Back

Socioeconomic disparities in perceived air quality and associated respiratory health outcomes among residents of Nairobi, Kenya

Otieno, E. A.; Mwitari, J. M.; Makalliwa, G. A.

2026-08-23 occupational and environmental health
10.64898/2026.08.19.26360866 medRxiv
Show abstract

Socioeconomic inequality in exposure to air pollution possess a significant public health challenge, yet little is known about how the disparities vary across the various economic status areas in Nairobi. Globally, studies have shown that exposure to air pollution is unequal across communities hence disparities in harm to human health. This study examined the association between socioeconomic characteristics and perceived air quality among residents of low- and high-socioeconomic status areas in Nairobi, Kenya. Two regions within Nairobi County were selected for this study: Mukuru kwa Njenga (representing the Low Socioeconomic Status) and Langata (representing the High Socioeconomic Status) with a sample size of 384 in HSES areas and 368 in LSES areas. A cross-sectional study was conducted among 752 respondents residing in selected LSES and HSES areas of Nairobi. Data was collected using a structured questionnaire assessing sociodemographic characteristics, income, education, employment, perceived air quality, and self-reported health outcomes associated with air pollution exposure. Descriptive statistics were used to summarize participant characteristics and perceived air quality. Chi-square tests were used to examine associations between residential area and categorical health outcomes, while ordinal logistic regression was used to assess the association between socioeconomic characteristics and perceived air-quality ratings. Perceived air quality differed significantly between residential socioeconomic groups. Respondents in LSES areas were more likely to rate air quality as poor or very poor, with 45.4% rating it as very poor, compared with only 1.6% of respondents in HSES areas. In contrast, 12.2% of HSES respondents rated air quality as good compared with 0.3% in LSES areas. The association between area of residence and perceived air-quality rating was statistically significant, {chi}2(3) = 282.672, p < 0.001. In the ordinal logistic regression model, HSES residence was associated with significantly lower odds of reporting poorer perceived air quality compared with LSES residence (OR = 0.135, 95% CI: 0.095-0.190, p < 0.001). Income was also significantly associated with perceived air quality, while respondents with no formal education had higher odds of reporting poorer perceived air quality compared with those with secondary education (OR = 3.254, 95% CI: 1.388-7.638, p = 0.007). Significant differences were also observed for several self-reported health outcomes. Respiratory problems were more prevalent among respondents in LSES areas than HSES areas (72.7% versus 50.4%; {chi}2(1) = 29.081, p < 0.001; Cramer's V = 0.224). However, allergies, eye irritation, and headaches were reported more frequently in HSES areas than in LSES areas, with significant associations observed for allergies ({chi}2(1) = 106.479, p < 0.001; Cramer's V = 0.429), eye irritation ({chi}2(1) = 136.577, p < 0.001; Cramer's V = 0.486), and headaches ({chi}2(1) = 149.180, p < 0.001; Cramer's V = 0.508). No statistically significant association was observed for cardiovascular problems, likely reflecting the very low number of reported cases. Substantial socioeconomic disparities in perceived air quality and self-reported respiratory health outcomes were observed. Residents of low-socioeconomic status areas consistently perceived poorer air quality and reported a higher burden of respiratory problems, highlighting the need for targeted interventions to reduce environmental health inequalities.

Matching journals

The top 6 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.