A Multi-site Mortality and Morbidity Assessment of Air Pollution in South Africa's Priority Areas: an Adapted Three-Stage Distributed Lag Non-linear Case-Crossover Framework for Parsimonious Datasets
1
Environmental Health Unit, Climate Change and Health, South African Medical Research Council, Pretoria, South Africa
2
School of Health Systems and Public Health, University Of Pretoria, Pretoria, South Africa
Popul. Med. 2026;8(Supplement Supplement 1):
ABSTRACT
BACKGROUND:
In LMICs, population-level evidence on the acute health effects of air pollution is limited by sparse data and underdeveloped surveillance systems. South Africa’s Air Quality Management Priority Areas (Highveld, Vaal Triangle, and Waterberg-Bojanala), identified for chronic air quality exceedances, provide an opportunity to examine exposure.
OBJECTIVES:
We assessed the short-term effects of five major air pollutants (NO₂, SO₂, O₃, PM₁₀, PM₂.₅) on cause-specific mortality and morbidity between 2005 and 2020. We adapted a distributed lag non-linear model (DLNM) embedded in a three-stage DL-CCO framework for mortality, and a pseudo-case-crossover design for morbidity validation.
METHODS:
Weekly mortality and monthly morbidity data (ICD-10 J(All), A(15-19)) were linked to ambient air pollutant concentrations across the Priority Areas. The modelling strategy was: (1) DLNM estimation of district-level risk functions, (2) pooling via random-effects meta-analysis, and (3) application of a distributed lag case-crossover (DL-CCO) approach using conditional logistic regression to validate findings. For morbidity, where matched control data were unavailable, a pseudo-case-crossover approach was applied as a sensitivity test.
RESULTS:
Increases of 10 µg/m³ in PM2.5 and NO₂ were associated with elevated respiratory and infectious mortality risks within a 3-week lag structure. Pooled estimates showed a significant cumulative relative risk (RR) of 1.17 (95% CI: 1.09–1.26) for pneumonia following NO₂ exposure in HPA, and 1.21 (95% CI: 1.10–1.34) for tuberculosis mortality associated with PM2.5 in VTAPA. DL-CCO validation confirmed consistent lag–response patterns for mortality, while pseudo-CCO analyses for morbidity showed parallel but attenuated associations. No significant associations were found for SO₂ or O₃.
CONCLUSIONS:
This study is the first to implement a DLNM framework for mortality and pseudo-CCO sensitivity test for morbidity in Southern Africa. The multi-pollutant, multi-region analysis confirms the acute health burden of NO₂ and PM2.5 and demonstrates the feasibility of applying advanced epidemiologic models in resource-constrained settings.
In LMICs, population-level evidence on the acute health effects of air pollution is limited by sparse data and underdeveloped surveillance systems. South Africa’s Air Quality Management Priority Areas (Highveld, Vaal Triangle, and Waterberg-Bojanala), identified for chronic air quality exceedances, provide an opportunity to examine exposure.
OBJECTIVES:
We assessed the short-term effects of five major air pollutants (NO₂, SO₂, O₃, PM₁₀, PM₂.₅) on cause-specific mortality and morbidity between 2005 and 2020. We adapted a distributed lag non-linear model (DLNM) embedded in a three-stage DL-CCO framework for mortality, and a pseudo-case-crossover design for morbidity validation.
METHODS:
Weekly mortality and monthly morbidity data (ICD-10 J(All), A(15-19)) were linked to ambient air pollutant concentrations across the Priority Areas. The modelling strategy was: (1) DLNM estimation of district-level risk functions, (2) pooling via random-effects meta-analysis, and (3) application of a distributed lag case-crossover (DL-CCO) approach using conditional logistic regression to validate findings. For morbidity, where matched control data were unavailable, a pseudo-case-crossover approach was applied as a sensitivity test.
RESULTS:
Increases of 10 µg/m³ in PM2.5 and NO₂ were associated with elevated respiratory and infectious mortality risks within a 3-week lag structure. Pooled estimates showed a significant cumulative relative risk (RR) of 1.17 (95% CI: 1.09–1.26) for pneumonia following NO₂ exposure in HPA, and 1.21 (95% CI: 1.10–1.34) for tuberculosis mortality associated with PM2.5 in VTAPA. DL-CCO validation confirmed consistent lag–response patterns for mortality, while pseudo-CCO analyses for morbidity showed parallel but attenuated associations. No significant associations were found for SO₂ or O₃.
CONCLUSIONS:
This study is the first to implement a DLNM framework for mortality and pseudo-CCO sensitivity test for morbidity in Southern Africa. The multi-pollutant, multi-region analysis confirms the acute health burden of NO₂ and PM2.5 and demonstrates the feasibility of applying advanced epidemiologic models in resource-constrained settings.
| eISSN: | 2654-1459 |
We process personal data collected when visiting the website. The function of obtaining information about users and their behavior is carried out by voluntarily entered information in forms and saving cookies in end devices. Data, including cookies, are used to provide services, improve the user experience and to analyze the traffic in accordance with the Privacy policy. Data are also collected and processed by Google Analytics tool (more).
You can change cookies settings in your browser. Restricted use of cookies in the browser configuration may affect some functionalities of the website.
You can change cookies settings in your browser. Restricted use of cookies in the browser configuration may affect some functionalities of the website.
