Comparing estimates of household expenditures between pictorial diaries and surveys in three low- and middle-income countries.

In most low- and middle-income countries (LMICs), household out-of-pocket (OOP) health spending constitutes a major source of healthcare financing. Household surveys are commonly used to monitor OOP health spending, but are prone to recall bias and unable to capture seasonal variation, and may underestimate expenditure-particularly among households with long-term chronic health conditions. Household expenditure diaries have been developed as an alternative to overcome the limitations of surveys, and pictorial diaries have been proposed where literacy levels may render traditional diary approaches inappropriate. This study compares estimates for general household and chronic healthcare expenditure in South Africa, Tanzania and Zimbabwe derived using survey and pictorial diary approaches. We selected a random sub-sample of 900 households across urban and rural communities participating in the Prospective Urban and Rural Epidemiology study. For a range of general and health-specific categories, OOP expenditure estimates use cross-sectional survey data collected via standardised questionnaire, and data from these same households collected via two-week pictorial diaries repeated four times over 2016-2019. In all countries, average monthly per capita expenditure on food, non-food/non-health items, health, and consequently, total household expenditure reported by pictorial diaries was consistently higher than that reported by surveys (each p<0.001). Differences were greatest for health expenditure. The share of total household expenditure allocated to health also differed by method, accounting for 2% in each country when using survey data, and from 8-20% when using diary data. Our findings suggest that the choice of data collection method may have significant implications for estimating OOP health spending and the burden it places on households. Despite several practical challenges to their implementation, pictorial diaries offer a method to assess potential bias in surveys or triangulate data from multiple sources. We offer some practical guidance when considering the use of pictorial diaries for estimating household expenditure.

Personal and household PM2.5 and black carbon exposure measures and respiratory symptoms in 8 low- and middle-income countries.

BACKGROUND: Household air pollution (HAP) from cooking with solid fuels has been associated with adverse respiratory effects, but most studies use surveys of fuel use to define HAP exposure, rather than on actual air pollution exposure measurements. OBJECTIVE: To examine associations between household and personal fine particulate matter (PM2.5) and black carbon (BC) measures and respiratory symptoms. METHODS: As part of the Prospective Urban and Rural Epidemiology Air Pollution study, we analyzed 48-h household and personal PM2.5 and BC measurements for 870 individuals using different cooking fuels from 62 communities in 8 countries (Bangladesh, Chile, China, Colombia, India, Pakistan, Tanzania, and Zimbabwe). Self-reported respiratory symptoms were collected after monitoring. Associations between PM2.5 and BC exposures and respiratory symptoms were examined using logistic regression models, controlling for individual, household, and community covariates. RESULTS: The median (interquartile range) of household and personal PM2.5 was 73.5 (119.1) and 65.3 (91.5) µg/m3, and for household and personal BC was 3.4 (8.3) and 2.5 (4.9) x10-5 m-1, respectively. We observed associations between household PM2.5 and wheeze (OR: 1.25; 95%CI: 1.07, 1.46), cough (OR: 1.22; 95%CI: 1.06, 1.39), and sputum (OR: 1.26; 95%CI: 1.10, 1.44), as well as exposure to household BC and wheeze (OR: 1.20; 95%CI: 1.03, 1.39) and sputum (OR: 1.20; 95%CI: 1.05, 1.36), per IQR increase. We observed associations between personal PM2.5 and wheeze (OR: 1.23; 95%CI: 1.00, 1.50) and sputum (OR: 1.19; 95%CI: 1.00, 1.41). For household PM2.5 and BC, associations were generally stronger for females compared to males. Models using an indicator variable of solid versus clean fuels resulted in larger OR estimates with less precision. CONCLUSIONS: We used measurements of household and personal air pollution for individuals using different cooking fuels and documented strong associations with respiratory symptoms.

Measuring and predicting personal and household Black Carbon levels from 88 communities in eight countries.

Black Carbon (BC) is an important component of household air pollution (HAP) in low- and middle- income countries (LMICs), but levels and drivers of exposure are poorly understood. As part of the Prospective Urban and Rural Epidemiological (PURE) study, we analyzed 48-hour BC measurements for 1187 individual and 2242 household samples from 88 communities in 8 LMICs (Bangladesh, Chile, China, Colombia, India, Pakistan, Tanzania, and Zimbabwe). Light absorbance (10-5 m-1) of collected PM2.5 filters, a proxy for BC concentrations, was calculated via an image-based reflectance method. Surveys of household/personal characteristics and behaviors were collected after monitoring. The geometric mean (GM) of personal and household BC measures was 2.4 (3.3) and 3.5 (3.9)·10-5 m-1, respectively. The correlation between BC and PM2.5 was r = 0.76 for personal and r = 0.82 for household measures. A gradient of increasing BC concentrations was observed for cooking fuels: BC increased 53% (95%CI: 30, 79) for coal, 142% (95%CI: 117, 169) for wood, and 190% (95%CI: 149, 238) for other biomass, compared to gas. Each hour of cooking was associated with an increase in household (5%, 95%CI: 3, 7) and personal (5%, 95%CI: 2, 8) BC; having a window in the kitchen was associated with a decrease in household (-38%, 95%CI: -45, -30) and personal (-31%, 95%CI: -44, -15) BC; and cooking on a mud stove, compared to a clean stove, was associated with an increase in household (125%, 95%CI: 96, 160) and personal (117%, 95%CI: 71, 117) BC. Male participants only had slightly lower personal BC (-0.6%, 95%CI: -1, 0.0) compared to females. In multivariate models, we were able to explain 46-60% of household BC variation and 33-54% of personal BC variation. These data and models provide new information on exposure to BC in LMICs, which can be incorporated into future exposure assessments, health research, and policy surrounding HAP and BC.

Multinational prediction of household and personal exposure to fine particulate matter (PM2.5) in the PURE cohort study.

INTRODUCTION: Use of polluting cooking fuels generates household air pollution (HAP) containing health-damaging levels of fine particulate matter (PM2.5). Many global epidemiological studies rely on categorical HAP exposure indicators, which are poor surrogates of measured PM2.5 levels. To quantitatively characterize HAP levels on a large scale, a multinational measurement campaign was leveraged to develop household and personal PM2.5 exposure models. METHODS: The Prospective Urban and Rural Epidemiology (PURE)-AIR study included 48-hour monitoring of PM2.5 kitchen concentrations (n = 2,365) and male and/or female PM2.5 exposure monitoring (n = 910) in a subset of households in Bangladesh, Chile, China, Colombia, India, Pakistan, Tanzania and Zimbabwe. PURE-AIR measurements were combined with survey data on cooking environment characteristics in hierarchical Bayesian log-linear regression models. Model performance was evaluated using leave-one-out cross validation. Predictive models were applied to survey data from the larger PURE cohort (22,480 households; 33,554 individuals) to quantitatively estimate PM2.5 exposures. RESULTS: The final models explained half (R2 = 54%) of the variation in kitchen PM2.5 measurements (root mean square error (RMSE) (log scale):2.22) and personal measurements (R2 = 48%; RMSE (log scale):2.08). Primary cooking fuel type, heating fuel type, country and season were highly predictive of PM2.5 kitchen concentrations. Average national PM2.5 kitchen concentrations varied nearly 3-fold among households primarily cooking with gas (20 µg/m3 (Chile); 55 µg/m3 (China)) and 12-fold among households primarily cooking with wood (36 µg/m3 (Chile)); 427 µg/m3 (Pakistan)). Average PM2.5 kitchen concentration, heating fuel type, season and secondhand smoke exposure were significant predictors of personal exposures. Modeled average PM2.5 female exposures were lower than male exposures in upper-middle/high-income countries (India, China, Colombia, Chile). CONCLUSION: Using survey data to estimate PM2.5 exposures on a multinational scale can cost-effectively scale up quantitative HAP measurements for disease burden assessments. The modeled PM2.5 exposures can be used in future epidemiological studies and inform policies targeting HAP reduction.

Household and personal air pollution exposure measurements from 120 communities in eight countries: results from the PURE-AIR study.

BACKGROUND: Approximately 2·8 billion people are exposed to household air pollution from cooking with polluting fuels. Few monitoring studies have systematically measured health-damaging air pollutant (ie, fine particulate matter [PM2·5] and black carbon) concentrations from a wide range of cooking fuels across diverse populations. This multinational study aimed to assess the magnitude of kitchen concentrations and personal exposures to PM2·5 and black carbon in rural communities with a wide range of cooking environments. METHODS: As part of the Prospective Urban and Rural Epidemiological (PURE) cohort, the PURE-AIR study was done in 120 rural communities in eight countries (Bangladesh, Chile, China, Colombia, India, Pakistan, Tanzania, and Zimbabwe). Data were collected from 2541 households and from 998 individuals (442 men and 556 women). Gravimetric (or filter-based) 48 h kitchen and personal PM2·5 measurements were collected. Light absorbance (10-5m-1) of the PM2·5 filters, a proxy for black carbon concentrations, was calculated via an image-based reflectance method. Surveys of household characteristics and cooking patterns were collected before and after the 48 h monitoring period. FINDINGS: Monitoring of household air pollution for the PURE-AIR study was done from June, 2017, to September, 2019. A mean PM2·5 kitchen concentration gradient emerged across primary cooking fuels: gas (45 µg/m3 [95% CI 43-48]), electricity (53 µg/m3 [47-60]), coal (68 µg/m3 [61-77]), charcoal (92 µg/m3 [58-146]), agricultural or crop waste (106 µg/m3 [91-125]), wood (109 µg/m3 [102-118]), animal dung (224 µg/m3 [197-254]), and shrubs or grass (276 µg/m3 [223-342]). Among households cooking primarily with wood, average PM2·5 concentrations varied ten-fold (range: 40-380 µg/m3). Fuel stacking was prevalent (981 [39%] of 2541 households); using wood as a primary cooking fuel with clean secondary cooking fuels (eg, gas) was associated with 50% lower PM2·5 and black carbon concentrations than using only wood as a primary cooking fuel. Similar average PM2·5 personal exposures between women (67 µg/m3 [95% CI 62-72]) and men (62 [58-67]) were observed. Nearly equivalent average personal exposure to kitchen exposure ratios were observed for PM2·5 (0·79 [95% 0·71-0·88] for men and 0·82 [0·74-0·91] for women) and black carbon (0·64 [0·45-0·92] for men and 0·68 [0·46-1·02] for women). INTERPRETATION: Using clean primary fuels substantially lowers kitchen PM2·5 concentrations. Importantly, average kitchen and personal PM2·5 measurements for all primary fuel types exceeded WHO's Interim Target-1 (35 µg/m3 annual average), highlighting the need for comprehensive pollution mitigation strategies. FUNDING: Canadian Institutes for Health Research, National Institutes of Health.

Household, Community, Sub-National and Country-level Predictors of Primary Cooking Fuel Switching in Nine Countries from the PURE Study.

INTRODUCTION: Switching from polluting (e.g. wood, crop waste, coal) to clean cooking fuels (e.g. gas, electricity) can reduce household air pollution (HAP) exposures and climate-forcing emissions. While studies have evaluated specific interventions and assessed fuel-switching in repeated cross-sectional surveys, the role of different multilevel factors in household fuel switching, outside of interventions and across diverse community settings, is not well understood. METHODS: We examined longitudinal survey data from 24,172 households in 177 rural communities across nine countries within the Prospective Urban and Rural Epidemiology (PURE) study. We assessed household-level primary cooking fuel switching during a median of 10 years of follow up (~2005-2015). We used hierarchical logistic regression models to examine the relative importance of household, community, sub-national and national-level factors contributing to primary fuel switching. RESULTS: One-half of study households (12,369) reported changing their primary cooking fuels between baseline and follow up surveys. Of these, 61% (7,582) switched from polluting (wood, dung, agricultural waste, charcoal, coal, kerosene) to clean (gas, electricity) fuels, 26% (3,109) switched between different polluting fuels, 10% (1,164) switched from clean to polluting fuels and 3% (522) switched between different clean fuels. Among the 17,830 households using polluting cooking fuels at baseline, household-level factors (e.g. larger household size, higher wealth, higher education level) were most strongly associated with switching from polluting to clean fuels in India; in all other countries, community-level factors (e.g. larger population density in 2010, larger increase in population density between 2005-2015) were the strongest predictors of polluting-to-clean fuel switching. CONCLUSIONS: The importance of community and sub-national factors relative to household characteristics in determining polluting-to-clean fuel switching varied dramatically across the nine countries examined. This highlights the potential importance of national and other contextual factors in shaping large-scale clean cooking transitions among rural communities in low- and middle-income countries.