Association between household air pollution due to unclean fuel use and depression among older adults in India: A cross-sectional study

Abstract

Background:

Depression is a major public concern among older adults. However, limited studies have examined the association of depression and household air pollution (HAP). Therefore, this study examines the association between HAP due to unclean fuel use and depression among older adults in India.

Methods:

Data from Longitudinal Aging Study of India (LASI), wave-1 (2017–18) were used (N = 50,206, ≥50 years). Depression measured by CIDI-SF (Composite International Diagnostic Interview-Short Form) was the outcome variable. Household unclean fuel use was considered as proxy of HAP. Bivariate analyses and multivariate logistic regression were used to fulfil the study objectives.

Results:

The prevalence of depression was greater among unclean fuel users than clean fuel users (9.6% vs 6.6%). Results showed that household unclean fuel use was associated with higher odds of depression. The interaction analyses showed that females and rural residents were at higher risk of depression due to HAP exposure. Furthermore, household cooking conditions also played an important role in the association between HAP exposure and depression. Unclean fuel use indoors without ventilation and with traditional chullah/stove was associated with higher odds of depression among older adults.

Conclusion:

The study concludes that exposure to HAP due to unclean fuel use should be considered as a potential risk factor of depression among older adults. Therefore, the study suggests an urgent need to create awareness about negative effects of unclean fuel use on mental health and promote clean fuel usage in households to ensure healthy aging.

INTRODUCTION

A rising public health concern is late-life depression as the global geriatric population is growing at an accelerated rate.[1] It is a common mental disorder characterized by persistent sadness, lack of interest, or pleasure in previously rewarding activities and can also disturb sleep and appetite.[2] According to the Global Burden of Diseases (GBD) report, depression is currently the third leading cause of global disease burden and is projected to become the leading cause by 2030.[3] According to the World Health Organization (WHO), an estimated 3.8% of the population experience depression, including 5.7% of adults older than 60 years.[2] Depression in elderly is frequently associated with a variety of risk factors, including sociodemographic, economic, and health-related, such as multiple comorbidities, which increases the risk of disability, neurocognitive disorders, and loss of social livelihood.[4,5] A recent study from India showed that the overall age-standardized prevalence of depression among older adults aged 45 years and above was around 6%, with a higher burden among females.[6] Furthermore, this study also confirms that the risk of depression was higher in those residing in rural areas, widowed, and with a lower socioeconomic status.[6] Considering the demographic shift toward an aging population in India, it is quite probable that depression would emerge as a significant public health challenge in the coming years. Hence, it is crucial to identify the risk factors associated with depression that may contribute to the onset of the condition among a significant portion of the older adults in India.

A recent study states that ambient air pollution is one of the leading causes of global burden of diseases, with particular higher estimates in low- and middle-income countries.[7] It is alarming to note that according to the world air quality report (WAQR) 2023, 22 of the top 30 most polluted cities across the world are from India.[8] Furthermore, previous epidemiological studies have found that ambient air pollution is associated with higher risk of depression in different countries.[9,10,11,12] However, contrary to ambient air pollution, globally little evidence is available on the association between depression and household air pollution (HAP) due to use of unclean fuel (coal, biomass, wood, crop residues, etc.). Globally, around 3 billion people use unclean fuel for their domestic needs, the majority of which are from low- and middle-income countries.[13]

But, recently, a few studies have documented about this association[14,15,16,17,18,19] among middle-aged and older adults and found a negative association between biomass fuel use and depression. Several plausible pathways could explain the association between unclean fuel use and depression. Exposure to particulate matter has damaging effects on the brain. They set off a chronic inflammatory process beginning with respiratory tracts, leading to increased production of proinflammatory markers (TNF-α), ROS, and ILβ, responsible for neurotoxicity. The ongoing production of ROS and inflammatory markers lead to the worsening of existing medical conditions.[20] In a hypothesized mechanism, these pollutants are also responsible for changes at the epigenetic level. For instance, methylation of CLOCK genes may trigger the onset or exacerbation of psychiatric symptoms. The same genes have been demonstrated to be implicated in the etiology of mood disorders.[21] Second, particulate matter produced by unclean fuels may lead to metabolic changes consistent with activation of the hypothalamus-pituitary-adrenal axis.[22] Activation of the hypothalamic-pituitary-adrenal axis is commonly reported in depressed individuals.[23] Third, exposure to particulate matter has been found to be significantly associated with higher levels of stress hormones.[24] The prolonged production of the stress hormone cortisol has been associated with the etiology of depression.[25]

In India, HAP is a significant contributor to the burden of diseases and accounted for 0.61 million deaths in 2019.[26] As per the NFHS report, still, more than 40% of the households depend on unclean fuel[27] and hence are exposed to a higher concentration of the toxic pollutants on a daily basis. Hence, it is pertinent to understand the effect of household unclean fuel use on depression in India. However, to the best of our knowledge, only two studies[20,28] have been conducted in India where higher air pollution (outdoor and indoor) levels generally exist. Therefore, building on previous research, this study primarily aims to assess the effect of HAP from unclean fuel use on depression among older adults in India, which is one of the most vulnerable age groups to both air pollution and depression. Additionally, we examine how gender and place of residence interact with depression and HAP exposure in our analysis. By exploring this relationship, we aim to provide insights into policy implications concerning energy, environment, and mental health.

DATA AND METHODS

The first wave (2017–18) of the Longitudinal Aging Study in India (LASI) data was analyzed in this study. LASI is a nationally representative survey encompassing a vast sample of 72,250 individuals aged 45 years and older, as well as their spouses, regardless of age, from all states and union territories across the country (except Sikkim). This survey gathered comprehensive insights into the physical, psychological, cognitive, and social well-being of the older adult population in India. To ensure a representative sample, the LASI employed a multistage stratified cluster sampling design, with three-stage sampling in rural areas and four-stage sampling in urban areas. The survey was conducted through face-to-face interviews using computer-assisted personal interview (CAPI) technology in the local language in addition to a direct health examination. The response rate at the individual level was exceptionally high at 95.6%. Detailed information regarding the sampling design, survey instruments, and data collection procedures can be found elsewhere.[29,30] In LASI, the initial 50+ sample consisted of 52,380 individuals. For this study, we included 50,206 participants with complete information on the outcome variable and household main cooking fuel after excluding missing samples (n = 2174).

Ethical considerations

The Indian Council of Medical Research (ICMR) provided the necessary guidance and ethical approval for conducting LASI, and all the procedures adhered to the relevant guidelines and regulations. Prior informed consent was obtained from all respondents before conducting the interviews.[30]

Variable’s description

Outcome variable

The outcome variable of the study is major probable depression, which was coded as 0 (‘No’) for those not diagnosed with depression and 1 (‘Yes’) for those diagnosed with depression. The Short Form Composite International Diagnostic Interview (CIDI-SF) was used in LASI to estimate the diagnostic symptom-based prevalence of depression. It comprises three screening questions on dysphoria and/or anhedonia over 2 weeks in the past year, along with seven symptom-based questions. A positive response to three or more of these symptoms leads to the classification of an individual as diagnosed with depression. The seven symptoms include loss of interest, fatigue, loss of appetite, difficulty concentrating, feelings of worthlessness, thoughts of death, and difficulty falling asleep.[31] This scale has been validated in field settings, particularly by nonclinicians in general population surveys[30,32] and in various cultural settings to estimate the probable psychiatric diagnosis of major depression.[6,31,33,34] This scale aligns with the criteria for a major depressive episode outlined in the Diagnostic and Statistical Manual of Mental Disorders (DSM).[31,35] The reliability of the CIDI-SF is considered acceptable based on the Cronbach’s alpha value (α = 0.70).

Exposure variable

To estimate indirect exposure, the study used several markers of household air pollution (HAP). Three exposure indicators were analyzed: (a) exposure to HAP from cooking fuel (clean fuel vs unclean fuel), (b) exposure to HAP from other purpose fuels (such as boiling water for bathing, lighting, etc.), and (c) the association between HAP exposure and depression, taking into account household cooking conditions. The LASI survey gathered data on the primary source of cooking fuel through the household questionnaire. The responses included liquified petroleum gas (LPG), biogas, kerosene, electricity, charcoal/lignite/coal, crop residue, wood/shrub, dung cake, do not cook at home, and others. Following previous literature studies,[14,17,36,37] we recoded responses as follows: cleaner fuel = 0 (if the response was electricity, LPG, or natural gas) and unclean fuel = 1 (if the response was charcoal, coal, crop residue, wood, shrub, dung cake, kerosene, or others). We excluded those who did not cook at home from the analyses (n = 110). We also applied the same categorization to unclean fuel use for other purposes. Households using unclean fuels were considered as having HAP exposure.

Furthermore, this study aims to estimate exposure to HAP resulting from unclean fuel use, taking into account other household factors. The variables were derived from survey questions regarding the type of stove used (mechanical stove/improved cook stove, traditional challah, open fire, others), the presence of ventilation (traditional chimney, electric chimney, exhaust fan, near window/door, none), and the location of cooking (inside the house, in a separate building, outdoors, others) in households that use unclean fuel for cooking. The presence of ventilation during cooking was coded as ‘yes’ if a traditional chimney, electric chimney, exhaust fan, or open window/door was used and ‘no’ otherwise. The place of cooking was recoded as ‘indoors’ if it took place inside a house or separate building and ‘outdoors’ otherwise. Therefore, to determine household cooking conditions, two composite variables were generated. The first variable, ‘Cooking fuel, ventilation and place of cooking’, is a combination of unclean fuel with cooking location and ventilation and is categorized into four groups: clean fuel (reference), indoor unclean fuel with ventilation, indoor unclean fuel without ventilation, and outdoor unclean fuel. The second variable was “Cooking fuel and stove type”, which estimated the effect of unclean fuel use with different types of stoves, categorized as clean fuel (reference), unclean fuel with improved stove, unclean fuel with traditional chullah/stove, and unclean fuel with open fire. The purpose of this classification was to evaluate the individual and combined effects of different markers of household air pollution (HAP) on depression risk.

Covariates

Several covariates were considered in this study based on prior literaturestudies,[15,20,36,38,39] including demographic, socioeconomic, behavioral, and health factors as well as exposure to passive smoking and region. Demographic factors included age groups categorized as 50-59, 60-69, 70-79, and 80+; gender (male, female); place of residence (urban, rural); and marital status, which was categorized as currently married or unmarried, which includes widowed, separated, divorced, and never married individuals. In the context of India, the health of an individual is significantly influenced by religion and the caste and social group divisions.[40] Schedule caste (SC) are the most disadvantaged group in the society, and schedule tribes (ST) are the indigenous group that is often geographically isolated from mainstream society. Hence, the study also accounted for religion (Hindu and others) and social groups/castes (recoded as SC/ST, Other Backward Classes (OBCs), Others) as potential confounding factors.

Socioeconomic variables comprised educational level (no schooling, primary, secondary, higher), work status (currently working, ever worked but not currently, never worked), and monthly per capita consumption expenditure (MPCE) quintile based on household consumption data. Data on food expenditure were collected over a 7-day period; nonfood expenditure was gathered over 30-day and 365-day periods. Both types of expenditures were standardized to the 30-day reference period. The MPCE variable is computed as a summary measure of consumption. Last, it was divided into five quintiles, ranging from the lowest quintile representing the poorest households to the highest quintile representing the richest households.[30] Behavioral factors included tobacco use (never smoked or used smokeless tobacco, ever smoked or used smokeless tobacco) and alcohol use (yes, no). Respondents’ physical activity was measured using the Global Physical Activity Questionnaire. A score of 1 was assigned if at least 150 minutes of moderate-intensity or at least 75 minutes of vigorous-intensity aerobic physical activity per week was reported.[41] The respondents’ body mass index (BMI) was calculated by dividing their weight (in kilograms) by the square of their height (in meters). The respondents were then categorized as underweight, normal weight, and overweight/obese based on their BMI. Multimorbidity was defined as the simultaneous presence of two or more diagnosed chronic health conditions or diseases. For this analysis, nine chronic health conditions were included: hypertension, diabetes, cancer or a malignant tumor, chronic lung disease, chronic heart diseases, stroke, arthritis/rheumatism/osteoporosis or other bone/joint problems, and high cholesterol. In this study, the variable was coded with three categories, where 0 means ‘no disease’, 1 denotes ‘1 disease’, and 2 denotes ‘2+ diseases’.[42] Exposure to passive smoking was coded ‘yes’ if any usual member of the household smoked inside the home. Additionally, the region of the country was coded as North, Central, East, Northeast, West, and South.

Statistical analyses

Bivariate statistics such as percentage distribution and Chi-square tests were used to describe the sample characteristics according to household unclean fuel use and depression. Appropriate survey weights (individual and national levels) were used to make the estimates nationally representative. Multivariate hierarchical logistic regression models were used to examine the associations between HAP due to unclean fuel use and depression. In addition, to study the interaction effects between HAP and place of residence and gender on depression, the multiplicative term of combinations of household cooking fuel use was included in separate logistic regression models. Results were presented as odds ratio with 95% confidence interval. Stata 17 was used to perform statistical analyses. P < 0.05 was considered statistically significant in the analyses.

RESULTS

Table 1 presents the sample characteristics of the study population. Around 8% of the population reported of depression. A large sample of population around 38% belonged to the 50–59 age group. More than half of the sample (53%) were females and from rural areas (70%). A higher proportion of the sample were currently married (71%) and belonged to Hindu religion (83%). Approximately 45% of the sample belonged to OBCs. In terms of education, over half of the study population had no schooling (53.28%). Regarding work status, around 43% were currently working, while 31% had worked before but were not currently employed. Health behaviors reveal that around 39% of the sample use tobacco, and 15% consumed alcohol. A significant portion of the population engaged in physical activity (62.71%) and had a normal BMI (47.02%). Around half of the sample had no multimorbidity (51.48%), and a large majority were not exposed to passive smoking (74.87%). A large sample of the study belonged to the east region (23.91%). Nearly 48% of the households were using unclean fuel for cooking.

Table 1.

Sample characteristics of the study population (50+), LASI 2017-18

		n		%
Depression
Yes		3,246		8.1
No		46,960		91.9
Age
50-59		20,280		38.55
60-69		18,287		36.55
70+		11,639		24.9
Gender
Male		23,522		46.79
Females		26,684		53
Residence
Rural		32,883		70.33
Urban		17,323		29.67
Marital status
Currently married		35,921		70.72
Unmarried		14,285		29.28
Religion
Hindu		36,931		82.69
Others		13,275		17.31
Social groups/Caste
SC/ST		16,912		27.98
OBC		18,970		45.11
Others		14,324		26.91
Educational Level
No schooling		24,876		53.28
Primary		12,334		22.9
Secondary		10,387		18.66
Higher		2,609		5.17
MPCE quintile
Poorest		9,830		21.22
Poorer		10,252		21.57
Middle		10,050		20.21
Richer		10,116		19.16
Richest		9,958		17.84
Work status
Currently working		21,082		42.94
Ever worked but not currently		15,319		31.5
Never worked		13,805		25.56
Tobacco use
No		31,255		61.25
Yes		18,951		38.75
Alcohol consumption
No		41,197		84.96
Yes		9,009		15.04
Physical Activity
No		19,175		37.29
Yes		31,031		62.71
BMI
Underweight		9,243		20.95
Normal		24,081		47.02
Overweight/Obese		16,882		32.02
Multimorbidity
None		25,549		51.48
Single morbidity		14,394		28.56
Multimorbidity		10,263		19.96
Exposure to passive smoking
No		37,487		74.87
Yes		12,719		25.13
Region
North		9,155		12.51
Central		6,879		20.81
East		9,179		23.91
Northeast		6,313		3.3
West		6,639		16.71
South		12,041		22.76
Cooking fuel
Clean fuel		26,758		52.02
Unclean fuel		23,448		47.98
Total		50,206

Note: n refers to sample size, is unweighted and % are weighted. SC=Scheduled Caste, ST=Scheduled Tribe; MPCE=Monthly per capita consumption expenditure; BMI=Body mass index

Table 2 presents the prevalence of depression among older adults by the type of cooking fuel use in the households. Depression was more prevalent among unclean fuel users (9.61%) compared to clean fuel users (6.66%). Both clean and unclean fuel users exhibited a slight increase in depression prevalence with advancing age. Females had higher depression prevalence than males for clean (5.95% vs 4.71%) and unclean fuel users (8.58% vs 6.74%). Rural residents reported higher prevalence of depression. Unmarried individuals had higher depression prevalence compared to married individuals. Hindus showed a higher depression prevalence compared to others in the unclean fuel group (8.59% vs 5.59%). In both fuel groups, individuals from the OBC caste group reported higher prevalence of depression.

Table 2.

Prevalence of depression by the use of cooking fuel in the household among older adults (50+) in India, LASI 2017-18

Depression		Clean fuel				Unclean fuel
		%		P		%				P
Age
50-59		5.12		0.04		7.23				0.076
60-69		5.25				8
70+		6				8.09
Gender
Male		4.71		<0.001		6.74				<0.001
Females		5.95				8.58
Residence
Rural		6.07		<0.001		7.94				<0.001
Urban		4.80				5.92
Marital status
Currently married		4.72		<0.001		6.73				<0.001
Unmarried		6.97				10.21
Religion
Hindu		5.48		0.152		8.52				<0.001
Others		5.03				5.59
Social groups/Caste
SC/ST		4.55		0.002		5.76				<0.001
OBC		5.79				9.67
Others		5.44				8.69
Educational Level
No schooling		6.4		<0.001		8.12				0.005
Primary		5.59				7.35
Secondary		4.49				6.25
Higher		3.25				7.62
MPCE quintile
Poorest		5.09		0.042		7.01				<0.001
Poorer		5.51				6.84
Middle		4.75				6.8
Richer		5.27				8.76
Richest		5.98				10.83
Work status
Currently working		4.28		<0.001		6.39				<0.001
Ever worked but not currently		6.32				10.05
Never worked		5.66				7.56
Tobacco use
No		5		<0.001		7.6				0.459
Yes		6.24				7.86
Alcohol consumption
No		5.35		0.529		8.06				<0.001
Yes		5.45				6.46
Physical Activity
No		5.84		0.004		8.12				0.088
Yes		5.04				7.50
BMI
Underweight		7		<0.001		8.18				0.042
Normal		5.31				7.29
Overweight/Obese		5.02				8.15
Multimorbidity
None		3.99		<0.001		6.20				<0.001
Single morbidity		5.44				9.26
Multimorbidity		7.48				11.61
Exposure to passive smoking
No		5.27		0.139		7.63				0.449
Yes		5.78				7.91
Region
North		5.91		<0.001		7.35				<0.001
Central		11.45				12.11
East		5.67				8.29
Northeast		1.78				3.13
West		5.80				8.34
South		3.85				6.46
Total		6.66				9.61

Note: P value estimated from Chi-square test; SC=Scheduled Caste, ST=Scheduled Tribe; MPCE=Monthly per capita consumption expenditure; BMI=Body mass index

Among clean fuel users, depression prevalence decreased significantly with higher education, while it varied little among unclean fuel users. Depression prevalence among clean fuel users remained consistent across different MPCE quintiles but significantly varied among unclean fuel users. Individuals who ever worked but were not currently working had higher depression prevalence in both fuel groups. Among clean fuel users, individuals who smoked or used smokeless tobacco had higher prevalence of depression (6.24%). Furthermore, individuals who do not consume alcohol reported higher prevalence of depression among unclean fuel users (8.06%). Individuals who were not physically active reported higher prevalence of depression among clean (5.84%) and unclean fuel users (8.12%). Among clean fuel users, underweight individuals had a significantly higher prevalence of depression. In contrast, among unclean fuel users, both underweight and overweight/obese individuals reported a higher prevalence of depression. Individuals with multimorbidity have higher prevalence of depression for both clean and unclean fuel users. Geographically, the central region displayed a higher prevalence of depression among clean (11.45%) and unclean fuel (12.11%) users.

Table 3 presents the logistic regression results of the association between household fuel use (for cooking and other purposes) and depression among older adults in India. Unclean fuel use for both cooking and other purposes was consistently associated with higher odds of depression in all the models. In the crude model (model 1), compared to clean fuel use for cooking, unclean fuel use for cooking fuel was associated with higher odds of depression (OR = 1.48, 95% CI: 1.37, 1.58, P < 0.001). In model 2, after adjustments for socioeconomic status and demographic factors, the higher odds of depression due to unclean fuel for cooking remained significant (OR = 1.46, 95% CI: 1.34, 1.59, P < 0.001). Furthermore, in model 3, unclean fuel use for cooking remained positively associated with a higher prevalence of depression; however, the odds accentuated but remained significant (OR = 1.35, 95% CI: 1.24, 1.48, P < 0.001). Similar associations were also observed for the association between unclean fuel use for other purposes and depression in all the models. In model 1, compared to clean fuel use, unclean fuel use for other purposes was positively associated with depression (OR = 1.44, 95% CI: 1.34, 1.55, P < 0.001). However, in model 2, after accounting for socioeconomic-demographic variables, the odds of depression decreased, but the association remained significant (OR = 1.31, 95% CI: 1.20, 1.42, P < 0.001). Finally, after accounting for all covariates in model 3, the association remained significant (OR = 1.29, 95% CI: 1.18, 1.41, P < 0.001).

Table 3.

Odds ratio of association between household fuel use and depression among older adults (50+) in India, LASI 2017-18

		Odds ratio		95% CI				P
Cooking fuel
Model 1
Clean fuel ®
Unclean fuel		1.48		1.37		1.58		<0.001
Model 2
Clean fuel ®
Unclean fuel		1.46		1.34		1.59		<0.001
Model 3
Clean fuel ®
Unclean fuel		1.35		1.24		1.48		<0.001
Other purpose fuel
Model 1
Clean fuel ®
Unclean fuel		1.44		1.34		1.55		<0.001
Model 2
Clean fuel ®
Unclean fuel		1.31		1.2		1.42		<0.001
Model 3
Clean fuel ®
Unclean fuel		1.29		1.18		1.41		<0.001

Note: Model 1: Crude model, Model 2: Model 1+ Adjusted for age and gender, marital status, residence, caste, religion, MPCE quintile, education, work status, Model 3: Model 2+ alcohol, tobacco use, physical activity, BMI, multimorbidity, exposure to passive smoking, regions

Table 4 presents the interaction effects of gender and place of residence with cooking fuel on depression. The results show significant interaction effects of gender and place of residence in the association of cooking fuel and depression. Compared to males using clean fuel, males (OR = 1.22, 95% CI: 1.07, 1.39, P < 0.001) and females (OR = 1.55, 95% CI: 1.35, 1.79, P < 0.001) using unclean fuel were at higher risk of depression. Rural residents using unclean fuel (OR = 1.47, 95% CI: 1.30, 1.66, P < 0.001) reported higher odds of depression than urban residents using clean fuel. Similarly, those living in urban areas using unclean fuel were also at higher risk of depression (OR = 1.27, 95% CI: 1.04, 1.55, P = 0.014).

Table 4.

Interaction effects of gender and place of residence on the association of household cooking fuel use and depression among older adults (50+) in India, LASI 2017-18

		Odds ratio		95% CI				P
Gender # cooking fuel
Male# clean fuel		ref
Male# unclean fuel		1.22		1.07		1.39		0.003
Female# clean fuel		1.19		1.04		1.36		0.013
Female# unclean fuel		1.55		1.35		1.79		<0.001
Residence# cooking fuel
Urban# clean fuel		ref
Rural# clean fuel		1.16		1.03		1.30		0.012
Rural # unclean fuel		1.47		1.30		1.66		<0.001
Urban# unclean fuel		1.27		1.04		1.55		0.014

Model adjusted for age, marital status, caste, religion, education level, work status, MPCE quintile, tobacco, alcohol use, physical activity, BMI, multimorbidity, exposure to passive smoking, other purpose fuel, region

This study also examined the association of household cooking conditions with depression presented in Table 5. In both crude and adjusted models, different household cooking conditions were significantly associated with depression. In the crude model, the results showed that compared to clean fuel use, use of unclean fuel indoors with ventilation, without ventilation, and outdoors was linked with higher likelihood of depression (P < 0.001). Similarly, unclean fuel use with traditional chullah/stoves and cooking outdoors was also positively associated with depression (P < 0.001). In the adjusted model, use of unclean fuel indoors with ventilation (OR = 1.27, 95% CI: 1.15, 1.40, P < 0.001), without ventilation (OR = 1.56, 95% CI: 1.38, 1.76, P < 0.001), and outdoors (OR = 1.38, 95% CI: 1.19, 1.60, P < 0.001) was significantly associated with increased odds of depression. Furthermore, considering cooking fuel and stove type, it was found that use of unclean fuel with traditional chullah was associated with higher odds of depression (OR = 1.37, 95% CI: 1.25, 1.50, P < 0.001) compared to clean fuel use. Similarly, unclean fuel with open fire also showed elevated odds of depression (OR = 1.34, 95% CI: 1.15, 1.56, P < 0.001). After adjustments of covariates, unclean fuel use with improved stove showed no statistically significant difference in the odds of depression (OR: 1.11, 95% CI: 0.82, 1.50, P = 0.486).

Table 5.

Odds ratio of association between household cooking conditions and depression among older adults (50+), LASI 2017-18

Crude model
		Odds ratio		95% CI				P
Cooking fuel, place of cooking, and ventilation
Clean fuel		ref
Unclean fuel indoors using ventilation		1.31		1.20		1.42		<0.001
Unclean fuel indoors no ventilation		1.76		1.59		1.94		<0.001
Unclean fuel outdoors		1.67		1.46		1.91		<0.001
Cooking fuel and type of stove
Clean fuel		ref
Unclean fuel improved stove		1.10		0.82		1.48		0.511
Unclean fuel with traditional chullah/stove		1.53		1.42		1.65		<0.001
Unclean fuel with open fire		1.27		1.11		1.45		<0.001
Adjusted model
Cooking fuel, place of cooking and ventilation
Clean fuel		ref
Unclean fuel indoors using ventilation		1.27		1.15		1.40		<0.001
Unclean fuel indoors no ventilation		1.56		1.38		1.76		<0.001
Unclean fuel outdoors		1.38		1.19		1.60		<0.001
Cooking fuel and type of stove
Clean fuel		ref
Unclean fuel improved stove		1.11		0.82		1.50		0.486
Unclean fuel with traditional chullah		1.37		1.25		1.50		<0.001
Unclean fuel with open fire		1.34		1.15		1.56		<0.001

Model adjusted for age, sex, residence, marital status, caste, religion, education level, work status, MPCE quintile, tobacco, alcohol use, physical activity, BMI, multimorbidity, exposure to passive smoking, region

DISCUSSION

This cross-sectional study examined the association between HAP due to unclean fuel use and depression among older adults in India. The study found that the prevalence of depression was higher among unclean fuel users, around 10%. The findings suggest that use of unclean fuel increases the odds of depression, even after adjusting for various factors. Gender and place of residence significantly influenced this association, with females and rural residents facing greater risks of depression due to HAP exposure. Household cooking conditions also played a significant role.

The findings of this study align with previous studies investigating the association between unclean fuel and depression in older adults in China and India.[14,16,17,20,28,36,38,43] After accounting for all the potential risk factors, the findings of the study suggest that unclean fuel use for cooking and other purposes was associated with 35% and 29%, respectively, higher chances of depression among older adults in India. Due to differences in the study population and definitions of unclean fuel use and depression, this value differed from previous studies.[20,28] In line with our study, one recent study in India examined the association between depressive symptoms and unclean fuel use among older adults aged 60 years and older and found that unclean fuel use was linked with an increased prevalence of depression.[28]

Our study provides important insights about the moderating effect of gender and place of residence on the association between HAP and depression. The results indicated that compared to males, females using unclean fuel for cooking were at higher risk of depression. A study found that premenopausal women cooking with biomass had higher odds of depression in India. Similarly, previous studies conducted in India and China found the risk of depressive symptoms higher for women compared to men due to unclean fuel use.[16,28] This could be possible because in the majority of Indian households, women are the primary cooks and caretakers of the house and hence are at higher exposure to unclean fuel pollutants.[44] Furthermore, individuals from rural areas were at greater risk of depression due to HAP exposure. Higher proportions of households continue to rely on unclean fuel due to affordability and a lack of reliable income sources; especially in rural areas, only 10% of households use clean fuel.[44,45] However, the use of clean cooking fuel has increased significantly over the period of time (from 2015-16 to 2019-21), particularly in rural areas due to the targeted interventions by the government. In 2016, the Government of India introduced the Pradhan Mantri Ujjwala Yojana (PMUY) to focus on economically disadvantaged households, particularly those from marginalized caste and class groups, by providing free LPG connections to make clean energy more accessible to them.[46]

Furthermore, this study provides compelling evidence about the association between household cooking conditions and depression. The findings suggested that unclean fuel use with traditional stove and open fire increased the risk of depression. Similarly, it was found that unclean fuel use with different combinations of ventilation and place of cooking was related to higher odds of depression. Particularly, the results highlighted that the risk of depression was higher for those using unclean fuel indoors without ventilation. The results are consistent with the findings of previous studies,[15,47] showing that the use of ventilation might reduce the harmful effects of unclean fuel use. However, additional research is warranted to explore the specific effect of household cooking conditions on the association of unclean fuel use and depression among older adults.

This study has several strengths. First, to the best of our knowledge, this is the first study to examine the association between HAP due to unclean fuel use and major depression among older adults in India. Second, this study is based on large-scale, nationally representative data in India, so the study can be generalized to a large population in similar settings. Third, this study investigates the separate and combined effects of household fuel use on depression, which is unique, and no study has done that so far in India. In addition, the study also examines the effect of cooking conditions, considering various household factors that might affect the level of exposure.

The present study provides essential findings about HAP due to unclean fuel use and depression among older adults in India. However, it is important to interpret the findings in the context of the study’s limitations. First, the study is cross-sectional in nature; therefore, causality cannot be established. Second, most of the measures used in this study were self-reported, which may introduce recall and social desirability biases into the results. Third, this study could not account for exposure to ambient air pollution, and occupational exposure to air pollution, which has been found to be associated with depression, may confound the findings.[18,48] Fourth, this study used proxy variables for exposure to HAP; therefore, the study could not control for accurate personal exposure (duration and frequency) due to data limitations. Also, detailed information regarding the change in patterns of fuel use over the years was lacking. Fifth, the impact of prior psychiatric illness, presence of neurocognitive disorder, history of trauma, and psychosocial stressors were not included in the analyses due to lack of in-depth information in the survey data. Sixth, this study measures depression using the internationally validated CIDI-SF scale, which provides a probable diagnosis rather than a confirmed clinical diagnosis. As such, the findings should be interpreted with caution, recognizing the potential limitations of relying on self-reported data and the absence of clinical confirmation by healthcare professionals. Last, since this study is based on secondary analyses of data that were not originally collected with the specific objectives of this research in mind, the findings should be considered acknowledging the potential constraints and limitations inherent in the data. To provide more insight, future studies should take these factors into account.

CONCLUSION

This study concluded that unclean fuel use for cooking and other purposes was associated with increased risk of depression among older adults in India. In addition, unclean fuel use for both cooking and other purposes was associated with higher risk of depression, especially in females and rural areas. Therefore, the findings of the study suggest that HAP due to unclean fuel use should be considered as a crucial risk factor for depression. The findings suggest that older adults should use clean fuel; however, additional research in future is needed to examine the causal association with future LASI waves in India.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.