Spatial epidemiology of tuberculosis diagnostic delays, healthcare access disparities, and socioeconomic inequities in Nairobi County, Kenya

Abstract

Introduction: Kenya ranks among the top 30 countries with a high tuberculosis (TB) burden globally. With a TB prevalence of 558 per 100,000, only 46% of TB cases are diagnosed and treated, leaving 54% undiagnosed and at risk of spreading the disease. This study analyzed the spatial distribution of tuberculosis diagnostic delays and their association with health care accessibility and socioeconomic inequalities in Nairobi County, Kenya.

Materials and methods: The cross-sectional study included 222 newly diagnosed bacteriologically confirmed Mycobacterium tuberculosis (Mtb) patients from Mbagathi County Hospital (MCH), Mama Lucy Kibaki Hospital (MLKH), and Rhodes Chest Clinic (RCC) in Nairobi County, Kenya. Patients were recruited consecutively through census sampling and categorized into two groups: delayed diagnosis (≥21 days from symptom onset) and non-delayed (<21 days) as defined by the WHO cutoff point. Patients' residential locations were georeferenced using handheld GPS devices and captured digitally via Kobo Collect. Spatial analyses were performed using ArcGIS Pro, version, where Global Moran's I statistic was used to assess spatial autocorrelation in the distribution of TB cases.

Result: Spatial analyses identified 28 statistically significant clusters of delayed TB diagnoses within Nairobi County. Spatial autocorrelation analysis using Moran's I revealed a significant clustered distribution (Moran's Index = 0.471, z-score = 3.370, p < 0.001). Hotspot analysis with the Getis-Ord Gi* statistic detected high-delay clusters (z > 2.58, p < 0.001) in informal settlements.

Discussion and conclusion: The study revealed significant spatial clustering of delayed TB diagnoses in Nairobi County, particularly in informal settlements. In contrast, timely diagnoses were predominantly clustered in high-income areas like Lang'ata and Karen. These clusters were significantly associated with lower household income and increased travel time to health facilities which underscored the need for targeted implementation of TB diagnostic services and control measures in the wards with the highest delays.

Fig 1. The map illustrates the distribution of delayed (red) and non-delayed (green) tuberculosis (TB) cases across Nairobi County.

Fig 2. Spatial autocorrelation report.

Fig 3. This map illustrates the hot and cold spots of delayed and non-delayed tuberculosis (TB) diagnoses across Nairobi County.

Fig 4. This map illustrates the clustering of delayed and non-delayed (TB) cases in Nairobi County, highlighting clustering patterns of delayed and non-delayed diagnoses.

Fig 5. This map illustrates the spatial distribution of tuberculosis (TB) cases in Nairobi County, highlighting clustering patterns of delayed and non-delayed diagnoses.

Fig 6. Cluster and Outlier scatterplot of delayed and non-delayed Tuberculosis cases in Nairobi County.

Fig 7. Spatial Analysis of Tuberculosis Diagnostic Delays and Distance to the diagnostic health facility.

Fig 8. Spatial Analysis of Tuberculosis Diagnostic Delays and Time travel to the diagnostic health facility.

Fig 9. Bivariate map of Tuberculosis Diagnostic Delays and Household income in Nairobi County.

Fig 10. Multivariate map of tuberculosis diagnostic delays, travel distance, travel time, and household income in Nairobi County.

Fig 11. service area analysis of public health facilities in Nairobi County, Kenya.

Fig 12. Location-allocation of new health facilities in Nairobi County, Kenya.

Fig 13. Closest Facility Route Network Analysis showing distance to the diagnosing health facilities in Nairobi County, Kenya.