The relationship between cooking fuel and health status from the perspective of income heterogeneity

doi:10.21203/rs.3.rs-1507796/v1

Download PDF

Research Article

The relationship between cooking fuel and health status from the perspective of income heterogeneity

https://doi.org/10.21203/rs.3.rs-1507796/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

This study estimates the potential relationship between household air pollution and health status via using data from the China Family Panel Studies (CFPS). Household air pollution is measured by different type of cooking fuels. Health status is evaluated from the perspective of respondents’ self-ratings of health, mental health, and by interviewer-assessed health. The data comprise more than 100,000 samples from 162 counties between 2012 and 2019. We employ a Hierarchical Linear Model to infer the potential impact of household air pollution caused by cooking fuel on residents’ health. After adjusting control variables, we find cleaner cooking fuels such as solar energy and biogas are more conducive to residents’ health status. Our findings manifest that cleaner fuels and lower exposure to household air pollution are better for residents’ health. Low-income households are more likely to be affected by household air pollution. This study emphasizes the consequence of the potential health effects on residents due to household air pollution from cooking fuels and the uneven distribution of those effects. It is recommended that such factors as household air pollution risks, health, and environmental fairness&justice should be considered in the policy formulation process.

Cooking fuel

Health status

Income heterogeneity

China

In recent decades, the rapid development of the global economy has brought about a huge increase in household energy consumption (Zhu et al., 1983; Yu et al., 2020). Household energy sources have turned diversified, and the traditional solid fuels such as firewood and other raw biomass have evolved into clean-energy sources such as solar energy and biogas (Wang et al., 2015; Wu et al., 2019; Yu et al., 2020). According to the latest data from the WHO, 63% of household energy consumption is provided by solid fuel, and the other 37% is clean energy (WHO, 2021). More than 2 billion people still rely on traditional biomass fuels to meet their daily energy needs. It is common knowledge that household energy consumption is an essential source of air pollution (Peabody et al., 2005), especially traditional solid fuels. Cooking fuels produce harmful gases and airborne particles, leading to household air pollution, which affects residents’ health. Straw burning, for example, produces particulate matter pollution and causes cardiorespiratory diseases; a 10 µg/m³ increase in PM_2.5 concentration increases mortality by 3.25% (He et al., 2020). Household air pollution from solid fuels is hence a major contributor to global burden of disease and a main obstacle to improving living standards (Carter et al., 2017). Despite use of various ventilation techniques during cooking, cooking-associated pollution still increases the risk of diseases such as chronic obstructive pulmonary disease, lung cancer, tuberculosis (Rehfuess et al., 2009 Kim et al., 2015; Balmes, 2019; Yu et al., 2020), respiratory ailments (Sapkota et al., 2013 Yasar et al., 2017; Qiu et al., 2019) and cardiovascular disease (Yasar et al., 2017; Yu et al., 2018; Arku et al., 2018ab; Qiu et al., 2019; Balmes, 2019; Li et al., 2020; Liu et al., 2020; Tian et al., 2021). In addition, air pollution exerts adverse effects on eye health and burns (Das et al., 2017), cognitive abilities (Qiu et al., 2019), contributes to all-cause death rates (Yu et al., 2018; Yu et al., 2020), premature death rates and disability-adjusted lifespan (Fischer, 2001; Chen et al., 2018). These impacts or losses are especially disastrous for vulnerable groups, such as women, children, seniors (Mishra, 2003; Dionisio et al., 2010; Agrawal et al., 2015; Qiu et al., 2019) and low-income countries or groups (Pant, 2012; Oguonu et al., 2018; Clasen et al., 2019).

According to the Global Burden of Diseases, Injuries, and Risk Factors Study (GBD) 2018, solid cooking fuels account for approximately 1.6 million deaths worldwide, as they produce air pollution during cooking (GBD 2017 Risk Factor Collaborators, 2018; Yu et al., 2020). Indoor air pollution is a major health problem in developing and low-income countries (Quinn et al., 2018; Silwal and McKay, 2015; Oguonu et al., 2018; Sidhu et al., 2017; Balmes, 2019). The importance of household air pollution as a public health threat varies drastically in accordance with the level of development: in low- and middle-income countries, household air pollution is responsible for almost 10% of mortality; globally, it is liable for 7.7% of mortality (WHO, 2021). Households in China, specifically, consume a huge amount of energy. Most of China’s energy is still based on solid fuels (64% and 18% of household energy consumption is solid fuel in rural and urban areas, respectively, WHO, 2021). The household air pollution is a serious issue, especially in underdeveloped and rural areas (Wang et al., 2015; Oguonu et al., 2018; Wu et al., 2019). Thousands of people are constantly exposed to indoor pollution at concentrations exceeding both domestic and international standards (Fischer, 2001). However, little attention is paid on the potential impact of household air pollution on health in China (Tao et al., 2014). Of particular interest is the health risk due to household air pollution from cooking fuel. Residents are exposed to this pollution regularly and systematically, so the observations in resident tracking surveys can be used to discern the relationship between cooking-induced household air pollution and health status (Nadimi and Tokimatsu, 2018; Wang et al., 2019; Taghizadeh-Hesary et al., 2020).

Residents can bring down ambient pollution by adding ventilation equipment, installing air conditioners or air purifiers, to reduce associated health risks (Hänninen et al., 2013). When cooking with liquefied petroleum gas, PM concentrations vary greatly depending on the location of the kitchen, fuel usage rate, and ventilation rate (Begum et al., 2009; Hänninen et al., 2013; Thomas et al., 2015). Good ventilation equipment can reduce household air pollution effectively and convert cooking fuel with low cleanliness (causing high household air pollution) into medium or high cleanliness (causing low household air pollution).

The contribution of this this research lies in three respects First, while most studies on cooking-fuel pollution hazards focus on the effects on all-cause death, morbidity and mortality of cardiovascular disease due and respiratory ailments, little discussion circle around the impact of household air pollution on self-evaluated health and mental health for individuals. The potential relationship between household air pollution and the evaluation of health indicators remains unknown. Second, most studies focus on developed countries or some developing countries, paying little attention to China, a country with large economic differences among its regions. Few studies have given thought to the effect of how health outcomes due to household air pollution exposure differ between people in dissimilar regions and with different income levels. The relationship between income, household air pollution and health still calls for further heed. Finally, previous studies focused on the health effects of household air pollution. Residents’ ability to avoid polluted household air remains an open question, as a result of the limitation of data availability. It is arduous to develop prospective countermeasures because the effectiveness of pollution-avoidance behaviors is rarely investigated, like the behavior of using air purifier or cleaner cooking fules.

We use CFPS data from 2012 to 2019 to construct indicators for household air pollution caused by cooking fuel, residents’ health indicators, and income indicators, as well as demographic indicators such as gender and age. Subsequently we discuss the relationship of household air pollution (caused by cooking fuel) on these health indicators to construct two indicators—exposure and cleanliness—based on the cooking fuel indicators. We deal with the relationship of these exposure and cleanliness indicators with residents’ health. Stratified regression and a dummy-variable regression model are used to discuss the relationship of income with household air pollution and residents’ health. Ultimately, a conclusion is drawn that differences in cooking fuels have a range of impacts on the health of residents.

2.1 Data sources

In this study, we used individual-level data from the China Family Panel Studies (CFPS) in 2012–2019. It’s collected by Institute of Social Science Survey, Peking University. CFPS reflects the changes of China’s society, economy, population, education and health by tracking and collecting data from individuals, families and communities covering 25 provinces, cities and autonomous regions. A total of 180,712 samples were included in the period from 2012 to 2019. To create a panel dataset, we only keep observations that are included in the 162 counties from the 2010 survey, obtaining 147,112 samples. In order to ensure the availability of the data, we excluded some individuals with missing indicators. A total of 45,400 samples were removed, leaving 101,712 samples. Men constituted 49.6% of the initial respondents; after winnowing, men constituted 48.31%, which leads to the conclusion that no significant change in the characteristics of the sample, and the sample can represent the original data set.

2.2 Health status

In our research, the dependent variable proves to be health. We use a series of health results to describe individuals’ health situation, including respondents’ self-rate health, interviewer-assessed health, and mental health.

Respondents’ self-rated health is comprehensive, including both physical and mental health (Yang and He, 2020). It’s widely applied in health-related research (Li and Liu, 2015). In CFPS research, each respondent was asked, “How do you think your health is?” and answered by selecting from “1. unhealthy”, “2. normal”, “3. relatively healthy”, “4. healthy” and “5. very healthy”.

On top of respondents’ self-rated health, we choose interviewer-assessed health, and mental health. Interviewer-assessed health is a key indicator (Clarke et al., 2003). In CFPS research, each interviewer answered the following question: “What’s the respondent’s health status?” by making a choice from 1 (worst) to 7 (best).

Mental health refers to people’s emotions, will, ability, pride and other psychological states, which are revealed through expression and behavior (Prince et al., 2007). The CFPS survey asked, “How often do you feel depressed?”. The options are “always” (5–7 days), “often” (3–4 days), “sometimes” (1–2 days) and “never” (less than 1 day) in 2011–2012 and 2016–2019, and “every day”, “often”, “half of the time”, “sometimes” and “never” in 2014–2015. As a means to ensure the uniformity of indexes in 2012–2019, we regroup the options: “1. Most of the time” (5–7 days), “2. often” (3–4 days), “3. sometimes” (1–2 days) and “4. never” (less than 1 day). The 2014–2015 responses, “often” and “half of the time” were reassigned to “often” (3–4 days). The four new categories are coded from 1 to 4, with higher codes corresponding to better mental health.

2.3 Cooking fuels and household air pollution

In this research, the key independent variables are cooking fuels. In CFPS research, respondents were asked to answer the question, “What kind of fuel is used for cooking?” among options “firewood”, “coal”, “gas/liquefied gas/natural gas”, “solar energy”, “biogas”, “electricity” and “other” in 2012–2013, and “firewood”, “coal”, “canned gas”, “pipeline gas”, “solar energy/biogas”, “electricity” and “other” in 2014–2019. In order to ensure the uniformity of indexes in 2012–2019, we regroup the choices: “1. firewood”, “2. coal”, “3. gas/liquefied gas/natural gas”, “4. solar energy/biogas”, “5. electricity” and “6. other”. Reference group contains firewood to avoid collinearity.

Cooking fuels can be divided into three categories in conformity with air cleanliness: firewood and coal (low cleanliness 0), gas, liquefied gas and natural gas (medium cleanliness 1), solar energy, biogas and electricity (high cleanliness 2); meanwhile, they can be divided into two categories according to air pollution exposure: firewood and coal (high exposure 0), and other fuels (low exposure 1). Reference group contains firewood to avoid collinearity.

2.4 Solution for household air pollution

We pay regard to the solution for household air pollution. In CFPS research 2017, respondents were asked, “Do you use a fresh air system or air purifier at home?” with answer options of yes = 1 or no = 0. This is applied as the residents’ evasion index to reflect their avoidance of air pollution in their households.

2.5 Other variables

We add individual characteristic control variables via using data from the CFPS, including gender, age, marriage, education, household per capita income, place of residence and health behaviors like smoking, drinking and exercise. Gender, marriage, residence and health behaviors are assigned dummy variables: men = 1, women = 0; urban = 1, rural = 0; yes = 1, no = 0. If the respondent has a smoking habit, he gets a score of “1” otherwise gets “0”. The same method operates in terms of drinking and exercise. Age, education, and household per capita income are continuous variables, representing the level of education and income.

We carry out a descriptive analysis of the population and split gender samples. For categorical variables, we take the frequency; for continuous variables, we give means with standard deviations. Since there are multiple respondents in each city, a hierarchical linear model of individual and city levels is implemented (Diez-Roux, 1998; Duncan et al., 1996).

In a bid to estimate the impact of household air pollution, measured by cooking fuels usage, on health status, we constructed the following Hierarchical Linear Model, as the baseline econometric model:

Level 2:${a}_{0} ={\beta }_{0}+{\beta }_{1}{\text{U}\text{r}\text{b}\text{a}\text{n}}_{c}+{r}_{0}$

$\text{H}\text{e}\text{a}\text{l}\text{t}{\text{h}}_{ict}$ stands as the dependent variable, representing the health status of respondent i in district c at time t. We use indicators such as respondents’ self-rated health, interviewer-assessed health and mental health to measure health status comprehensively. Cooking_fuels_ict is the key independent variable, betokening the fuel type used by respondent i in district c at time t, coded into the five categories mentioned earlier: “1. firewood”, “2. coal”, “3. gas/liquefied gas/natural gas”, “4. solar energy/biogas”, “5. electricity” (j = 1, 2, 3, 4, 5). Firewood is the control group. Z_ict is a vector of control variables indicating residents’ gender, age, marital status, years of education, smoking, drinking, exercise and residence (whether in the city) of respondent i in district c at time t, and the household per capita income of the respondent’s household. In Level 1, ${a}_{1}^{j}$ and ${a}_{2}$ are estimated coefficients. a₀ and e_ict stand forthe average health status of the respondent and the random errors, respectively. η_y × η_m and γ_i represent time fixed effects (year-by-month fixed effects) and the individual’s fixed effects, respectively (Bai, 2013).

In Level 2, β₀ reflects the average health status of individual, and r₀ represents individual’s random error. We can combine Level 1 and Level 2 in Eq. (1):

$$\text{H}\text{e}\text{a}\text{l}\text{t}{\text{h}}_{ict} =\varSigma {a}_{1}^{j} {\text{C}\text{o}\text{o}\text{k}\text{i}\text{n}\text{g}\_\text{f}\text{u}\text{e}\text{l}\text{s}}_{ict}^{j}+{a}_{2}{Z}_{ict}+{\beta }_{0}+{\beta }_{1}{\text{U}\text{r}\text{b}\text{a}\text{n}}_{c}+{r}_{0}+{\eta }_{y}\times {\eta }_{m}+{\gamma }_{i}+{e}_{ict}$$

Morever, we re-categorize cooking fuels in connection with cleanliness and exposure. Residents’ avoidance of household air pollution is investigated by a layered test to examine the effectiveness of air purification and ventilation.

Income can be regarded as a moderating variable, and high income can help residents to reduce the problems due to kitchen air pollution. In order to estimate the potential impact of income on the relationship between cooking fuels and health status, we incorporate income into the regression model, and it functions as follows:

Level 1: ${\text{H}\text{e}\text{a}\text{l}\text{t}\text{h}}_{ict} ={a}_{0} +\varSigma {a}_{1}^{j} {\text{C}\text{o}\text{o}\text{k}\text{i}\text{n}\text{g}\_\text{f}\text{u}\text{e}\text{l}\text{s}}_{ict}^{j}+{a}_{2}{\text{I}\text{n}\text{c}\text{o}\text{m}\text{e}\_\text{g}\text{r}\text{o}\text{u}\text{p}}_{ict}+\varSigma {a}_{3}^{j} {\text{C}\text{o}\text{o}\text{k}\text{i}\text{n}\text{g}\_\text{f}\text{u}\text{e}\text{l}\text{s}}_{ict}^{j}\times {\text{I}\text{n}\text{c}\text{o}\text{m}\text{e}\_\text{g}\text{r}\text{o}\text{u}\text{p}}_{ict}+{a}_{4}{Z}_{ict}+{\eta }_{y}\times {\eta }_{m}+{\gamma }_{i}+{e}_{ict}$ (3)

Level 2:${a}_{0} ={\beta }_{0}+{\beta }_{1}{\text{U}\text{r}\text{b}\text{a}\text{n}}_{c}+{r}_{0}$

Cooking_fuels_ict×Income_group_ict represents the interaction between cooking fuels and income. Respondents whose household per capita income exceeded the average were sorted into the high-income group (Income_group_ict=1); those below the average, into the low-income group (Income_group_ict=0). Evaluating the interaction between cooking fuel and income will demonstrate whether income accounts for the relationship between cooking-induced air pollution and health status.

Similarly, we combine Level 1 and Level 2 from Eq. (3) as follows:

$${\text{H}\text{e}\text{a}\text{l}\text{t}\text{h}}_{ict}=\varSigma {a}_{1}^{j} {\text{C}\text{o}\text{o}\text{k}\text{i}\text{n}\text{g}\_\text{f}\text{u}\text{e}\text{l}\text{s}}_{ict}^{j}+{a}_{2}{\text{I}\text{n}\text{c}\text{o}\text{m}\text{e}\_\text{g}\text{r}\text{o}\text{u}\text{p}}_{ict}+\varSigma {a}_{3}^{j} {\text{C}\text{o}\text{o}\text{k}\text{i}\text{n}\text{g}\_\text{f}\text{u}\text{e}\text{l}\text{s}}_{ict}^{j}\times {\text{I}\text{n}\text{c}\text{o}\text{m}\text{e}\_\text{g}\text{r}\text{o}\text{u}\text{p}}_{ict}+{a}_{4}{Z}_{ict}+{\beta }_{0}+{{\beta }_{1}\text{U}\text{r}\text{b}\text{a}\text{n}}_{c}+{r}_{0}+{\eta }_{y}\times {\eta }_{m}+{\gamma }_{i}+{e}_{ict}$$

All the aforementioned regression models are applied to the overall sample, and separately to the subsamples of men and women.

4.1 Statistical description

Descriptive characteristics are shown in Supplementary Table 1. The research sample includes 101,712 participants from 2012 to 2019 (see Supplementary Table 2). The literature on issues with self-rated health measures is vast (e.g. Butler, et al., 1987; Clarke and Ryan, 2006). First, we focus on the statistical description of health. The respondents’ mean self-rating is 2.901 (SD = 1.230), revealing that respondents generally think they are relatively healthy. Among other health indicators, the mean of the interviewer-assessed health scores is 5.470 (SD = 1.268), and the mean mental health score is 3.331 (SD = 0.753). In CFPS, residents’ cooking fuel usage was analyzed from 2012 to 2019. Among all the respondents, 44.01% use gas, liquefied gas or natural gas; 29.24% use firewood; 20.84% use electricity; 5.18% use coal; and less than 1% use solar energy or biogas. Firewood is still an important cooking fuel in China especial in rural areas(Liao,et al. 2018, Peng et al., 2016). The participants’ mean age is 47 years; 48.32% are men and 51.68% women; 80.20% are married or cohabiting. Their average education proves to be 6.880 years (SD = 4.995), and the mean household per capita income is 15,364 RMB (SD = 24,154 RMB) (2362 US, SD = 3716 US). In addition, 28.23% of participants smoke, 15.23% drink and 43.24% exercise. 48.11% live in urban areas, and 51.89% live in rural areas.

4.2 Associations between cooking fuels and self-rated health

We collected data from 162 cities in 2012–2019 and structured the sample into two levels, individual level (Level 1) nested inside the city level (Level 2). Given the complexity of design, we employ a multivariable linear regression model to estimate the correlation between cooking fuels and health status. In Table 1 (see Supplementary Table 3), Column (1) to Column (3) indicate the regression results for the overall dataset, and for men and women, respectively. Following controlling for individual characteristics and time effects, we observe that coal, gas/liquefied gas/natural gas, solar energy/biogas and electricity have a positive relationship with respondents’ self-rated health, compared with firewood (Capuno, 2018; Quinn et al., 2018), in agreement with previous research (Silwal and McKay, 2015). When using coal, coal gas/Liquefied gas/natural gas, solar energy/biogas and electricity as cooking fuels, Self-rated health scores increased by 0.068 (SE = 0.018), 0.097 (SE = 0.097), 0.123 (SE = 0.043), and 0.091 (SE = 0.012) respectively. Nearly one third of individuals use firewood as a cooking fuel, indicating the scope of the potential health risk that it poses via household air pollution in China. Moreover, within the four cooking fuels, solar energy/biogas has the greatest positive impact on respondents’ self-rated health, followed by gas/liquefied gas/natural gas and electricity. As renewable energy sources, solar energy and biogas yield the benefits of both low household emissions and low household pollution (Baul et al., 2018). Consider that these renewable resources account for less than 1% of cooking, their usage could be expanded to ease household air pollution and its associated health risks.

Column (2) and Column (3) in Table 1 are the results of hierarchical regression according to gender. The relationship of coal usage with the respondents’ self-rating is smaller for women, while gas/liquefied gas/natural gas, solar energy/biogas, and electric cooking fuel appear relatively large. This situation can be explained by social divisions: women are more engaged in cooking activities, so household air pollution caused by cooking fuel has a larger impact on their health (Jones et al., 1983; McLean et al., 2019ab).

These results strongly imply that policy formulation must take into consideration the health risks due to air pollution caused by cooking fuels. We have established a theoretical basis for the implementation of the “coal-to-gas” and “coal-to-electricity” transition policies in China from the health risks perspective.

Table 1 Association of cooking fuels and self-rated health.
Variable	Dependent variable: Self-rated health
	(1)		(2)		(3)
	Overall		Men		Women
	Coef.	Std. Err.	Coef.	Std. Err.	Coef.	Std. Err.
Coal	0.068***	0.018	0.072***	0.026	0.068***	0.026
Coal gas/Liquefied petroleum gas/Pipeline natural gas	0.097***	0.011	0.076***	0.016	0.116***	0.016
Solar/Biogas	0.123***	0.043	0.089	0.061	0.154**	0.060
Electricity	0.091***	0.012	0.086***	0.017	0.097***	0.016
Control variable	Yes		Yes		Yes
Year-by-month FE	Yes		Yes		Yes
County EF	Yes		Yes		Yes
Log likelihood	–154,883		–74,283		–80,163
N	101,712		49,142		52,570
Notes: *p < 0.01; p < 0.05; *p < 0.1. Estimated coefficient and robust standard errors were reported for variables.

4.3 Robustness checks

Index of self-rated health only involves 5 levels. It does not include data below 0 or over 5. In Table 2 part A, Given that the scores on the self-rated health scale were truncated, we used the Tobit model. Coal (women) = 0.038 is not significant in comparison with coal (women) = 0.068 in Table 1. however, the remain result of regression appears unchanged.

In a bid to ensure the integrity of data and reliability of the conclusion, multiple samples with missing metrics are removed. To include more samples, control variables do not include serious missing metrics such as income, maritality, city, smoking, alcohol consumption, and exercise. See Table 2 part B. The number of samples increases from 101,712 to 116,467. The result of regression do not change compared with original data.

Panel data has two dimensions, time series and cross-section. When such data is arranged in two dimensions, it is arranged on a plane; moreover, both individual fixed effect and time fixed effect are explored. Compared with Hierarchical Linear Model, Panel data model can sustain more information. See Table 2 part C. Coal (women) = 0.039 is not significant compared with coal (women) = 0.068 in Table 1. Similar to part A, the remain result of regression endures.

Among the factors which could cause health problem, cooking fuels can not be found. For some reasons, low income group face more health problem. In order to solve problem, we remove samples with income below quartiles and subjective self-assessment of health as unhealthy, so that prevents cooking fuel selection from potentially affecting by income and health issues. See Table 2 part D. The number of samples shrinks from 101,712 to 99,415. The result of regression does not vary compared with original data.

Table 2 Robustness checks.
Variable	Dependent variable: Self-rated health
	(1)		(2)		(3)
	Overall		Men		Women
	Coef.	Std. Err.	Coef.	Std. Err.	Coef.	Std. Err.
A. Tobit model
Coal	0.044**	0.018	0.053**	0.026	0.038	0.025
Coal gas/Liquefied petroleum gas/Pipeline natural gas	0.063***	0.011	0.050***	0.016	0.075***	0.016
Solar/Biogas	0.115***	0.042	0.051	0.059	0.177***	0.058
Electricity	0.064***	0.012	0.056***	0.016	0.073***	0.016
Log likelihood	–148,391		–71,167		–77,877
N	101,712		49,142		52,570
B. Adjust sample size
Coal	0.068***	0.017	0.079***	0.025	0.060**	0.024
Coal gas/Liquefied petroleum gas/Pipeline natural gas	0.092***	0.010	0.079***	0.015	0.108***	0.014
Solar/Biogas	0.116***	0.041	0.084	0.058	0.151***	0.057
Electricity	0.082***	0.011	0.073***	0.015	0.085***	0.015
Log likelihood	–177838		–86,118		–91,234
N	116,467		56,661		59,806
C. Panel data model
Coal	0.045**	0.018	0.054**	0.026	0.039	0.025
Coal gas/Liquefied petroleum gas/Pipeline natural gas	0.065***	0.011	0.051***	0.016	0.077***	0.016
Solar/Biogas	0.115***	0.042	0.052	0.059	0.176***	0.059
Electricity	0.065***	0.012	0.057***	0.016	0.074***	0.016
R − squared	0.270		0.257		0.281
N	101,712		49,142		52,570
D. Remove samples to prevent endogenesis
Coal	0.050***	0.019	0.057**	0.026	0.045*	0.026
Coal gas/Liquefied petroleum gas/Pipeline natural gas	0.088***	0.011	0.072***	0.016	0.103***	0.016
Solar/Biogas	0.112***	0.043	0.107*	0.061	0.117**	0.059
Electricity	0.077***	0.012	0.077***	0.017	0.077***	0.017
Log likelihood	–150,235		–72,327		–77,475
N	99,415		48,257		51,158
Control variable	Yes		Yes		Yes
Year-by-month FE	Yes		Yes		Yes
County EF	Yes		Yes		Yes
Notes: *p < 0.01; p < 0.05; *p < 0.1. Estimated coefficient and robust standard errors were reported for variables. Part A: Estimate by tobit model; Part B: To include more samples, control variables do not include serious missing metrics such as income, maritality, city, smoking, alcohol consumption, and exercise; Part C: Estimate by panel data models; Part D: Remove samples with income less than quartiles and subjective self-assessment of health as unhealthy. It prevents cooking fuel selection from potentially affecting by income and health issues.

4.4 Associations between cooking fuels and other health status

Individual health comprises two dimensions, both physical and mental. We utilize mental health indicators to reflect residents’ mental health status and interviewer-assessed health index to find the relationship between cooking fuels and individuals’ health status. In Table 3, Column (1) to Column (3) show the regression results of the overall sample, men, and women respectively. Part A and Part B represent the regression results of the interviewer-observed health status and the residents’ mental health indicators.

In conclusion, all indexes are consistent with the main results: compared with firewood, all of coal, coal gas/liquefied gas/natural gas, solar energy/biogas are positively associated with interviewer-observed health status and the residents’ mental health, albeit to differing extents. In Parts A–B (Table 3), Column (1) briefs the impact of different cooking fuels on the interviewer’s observation health status and the residents’ mental health. It has consistent results with Table 2 Column (1). From the perspective of mental health, the cooking fuels, ranked from the highest to the lowest positive relationships, are solar energy/biogas, coal gas/liquefied gas/natural gas, electricity, and coal. Give priority to interviewer-assessed health, the rank is solar energy/biogas, coal gas/liquefied gas/natural gas and electricity. A gender-stratified regression for other health indicators produces conclusions basically consistent with the main results above.

Table 3 Association of cooking fuels and other health status.
Variable	(1)		(2)		(3)
	Overall		Men		Women
	Coef.	Std. Err.	Coef.	Std. Err.	Coef.	Std. Err.
A. Dependent variable: Interviewer-assessed health
Coal	0.077***	0.019	0.093***	0.027	0.066***	0.026
Coal gas/Liquefied petroleum gas/Pipeline natural gas	0.161***	0.012	0.149***	0.016	0.172***	0.016
Solar/Biogas	0.286***	0.043	0.253***	0.061	0.318***	0.059
Electricity	0.093***	0.012	0.094***	0.017	0.094***	0.017
Control variable	Yes		Yes		Yes
Year-by-month FE	Yes		Yes		Yes
County EF	Yes		Yes		Yes
Log likelihood	–145,586		–69,485		–75,891
N	96,405		46,328		50,077
B. Dependent variable: Mental health
Coal	–0.014	0.012	–0.031*	0.017	0.003	0.018
Coal gas/Liquefied petroleum gas/Pipeline natural gas	0.047***	0.008	0.036***	0.011	0.059***	0.011
Solar/Biogas	0.074**	0.029	0.038	0.040	0.110***	0.041
Electricity	0.030***	0.008	0.022**	0.011	0.039***	0.011
Control variable	Yes		Yes		Yes
Year-by-month FE	Yes		Yes		Yes
County EF	Yes		Yes		Yes
Log likelihood	–110,809		–51,902		–58,433
N	100,033		48,489		51,544
Notes: *p < 0.01; p < 0.05; *p < 0.1. Estimated coefficient and robust standard errors were reported for variables.

4.5 Association between household air pollution and self-rated health

To examine the effects of individual exposure to household air pollution caused by cooking fuel, we use two indicators (cleanliness and exposure) to examine the health effects by different cooking fuels. In Table 4, Column (1) to Column (3) illustrate the regression results for the overall sample, men, and women, respectively, and Part A and Part B represent the regression results via using the cleanliness and exposure indicators, respectively.

After adjusting individual characteristics and other variables, results reveal that higher cleanliness is associated with more positive measures of residents’ health. The reason behind may be that cleaner fuels produces fewer harmful gas emissions, thereby reducing household air pollution (Armah et al., 2019). By the same token, after adjusting for individual characteristics and other variables, lower exposure was found to associate with more positive measures of residents’ health. A plausible explanation is that cooking fuels with lower exposure level produce less household air pollution, and then following less adverse effects on residents’ health.

We use cleanliness and exposure to construct gender-stratified regressions (Table 4, Columns (2)–(3), Parts A and B). The results are consistent with the basic results (Table 4, Column (1), Parts A and B).

The above research results accord with the results of Rosenthal et al.’s (2018), who claimed that cooking fuels with high cleanliness and low pollution exposure brought not only greater potential benefits to health, but also contributed more in achieving climate and other sustainable development goals.

Table 4 Association of exposure to household air pollution and self-rated health.
Variable	Dependent variable: Self-rated health
	(1)		(2)		(3)
	Overall		Men		Women
	Coef.	Std. Err.	Coef.	Std. Err.	Coef.	Std. Err.
A. Three category household air pollution exposure variable
Medium cleanliness of the fuel	0.086***	0.011	0.065***	0.016	0.104***	0.015
High cleanliness of the fuel	0.080***	0.011	0.074***	0.016	0.086***	0.016
Control variable	Yes		Yes		Yes
Year-by-month FE	Yes		Yes		Yes
County EF	Yes		Yes		Yes
Log likelihood	–154,890		–74,287		–80,167
N	101,712		49,142		52,570
B. Binary exposure variable
Low exposure	0.083***	0.010	0.069***	0.014	0.095***	0.013
Control variable	Yes		Yes		Yes
Year-by-month FE	Yes		Yes		Yes
County EF	Yes		Yes		Yes
Log likelihood	–154,890		–74,287		–80,167
N	101,712		49,142		52,570
Notes: *p < 0.01; p < 0.05; *p < 0.1. Estimated coefficient and robust standard errors were reported for variables. High cleanliness of the fuel, Solar/Biogas/Electricity; Medium cleanliness of the fuel, Coal/Coal gas/Liquefied petroleum gas/Pipeline natural gas; Low cleanliness of the fuel, Firewood; High exposure, Firewood; Low exposure, Coal gas/Liquefied petroleum gas/Pipeline natural gas/Solar/Biogas/Electricity

4.6 Subgroup analysis of the relationship between cooking fuels and self-rated health, by income

We performed a stratified regression based on the level of household per capita income. To avoid the impact of extreme values on sample, we take the median (($10,060) as the dividing line between low and high income (Owens, 2018; Bell et al., 2019) (see Table 5). Furthermore, we also try to take the average ($15,364) as the dividing line between low and high income (see Supplementary Table 4). In Table 5, Column (1) to Column (3) show the regression results of the overall sample, men, and women, respectively. Part A and Part B represent the regression results for the low-income group and the high-income group, respectively. We find no significant difference for this association between low-income group and the overall sample (Table 5 Column (1) Part A), while the coefficients for all types of cooking fuel in high-income group are lower than those of the low-income group (Table 5 Column (1) Part B). These differences are not statistically noticeable (Table 5 Column (1) Part B), which testifies that there exists a weaker relationship between cooking fuel choice and self-rated health among the high-income group. One rationable might be that high-income people are more aware of household pollution and able to avoid it. 2nd Column and 3rd Column (see Table 5) are the results of a gender-stratified regression. In the high-income group, only gas/liquefied gas/natural gas in the male sub-sample exhibit a negative association with subjective self-rated health, with other types of cooking fuel showing no significant associations. However, in low-income group, all cooking fuels in women’s sub-sample are positively associated with subjective self-rated health. This result reflects the characteristic social division of labor between men and women. In traditional Chinese context, women usually do the home cooking, so they face more risks of cooking-associated air pollution (Peng et al., 2016).

In order to compare the relationship between cooking fuel use and subjective self-rated health in different income groups, we introduce an interaction term between income groups and cooking fuel, and use dummy variable regression to further examine the relationships among income, cooking fuel and residents’ health. Table 6, Column (1) to Column (3) illustrate the regression results of the overall sample, men, and women, respectively. There are considerable differences in the size and significance of the interaction coefficients between the income group dummy variables and the cooking fuel use variables. Interestingly, income has a more positive association with the health risks of firewood and electricity. Notably, in Table 6 Column (1), the health self-rating coefficients for high-income families are 0.140 for firewood, and 0.156 for electricity. However, in low-income group, they are 0 and 0.119. High income is therefore associated with less adverse health effects due to solid fuels, and greater health benefit from clean energy. This result suggests that increasing household income can help lower the health risks caused by cooking fuels.

Table 5 Association of cooking fuels and self-rated health: By income group.
Variable	Dependent variable: Self-rated health
	(1)		(2)		(3)
	Overall		Men		Women
	Coef.	Std. Err.	Coef.	Std. Err.	Coef.	Std. Err.
A. Low income group
Coal	0.102***	0.024	0.101***	0.034	0.104***	0.032
Coal gas/Liquefied petroleum gas/Pipeline natural gas	0.135***	0.016	0.124***	0.023	0.142***	0.022
Solar/Biogas	0.136***	0.057	0.072	0.082	0.199**	0.079
Electricity	0.112***	0.016	0.132***	0.022	0.092***	0.022
Control variable	Yes		Yes		Yes
Year-by-month FE	Yes		Yes		Yes
County EF	Yes		Yes		Yes
Log likelihood	–79,222		–37,453		–41,509
Observations	50,854		24,209		26,645
B. High income group
Coal	–0.023	0.031	–0.013	0.043	–0.029	0.043
Coal gas/Liquefied petroleum gas/Pipeline natural gas	0.007	0.017	–0.040*	0.024	0.055**	0.024
Solar/Biogas	0.059	0.065	0.073	0.091	0.041	0.092
Electricity	0.018	0.018	–0.033	0.026	0.069***	0.026
Control variable	Yes		Yes		Yes
Year-by-month FE	Yes		Yes		Yes
County EF	Yes		Yes		Yes
Log likelihood	–75,134		–36,465		–38,325
N	50,858		24,933		25,925
Notes: *p < 0.01; p < 0.05; *p < 0.1. Estimated coefficient and robust standard errors were reported for variables. High and low income are grouped by median household income per capita ($10,060).

Table 6 Association of cooking fuels, income group (median) and self-rated health.
Variable	Dependent variable: Self-rated health
	(1)		(2)		(3)
	Overall		Men		Women
	Coef.	Std. Err.	Coef.	Std. Err.	Coef.	Std. Err.
Coal	0.101***	0.022	0.101***	0.032	0.101***	0.030
Coal gas/Liquefied petroleum gas/Pipeline natural gas	0.143***	0.014	0.138***	0.020	0.146***	0.019
Solar/Biogas	0.134**	0.054	0.069	0.078	0.194**	0.075
Electricity	0.119***	0.014	0.133***	0.020	0.106***	0.020
High income×Firewood	0.140***	0.016	0.184***	0.022	0.097***	0.022
High income×Coal	0.018	0.033	0.067	0.047	–0.021	0.046
High income×(Coal gas/Liquefied petroleum gas/Pipeline natural gas)	0.010	0.014	0.011	0.020	0.010	0.019
High income×(Solar/Biogas)	0.081	0.084	0.194*	0.119	–0.027	0.118
High income×Electricity	0.037***	0.017	0.0328	0.024	0.045*	0.024
Control variable	Yes		Yes		Yes
Year-by-month FE	Yes		Yes		Yes
County EF	Yes		Yes		Yes
Log likelihood	–154,841		–74,247		–80,152
N	101,712		49,142		52,570
Notes: *p < 0.01; p < 0.05; *p < 0.1. Estimated coefficient and robust standard errors were reported for variables. High and low income are grouped by median household income per capita ($10,060).

On the basis of investigation, this research exists as the very first study to address the potential relationship between household air pollution and health using data from a large micro-tracking survey in China, with over 100,000 samples covering 162 counties from CFPS 2012–2019. We evaluated the potential impact of household air pollution caused by cooking fuels on the health of residents and deal with methods to mitigate the adverse effects on residents’ health by improving indoor air quality and increasing household per capita income. This study finds that in compason with firewood, solar energy and biogas are more conducive to residents’ health—the cleaner the cooking fuel and the lower the exposure, the better the health of residents.

Previous studies have also demonstrate that solid fuels increase residents’ health risks, while clean fuels reduce residents’ health risks (Rosenthal et al., 2018; Foell et al., 2011; Smith, 2015). The main reasons for household air pollution involve carbon monoxide gas, smoke and other tiny particles caused by incomplete combustion. These substances are mixed in the air. Long-term exposure to this environment will intensifies the risks of heart disease, high blood pressure, and cancer (Yu et al., 2020).

Existing research focuses mainly on developing countries or regions. To cite an example, Ravilla (2016) used population data from northern and southern India to discuss the impact of cooking fuel on women’s cataracts, and provided strong evidence for a relationship between biomass fuels and women’s cataracts; Das (2017) utilized a cross-sectional sample of 655 households in Malawi to examine the relationship between biomass-powered cooking and women’s health. The study found that high-quality firewood was associated with significantly lower odds of persistent phlegm. There are other studies that focus on vulnerable groups including children. For example, Kilabuko and Nakai (2007) used data from the Tanzania Demographic and Health survey 2004–2005 to shed light on charcoal and kerosene use on ARI in children under five years old in Tanzania. The findings indicate that it is essential to actualize a shift from the use of biomass fuels, charcoal and kerosene for cooking to clean sources such as gas and electricity so as to achieve meaningful reduction of Acute Respiratory Infection (ARI) prevalence in Tanzania. Wong et al. (2013) used data collected between 1999 and 2004, from 512,707 primary and secondary school children in 47 countries, to investigate the association between asthma and a range of cooking fuels around the world, and found that open-fire cooking was associated with an increased risk of asthma symptoms.

As a developing country with a vast, diverse territory and a huge population, China is driven to explore potential relationships between household air pollution, domestic practices, and residents’ health. Furthermore, the health effects of household air pollution are heterogeneous in that the health of low-income people is more susceptible to household air pollution. In order to identify the relationship between income, household air pollution and health status, we introduced interaction terms between income groups and cooking fuels. We find that high-income households benefit more from good cooking fuels. The high-income group can not only reduce the adverse effects due to firewood combustion, but also plays benefit more from electric cooking. This conclusion is significant for formulating energy-related and fuel-related policies that consider income heterogeneity but also contributes to the realization of environmental justice.

We operate a hierarchical linear model to examine the potential impact of household air pollution caused by cooking fuel on residents’ health, while there also exists some limitations that should be considered when interpreting our findings. First of all, we take the cooking fuel type as indicative of household air pollution in our study, but the actual degree of household air pollution is tasking to measure. The investigation can be facilitated by using an instrument better able to measure household air pollution, and to associate overall household air pollution with the pollution caused specifically by cooking. Second, we fail to prove the biological or medical mechanisms of household air pollution on the health of residents. only statistical analysis and quantitative regression methods are adapted to verify the potential impact of household air pollution on health.

At present, about 30% of Chinese households still use low-cleanliness cooking fuel, and high-cleanliness cooking fuel accounts for less than 25% of the total (Tang and Liao, 2014). Clean cooking fuels, especially high-cleanliness and environmentally friendly fuels, have wide scope for popularized application. We recommend speeding up and expanding the implementation of the “coal-to-gas” and “coal-to-electricity” energy transition policies, to promote the improvement of residents’ health and to enhance the even distribution of health resources. Moreover, we propose that government strengthen health education for low-income households to raise awareness of household air pollution prevention and to encourage cleaner cooking fuel choices.

Ethics approval and consent to participate

Ethics approval was obtained from the Institute of Social Science Survey (Peking University).

Consent to Publish

All authors agree to publish

Authors’ contributions

ZY contributed study design. ZY and XZ conducted all data processing, analyses and initial drafting. JL, GZ and LF contributed to the interpretation of the results. GZ, LF and CL were involved in the study supervision. All authors contributed in drafting the manuscript and revision. All authors read and approved the final manuscript.

Funding

This work was supported by the Fundamental Research Funds for the Central Universities (No. FRF-BR-20-04A).

Competing interests

The authors declare that they have no conflict of interest.

Availability of data and materials

China Family Panel Studies (CFPS):

The datasets generated and/or analysed during the current study are available in the Institute of social science survey, Peking University repository, https://doi.org/10.18170/DVN/45LCSO.

Acknowledgements

We would like to acknowledge the China Family Panel Studies (CFPS) team for providing data and the training of using the dataset.

Agrawal S, Yamamoto S (2015) Effect of indoor air pollution from biomass and solid fuel combustion on symptoms of preeclampsia/eclampsia in Indian women. Indoor Air 25(3):341–352
Arku RE, Birch A, Shupler M, Yusuf S, Hystad P, Brauer M (2018a) Characterizing exposure to household air pollution within the Prospective Urban Rural Epidemiology (PURE) study. Environ Int 114:307–317
Arku RE, Ezzati M, Baumgartner J, Fink G, Zhou B, Hystad P, Brauer M (2018b) Elevated blood pressure and household solid fuel use in premenopausal women: Analysis of 12 Demographic and Health Surveys (DHS) from 10 countries. Environ Res 160:499–505
Armah FA, Ekumah B, Yawson DO (2019) Predictive Probabilities of Access to Clean Cooking: Evidence from the Demographic and Health Surveys in 31 Countries in Sub-Saharan Africa. Environ Justice 12(3):118–131
Bai J (2013) Fixed-effects dynamic panel models, a factor analytical method. Econometrica 81(1):285–314
Balmes JR (2019) Household air pollution from domestic combustion of solid fuels and health. J Allergy Clin Immunol 143(6):1979–1987
Baul TK, Datta D, Alam A (2018) A comparative study on household level energy consumption and related emissions from renewable (biomass) and non-renewable energy sources in Bangladesh. Energy Policy 114:598–608
Begum BA, Paul SK, Hossain MD, Biswas SK, Hopke PK (2009) Indoor air pollution from particulate matter emissions in different households in rural areas of Bangladesh. Build Environ 44(5):898–903
Bell A, Chetty R, Jaravel X, Petkova N, Van Reenen J (2019) Who becomes an inventor in America? The importance of exposure to innovation. Q J Econ 134(2):647–713
Capuno JJ, Tan CAR Jr, Javier X (2018) Cooking and coughing: Estimating the effects of clean fuel for cooking on the respiratory health of children in the Philippines. Glob Public Health 13(1):20–34
Carter E, Norris C, Dionisio KL, Balakrishnan K, Checkley W, Clark ML, Ghosh S, Jack DW, Kinney PL, Marshall JD, Naeher LP, Peel JL, Sambandam S, Schauer JJ, Smith KR, Wylie BJ, Baumgartner J (2017) Assessing Exposure to Household Air Pollution: A Systematic Review and Pooled Analysis of Carbon Monoxide as a Surrogate Measure of Particulate Matter.Environmental Health Perspectives, 125(7),076002
Chen Y, Shen H, Smith KR, Guan D, Chen Y, Shen G, Liu J, Cheng H, Zeng EY, Tao S (2018) Estimating household air pollution exposures and health impacts from space heating in rural China. Environ Int 119:117–124
Chowdhury Z, Campanella L, Gray C, Masud AA, Marter-Kenyon J, Pennise D, Charron D, Xia Z (2013) Measurement and modeling of indoor air pollution in rural households with multiple stove interventions in Yunnan, China. Atmos Environ 67:161–169
Clarke P, Sproston K, Thomas R (2003) An investigation into expectation-led interviewer effects in health surveys. Soc Sci Med 56(10):2221–2228
Clasen T, Smith KR (2019) Let the “A” in WASH Stand for Air: Integrating Research and Interventions to Improve Household Air Pollution (HAP) and Water, Sanitation and Hygiene (WaSH) in Low-Income Settings. Environ Health Perspect 127(2):025001
Das I, Jagger P, Yeatts K (2017) Biomass Cooking Fuels and Health Outcomes for Women in Malawi. EcoHealth 14:7–19
Diez-Roux AV (1998) Bringing context back into epidemiology: variables and fallacies in multilevel analysis. Am J Public Health 88(2):216–222
Dionisio KL, Howie S, Fornace KM, Chimah O, Adegbola RA, Ezzati M (2010) Measuring the exposure of infants and children to indoor air pollution from biomass fuels in The Gambia. Indoor Air 18(4):317–327
Duncan JM (1996) State of the Art: Limit Equilibrium and Finite-Element Analysis of Slopes. J Geotech Eng 123(7):577–596
Fischer SL (2001) Biomass-derived liquid cooking fuels for household use in rural China: potential for reducing health costs and mitigating greenhouse gas emissions. Energy Sustain Dev 5(1):23–30
Foell W, Pachauri S, Spreng D, Zerriffi H (2011) Household cooking fuels and technologies in developing economies. Energy Policy 39(12):7487–7496
GBD 2017 Risk Factor Collaborators (2018) Global, regional, and national comparative risk assessment of 84 behavioural, environmental and occupational, and metabolic risks or clusters of risks for 195 countries and territories, 1990–2017: a systematic analysis for the Global Burden of Disease Study 2017. Lancet 392(10159):1923–1994
Hänninen O, Asikainen A, Leal V, Santos HRR, Wargocki P, Carrer P, Jantunen M, Fernandes EDO (2013) European Guidelines for Ventilation (HealthVent –project): Implications on Energy and Health. Environment and Health–Bridging South, North, East and West. Conference of ISEE, ISES and ISIAQ
He G, Liu T, Zhou M (2020) Straw burning, PM2.5, and death: Evidence from China. J Dev Econ 145:102468
Hystad P, Duong ML, Brauer M, Larkin A, Arku R, Kurmi OP, Fan WQ, Avezum A, Azam I, Chifamba J, Dans A, du Plessis JL, Gupta R, Kumar R, Lanas F, Liu Z, Lu Y, Lopez-Jaramillo P, Mony P, Mohan V, Mohan D, Nair S, Puoane T, Rahman O, Lap AT, Wang Y, Wei L, Yeates K, Rangarajan S, Teo K, Yusuf S (2019) Health Effects of Household Solid Fuel Use: Findings from 11 Countries within the Prospective Urban and Rural Epidemiology Study. Environ Health Perspect 127(5):057003
Jones JR, Higgins IT, Higgins MW, Keller JB (1983) Effects of Cooking Fuels on Lung Function in Nonsmoking Women. Arch Environ Health 38(4):219–222
Kan X, Chiang CY, Enarson DA, Chen W, Yang J, Chen G (2011) Indoor solid fuel use and tuberculosis in China: a matched case-control study. BMC Public Health 11:498
Khan MSB, Lohano HD (2018) Household air pollution from cooking fuel and respiratory health risks for children in Pakistan. Environ Sci Pollut Res 5:24778–24786
Kilabuko JH, Nakai S (2007) Effects of Cooking Fuels on Acute Respiratory Infections in Children in Tanzania. Int J Environ Res Public Health 4(4):283–288
Kim C, Gao YT, Xiang YB, Barone-Adesi F, Zhang Y, Hosgood HD, Ma S, Shu X, Ji BT, Chow WH, Seow WJ, Bassig B, Cai Q, Zheng W, Rothman N, Lan Q (2015) Home kitchen ventilation, cooking fuels, and lung cancer risk in a prospective cohort of never smoking women in Shanghai, China. Int J Cancer 136(3):632–638
Kurmi OP, Lam KBH, Ayres JG (2012) Indoor air pollution and the lung in low- and medium-income countries. Eur Respir J 40(1):239–254
Lakshmi PVM, Virdi NK, Sharma A, Tripathy JP, Smith KR, Bates MN, Kumar R (2013) Household air pollution and stillbirths in India: Analysis of the DLHS-II National Survey. Environ Res 121:17–22
Li L, Yang A, He X, Liu J, Ma Y, Niu J, Luo B (2020) Indoor air pollution from solid fuels and hypertension: A systematic review and meta-analysis. Environ Pollut 259:113914
Li X, Liu B (2015) The effect of health changes on self-reported quality of life of the elderly in China: Fixed effects model analysis based on CLHLS data. Popul Econ 6:1–11 (in Chinese)
Liu Z, Li J, Rommel J, Feng S (2020) Health impacts of cooking fuel choice in rural China. Energy Econ 89:104811
McLean EV, Bagchi-Sen S, Atkinson JD, Ravenscroft J, Hewner S, Schindel A (2019a) Country-level analysis of household fuel transitions. World Dev 114:267–280
McLean EV, Bagchi-Sen S, Atkinson JD, Schindel A (2019b) Household dependence on solid cooking fuels in Peru: An analysis of environmental and socioeconomic conditions. Glob Environ Change 58:101961
Mestl HES, Aunan K, Seip HM (2007) Health benefits from reducing indoor air pollution from household solid fuel use in China—Three abatement scenarios. Environ Int 33(6):831–840
Mishra V (2003) Effect of indoor air pollution from biomass combustion on prevalence of asthma in the elderly. Environ Health Perspect 111(1):71–78
Nadimi R, Tokimatsu K (2018) Energy use analysis in the presence of quality of life, poverty, health, and carbon dioxide emissions. Energy 153(15):671–684
Oguonu T, Obumneme-Anyim LN, Eze JN, Ayuk AC, Okoli CV, Ndu LK (2018) Prevalence and determinants of airflow limitation in urban and rural children exposed to cooking fuels in South-East Nigeria. Paediatrics and International Child Health 38(2):121–127
Owens A (2018) Income segregation between school districts and inequality in students’ achievement. Sociol Educ 91(1):1–27
Pant KP (2012) Cheaper Fuel and Higher Health Costs Among the Poor in Rural Nepal. Ambio 41:271–283
Prince M, Patel V, Saxena S, Maj M, Maselko J, Phillips MR, Rahman A (2007) No health without mental health. The Lancet 370(9590):859–877
Qiu Y, Yang FA, Lai W (2019) The impact of indoor air pollution on health outcomes and cognitive abilities: empirical evidence from China. Popul Environ 40(4):388–410
Quinn A, Bruce N, Puzzolo E, Dickinson K, Sturke R, Jack DW, Mehta S, Shankar A, Sherr K, Rosenthal J (2018) An analysis of efforts to scale up clean household energy for cooking around the world. Energy Sustain Dev 46:1–10
Ravilla T (2016) Inequities in eye care in South Asia. Community Eye Health Journal 29(95):S01–S03
Rehfuess E, Mehta S, Prüss-Ustün A (2006) Assessing household solid fuel use: multiple implications for the millennium development goals. Environ Health Perspect 114(3):373–378
Rehfuess EA, Tzala L, Best N, Briggs DJ, Joffe M (2009) Solid fuel use and cooking practices as a major risk factor for ALRI mortality among African children. J Epidemiol Community Health 63(11):887–892
Rosenthal J, Quinn A, Grieshop AP, Pillarisetti A, Glass R (2018) Clean cooking and the SDGs: Integrated analytical approaches to guide energy interventions for health and environment goals. Energy Sustain Dev 42:152–159
Sapkota A, Zaridze D, Szeszenia-Dabrowska N, Mates D, Fabiánová E, Rudnai P, Janout V, Holcatova L, Brennan P, Boffeta P, Hashibe M (2013) Indoor air pollution from solid fuels and risk of upper aerodigestive tract cancers in Central and Eastern Europe. Environ Res 120:90–95
Shupler M, Hystad P, Birch A, Miller-Lionberg D, Jeronimo M, Arku RE, Chu YL, Mushtaha M, Heenan L, Rangarajan S, Seron P, Lanas PF, Cazor F, Lopez-Jaramillo P, Camacho PA, Perez M, Yeates K, West N, Brauer M (2020) Household and personal air pollution exposure measurements from 120 communities in eight countries: results from the PURE-AIR study. Lancet Planet Health 4(10):451–462
Sidhu MK, Ravindra K, Mor S, John S (2017) Household air pollution from various types of rural kitchens and its exposure assessment. Sci Total Environ 586:419–429
Silwal AR, McKay A (2015) The Impact of Cooking with Firewood on Respiratory Health: Evidence from Indonesia. J Dev Stud 51(12):1619–1633
Smith KR (2015) Changing Paradigms in Clean Cooking. EcoHealth 12(1):196–199
Taghizadeh-Hesary F, Rasoulinezhad E, Yoshino N, Chang Y, Taghizadeh-Hesary F, Peter M (2020) The energy-pollution-health nexus: A Panel data analysis of low-and middle-income Asian nations. ADB Economics Working Paper Series. Available at: https://www.adb.org/sites/default/files/publication/571736/adbi-wp1086.pdf [Accessed 26 June. 2021]
Tang X, Liao H (2014) Energy poverty and solid fuels use in rural China: Analysis based on national population census. Energy Sustain Dev 23:122–129
Tao MH, Zhou J, Rialdi AP, Martinez R, Dabek J, Scelo G, Lissowska J, Chen J, Boffetta P (2014) Indoor air pollution from solid fuels and peripheral Blood DNA methylation: Findings from a population study in Warsaw, Poland. Environ Res 134:325–330
Thomas E, Wickramasinghe K, Mendis S, Roberts N, Foster C (2015) Improved stove interventions to reduce household air pollution in low and middle income countries: a descriptive systematic review. BMC Public Health 15(1):650
Tian Z, Tian Y, Shen L, Shao S (2021) The health effect of household cooking fuel choice in China: An urban-rural gap perspective. Technol Forecast Soc Chang 173:121083
van der Kroon D, Brouwer R, van Beukering PV (2013) The energy ladder: Theoretical myth or empirical truth? Results from a meta-analysis. Renew Sustain Energy Rev 20:504–513
Wang Q, Kwan MP, Zhou K, Fan J, Wang Y, Zhan D (2019) Impacts of residential energy consumption on the health burden of household air pollution: Evidence from 135 countries. Energy Policy 128:284–295
Wang Q, Liu P, Yuan X, Cheng X, Ma R, Mu R, Zou J (2015) Structural Evolution of Household Energy Consumption: A China Study. Sustainability 7(4):3919–3932
WHO (2021) Household air pollution data. Available at: https://www.who.int/data/gho/data/themes/air-pollution/household-air-pollution [Accessed 26 June. 2021]
Wong GWK, Brunekreef B, Ellwood P, Anderson HR, Asher MI, Crane J, Lai CKW, for the ISAAC Phase Three Study Group (2013) Cooking fuels and prevalence of asthma: a global analysis of phase three of the International Study of Asthma and Allergies in Childhood (ISAAC). The Lancet Respiratory Medicine 1(5):386–394
Wu C, Zheng X, You C, Wei C (2019) Household energy consumption in rural China: Historical development, present pattern and policy implication. J Clean Prod 211:981–991
Yang KW, He H (2020) The Impact of Internet Usage on Health of Residents: A Study from the China Labor-force Dynamic Survey 2016. Nankai Economic Studies 3:182–203 (Chinese)
Yasar A, Nazir S, Tabinda AB, Nazar M, Rasheed R, Afzaal M (2017) Socio-economic, health and agriculture benefits of rural household biogas plants in energy scarce developing countries: A case study from Pakistan. Renewable Energy 108:19–25
Yu K, Lv J, Qiu G, Yu G, Guo Y, Bian Z, Yang L, Chen Y, Wang C, Pan A, Liang L, Hu FB, Chen Z, Li L, Wu T (2020) Cooking fuels and risk of all-cause and cardiopulmonary mortality in urban China: a prospective cohort study. The Lancet Global Health 8(3):430–439
Yu K, Qiu G, Chan KH, Chan KH, Lam KBH, Kurmi OP, Bennett AD, Yu C, Pan A, Lv J, Guo Y, Bian Z, Yang L, Chen Y, Hu FB, Chen Z, Li L, Wu T (2018) Association of Solid Fuel Use With Risk of Cardiovascular and All-Cause Mortality. Rural China JAMA 319(13):1351–1361
Zhou Y, Zou Y, Li X, Chen S, Zhao Z, He F, Zou W, Luo Q, Li W, Pan Y, Deng X, Wang X, Qiu R, Liu S, Zheng J, Zhong N, Ran P (2014) Lung Function and Incidence of Chronic Obstructive Pulmonary Disease after Improved Cooking Fuels and Kitchen Ventilation: A 9-Year Prospective Cohort Study. PLoS Med 11(3):e1001621
Zhu H, Brambley MR, Morgan RP (1983) Household energy consumption in the People’s Republic of China. Energy 8(10):763–774

SupplementaryTables.docx

Download PDF

Version 1

posted

You are reading this latest preprint version

The relationship between cooking fuel and health status from the perspective of income heterogeneity

Status:

Version 1

Abstract

1 Introduction

2 Data And Variables

2.1 Data sources

2.2 Health status

2.3 Cooking fuels and household air pollution

2.4 Solution for household air pollution

2.5 Other variables

3 Empirical Strategy

4 Results

4.1 Statistical description

4.2 Associations between cooking fuels and self-rated health

4.3 Robustness checks

4.4 Associations between cooking fuels and other health status

4.5 Association between household air pollution and self-rated health

4.6 Subgroup analysis of the relationship between cooking fuels and self-rated health, by income

5 Discussion

Declarations

References

Supplementary Files

Status:

Version 1