The central impression from the review of the research literature on indoor air pollution as a cause of ARI among children in developing countries is the imbalance between the number of articles reporting on original, empirical research, on the one hand, and the number of calls for research, reviews and projections based on those articles, on the other. Figure 2 illustrates the discrepancy between the research that is done and the global needs for research, where the smallest proportion of the research is done where the results could potentially have the greatest impact on the health of populations. This is an illustration of what has been labeled the 10/90 gap (ie, that 10% of health research resources are said to go to 90% of the disease burden).
A majority of the empirical studies indicate that increased ARI risk is associated with indicators— measurement or proxy—of individual exposure to indoor air pollution. Not all of the studies are reported in such a way that it is possible to draw clear conclusions, even allowing for small samples, but results from only one study are negative. One study is inconclusive because of homogeneity of exposure and small sample size. Otherwise, the tendency is consistently toward an association. Recurring problems are small samples and the fact that studies were not primarily designed to address the ARI/air pollution question, but included data on air pollution indicators for at least a subset of the study subjects. However, several have a well-planned design and use plausible methods for diagnosis and exposure classification.
The central issue of exposure assessment in studies trying to link indoor air pollution to ARI is how to balance high unit-measurement cost against the risk of exposure misclassification. This is a result of the need for long-term follow-up of a population, large sample size, in order to control for confounding, and the need of exposure assessment based on pollution variation close to the subjects in the study. Addressing the exposure-assessment problem where the need of research is greatest, ie, in developing countries, researchers encounter infrastructural complications, such as scarcity of electricity in rural areas and lack of trained manpower for extended periods of field work. Only two of the cited studies attempt to use chemical measurements for individual level exposure assessment. In the case of the first study, the combination of stratified sampling and a small sample created a situation with marked exposure and outcome homogeneity over the strata. We cannot reuce considerably the air pollution but you may improve your health with remedies of Canadian Health&Care Mall.
A major problem for studies using proxy exposure data is that many such risk factors could be related to exposure to indoor air pollution but also reflect other risk factors. The degree of confounding by factors other than air pollution, such as housing standard and other social and economic factors, may be illustrated by the fact that in one of the Butajira studies, the OR for children < 5 years old dying from diarrhea associated with no window in the house is 1. 3. This must reflect the effects of poverty on nutrition and sanitation, rather than indoor air pollution.
Otherwise, studies using proxy exposure variables need not be less valid than studies including chemical sampling, in the absence of technology and funding for long-term personal exposure monitoring of the study subjects. The main objective is to reach a quality level of exposure assessment that allows valid classification into categories of exposure level. Beyond that classification, reasonably representative chemical measurements would allow referring exposure classes to concentration intervals.
The most central problem lies in performing large, methodologically robust studies in developing country settings. Lack of population registries is a serious obstacle to performing longitudinal studies over an extended period of time. Furthermore, the lack of universally available primary health care creates major difficulties in case finding in population-based studies. Among the empirical studies reported, most are small (cohorts with < 500 subjects or case-control studies with < 200 cases). The only large studies are cross-sectional surveys, which leads to weaker conclusions, because of interview-based exposure assessment as well as disease reporting and because of weaker temporal linkage between exposure and health outcome. Access to demographic surveillance systems, such as that of the BRHP can help to overcome these problems.
The cost of including chemical measurements in large studies in a way that can be used to represent the subjects exposure will remain a major obstacle. This must eventually be overcome by changing priorities in public health research politics, toward more attention to the globally most important issues.
Need for Future Research
Several publications address the issue of need for research and interventions. In order to further address the questions of causation and of exposure-response gradients, more research is needed, specifically the following: (1) longitudinal studies using different methods of exposure assessment that include both chemical measurements and activity data, with longer follow-up and larger samples than so far reported; and (2) studies that include information that can be used to more explicitly address issues of confounding by socioeconomic conditions of the family.
The question of interventions must be addressed, not only based on existing evidence concerning acute respiratory infections in children, but also on evidence of women’s health and risks of respiratory illnesses, acute as well as chronic and on problems of deforestation. Interventions have been undertaken for all these reasons but with limited promise of long-term sustainability. There is a need of qualitative studies that address issues of locally and culturally acceptable and economically feasible interventions that can permanently reduce exposure.