantial dataset comprising 1103 valid responses [1].This survey collected a wealth of information related to gas usage for cooking, encompassing parameters such as cooking frequency, time periods for cooking activities, preferred cooking methods, and other factors that could influence indoor NO2 concentrations during gas cooking, such as ventilation.Using this comprehensive dataset, we were able to simulate how individuals used gas for cooking daily, including when they cooked and at what emission rates.These data served as crucial input parameters for the model presented in Figure S1, forming foundation for NO2 exposure concentration estimation.

DALY rates and population in China
The equation used to calculate the DALY rate for a specific age group and sex in a specific province in China in 2019 was as follows: ,, =   ,    ×   ℎ,,   ℎ,    (S1) The DALY rates for all ages and both sexes in 31 Chinese provinces were obtained from a dataset provided by National Centre for Chronic and Noncommunicable Disease Control and Prevention (Table S3) [4].The DALY rates for people of different age and sexes groups were from the Global Burden of Disease Study 2019 (GBD 2019) (Table S4)[5].
The age and sex-specific population size in urban areas in 31 Chinese provinces in 2019 was calculated based on the urban population of each city in 2019 [6] (Table S5) and the age and gender composition of the population of each province [7] (Table S6).

Figure S2
Figure S2 DALYs attributable to NO2 by age and sex in 2019.