Rifampicin-resistant tuberculosis in Nangarhar, Afghanistan: prevalence and risk factors

Abstract

Background

Afghanistan bears a high burden of tuberculosis (TB). Rifampin-resistant TB (RR-TB) poses a serious threat to disease control, and province-level data remain limited. This study aimed to identify associated risk factors and estimate the prevalence of Rifampin-Resistant TB among adult in Nangarhar Province.

Methods

We conducted an observational study with two components, facility based cross-sectional study to estimate the prevalence of rifampin-resistant TB, with a case–control analysis to evaluate associated risk factors, at three TB centers in Nangarhar province, from 1 September 2023 to 27 June 2024. A cross-sectional study estimated the prevalence of Rifampin-resistant TB among adults with Xpert-confirmed TB. A case–control analysis evaluated associated risk factors compared rifampicin-resistant TB cases with rifampicin-susceptible controls selected from the same centers. Eligibility and classification were determined by laboratory testing, including Xpert MTB/RIF, culture, and phenotypic drug susceptibility testing; clinical assessment and structured interviews were subsequently conducted. Multivariable logistic regression was used to identify independent risk factors for RR-TB.

Results

The prevalence of rifampicin-resistant TB was 3.3% (67/2,038; 95% CI: 2.6–4.2). Prevalence was higher among previously treated patients than among new patients (30/273, 11.0%; 95% CI: 7.6–15.4 vs. 37/1,765, 2.1%; 95% CI: 1.5–2.8). Independent predictors of rifampicin resistance included prior TB treatment (aOR 4.0, 95% CI 1.5–10.7), household crowding (family size 5–10: aOR 5.1, 95% CI 2-12.6); sleeping density ≥ 4 persons/room: (aOR 20.8, 95% CI 2.67–162.8), TB exposure (any contact: aOR 15, 95% CI 3.2–69.6); exposure > 6 months: aOR 27.0, 95% CI 5-157), and smear positivity (aOR 2.5, 95% CI 1.2–5.15). Among the 60 RR TB cases with phenotypic results, 57/60 (95%) also had isoniazid resistance, and 10/60 (17%) had fluoroquinolone resistance.

Conclusion

RR-TB in Nangarhar is concentrated among retreatment patients and crowded households. The findings highlight the importance of universal rapid drug susceptibility testing, adherence support, and household-centred contact investigation in this setting.

Trial registration

Clinical trial number not applicable.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12879-026-12795-9.

Introduction

Tuberculosis (TB) remains a leading cause of death from infectious disease worldwide. In 2023, an estimated 10.8 million people developed TB and about 1.25 million died, while multidrug- and rifampin-resistant TB (MDR/RR-TB) continued to undermine control efforts, affecting approximately 3.2% of new and 16% of previously treated patients globally. These data underscore the urgency of rapid diagnosis and effective treatment within the WHO End TB Strategy [1, 2].

Afghanistan is a high TB-burden setting where protracted conflict, population displacement, and health-system constraints complicate case detection, continuity of care, and infection prevention and control. Although scale-up of molecular testing (e.g., Xpert MTB/RIF) has improved rifampin resistance detection, programmatic management of drug-resistant TB (DR-TB) remains uneven, and provincial evidence or research to guide targeted interventions is limited. Prior programmatic experience from Kandahar highlights the real-world challenges of delivering DR-TB care amid instability [1, 3].

Risk factors for MDR/RR-TB are well described across settings. Consistent determinants include previous TB treatment, close or prolonged exposure to infectious cases, and markers of household crowding and deprivation—factors that amplify transmission and the likelihood of resistant disease [4−7]. In eastern Afghanistan, where multigenerational households and dense living conditions are common, these pathways may be particularly salient.

To address critical local evidence gaps, including the lack of recent provincial estimates of RR-TB, limited data on household and exposure-related risk factors, and scarce information on additional first- and second-line drug resistance among RR-TB cases in eastern Afghanistan, we conducted an observational study in Nangarhar Province. The study aimed to estimate the Prevalence of RR-TB among adult using routine Xpert MTB/RIF data, identify independent risk factors through a case–control analysis, and describe phenotypic resistance patterns among rifampicin-resistant cases. This provincial focus aims to inform practical, household- and facility-level interventions to interrupt transmission and improve outcomes.

Methods

Study design and reporting

We conducted an observational study with two analytic components. First, a cross-sectional analysis of routinely collected Xpert MTB/RIF data was used to estimate the facility-based prevalence of rifampicin-resistant tuberculosis (RR-TB). Second, a case–control study was undertaken to identify factors associated with RR-TB. Reporting follows the STROBE guidelines (Appendix 1) [8].

Setting and period

The study was conducted at three major TB diagnostic and treatment centres in Jalalabad, Nangarhar Province, Nangarhar Regional Hospital, Fatema Zuhra Provincial Hospital, and the Regional TB Laboratory/MDR ward, between 1 September 2023 and 27 June 2024. These centres are the largest in the province, collectively managing approximately 2,400 patients annually. Of the six GeneXpert-equipped TB facilities in Nangarhar, the remaining three are district-level centres serving smaller populations. This selection provides a representative sample of the regional TB population.

Participants and eligibility

For the cross-sectional component, all adults aged ≥ 15 years with Xpert MTB/RIF-confirmed tuberculosis recorded at the three centres during the study period were included. Rifampin resistance was defined by Xpert-detected Rifampin-resistance, and “new” and “previously treated” TB categories followed WHO case definitions [9, 10].

For the case–control analysis, Rifampin-resistant TB cases were enrolled consecutively as they were diagnosed. Rifampin-susceptible TB controls were randomly selected from the same diagnostic registers during the same study period (1 September 2023–27 June 2024), ensuring that cases and controls were contemporaneous and comparable. Exclusion criteria were age < 15 years, negative Xpert, pregnancy, and inability to provide informed consent.

Sample size and sampling

The target sample for the case–control analysis was calculated in Epi-Info v7.2.4.0 assuming an odds ratio (OR) of 6.0, 80% power, 95% confidence, a 1:2 case-control ratio, and 10% non-response, yielding 65 cases and 130 controls.¹³ During the study period, Xpert MTB/RIF testing was routinely performed for all adults (≥ 15 years) with presumptive pulmonary TB attending the three study centers, Cases were enrolled consecutively as diagnosed; controls were selected by simple random sampling from the same diagnostic registers. Controls were selected using a manual simple random (lottery) method from all eligible rifampicin-susceptible TB (RS-TB) patients diagnosed in the same week and at the same center as each rifampicin-resistant TB (RR-TB) case. For each case, two slips were drawn blindly from the mixed pool of patient names, ensuring random selection and temporal comparability. This approach was repeated throughout the study period until the target sample of (130/1971) controls was reached. Controls were also frequency-matched to cases by age and sex; although minor differences existed, the groups were generally well balanced, minimizing confounding and supporting the validity of adjusted estimates in the logistic regression analysis. Both studies conducted in same study period.

Data collection

Data were collected by three trained medical doctors who conducted face-to-face interviews using a structured questionnaire developed specifically for this study, performed clinical assessments, and abstracted relevant data from patient records. Eligibility and case–control classification were determined by Xpert MTB/RIF testing; for Rifampin-resistant TB cases, additional sputum samples were collected for culture anGLId first- and second-line phenotypic drug susceptibility testing (DST). Interviews and clinical assessments were conducted after laboratory confirmation. Interviewers were blinded to participants’ Rifampin resistance status during questionnaire administration. Laboratory and clinical procedures followed World Health Organization (WHO) and Global Laboratory Initiative (GLI) guidance [10−12]. In addition to interview data, the following variables were abstracted from patient records: age, sex, TB treatment history (new or previously treated), sputum smear microscopy results, Xpert MTB/RIF results, HIV status, and treatment registration details.

The structured questionnaire was developed from published TB risk-factor studies and reviewed for content and clarity by six Mahidol University experts, local and international TB specialists, and the Institutional Ethics Committee. It was translated into national language and pretested among 10 TB patients to ensure comprehension and cultural appropriateness. Although formal psychometric validation was not performed, the extensive expert review, ethical approval, and pilot testing support its suitability for this study. An English version of the questionnaire is provided as Supplementary File 1.

Laboratory procedures

Molecular testing was performed using the GeneXpert^® Dx System (version 6.4) with the Xpert MTB/RIF assay, in accordance with WHO recommendations for rapid detection of Mycobacterium tuberculosis and rifampicin resistance [10]. Smear microscopy was conducted using standard acid-fast bacilli (AFB) staining procedures following Global Laboratory Initiative (GLI) guidance [11]. For patients with rifampicin-resistant TB, additional sputum samples were collected for solid culture on both Ogawa and Lowenstein–Jensen media. Phenotypic drug susceptibility testing (DST) was performed for first- and second-line anti-tuberculosis drugs, including isoniazid, rifampicin, ethambutol, amikacin, kanamycin, capreomycin, ofloxacin, and levofloxacin, following WHO technical manuals [12].

Outcomes and definitions

The primary outcome was the prevalence of Rifampin-resistant tuberculosis (RR-TB) among Xpert MTB/RIF-confirmed TB cases. Among Rifampin-resistant TB cases with available phenotypic drug susceptibility testing results, resistance profiles were further classified based on the tested drug panel. Multidrug-resistant TB (MDR-TB) was defined as resistance to both Rifampin and isoniazid. Pre–extensively drug-resistant TB (pre-XDR-TB) was defined as MDR-TB with additional resistance to any fluoroquinolone (ofloxacin or levofloxacin). Extensively drug-resistant TB (XDR-TB) was defined as MDR-TB with resistance to any fluoroquinolone plus at least one second-line injectable drug (amikacin, kanamycin, or capreomycin) [10, 14], for transparency, these categories differ from newer WHO definitions that incorporate resistance to bedaquiline and/or linezolid; this distinction is noted when interpreting findings [14]. Family size was defined as the total number of individuals routinely living in the same household and sharing meals. For analysis, family size was categorized as ≤ 4, 5–10, and > 10 household members.

Statistical analysis

We summarized variables using counts and percentages. Bivariate associations with RR-TB were evaluated using Chi square or Fisher’s exact tests as appropriate. Variables with p < 0.02 in bivariate analyses were entered into a multivariable logistic regression model to estimate adjusted odds ratios (AORs) with 95% confidence intervals (CIs). Model calibration was assessed using the Hosmer–Lemeshow test [15], multicollinearity using variance inflation factors (VIFs) [16], and internal stability via bootstrap resampling (1,000 iterations) [17]. Two-sided p < 0.05 was considered statistically significant. Analyses were performed in IBM SPSS Statistics v27 [18].

Ethics

The study protocol was approved by Mahidol University (MUTM 2023-068-01/02) and the Research Ethics Committee of Nangarhar Medical Faculty (NMF-REC-014-2023. Written informed consent was obtained from all participants or their caregivers, as applicable. Procedures adhered to the Declaration of Helsinki [19], for participants unable to provide a written signature due to illiteracy, informed consent was documented using a thumbprint in the presence of an independent witness after ensuring that the study procedures were clearly explained and voluntarily agreed upon. All study procedures were conducted in accordance with the Declaration of Helsinki and national ethical requirements.

Results

Study profile

Between 1 September 2023 and 27 June 2024, 2,038 adults with Xpert-confirmed TB were recorded across three sites; 67 had Rifampin resistance and 1,971 were Rifampin-susceptible. The analytic dataset comprised 65 RR-TB cases and 130 controls; phenotypic first/second-line DST was available for 60/65 RR-TB cases. Of 67 RR-TB patients, two were excluded due to a lack of consent.

Prevalence of Rifampin-resistant TB

Overall, Rifampicin-resistant TB prevalence among adults with Xpert-confirmed TB was 3.3% (67/2,038; 95% CI: 2.6–4.2). Prevalence was substantially higher among previously treated patients compared with new patients (30/273, 11.0%, 95% CI: 7.6–15.4 vs. 37/1,765, 2.1%, 95% CI: 1.5–2.8). In the analytic dataset, 63.1% (41/65) of RR-TB cases were female. Table 1 summarizes prevalence by treatment history.

Table 1.

RR-TB prevalence and distribution by treatment history

Population	Total TB, n	RR-TB, n (%, 95% CI)	RS-TB, n (%)
All Xpert-positive adults	2,038	67 (3.3%) (95% CI:2.6–4.2)	1,971 (96.7)
New patients	1,765	37 (2.1%) (95% CI:1.5–2.8)	1,728 (97.9)
Previously treated patients	273	30 (11%) (95% CI:7.6–15.4)	243 (89.0)

Abbreviations: RR-TB, Rifampin-resistant TB; RS-TB, Rifampin-susceptible TB

Risk factors for Rifampin resistance

In bivariate analyses, household crowding (family size 5–10, sleeping density ≥ 4 persons/room), TB exposure (any contact, contact > 6 months, contact age 35–44 y, ≥ 55 y), family history of MDR TB, and smear positivity were associated with RR TB; retreatment showed a strong association. In adjusted analyses (Table 2; Fig. 1), retreatment remained a key predictor (aOR 4.0, 95%CI 1.5–10.7; P = 0.005). Among RR TB cases, 43.1% were retreatment patients. Most had prior treatment at public facilities with the standard RHZE regimen for six months. Previous outcomes included cure (60.7%), treatment completion (21.4%), and default (17.9%), with Direct observation Treatment Short course (DOTS) mainly community-based (78.6%), subcategory analyses showed no significant differences between RR and RS patients, indicating that retreatment overall is a strong predictor of Rifampin resistance, while specific subcategory characteristics did not independently affect risk. Household crowding showed graded associations (family size 5–10: aOR 5.1, 95%CI 2.0–12.6; sleeping density ≥ 4: aOR 20.8, 95%CI 2.67–162.8). A family history of MDR-TB (aOR 18.0, 95%CI 1.93–153) and TB exposure (any contact: aOR 15.0, 95%CI 3.2–69.6; >6 months: aOR 27.0, 95%CI 5.0–157) had the largest effects. Notably, smear-positive microscopy independently predicted RR-TB (aOR 2.5, 95%CI 1.2–5.15). Although all participants were bacteriologically confirmed by Xpert/MTB RIF, smear microscopy positivity varied across cases and controls and reflects bacillary burden rather than diagnostic inclusion. Smear-positive patients likely represent higher organism load and prolonged or intense transmission, which may facilitate the selection or acquisition of drug-resistant strains. The independent association between smear positivity and Rifampin resistance therefore suggests a biological and epidemiological relationship rather than an automatic or selection-driven association. Among participants with a reported TB contact, contact with older index cases was independently associated with RR TB. Compared with contacts aged 15–24 years, exposure to contacts aged 35–44 years (aOR 21.4, 95% CI 1.6–295) and ≥ 55 years (aOR 23.0, 95% CI 1.1–486) showed higher odds of Rifampin resistance. Age/sex were not independent predictors.

Table 2.

Key determinants of RR-TB: crude versus adjusted effects (N = 195)

Factor (reference)	Crude OR (95%CI)	p	Adjusted OR (95%CI)	p
Retreated TB (reference: new TB)	3.71 (1.9–7.27)	< 0.001	4.0 (1.5–10.7)	0.005
Family size 5–10 (reference: >10)	3.2 (1.6–6.1)	< 0.001	5.1 (2.0–12.6)	< 0.001
Sleeping density 2–3 person/room (reference: 1)	2.76 (0.9–8.6)	0.079	6.5 (1.02–41)	0.047
Sleeping density ≥ 4 person/room (reference: 1)	3.34 (1.05–11.10)	0.049	20.8 (2.67–162.8)	0.004
Family history MDR-TB (reference: none)	16.33 (1.93–138)	0.010	18.0 (1.93–153)	0.020
Any TB contact (reference: none)	5.15 (2.66–9.95)	< 0.001	15.0 (3.2–69.6)	0.001
Contact > 6 months† (reference: ≤6 months)	5.77 (1.85–18.02)	0.003	27.0 (5.0–157)	< 0.001
Smear-positive (reference: negative)	2.58 (1.39–4.80)	0.003	2.5 (1.2–5.15)	0.013
Contact age 35–44 y† (reference: 15–24 y)	10.0 (1.81–56)	0.008	21.4 (1.6–295)	0.022
Contact age ≥ 55 y† (reference: 15–24 y)	9.6 (1.37–67.2)	0.022	23.0 (1.1–486)	0.040

†Contact-stratified analyses restricted to participants reporting TB contact (subsample)

Fig. 1.

Forest plot of adjusted predictors of rifampin resistance (N = 195). A log-scale forest plot visualizing the adjusted odds ratios and 95% CIs for the primary predictors listed in Y-axis

Additional resistance among RR-TB

Among the 65 RR-TB cases, two isolates were contaminated and three showed insufficient growth, resulting in incomplete phenotypic DST. Among the 60 RR-TB with phenotypic results, 95% (57/60) also had isoniazid resistance (MDR-TB); 16.7% (10/60) had fluoroquinolone resistance (pre-XDR-TB) and 1.7% (1/60) met XDR-TB criteria under the operational definitions used. Table 3 shows phenotypic susceptibility patterns. Drug-level resistance: ethambutol 16.7%, ofloxacin 10%, levofloxacin 15%, amikacin 3.3%; kanamycin 0%, capreomycin 0%. Classification reflects the operational World Health Organization (WHO) categories in use for the panel tested during the study [13]. Although the missing DST results may slightly underestimate pre-XDR and XDR proportions, the overall resistance patterns remain largely representative.

Table 3.

Phenotypic drug susceptibility patterns among RR-TB cases (N = 60 with results)

Drug / profile	Sensitive, n	Resistant, n	Not available, n	% resistant among tested
Isoniazid (INH)	3	57	5	95
Rifampin (RIF)*	0	60	5	100
Ethambutol (EMB)	50	10	5	16.7
Ofloxacin (OFX)	54	6	5	10
Levofloxacin (LFX)	51	9	5	15
Amikacin (AMK)	58	2	5	3.3
Kanamycin (KAN)	60	0	5	0.0
Capreomycin (CAP)	60	0	5	0.0
Composite profiles
MDR-TB (INH + RIF)	3	57	5	95
Pre-XDR-TB†	50	10	5	16.7
XDR-TB†	59	1	5	1.7

Composite profiles: MDR-TB (isoniazid + Rifampin) 95%; pre-XDR-TB† 16.7%; XDR-TB† 1.7%

*Rifampin resistance by Xpert; phenotypic confirmation available in 60/65

†Operational definitions as per study panel (see Methods) ¹³

Denominators exclude “Not available.”

Model diagnostics and internal validation

The multivariable model showed acceptable calibration on the Hosmer–Lemeshow test. Variance inflation factors (VIFs) indicated multicollinearity among exposure variables, particularly TB contact history (VIF 40.7) and contact type (drug-sensitive vs. drug-resistant) (VIF 21.6), consistent with overlapping exposure information; estimates for these covariates should be interpreted with caution. Bootstrap resampling (1,000 iterations) supported the direction and significance of the principal predictors (retreatment, crowding measures, TB contact metrics, family history of MDR-TB, and smear positivity), indicating stable estimates despite wide CIs in sparse strata. All variables were retained due to their distinct epidemiological relevance, and sensitivity analyses excluding correlated variables confirmed that effect estimates for key predictors remained stable.

Discussion

The study highlights that rifampin-resistant TB in Nangarhar is strongly associated with prior TB treatment and household transmission, with crowding and prolonged TB contact as important drivers. These findings underscore the need for targeted interventions, including universal rapid drug susceptibility testing, robust adherence support for retreatment patients, and household-centered contact investigation and prevention measures. The high prevalence of MDR and pre-XDR TB among RR cases further emphasizes early detection and tailored therapy to prevent amplification of resistance. These findings are consistent with increased risk in high-exposure household settings, but genomic or temporal data were not available to confirm direct transmission or clustering.

The overall and stratified prevalence estimates align with global and regional patterns showing far higher resistance among previously treated patients. At the population level, WHO estimates MDR/RR-TB in ~ 3.2% of new and 16% of previously treated patients globally, consistent with the gradient observed here [1, 2]. Nationally representative estimates of rifampicin-resistant TB prevalence in Afghanistan are limited. Routine surveillance data suggest that 512 RR-TB cases were identified among bacteriologically confirmed TB patients in a recent year, of whom 458 were started on treatment, reflecting both the presence of drug resistance and gaps in case detection and management [1]. Provincial experience from Kandahar likewise documents substantial programmatic challenges in DR-TB care [3]. Our estimates sit within the range reported in the region, Pakistan (new 3.7%, retreatment 18.1%), Iran (new ~ 1%, retreatment ~ 12%), and multi-country summaries, while differing from some subnational hotspots [20−25]. Together, these comparisons suggest a moderate provincial burden with marked heterogeneity across South and Central Asia.

Prior treatment, a consistent driver of MDR/RR-TB globally remained independently associated with RR-TB here, echoing the causal pathway of resistance selection under inadequate or interrupted therapy Prior TB treatment was predominantly with the standard RHZE regimen, administered mainly through community-based DOTS (78.6%), with a smaller proportion of patients lacking structured adherence support. No participants had previously received treatment specifically for drug-resistant TB [4]. The strong signals for household crowding and sustained exposure (> 6 months) support transmission-dominant dynamics: multigenerational households with high sleeping density facilitate repeated close contact with infectious cases [6, 7]. The large effect sizes observed for family history of MDR TB and for any TB contact are consistent with increased risk in shared household settings and support the importance of systematic contact investigation and tailored household-level interventions, without implying confirmed transmission pathways [7]. The stronger association observed for medium-sized households (5–10 members) compared with very large households (> 10) appears counter-intuitive but likely reflects differences in effective crowding rather than household size alone. In this setting, medium-sized households may involve more intense and prolonged close contact within fewer rooms, a hypothesis supported by the very strong independent association with sleeping density (≥ 4 persons per room). Larger households may occupy more rooms or compounds, reducing per-room exposure despite higher total household size. These findings suggest that contact intensity and sleeping density are more relevant determinants of RR-TB transmission than household size per se.

Given the strong associations with household crowding and prolonged exposure, systematic household contact investigation and tuberculosis preventive treatment (TPT) are critical to interrupt transmission. The WHO Global Tuberculosis Report 2023 emphasizes TPT as a core intervention to prevent progression from infection to disease and recommends TPT for household contacts after active TB is excluded [1, 2]. Smear positivity was independently associated with Rifampin-resistant TB, potentially reflecting more advanced disease at diagnosis rather than a direct measure of bacillary load or infectiousness [5].

The higher proportion of females among RR-TB cases in this study contrasts with the global male predominance in TB notifications but is consistent with findings from Afghanistan’s Kandahar program. This pattern may reflect context-specific sociocultural factors, including women’s predominant caregiving roles, prolonged household exposure in crowded settings, and spending more time indoors with infectious family members. Differences in healthcare-seeking behavior and access may also contribute [3], whether this reflects true differential risk or selection/ascertainment requires further study.

Drug-resistance profiles and clinical implications. The very high proportion of MDR-TB among RR-TB (95%) and non-trivial pre-XDR (16.7%) underscore the need for universal drug susceptibility testing (DST) at baseline and during care. The pre-XDR/XDR frequencies are comparable to reports from Pakistan, India, and Iran and underline the importance of early detection of fluoroquinolone resistance to avoid functional monotherapy and amplification of resistance [1, 26−28]. While classification here used operational definitions aligned to the tested panel, clinicians should be mindful that newer WHO definitions incorporate resistance to bedaquiline and/or linezolid, which were not phenotypically tested in this study [13].

Collectively, the study findings highlight five evidence-based priorities relevant to rifampicin-resistant TB in Nangarhar (Fig. 2). First, the high proportion of MDR-TB and pre-XDR among RR-TB cases supports expanded access to rapid drug susceptibility testing to detect resistance beyond rifampicin at diagnosis, consistent with WHO guidance [1, 10−13]. Second, the strong association between RR-TB and prior treatment underscores the need for focused monitoring of retreatment patients, who represent a key high-risk group [1, 2]. Third, the large effect sizes observed for household crowding, prolonged exposure, and family history of MDR-TB emphasize the importance of household-centered contact investigation in crowded settings; where feasible under national policy, consider preventive options for high-risk DR-TB contacts [1, 7]. Fourth, the independent association with smear positivity highlights the role of early identification of infectious cases to reduce transmission [1]. Finally, build linked data systems that capture and connect diagnostic, treatment, and contact-tracing information across facilities to identify transmission clusters and steer targeted responses.

Fig. 2.

Programmatic implications for RR-TB control

This study has several strengths which include two design (prevalence plus risk-factor analysis), standardized diagnostics across three centres, rigorous multivariable modelling with internal validation via bootstrap resampling, and comprehensive first-/second-line phenotypic DST among RR-TB isolated. The study has several limitations; the relatively small number of Rifampin-resistant TB cases resulted in wide confidence intervals for some associations and limited statistical power. Some exposure variables showed collinearity (e.g., household TB contact measures), requiring cautious interpretation. The operational definitions for pre-XDR and XDR TB were based on the available testing panel and may not fully align with updated WHO classifications. Drug-susceptibility testing was incomplete for five RR-TB isolates. Because the study was limited to a single province, the findings may not be generalizable to other regions of Afghanistan. Residual confounding by unmeasured factors cannot be completely excluded; however, all participants were screened for HIV (all were negative) and assessed for major comorbidities including diabetes, liver and renal disease. Although interviewers were blinded to Rifampin status during data collection, laboratory results were available within the clinical setting; therefore, complete blinding at all levels was not possible. This may have introduced limited information bias, although the use of standardized questionnaires and trained interviewers likely minimized this risk. Several exposure variables were self-reported, including TB contact history, duration of exposure, and household characteristics, which may be subject to recall bias and information bias. Misclassification of exposures is therefore possible, although standardized questionnaires and trained interviewers were used to minimize these biases. Finally, potential information bias from self-reported questionnaire items, cannot be ruled out.

Conclusion

Rifampicin-resistant tuberculosis in Nangarhar is primarily associated with prior TB treatment and household-level transmission, particularly in crowded and multigenerational settings. The considerable prevalence of multidrug-resistant and pre-extensively drug-resistant TB among RR-TB cases highlights the urgent need to expand access to rapid and comprehensive drug-susceptibility testing at diagnosis and to ensure timely initiation of effective treatment. Strengthening household-based prevention and improving awareness, especially among women and multigenerational families may help reduce transmission in this setting. These coordinated measures, supported by integrated diagnostic, treatment, and surveillance systems, are important for reducing the burden of drug-resistant TB in this high-burden setting.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

CI

Confidence interval

DOTS

Directly Observed Treatment, Short-course

DST

Drug susceptibility testing

HIV

Human immunodeficiency virus

MDR-TB

Multidrug-resistant tuberculosis

MTB

Mycobacterium tuberculosis

MTB/RIF

Mycobacterium tuberculosis / Rifampicin

NTP

National Tuberculosis Program

OR

Odds ratio

RHZE

Rifampicin, Isoniazid, Pyrazinamide, and Ethambutol

RR-TB

Rifampicin-resistant tuberculosis

RS-TB

Rifampicin-sensitive tuberculosis

TB

Tuberculosis

VIF

Variance inflation factor

WHO

World Health Organization

XDR-TB

Extensively drug-resistant tuberculosis

Author contributions

Shah Agha Salehi conceived and designed the study, acquired and analyzed the data, interpreted the findings, and drafted the manuscript. Wiwat Chancharoenthana provided conceptual oversight, critical intellectual input, and extensive review and final editing. U.S., S.M., J.D., J.T., and N.S. critically reviewed the work and contributed to manuscript revisions. All authors approved the final manuscript and agreed to be accountable for all aspects of the work.

Funding

Open access funding provided by Mahidol University. No specific funding was received.

Data availability

Data are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

The study protocol was approved by Mahidol University (MUTM 2023-068-01/02) and the Research Ethics Committee of Nangarhar Medical Faculty (NMF-REC-014-2023; Written informed consent was obtained from all participants or their caregivers, as applicable. for participants unable to provide a written signature due to illiteracy, informed consent was documented using a thumbprint in the presence of an independent witness after ensuring that the study procedures were clearly explained and voluntarily agreed upon. The study was conducted in accordance with the ethical principles of the Declaration of Helsinki.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.