Diagnostic accuracy of CXR CAD software for detection of TB in household contacts [TBN 087]

A prospective cohort study evaluated digital chest X-ray computer-aided detection (CAD) among adult household contacts of patients with rifampicin-resistant tuberculosis (RR-TB) in Khayelitsha, South Africa. Recruitment occurred from November 2014 to September 2017, with follow-up until May 2021. The objectives were to assess the diagnostic accuracy of three CAD packages for prevalent and incident pulmonary TB, evaluate recommended CAD thresholds, and compare or combine CAD scores with blood-based biomarkers.

Eligible participants were household contacts aged 18 years or older. Pregnant participants, those already on TB treatment, and those without CAD readings were excluded. At baseline, all participants underwent symptom screening, HIV testing, physical examination, digital posterior-anterior CXR, and microbiological testing regardless of symptoms, using spontaneous and induced sputum samples processed by smear microscopy, Xpert MTB/RIF, and MGIT culture. Three CAD tools were evaluated: CAD4TBv7, qXRv3, and Lunitv3, using thresholds of 50, 0.5, and 0.15, respectively. No participants received preventive therapy, consistent with guidelines at the time. A nested subgroup of HIV-uninfected, asymptomatic participants also underwent CRP, ESR, QuantiFERON-Gold, and 3-gene RNA MTB-HR testing.

Among 483 analyzed participants, median age was 33 years, 61% were female, 23% had previous TB, and 28% were people with HIV. Median follow-up was 4.6 years. Prevalent bacteriologically confirmed TB was found in 23 participants (4.7%), and 38 of 460 participants without prevalent TB later developed incident TB (8.3%). CAD tools performed well for prevalent TB, with AUCs of 0.87 to 0.91 for all prevalent cases, but were less accurate for predicting incident TB from baseline CXR, with AUCs of 0.60 to 0.65. At recommended thresholds, sensitivity and specificity for all prevalent TB were 0.70/0.93 for CAD4TBv7, 0.57/0.94 for qXRv3, and 0.87/0.86 for Lunitv3, compared with 0.61/0.87 for human CXR reading. CAD accuracy was better in participants without previous TB. In the biomarker subgroup, CAD outperformed blood biomarkers for asymptomatic prevalent TB, and adding blood biomarkers did not significantly improve detection of prevalent or incident TB.

Overall, CAD-based CXR screening was useful for detecting prevalent TB among adult RR-TB household contacts, including asymptomatic cases, but had limited ability to predict future incident TB. Key limitations include a single high-burden setting, exclusion of children and pregnant participants, incomplete follow-up sputum rescreening, and reduced generalizability to populations receiving preventive therapy.

Source: Macpherson L, Kik SV, Quartagno M, Lakay F, Jaftha M, Yende N, Galant S, Aziz S, Daroowala R, Court R, Taliep A. Diagnostic accuracy of chest X-ray computer-aided detection software for detection of prevalent and incident tuberculosis in household contacts. Clinical Infectious Diseases. 2025 Mar 15;80(3):626-36.

Reducing Household Tuberculosis Transmission [TBN 086]

A pilot cluster-randomized controlled trial used a hybrid type 1 effectiveness-implementation design to evaluate whether a targeted respiratory bundle could reduce acquisition of Mycobacterium tuberculosis (Mtb) infection among household contacts (HHCs) of patients with pulmonary tuberculosis. The study was conducted in Santiago, Chile, between October 2021 and April 2024 across three healthcare districts comprising 44 primary healthcare clinics. Healthcare districts were randomized at the cluster level to either the intervention arm (2 districts, 25 clinics) or control arm (1 district, 19 clinics) to minimize contamination and facilitate real-world implementation.

Eligible index patients had newly diagnosed pulmonary tuberculosis confirmed by culture, acid-fast bacillus smear, or Xpert MTB/RIF Ultra PCR and had received no more than three daily doses of anti-tuberculosis therapy. Household contacts aged >5 years were invited to participate. The intervention consisted of a two-week respiratory bundle: KN95/FFP2 mask use by both patients and household contacts when sharing indoor spaces, sleeping separately for the index patient, improved ventilation through open windows, and educational materials. Controls received routine tuberculosis care. Household contacts underwent symptom screening, chest radiography, and QuantiFERON-TB Gold Plus (QFT) testing at baseline and, if initially QFT-negative, again after 12 weeks. The primary outcome was incident tuberculosis infection, defined by QFT conversion.

A total of 157 index patients and 384 household contacts were included in the analysis. Among household contacts, 32.3% had positive baseline QFT results and 67.7% were QFT-negative. Of the 216 QFT-negative contacts assigned to intervention or control groups, 179 (82.9%) completed 12-week follow-up. QFT conversion occurred in 10.8% (10/93) of controls and 12.8% (11/86) of intervention participants, yielding a risk ratio (RR) of 1.10 (95% CI, 0.71-1.71; P = .68), indicating no significant reduction in new tuberculosis infection with the intervention. In the per-protocol analysis, participants with good adherence to the respiratory bundle at both day 7 and day 14 had a QFT conversion rate of 6.7%, compared with 10.8% in controls (RR 0.69, 95% CI 0.25-1.91; P = .47), although this difference was not statistically significant. Factors independently associated with increased risk of QFT conversion included high sputum bacillary burden in index patients (adjusted RR [adjRR] 12.10, 95% CI 2.52-55.81), drug use by the index patient (adjRR 10.02, 95% CI 2.70-36.33), suboptimal treatment adherence (adjRR 3.56, 95% CI 1.17-10.74), and household contact age below 45 years (adjRR 7.56, 95% CI 1.57-35.37). The intraclass correlation coefficient for QFT conversion within households was 0.085 (95% CI 0.005-0.360).

In this pilot cluster-randomized trial, the respiratory bundle did not significantly reduce incident Mtb infection among household contacts in the intention-to-treat analysis. However, lower infection rates among participants with good adherence suggest that adherence may influence effectiveness and warrants further investigation in larger trials. Important implementation barriers included household overcrowding, limited ability to isolate index patients, family social dynamics around meals, and stigma related to tuberculosis disclosure. As a pilot study, statistical power was limited, and confidence intervals were wide. The study provides moderate-level evidence from a randomized design regarding feasibility and implementation challenges in real-world household tuberculosis prevention.

Source: Ruiz-Tagle C, Seguel R, Villarroel L, Bernales M, Vargas-García S, Pizarro A, Peña C, Neira V, García P, Allel K, Nathavitharana RR. Reducing Household Tuberculosis Transmission: A Pilot Cluster-Randomized Controlled Trial. Clinical Infectious Diseases. 2026 Feb 15;82(2):291-8.

Gendered Patterns of Suboptimal Care Engagement Among TB Patients Who “Successfully” Completed Treatment [TBN 085]

A prospective cohort study examined patterns of care engagement among adults with drug-susceptible pulmonary TB who were programmatically classified as having treatment success. The study used latent class trajectory modeling of medication refill data and was conducted from February 2021 to August 2022 in 21 government healthcare facilities in Buffalo City Metro Health District, Eastern Cape Province, South Africa.

The analysis included 548 of 657 enrolled adults (83.4%) who were classified as cured or treatment completed. Eligible participants were aged 18 years or older, spoke English or isiXhosa, lived in a participating clinic catchment area, and gave informed consent; people with extrapulmonary TB without lung involvement or drug-resistant TB were excluded. Participants completed staff-administered questionnaires on sociodemographic factors, health status, PHQ-9 depression symptoms, AUDIT alcohol use, GAD-7 anxiety symptoms, TB knowledge, attitudes, and beliefs. Refill dates, scheduled visits, treatment start dates, and outcomes were abstracted from medical records. The main outcome was cumulative missed TB medication refill days during treatment, analyzed using latent class trajectory modeling. Level of evidence: prospective observational cohort.

Among those with treatment success, median age was 38 years (IQR, 30 to 47), 67% were men, 78.3% were unemployed, 46.2% were living with HIV, 28.1% had previous TB, and 38.9% screened positive for moderate to severe depression. Three overall engagement trajectories were identified: consistent engagement (84.1%), suboptimal engagement after 2 months (7.7%), and suboptimal engagement from onset (8.2%). By treatment completion, predicted cumulative missed refill days were 9.68 (95% CI, 7.41 to 11.83), 68.42 (95% CI, 60.35 to 76.92), and 55.47 (95% CI, 48.05 to 62.66), respectively. Men had three classes, while women had two; overall suboptimal engagement was higher among men than women (16.9% vs 10.5%). Recent TB within the past 2 years was strongly associated with suboptimal engagement overall (aOR, 4.38; 95% CI, 2.29 to 8.36), among men, and among women. Among men, HIV-negative status was also associated with suboptimal from-initiation engagement (aOR, 2.72; 95% CI, 1.13 to 6.54).

In conclusion, many patients labeled as having TB treatment success still had meaningful refill delays, especially men and those with recent prior TB. Key limitations include use of refill timing as a proxy for adherence, restriction to patients classified as treatment success, and generalizability mainly to similar public-sector TB settings in South Africa.

Source: Medina-Marino A, Arua E, de Vos L, Fiphaza K, Bezuidenhout D, Ngcelwane N, Charalambous S, Daniels J. Hidden in Success: Gendered Patterns of Suboptimal Care Engagement Among Tuberculosis Patients Who “Successfully” Completed Treatment in South Africa. Clinical Infectious Diseases. 2025 Dec 19:ciaf714.

Characterizing Treatment Adherence Trajectories in the endTB Multisite Cohort of DR-TB Patients [TBN 084]

A study analyzed adherence patterns and their relationship with treatment outcomes among patients with multidrug-resistant or rifampicin-resistant tuberculosis (MDR/RR-TB) enrolled in the endTB Observational Study, a prospective multicountry cohort conducted between April 2015 and December 2019. The study included patients treated with regimens containing at least bedaquiline and/or delamanid across 12 countries. The objective was to identify distinct adherence trajectories during treatment and assess how these trajectories were associated with unsuccessful treatment outcomes, defined as treatment failure, death, or loss to follow-up.

A total of 1,787 patients were included from an original cohort of 2,803 consenting participants. Eligible patients had started an endTB regimen after enrollment, had at least one month of adherence data, a recorded final treatment outcome, and complete covariate information. Monthly adherence was calculated as the proportion of prescribed treatment days on which all medications were taken as prescribed. Adherence data were collected through directly observed therapy (DOT), self-report, or pill counts, depending on treatment delivery. The investigators applied a joint latent class mixed model consisting of a multinomial logistic model for class membership, a class-specific linear mixed model for adherence trajectories, and a class-specific survival model for time to unsuccessful treatment outcome. The survival model adjusted for age, sex, previous TB treatment, HIV/antiretroviral therapy status, hepatitis C virus (HCV) infection, diabetes, extensive disease, low BMI, fluoroquinolone resistance, baseline regimen drugs, and study site.

The median age was 35 years (IQR 26-45), 36.9% were female, 65.0% had fluoroquinolone resistance, and 65.7% had extensive disease. Median treatment duration was 20 months, and median monthly adherence was 95.9% (IQR 88.8%-100%). Overall, 19.0% of patients experienced an unsuccessful outcome, including 7.6% deaths, 3.3% treatment failures, and 8.1% loss to follow-up. Four adherence trajectory classes were identified: "consistently high" (72.5%), "high to low" (14.3%), "low to high" (7.3%), and "consistently low" (5.9%). Median adherence ranged from 98.0% in the consistently high group to 42.1% in the consistently low group. Unsuccessful outcomes occurred in 74.3% of the consistently low group, 1.5% of the low-to-high group, and 6.8% of the consistently high group. Compared with the consistently high group, the adjusted hazard ratio (aHR) for unsuccessful outcomes was 23.2 (95% CI 15.7-24.3) in the high-to-low group and 43.2 (95% CI 26.2-71.5) in the consistently low group. The low-to-high group did not have a significantly different risk (aHR 0.7, 95% CI 0.1-3.8). Adherence trajectory classification predicted unsuccessful outcomes substantially better than conventional adherence measures, with an AUROC of 0.84 (95% CI 0.82-0.86) versus approximately 0.65 for classifications based on overall adherence rates.

Distinct longitudinal adherence trajectories were strongly associated with MDR/RR-TB treatment outcomes, and trajectory-based classification predicted unsuccessful outcomes more accurately than conventional summary adherence measures. These findings suggest that patterns of adherence over time may be more clinically informative than overall adherence percentages alone. Limitations include exclusion of several study sites because of adherence data quality concerns, reliance on adherence measures that partly used self-report or pill counts, and inclusion only of patients with complete data. As an observational cohort study, residual confounding cannot be excluded. 

Source: Law S, Fulcher I, Ashraf S, Bastard M, Docteur W, Franke MF, Guerra D, Hewison C, Huerga H, Khan M, Khan P. Characterizing Treatment Adherence Trajectories in the endTB Multisite Cohort of Drug-Resistant Tuberculosis Patients: An Application of Group-Based Trajectory Modeling. Clinical Infectious Diseases. 2026 Mar 15;82(3):e571-9.

TPT for Household Contacts at Health Facility and Community Settings in Pakistan [TBN 083]

A study assessed whether adding community-based services to fixed health facilities improved completion of the tuberculosis preventive treatment (TPT) cascade among household contacts of individuals with TB. It was a programmatic cascade analysis within the Zero TB Initiative conducted from January 2018 to March 2021 in Karachi and Peshawar, Pakistan, using 8 health facilities in Karachi, 6 in Peshawar, and community-based mobile X-ray van services that began in May 2019.

Household contacts were first invited to fixed health facilities for evaluation. After 2 phone reminders and a household visit, contacts who did not attend but were reachable were offered community-based screening near the patient’s home. All contacts evaluated at facilities or mobile vans received symptom screening, clinical evaluation, chest radiography, and Xpert MTB/RIF testing using sputum or stool samples when indicated. Contacts in whom TB disease was ruled out were offered TPT regardless of TB infection status. Contacts aged 2 years or older received weekly isoniazid–rifapentine for 12 doses (3HP), while children younger than 2 years received 6 months of daily isoniazid (6H). Completion was defined as at least 11 of 12 3HP doses within 16 weeks or about 160 6H doses within 7 months, assessed using pharmacy records and self-report. The cascade included household contact enumeration, TB evaluation, TB diagnosis, TPT eligibility, TPT initiation, and TPT completion. The program did not capture the number prescribed TPT between eligibility and initiation. Level of evidence: observational programmatic implementation evidence.

Overall, 24,369 of 28,443 household contacts (85.7%) completed clinical evaluation; 20,855 (85.6%) were evaluated at health facilities and 3,514 (14.4%) in community settings. TB was diagnosed in 612 of 24,369 evaluated contacts (2.5%). Among 23,757 TPT-eligible contacts, 14,436 (60.8%) initiated TPT, and 10,879 of those initiating treatment (75.4%) completed it. Adding community-based services increased clinical evaluation by 12.4 percentage points (95% CI, 11.7 to 13.0), treatment completion by 11.6 percentage points (95% CI, 10.6 to 12.7), and overall cascade completion by 5.9 percentage points (95% CI, 5.1 to 6.7). In Karachi, community-based services increased cascade completion by 4.6 percentage points (95% CI, 3.7 to 5.4); in Peshawar, the increase was 10.6 percentage points (95% CI, 8.9 to 12.3).

In conclusion, adding community-based screening and TPT follow-up to fixed facility services improved evaluation, treatment completion, and overall TPT cascade completion among household contacts in two Pakistani cities. Key limitations include the observational programmatic design, reliance partly on self-reported treatment completion, absence of captured data on TPT prescription, and limited generalizability beyond urban Zero TB Initiative settings with mobile X-ray capacity.

Source: Jaswal MR, Martinez L, Brooks M, Farooq S, Safdar N, Shah JA, Islam Z, Nasir K, Fareed U, Manzar S, Maniar R. Tuberculosis-Preventive Treatment for Household Contacts at Health Facility and Community Settings in Pakistan. Clinical Infectious Diseases. 2025 Jun 15;80(6):1290-2.

Preferences for TB Point-of-Care Tests among Tuberculosis-Affected Individuals [TBN 082]

The study aimed to quantify preferences among TB-affected adults for point-of-care (POC) TB tests compared with standard facility-based testing. This survey-based discrete choice experiment (DCE) was conducted from December 2022 to September 2024 in outpatient public health facilities in the Philippines, Vietnam, South Africa, Uganda, and India, with an additional community-based sample from Uganda.

Participants were adults aged 18 years or older with presumptive or confirmed TB. The analysis included 1149 participants who met data quality standards: 207 from India, 210 from the Philippines, 219 from South Africa, 305 from Uganda, and 208 from Vietnam. The DCE tested 5 diagnostic attributes: sample type, sensitivity, cost, location, and time to result. Participants completed 12 randomized choice tasks comparing 2 TB test options and stated whether they would actually use their preferred option. The study population had a median age of 42 years, 50.4% were female, 9.1% were HIV-positive, 13.1% had diabetes, and 32.7% had current or prior TB.

POC testing was preferred over standard-of-care testing when sensitivity was equal, with preference shares of 78.8% (95% CI, 77.9–79.8) for facility-based POC testing, 70.0% (95% CI, 68.5–71.5) for community-based POC testing, and 64.6% (95% CI, 62.7–66.5) for home-based POC testing. Preferences declined when POC sensitivity was lower: at 10% lower sensitivity, preference shares were 58.1%, 50.7%, and 47.8%, respectively; at 20% lower sensitivity, they were 41.9%, 36.3%, and 34.4%. Fifteen-minute POC testing was consistently preferred over same-day 3-hour testing, increasing preference shares by 8.6 to 11.9 percentage points. Fewer than 1.5% chose neither test, and sample type had little effect on preferences. Vietnam showed the strongest overall preferences, while country differences were more visible at lower accuracy levels. In Uganda, community-enrolled participants showed higher preferences for community and home testing than facility-enrolled participants.

Overall, TB-affected individuals generally preferred rapid POC TB testing, especially when accuracy matched standard testing, but willingness declined as sensitivity decreased. Key limitations include reliance on stated preferences rather than observed testing behavior and possible limits to generalizability beyond the included high-burden countries and enrollment settings. 

Source: Nalugwa T, Shah KM, Marcelo D, Nakawunde R, Trinh T, Emmanuel J, Nakaweesa A, Schraufnagel A, Andama A, Christopher DJ, Van Luong D. Predicted Preferences for Tuberculosis Point-of-Care Tests among Tuberculosis-Affected Individuals in Five High Burden Countries. Clinical Infectious Diseases. 2026 Jan 14:ciag022.

Undernourished household contacts are at increased risk of TB disease, but not TB infection [TBN 081]

A prospective cohort study assessed whether undernutrition increased risk of TB infection and TB disease among household contacts of persons with TB in India. Participants were recruited within 2 months of index TB diagnosis from 5 diverse RePORT India sites and followed for a median of 24 months.

The study enrolled 857 household contacts after excluding those with microbiologically confirmed prevalent TB at baseline. Undernutrition was defined as BMI <18.5 kg/m². Incident TB was diagnosed microbiologically or clinically during follow-up, with a stricter analysis excluding cases diagnosed within 90 days. IGRA conversion was assessed among baseline IGRA-negative participants using Quantiferon Gold Plus, with standard (>0.35) and stringent (>0.70) conversion thresholds.

Among 857 participants, 239 (27.9%) were undernourished. There were 18 incident TB cases, including 10 among undernourished participants. Undernutrition was associated with higher TB disease risk (HR 3.16, 95% CI 1.25 to 8.02), but this weakened under the stricter TB definition (HR 1.88, 95% CI 0.65 to 5.43). Each 1 kg/m² higher BMI was associated with lower TB incidence risk (adjusted HR 0.85, 95% CI 0.73 to 0.98). Among 377 baseline IGRA-negative contacts, 264 had repeat testing; 56 had standard IGRA conversion and 43 had stringent conversion. BMI was not significantly associated with IGRA conversion.

Lower BMI was associated with progression to TB disease, but not clearly with new TB infection. Key limitations include few incident TB cases, possible co-prevalent disease despite sensitivity analysis, and incomplete repeat IGRA testing, which may limit precision and generalizability.

Source: Sinha P, Ezhumalai K, Du X, Ponnuraja C, Dauphinais MR, Gupte N, Sarkar S, Gupta A, Gaikwad S, Thangakunam B, Paradkar M. Undernourished household contacts are at increased risk of tuberculosis (TB) disease, but not TB infection—a multicenter prospective cohort analysis. Clinical Infectious Diseases. 2024 Jul 15;79(1):233-6.

Understanding delayed diagnosis and treatment of TB in Semarang, Indonesia [TBN 080]

Who

TB patients aged >15 years, residing in Semarang, Indonesia, already recorded as TB patients, and consenting to participate. Three extrapulmonary TB patients were excluded. Mean age was 41 years; 53.1% were female; 85% had primary–high school education; more than half had income below Semarang minimum income standard.

What

The study explored health-seeking behaviors and factors associated with diagnostic delay among TB patients. Diagnostic delay occurred in 48% of participants. Median delay was 19 days, ranging from 3 to 115 days. Almost half were diagnosed on their third care-seeking visit, and only 14.6% were diagnosed at the first health service visited. Self-treatment was common, including pharmacy use, home remedies, local drug stores, and traditional medicine.

Education level was significantly associated with delay (p = 0.014), as were poor attitudes toward TB (p = 0.027). Beliefs that TB is hereditary (p < 0.001), traditional medicine is more effective (p = 0.003), and prolonged cough is not concerning if daily activities continue (p = 0.033) were linked to delay. Multivariable results were incompletely provided, but higher education was reported with AOR = 0.348; 95% CI: 0.127–0.951; p = 0.042, though the wording appears internally inconsistent.

When

Data collection was conducted in 2022.

Where

Semarang, Central Java Province, Indonesia. Ten primary healthcare services with the highest TB incidence were purposively selected.

Why

The study addressed a knowledge gap regarding TB patients’ health-seeking behaviors and contributors to diagnostic delay in Semarang.

How

Cross-sectional study. Data were collected by trained enumerators using questionnaires at participants’ residences. Questionnaire reliability was tested among 30 patients before implementation, with acceptable to good reliability across knowledge, attitudes, beliefs, health staff support, and family support scales.

Source: Handayani S, Isworo S, Hinchcliff R, Wahyudi F, Aryani L, Triyono A. Understanding delayed diagnosis and treatment of tuberculosis: a cross-sectional study in Semarang, Indonesia. The Journal of Infection in Developing Countries. 2026 Jan 31;20(01):104-10.

Most patients with TB at Medan Pulmonary Specialty Hospital had low medication adherence [TBN079]

A study was conducted from April to May 2024 at Medan Pulmonary Specialty Hospital, North Sumatra. The study population consisted of patients with pulmonary tuberculosis receiving treatment at the DOTS Clinic. A total of 95 respondents were selected using purposive sampling. Data were collected using a medication adherence questionnaire, namely the Morisky Medication Adherence Scale-8 (MMAS-8).

Among the 95 respondents, the largest age group was 46–55 years, with 27 respondents (28.4%), while the smallest age group was >65 years, with 2 respondents (2.1%). There were 54 male respondents (56.8%) and 41 female respondents (43.2%). The most common occupation was self-employed, with 42 respondents (44.2%), while the least common was police/military, with 1 respondent (1.1%). The most common education level was senior high school/vocational high school, with 44 respondents (46.3%), while the least common was no formal education, with 2 respondents (2.1%).

Medication adherence was most commonly categorized as low, reported in 59 respondents (62.1%). Moderate adherence was reported in 21 respondents (22.1%), while high adherence was the least common category, reported in 15 respondents (15.8%).

Based on the findings, most patients with pulmonary tuberculosis at Medan Pulmonary Specialty Hospital had low medication adherence.

Source: Ginting AA, Pakpahan RE, Zebua SE. Gambaran Kepatuhan Minum Obat Penderita Tuberkulosis Paru Di Rumah Sakit Khusus Paru Medan Sumatera Utara. INNOVATIVE: Journal Of Social Science Research. 2024 Oct 17;4(5):7819-33.

Age differences in factors associated with pulmonary tuberculosis [TBN 078]

1. Who

• Population: 715,394 Indonesian participants aged ≥16 years from RISKESDAS 2018.
• Age groups:
  • 16–45 years: 469,517 participants
  • 46–64 years: 191,732 participants
  • ≥65 years: 54,145 participants
• Key conditions assessed: Pulmonary tuberculosis (PTB), diabetes mellitus (DM), heart disease, smoking status, BMI, education, employment, sex, residence, and family size.

2. What

• Study focus: Prevalence of PTB and factors associated with PTB across different age groups in Indonesia.
• PTB prevalence:
  • 16–45 years: 3.5‰
  • 46–64 years: 6.8‰
  • ≥65 years: 9.6‰
• Independent factors associated with PTB:
  • Age 16–45: education ≤6 years, former smoking, underweight, DM, heart disease.
  • Age 46–64: male sex, large family size, education ≤6 years, unemployment, former smoking, underweight, DM, heart disease.
  • Age ≥65: male sex, education ≤6 years, former smoking, underweight, DM, heart disease.
• Strongest associations:
  • DM in age 16–45: aOR 6.23, 95% CI 4.37–8.89.
  • Underweight in age 46–64: aOR 3.64, 95% CI 3.02–4.39.
  • Underweight in age ≥65: aOR 2.72, 95% CI 2.09–3.55.
• Interaction findings: Associations between PTB and former smoking, DM, and heart disease differed significantly by age group.

3. When

• Survey year: 2018.
• PTB definition timeframe: Diagnosed by a healthcare professional within the past year.
• Survey frequency: RISKESDAS is conducted every five years.

4. Where

• Location: Indonesia.
• Coverage: All 34 provinces, 416 districts, and 98 cities.
• Data source: Nationally representative RISKESDAS 2018 survey.

5. Why

• Rationale: TB-DM comorbidity is linked to treatment failure, recurrence, and drug resistance.
• Knowledge gap: Indonesia’s national TB program does not currently integrate age-specific grouping in diagnosis and treatment strategies.
• Objective: To investigate PTB prevalence and age-specific factors associated with PTB.

6. How

• Study design: Cross-sectional analysis of nationally representative survey data.
• Level of evidence: Observational, cross-sectional evidence.
• Sampling: Two-stage sampling using probability proportional to size and systematic household selection.
• Data collection: Face-to-face interviews, structured questionnaires, visual aids, anthropometric measurements, and blood glucose testing.
• Statistical methods:
• Rao-Scott Chi-square tests
• Univariable and multivariable binary logistic regression
• Adjusted odds ratios with 95% confidence intervals
• Interaction analysis by age group

Source: Susanti EW, Wiratama BS, Hsieh FI. Age differences in factors associated with pulmonary tuberculosis: a cross-sectional study of Indonesian Basic Health Research (RISKESDAS) 2018. Infectious Diseases. 2026 Feb 1;58(2):221-32. https://benangmerah.net/record/97/age-differences-in