**Background:** The World Health Organization estimated 450,000 cases of rifampin-resistant (RR) or multidrug-resistant TB (MDR-TB) in 2021. Household contacts (HHCs) of patients with bacteriologically confirmed pulmonary TB are at substantial risk of infection and disease, yet few studies have assessed risk factors specifically among HHCs of patients with drug-resistant TB. This study aimed to identify factors associated with prevalent Mycobacterium tuberculosis (Mtb) infection and prevalent TB disease in HHCs aged ≥15 years exposed to RR-TB in the household.
**Methods:** This was a secondary analysis of data from the PHOENIx Feasibility Study, a cross-sectional observational study conducted from October 2015 through April 2016 across 16 sites in 8 countries: Botswana, Brazil, Haiti, India, Kenya, Peru, South Africa, and Thailand. Adult index participants (≥18 years) with pulmonary MDR-TB confirmed by culture, molecular testing, or Xpert MTB/RIF rpoB mutation who had started appropriate TB treatment within the past 6 months were enrolled along with their HHCs. A HHC was defined as anyone living or having lived in the same dwelling or plot with shared housekeeping arrangements who reported exposure within 6 months before the index participant started MDR-TB treatment. For this analysis, only HHCs ≥15 years were included. Mtb infection was determined by interferon-gamma release assay (IGRA; QuantiFERON TB Gold or Gold In-Tube) among HHCs without current active TB diagnosed by the local TB program. TB disease was centrally adjudicated by a blinded outcome review group using modified ACTG criteria (confirmed TB required a positive MTBC culture, Xpert MTB/RIF, or HAIN GenoType MTBDRplus plus compatible clinical symptoms; probable pulmonary TB required compatible symptoms plus positive sputum smear, abnormal chest imaging, or caseating granulomata). Logistic regression models using generalized estimating equations (GEE) with exchangeable working correlation were fitted to account for household clustering. Multivariable models included age and HIV status a priori; other covariates significant at p<0.25 in univariable analysis were considered using manual forward selection and retained if p<0.05.
**Key Results:** A total of 712 HHCs ≥15 years from 279 households were enrolled. Median age was 34 years (range 15–90), 63% were female, 22% were current smokers and 8% previous smokers, 8% were HIV-positive, and 11% had been previously treated for TB. Of 686 HHCs with determinate IGRA results, 471 tested positive, yielding a prevalence of Mtb infection of 68.8% (95% CI: 64.6%–72.8%). In multivariable analysis, IGRA positivity was significantly associated with age 25–49 years (aOR 1.6 vs. ≥50 years, 95% CI: 1.0–2.5), prior TB treatment (aOR 2.1, 95% CI: 1.0–4.2), incarceration in the past 12 months (aOR 7.3, 95% CI: 1.5–34.4), substance and/or alcohol use without incarceration (aOR 2.6, 95% CI: 1.3–5.5), sharing a sleeping room with the index participant (aOR 2.4, 95% CI: 1.4–4.2), spending 6–7 evenings/week with the index participant even if in a different room (aOR 2.1, 95% CI: 1.2–3.5), living in a household with smokers (aOR 1.6, 95% CI: 1.2–2.5), and living in a home with exterior walls made of adobe, rammed earth, sticks/stones and mud, wood, straw, tin, or other materials (aOR 1.9, 95% CI: 1.2–2.9).
Forty-six HHCs (6.5%, 95% CI: 4.6%–9.0%) had prevalent TB disease (8 diagnosed before study entry, 14 confirmed, 24 probable). Among HHCs with prior TB treatment, prevalence was 30.0% (95% CI: 20.9%–41.0%) vs. 3.5% (95% CI: 2.2%–5.5%) among those without. In multivariable analysis (excluding 8 HHCs diagnosed before study entry), TB disease was significantly associated with age ≥50 years (aOR 0.3 for age 25–49 vs. ≥50, 95% CI: 0.1–0.6; aOR 0.5 for age 15–24 vs. ≥50, 95% CI: 0.2–1.1), current smoking (aOR 2.9, 95% CI: 1.3–6.4), previous smoking (aOR 4.2, 95% CI: 1.5–11.7), daily alcohol use in the past 12 months (aOR 3.1, 95% CI: 1.3–7.5), and prior TB treatment (aOR 9.7, 95% CI: 4.5–21.1). Notably, among 12 HHCs with confirmed TB who had drug-susceptibility testing for both isoniazid and rifampin, 8 (67%) had no resistance to either drug, discordant with their index participant's resistance profile.
**Clinical Implications:** This study demonstrates that HHCs of patients with RR-TB have a high burden of both Mtb infection (68.8%) and TB disease (6.5%), yet very few were receiving TB preventive treatment (only 0.1% were on TPT at enrollment). The identification of overlapping risk factors for infection (exposure intensity, incarceration, substance/alcohol use, lower SES housing, household smoking) and distinct risk factors for disease (older age, personal smoking, heavy alcohol use, prior TB treatment) can help TB program staff prioritize HHCs for screening and preventive therapy. The high rate of discordant drug-susceptibility profiles between index participants and HHCs suggests that transmission may occur from sources outside the household or that drug-susceptible strains may be selectively transmitted, underscoring the importance of comprehensive diagnostic evaluation rather than empiric treatment based on the index patient's resistance pattern. Limitations include the cross-sectional design (which cannot establish temporality), potential underestimation of infection due to imperfect IGRA sensitivity, and variable time between index participant treatment initiation and HHC enrollment (median 9 weeks, range 0–33 weeks), which may have affected prevalence estimates.