Should neighbours of tuberculosis (TB) cases be prioritised for active case finding in high TB-burden settings? A prospective molecular epidemiological study.
Raspati C Koesoemadinata, Susan McAllister, Chuan-Chin Huang, Sri Hartati, Hanif Djunaedy, Nury Fitri Dewi, Lidya Chaidir, Bony Wiem Lestari, et al. (13 authors)
BMJ global health · 2025-09
Abstract
INTRODUCTION: In high tuberculosis (TB)-burden countries, considerable transmission of() likely occurs outside of households. We aimed to estimate the TB prevalence and incidence in households and neighbourhoods around known TB cases and to understand transmission patterns.
METHODS: Household and neighbourhood contacts of pulmonary TB index cases from contiguous areas in Bandung, Indonesia, were screened and followed up for 12 months. Sputum samples underwent smear microscopy,culture, Xpert MTB/RIF, DNA isolation and whole-genome sequencing (WGS). Pairwise single-nucleotide polymorphism (SNP) distance ≤12 defined transmission for pairs with known epidemiological links, or SNP≤3 for pairs without epidemiological link. An SNP=12 cut-off was used to characterise transmission clusters.
RESULTS: From 213 index cases, 514 household and 4141 neighbourhood contacts underwent TB screening: 19 household (3.70%, 95% CI 2.24 to 5.71) and 45 neighbourhood (1.09%, 95% CI 0.79 to 1.45) contacts were identified with TB, of whom 18 (3.50%, 95% CI 2.20 to 5.48) and 38 (0.92%, 95% CI 0.65 to 1.13) respectively, were bacteriologically confirmed. During follow-up, 11 household and 13 neighbourhood contacts were identified with TB (incidence per 100 000 person-years: 2286 (95% CI 1286 to 4148) and 350 (95% CI 190 to 563)), of whom 6 and 8, respectively, were bacteriologically confirmed (incidence per 100 000 person-years: 1247 (95% CI 560 to 2776) and 201 (95% CI 101 to 402)). A total of 223 patientisolates underwent WGS. Of 15 intra-household pairs, 8 (53.3%) were transmission pairs. Of 24 neighbour to index case pairs, 1 (4.2%) was a transmission pair. 11 of 19 transmission pairs shared no epidemiological link. We identified 25.genetic clusters from 205 mono-TB isolates overall.
CONCLUSION: Neighbours have lower prevalence and incidence of TB than household contacts, but twice as many cases. Very few receivedfrom their index case, suggesting uncontrolled community-wide transmission. Whole population active case finding may be necessary in high TB-burden settings.
MeSH terms
- Humans
- Male
- Adult
- Female
- Prospective Studies
- Indonesia
- Mycobacterium tuberculosis
- Molecular Epidemiology
- Contact Tracing
- Incidence
- Middle Aged
- Tuberculosis, Pulmonary
- Prevalence
- Young Adult
- Adolescent
- Family Characteristics
- Residence Characteristics
- Child
- Polymorphism, Single Nucleotide
- Child, Preschool