Spatial distribution and determinants of tuberculosis incidence in Mozambique: A nationwide Bayesian disease mapping study

Autores da FMUP

• Luís Filipe Ribeiro De Azevedo

  Autor

Participantes de fora da FMUP

• Cuboia, N
• Reis-Pardal, J
• Pfumo-Cuboia, I
• Manhi?a, I
• Mutaquiha, C
• Nitrog?nio, L
• Zindoga, P

Unidades de investigação

• Laboratório Associado RISE- Rede de Investigação em Saúde

  

  Investigador

  Fernando Carlos De Landér Schmitt

• Medicina da Comunidade, Informação e Decisão em Saúde

• RISE-Health

  

  Investigador

  Fernando Carlos De Landér Schmitt

Abstract

Introduction: Mozambique is a high-burden country for tuberculosis (TB). International studies show that TB is a disease that tends to cluster in specific regions, and different risk factors (HIV prevalence, migration, overcrowding, poverty, house condition, temperature, altitude, undernutrition, urbanization, and inadequate access to TB diagnosis and treatment) are reported in the literature to be associated with TB incidence. Although Mozambique has a higher burden of TB, the spatial distribution, and determinants of TB incidence at the subnational level have not been studied yet for the whole country. Therefore, we aimed to analyze the spatial distribution and determinants of tuberculosis incidence across all 154 districts of Mozambique and identify the hotspot areas. Method: We conducted an ecological study with the district as our unit of analysis, where we included all cases of tuberculosis diagnosed in Mozambique between 2016 and 2020. We obtained the data from the Mozambique Ministry of Health and other publicly available open sources. The predictor variables were selected based on the literature review and data availability at the district level in Mozambique. The parameters were estimated through Bayesian hierarchical Poisson regression models using Markov Chain Monte Carlo simulation. Results: A total of 512 877 people were diagnosed with tuberculosis in Mozambique during our five-year study period. We found high variability in the spatial distribution of tuberculosis incidence across the country. Sixtytwo districts out of 154 were identified as hotspot areas. The districts with the highest incidence rate were concentrated in the south and the countrys central regions. In contrast, those with lower incidence rates were mainly in the north. In the multivariate analysis, we found that TB incidence was positively associated with the prevalence of HIV (RR: 1.23; 95 % CrI 1.13 to 1.34) and negatively associated with the annual average temperature (RR: 0.83; 95 % CrI 0.74 to 0.94). Conclusion: The incidence of tuberculosis is unevenly distributed across the country. Lower average temperature and high HIV prevalence seem to increase TB incidence. Targeting interventions in higher-risk areas and strengthening collaboration between HIV and TB programs is paramount to ending tuberculosis in Mozambique, as established by the WHOs End TB strategy and the Sustainable Development Goals.

Keywords

• Tuberculosis; Mozambique; Spatial analysis; Predictors; Incidence; HIV