Factors associated with new leprosy diagnosis in Kwale County, Kenya
Data files
Oct 22, 2025 version files 51.04 KB
Abstract
Leprosy, a chronic bacterial disease caused by Mycobacterium leprae, is curable yet neglected. Approximately 200,000 new cases are reported globally each year, with India contributing 60%. In 2020, the African WHO regions had a leprosy burden of 14.9 per 1,000,000 population. Despite maintaining the global elimination target of <1/10000 population, Kenya reported a six-fold increase in cases(63-163) from 2011 to 2021, with Kwale County contributing 24.3%. This study aimed to determine the factors associated with new leprosy diagnoses. We conducted a 1:3 case-control study in the Kwale from June to September 2023. Cases included people who were treated for leprosy based on clinical/laboratory and epidemiological criteria between January 2022 and May 2023. Controls were persons with no signs or symptoms and were a neighboring household to a case in another or nearby plot, matched by sex, age group of ±10 years, and village. Questionnaires were administered to both groups. Factors associated with Leprosy were evaluated using multivariable logistic regression. Stepwise backward elimination was used to build a final model; p-values of ≤0.05 were considered significant. A total of 65 cases and 195 controls were enrolled. The mean age was 55 years (SD±16) for the cases and 54 years (SD±15) for the controls. Among cases,56.6% (n=37) were married, compared to 71.1% (n=139) of controls. 55% (n=36) of the cases and 41% (n=81) of the controls were illiterate. The odds of being diagnosed with leprosy were seven times higher among patients with a family size ≥5 members (aOR=6.99, 95% CI: 2.71–18.06) and four times higher among those with a family contact (aOR=4.33, 95% CI: 2.18–8.58). Social contact (aOR=2.24,95% CI: 1.16–4.32) and non-vaccination with BCG (aOR=2.24,95% CI: 1.11–4.53) are associated with double odds of new leprosy diagnosis. Leprosy prevention efforts in high-burden counties should prioritize early detection, prompt treatment of cases, enhanced community-based surveillance, and household contact tracing to reduce transmission. In addition, the Ministry of Health should sustain and expand the BCG vaccination coverage among all eligible persons.
https://doi.org/10.5061/dryad.pg4f4qs14
Description of the data and file structure
File: Factors_Associated_with_New_Leprosy_Diagnosis_in__Kwale_CountyVersion_III.xlsx
The repository contains data and supporting files for the manuscript titled: “Factors Associated with New Leprosy Diagnosis in Kwale County, Kenya.”
Data Structure:
The dataset consists of de-identified data collected from participants in Kwale County, Kenya, focusing on Factors Associated with New Leprosy Diagnosis
Dataset:
- data_leprosy_kwale_Kenya. Xlsx-Contains anonymized individual-level data with demographic, clinical, behavioural, and environmental risk factors.
- Metadata text -Explains the data collection methods, inclusion/exclusion criteria, and ethical considerations.
File Structure:
- Main datasets, variable descriptions, data collection details, descriptive analysis results, manuscript for submission
Description of key variables:
- Demographic factors: Age, gender, occupation, residence
- Clinical factors: Contact history, BCG vaccination
- Behavioural Factors: Frequency of changing line
- Environmental factors: Family size, overcrowding
Data access & ethical considerations
- The data is de-identified to protect participant privacy.
- Ethical approval was obtained from Moi University Institutional Review and Ethics Committee (Reference number: IREC/405/2023)
Data were analyzed using MS Excel and STATA version 16.1. Descriptive analysis summarized continuous variables as means and standard deviations. Categorical variables as proportions. Bivariate logistic regression assessed the associations between independent variables and the leprosy disease. Variables with p-values ≤0.2 were included in the multivariable analysis. Stepwise backward elimination was applied to build the best-fitting model, given the multiple variables under consideration and the nature of the study design, where the outcome is already known. This technique allowed the efficient selection of the statistically significant risk factors while controlling for potential confounders. Factors with p-values ≤0.05 were retained in the final model. Adjusted odds ratios with 95% confidence intervals were reported to determine the significance of associations.
