Air Pollution Spikes Elevate Cardiorespiratory Risks in Kinshasa Study

In an ecological investigation conducted across Kinshasa, researchers determined that short-term exposure to elevated concentrations of fine particulate matter, known as PM2.5, significantly increases the risk of hospital admissions for respiratory and cardiovascular diseases. Fine particulate matter consists of airborne particles with an aerodynamic diameter smaller than 2.5 micrometers, which can penetrate deep into pulmonary alveoli and enter the systemic circulation to induce both localized pulmonary inflammation and broader systemic complications. Throughout the evaluated multi-year period from March 2018 to August 2020, in situ air quality monitoring within the Democratic Republic of Congo’s capital revealed that daily mean concentrations persistently exceeded the World Health Organization’s air quality guideline threshold of 15 micrograms per cubic meter. The study identified pronounced temporal dynamics in these exposure levels, demonstrating that the health risks faced by urban populations are heavily concentrated during recurring dry season periods when pollution reaches critical peaks.

The investigation addressed a prominent gap in environmental health research across sub-Saharan Africa, where empirical data tracking the immediate health effects of short-term particulate matter fluctuations have remained exceptionally scarce despite rapid urban expansion. Kinshasa, a megacity home to approximately 13.9 million residents according to 2019 demographic records, experiences substantial environmental pressures driven by heavy vehicular traffic, domestic biomass and waste combustion, road dust, industrial activity, and an absence of national regulatory air quality standards. Previous monitoring efforts documented substantial baseline pollution levels, including an average daily PM2.5 concentration of 94.7 micrograms per cubic meter in 2017, but few regional analyses have quantitatively modeled the associated acute epidemiological burdens using specialized software tools. By coupling real-time optical sensor data with standardized risk modeling frameworks, the research team sought to evaluate the Air Quality Index across urban municipalities and estimate the specific cardiorespiratory risks confronting the general population across all age demographics.

To capture particulate matter concentrations, the researchers analyzed continuous environmental measurements obtained from an established monitoring network deployed in collaboration with Columbia University’s Lamont-Doherty Earth Observatory. This ground-level network utilized PurpleAir PA-II-SD low-cost sensors distributed across four representative urban and academic locations, including the United States Embassy in Gombe, Cité Bella Vue in Kintambo, the National Pedagogical University in Ngaliema, and the University of Kinshasa campus in Lemba. Each device incorporated twin Plantower optical modules designed to record particle counts across six distinct size fractions every sixty seconds, alongside integrated meteorological sensors measuring ambient temperature, atmospheric pressure, and relative humidity. Because low-cost optical counters exhibit systematic measurement biases under varied environmental conditions, the sensor outputs were calibrated against a reference BAM-1020 regulatory monitor using multiple linear regression and random forest models. This correction adjusted the mean absolute error from an initial 14 micrograms per cubic meter down to 3.4 micrograms per cubic meter while raising the coefficient of determination to 0.96.

Spatial consistency across the sensor network was evaluated through cross-correlation analyses and spatial autocorrelation metrics to verify how effectively the monitors represented regional air masses. Daily particulate readings between the four distinct sites demonstrated positive correlation coefficients ranging from 0.44 to 0.78, indicating that broad meteorological phenomena govern pollution over the entire metropolis while local urban microenvironments introduce observable variations. Spatial clustering was confirmed through Moran’s I autocorrelation testing, which yielded a value of 0.63 with a statistically significant score confirmed by Monte Carlo simulation. Because the monitoring station located at the U.S. Embassy provided the most complete continuous daily time series alongside the strongest correlations with neighboring stations, its calibrated record was selected as the primary exposure dataset for the subsequent health impact assessments.

Daily PM2.5 mass concentrations were converted into the standardized U.S. Environmental Protection Agency Air Quality Index, which categorizes atmospheric purity into numbered intervals and health risk bands. Over the entirety of the monitoring timeframe between 2018 and 2020, the analysis found zero days that met the criteria for the “Good” air quality category, which corresponds to an index value between 0 and 50 or concentrations below 9.0 micrograms per cubic meter. The temporal distribution revealed that air quality predominantly fluctuated between moderate, unhealthy for sensitive groups, and unhealthy classifications. In 2018, “Moderate” conditions accounted for 30 percent of days, while “Unhealthy for Sensitive Groups” represented 36 percent, “Unhealthy” comprised 33 percent, and “Very Unhealthy” conditions were recorded on 1 percent of days. The proportion of moderate air quality days expanded notably to 47 percent in 2019 and climbed to 81 percent in 2020, a year during which the unhealthy category was entirely absent and days classified as unhealthy for sensitive groups dropped to 19 percent, reflecting shifts associated with anthropogenic emission reductions during COVID-19 pandemic restrictions.

To estimate the relative risks of hospital admissions for respiratory and cardiovascular conditions, the researchers employed the World Health Organization’s AirQ+ software package. This epidemiological assessment tool applies a log-linear model where relative risks scale exponentially with increasing pollutant concentrations above an established baseline cut-off of 8.8 micrograms per cubic meter. The modeling parameters utilized beta coefficients of 0.0019 for respiratory diseases, carrying a 95 percent confidence interval from 0 to 0.0039, and 0.0009 for cardiovascular diseases, with a 95 percent confidence interval from 0.0002 to 0.0016. Because localized hospital admission registries were not available across the metropolis, baseline national admission rates and disease proportions were derived from the Institute for Health Metrics and Evaluation Global Burden of Disease database and scaled to Kinshasa’s urban population projections.

The AirQ+ simulations demonstrated an upward, monotonic relationship between PM2.5 concentrations and the relative risk of hospital admissions across all analyzed categories. In 2018, when mean daily PM2.5 concentrations reached 45.4 micrograms per cubic meter under conditions unhealthy for sensitive groups, 78.35 micrograms per cubic meter on unhealthy days, and 136.85 micrograms per cubic meter during very unhealthy events, estimated respiratory admission risks increased by 5.9 percent, 12.7 percent, and 25.8 percent, respectively. For cardiovascular diseases during that same year, corresponding relative risks reached 1.028, 1.059, and 1.117, indicating excess admission increases of 2.8 percent, 5.8 percent, and 11.78 percent. In 2019, exposures associated with moderate to unhealthy levels corresponded to respiratory admission increases of 5.8 percent and 9.6 percent, alongside cardiovascular admission increases of 2.8 percent and 4.5 percent. By 2020, lower mean pollution concentrations of 39.35 micrograms per cubic meter on sensitive-group alert days were tied to risk increases of 4.7 percent for respiratory admissions and 2.2 percent for cardiovascular admissions.

A strong seasonal dynamic characterized the health burden throughout the study, showing that adverse impacts peaked sharply during the regional dry season extending from June through September. During these dry months, elevated concentrations of fine particulate matter coincided with meteorological factors such as reduced precipitation that eliminated atmospheric pollutant washout, dry winds that resuspended surface mineral dust, and atmospheric temperature inversions that limited vertical dispersion. These environmental variables combined with intensified seasonal biomass combustion and savannah burning, driving air quality indices into the unhealthy and very unhealthy tiers. The model confirmed that relative risks for cardiorespiratory hospital admissions were highest during June and July, reaching a peak relative risk of 1.258 for respiratory diseases and 1.117 for cardiovascular disorders, illustrating the heightened seasonal vulnerability of residents across the metropolitan area.

The study also examined the underlying distribution of diagnostic categories in the health data, uncovering disease profiles that differ notably from high-income urban environments. In Kinshasa, records derived from the IHME database revealed that 97.3 percent of all respiratory hospital admissions were associated with upper respiratory tract infections such as rhinitis, with lower respiratory tract infections such as bronchitis and pneumonia comprising 2.3 percent, and chronic noncommunicable respiratory ailments like asthma and chronic obstructive pulmonary disease accounting for only 0.3 percent. Among cardiovascular hospitalizations, ischemic heart disease constituted 48.6 percent and stroke represented 27.9 percent. Researchers noted that the acute nature of upper respiratory infections dominating the local burden contrasts sharply with Western datasets, where chronic illnesses like COPD drive the majority of pollution-linked hospital encounters, potentially explaining why short-term particulate spikes produced such pronounced relative risk surges in this population.

Several methodological constraints were highlighted by the authors to provide necessary context for interpreting the findings. Primary among these is the reliance on particulate data from a single long-term monitoring station at the U.S. Embassy, which may not capture localized intra-urban exposure variations across informal settlements or high-density traffic corridors. Furthermore, health data were derived from generalized IHME national modeling estimates rather than direct, real-time hospital admissions logs from Kinshasa clinics, potentially masking local diagnostic nuances. The log-linear beta coefficients implemented in AirQ+ originated from international epidemiological literature characterized by lower baseline air pollution environments, which could underestimate the true physiological consequences of extreme exposures. Finally, chemical speciation was not performed to measure specific toxic constituents such as elemental carbon, heavy metals, or secondary organic aerosols. Despite these limitations, the authors concluded that the findings provide essential baseline evidence demonstrating substantial health hazards from urban PM2.5 in sub-Saharan Africa, supporting the urgent implementation of continuous air monitoring networks, targeted dust suppression, clean household energy programs, and seasonal healthcare preparedness.

Subject of Research: Short-term fine particulate matter exposure and associated cardiorespiratory hospitalization risks in Kinshasa

Article Title: Short-term effects of PM2.5 exposure and cardiorespiratory health risks in a Sub-Saharan African urban setting

Article References: Ngutuka, M. K., Matondo, D. M., Bangelesa, F., Kianu, B. P., Kabengele, B. O., Tangou, T. T., Konde, J. N., Irankunda, E., Kiyombo, G. M., & Kayembe, J. M. N. (2025). Short-term effects of PM2.5 exposure and cardiorespiratory health risks in a Sub-Saharan African urban setting. Discover Atmosphere, 3(1), Article 42. https://doi.org/10.1007/s44292-025-00068-0

Image Credits: AI Generated

DOI: 10.1007/s44292-025-00068-0

Keywords: air pollution, PM2.5, Kinshasa, cardiorespiratory health, environmental health, public health, respiratory disease, cardiovascular disease, AirQ+, sub-Saharan Africa, Short-term, effects