The social determinants of tuberculosis in Peru

---

Who

• Cases: 2,337 individuals aged ≥15 years diagnosed with pulmonary or extrapulmonary tuberculosis (with or without bacteriological confirmation). Median age 31 years (IQR 23–47); 64% male.

• Controls: 981 individuals aged ≥15 years from randomly selected households in the same communities. Median age 38 years (IQR 25–54); 40% male.

• Setting population: Residents of 32 high–tuberculosis-burden communities (~900,000 people).

• Healthcare context: Communities served by Ministry of Health (MINSA)-run health posts.

---

What

• The study examined how household-level poverty and interrelated personal risk factors (e.g., smoking, alcohol use, undernutrition, education, incarceration, social capital) increase the risk of tuberculosis.

• Key findings:

  • Household poverty was strongly associated with tuberculosis (adjusted odds ratio [aOR] 3.1 for poorer vs. less poor households).

  • Tuberculosis risk increased non-linearly with worsening poverty; 21% of cases were in the poorest poverty decile.

  • Population attributable fractions (PAFs) suggested that nearly 47% of tuberculosis burden could be reduced if poorer households achieved poverty levels comparable to the less poor.

  • Several personal risk factors independently contributed to tuberculosis risk even after adjusting for poverty, including low education, alcohol excess, underweight, smoking, HIV, diabetes, prior tuberculosis, incarceration, and low social capital.

  • Most personal risk factors showed clear social gradients, being more prevalent among poorer households, except HIV (no gradient) and diabetes/other immunosuppression (more prevalent in less poor households).

---

When

• Communities were followed from 2013 onward.

• Recruitment and detailed data collection occurred during the study period up to 2019.

• Tuberculosis notification data refer to 2019.

---

Where

• Callao, Peru, a metropolitan area bordering Lima.

• Specifically, 32 of 45 communities in Callao with high tuberculosis rates.

---

Why

• To address gaps in understanding how household poverty and downstream personal risk factors interact to shape tuberculosis risk.

• The study aimed to move beyond single risk factors and explicitly apply a social epidemiological framework to tuberculosis transmission and vulnerability.

---

How

• Study design: Case–control study nested within the PREVENT TB study.

• Case identification: Passive case finding through MINSA-run health posts; cases recruited at diagnosis or during treatment.

• Control selection: Randomly selected households using satellite mapping and random number tables; all household members invited after adult consent.

• Data collection: Structured questionnaires administered by trained research nurses.

  • Household poverty assessed across physical, human, and financial capital dimensions using the Sustainable Livelihood Framework and principal component analysis (PCA).

  • Personal risk factors grouped into five domains: education/behavioural, exposure, biological, nutritional, and psychosocial.

• Analysis: Directed acyclic graphs (DAGs) guided causal assumptions; multivariable regression estimated adjusted odds ratios and population attributable fractions.

Source: Saunders, M.J., Montoya, R., Quevedo, L., Ramos, E., Datta, S. and Evans, C.A., 2025. The social determinants of tuberculosis: a case-control study characterising pathways to equitable intervention in Peru. Infectious diseases of poverty, 14(1), p.53.