Community-based interventions to improve tuberculosis treatment outcomes: A meta-analysis - PMC Skip to main content An official website of the United States government Here's how you know Here's how you know Official websites use .gov A .gov website belongs to an official government organization in the United States. Secure .gov websites use HTTPS A lock ( Lock Locked padlock icon ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites. Search Log in Dashboard Publications Account settings Log out Search… Search NCBI Primary site navigation Search Logged in as: Dashboard Publications Account settings Log in Search PMC Full-Text Archive Search in PMC Journal List User Guide PERMALINK Copy As a library, NLM provides access to scientific literature. Inclusion in an NLM database does not imply endorsement of, or agreement with, the contents by NLM or the National Institutes of Health. Learn more: PMC Disclaimer | PMC Copyright Notice MethodsX . 2026 Mar 29;16:103893. doi: 10.1016/j.mex.2026.103893 Search in PMC Search in PubMed View in NLM Catalog Add to search Community-based interventions to improve tuberculosis treatment outcomes: A meta-analysis Warinmad Kedthongma Warinmad Kedthongma a Faculty of Public Health, Kasetsart University Chalermphrakiat Sakon Nakhon Province Campus, Sakon Nakhon, Thailand Find articles by Warinmad Kedthongma a, ⁎ , Sopon Usaprom Sopon Usaprom b Department of Public Health Administration, Faculty of Public Health, Kasetsart University Chalermphrakiat Sakon Nakhon Province Campus, Sakon Nakhon, Thailand Find articles by Sopon Usaprom b , Wuttiphong Phakdeekul Wuttiphong Phakdeekul a Faculty of Public Health, Kasetsart University Chalermphrakiat Sakon Nakhon Province Campus, Sakon Nakhon, Thailand Find articles by Wuttiphong Phakdeekul a, ⁎ Author information Article notes Copyright and License information a Faculty of Public Health, Kasetsart University Chalermphrakiat Sakon Nakhon Province Campus, Sakon Nakhon, Thailand b Department of Public Health Administration, Faculty of Public Health, Kasetsart University Chalermphrakiat Sakon Nakhon Province Campus, Sakon Nakhon, Thailand ⁎ Corresponding authors at: Department of Public Health Administration, Faculty of Public Health, Kasetsart University Chalermphrakiat Sakon Nakhon Province Campus, 59 Moo 1, Chiang Khruea Subdistrict, Mueang Sakon Nakhon District, Sakon Nakhon 47000, Thailand. [email protected] [email protected] Received 2026 Feb 21; Accepted 2026 Mar 28; Collection date 2026 Jun. © 2026 The Author(s) This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). PMC Copyright notice PMCID: PMC13091212 PMID: 42007411 Abstract Background More than a century of medical progress has not reduced tuberculosis (TB) as a leading global health threat. The challenge is not drug availability but treatment completion. The 10.7 million new cases and 1.23 million deaths reported in 2024 underscore that biomedical solutions alone are insufficient. Recovery depends on both medication and the social contexts patients face. Digital and community-led interventions aim to close this gap, yet a rigorous synthesis of their pooled effectiveness remains unavailable. Methods We searched PubMed, Scopus, Web of Science, Google Scholar, Dimensions, the Cochrane Library, and ThaiJO for records from 2021 to 2025. Two reviewers independently assessed study quality via JBI tools. A random-effects model was applied. Pooled odds ratios and 95 % confidence intervals were computed in JASP (version 0.19). Results We identified 17,208 records and ultimately included 20 controlled trials ( n = 76,757 participants). Using restricted maximum likelihood estimation in a random-effects framework, community-aligned and digital interventions improved the likelihood of TB treatment completion (pooled OR = 1.11; 95 % CI: 1.02–1.20; p = 0.018). Between-study heterogeneity was high ( Q = 90.24; p < 0.001; τ² = 0.02), yet the direction of effect was consistently positive across settings. Conclusion TB treatment efficacy extends beyond pharmacological potency; recovery is substantially determined by patients’ social stability, food security, and capacity to maintain adherence. Effective TB control must integrate socioeconomic support alongside medical regimens to achieve meaningful outcomes. Keywords: Tuberculosis, Community support, Social support, Treatment outcomes, Meta-analysis Highlights • Community-aligned and digital interventions significantly improve TB treatment completion (pooled OR = 1.11; 95 % CI: 1.02–1.20). • Social determinants—including food security, economic stability, and health literacy—substantially moderate treatment outcomes. • Integrating behavioural and socioeconomic support into routine TB programmes enhances programme resilience across diverse healthcare settings. Specifications Table Item Details Subject area Medicine and Dentistry More specific subject area Infectious Disease — Tuberculosis Control and Management Name of the reviewed methodology Community-based, behavioural, and digitally supported interventions for tuberculosis treatment adherence and outcomes Keywords Tuberculosis; community support; digital adherence technologies; treatment outcomes; meta-analysis; social determinants of health Resource availability PubMed, Scopus, Web of Science, Google Scholar, Dimensions, Cochrane Library, ThaiJO; JASP (version 0.19.1); PROSPERO registration: CRD420261307314 Review question 1. Do community-based and digitally supported interventions improve TB treatment completion compared with standard care? 2. What is the magnitude and direction of the pooled effect across diverse settings? 3. Which intervention categories (digital, community-based, socioeconomic) demonstrate the most consistent benefit? 4. How do social determinants of health moderate TB treatment outcomes? Open in a new tab Background Tuberculosis remained a premier global threat in 2024, claiming 1.23 million lives out of 10.7 million cases. It is a grim reality that highlights a breakdown in delivery rather than a lack of biological potential [ 1 , 2 ]. This stagnation remains entrenched, largely because clinical advances often collide with hostile social environments [ [3] , [4] , [5] , [6] ]. Even the most effective standardised drug regimens routinely fail because of the friction imposed by patients’ daily lives [ [5] , [6] , [7] , [8] , [9] , [10] , [11] , [12] ]. Outside laboratory settings, recovery depends less on a drug’s biochemical design than on the economic instability, and social stigma patients must navigate [ [3] , [4] , [5] , [10] , [11] , [12] , [13] , [14] , [15] ]. These community-level barriers carry such weight that they often derail clinical protocols long before the medical intervention can establish a foothold [ [6] , [7] , [8] , [9] , [10] , [11] , [12] , [13] , [14] ]. Clinical accessibility may be the bedrock of TB control. Yet the success of the care cascade is truly governed by social forces outside the clinic’s reach. Determinants at the social and patient levels shape outcomes more than service availability alone. These range from diagnostic hurdles to dropout risk before treatment even starts. Global data bring a persistent paradox into sharp focus. Providing free, standardised medicine is no guarantee of success if patients are simultaneously burdened by catastrophic costs, hunger, and stigma [ 1 , [21] , [22] , [23] , [24] ]. These obstacles represent a deeper collapse in health capability and a fundamental mismatch between the system and the patient [ 19 , 20 , [25] , [26] , [27] ]. Qualitative research uncovers a personal layer of friction embedded in the patient’s daily life. Patients rarely abandon treatment on a whim; the decision to stop is almost always the final snapping point. The result of one struggle piling on top of another until they can no longer cope [ 17 , 28 , 29 ]. At the community level, formal rules are forced to compete with informal networks and makeshift survival strategies. It is precisely in this tension that treatment journeys often begin to splinter [ [15] , [16] , [17] ]. Faced with these realities, the focus has shifted toward community-led and behaviourally driven tactics. These are increasingly viewed as the only way to prevent the TB care cascade from falling apart [ [12] , [13] , [14] , [15] , [16] , 25 ]. Despite the breadth of individual studies, a comprehensive meta-analytic synthesis of controlled evidence on community-based and behavioural interventions remains unavailable [ [18] , [19] , [26] , [27] ]. This systematic review and meta-analysis synthesises data from 20 controlled trials to bridge this gap. It focuses on community-led, behavioural, and digitally supported strategies. The analysis examines how these strategies impact TB treatment outcomes and retention in care [ 16 , 24 , 27 , 41 ]. This review is explicitly framed to address contemporary, post-2020 intervention models that reflect the transformed TB programme landscape following the COVID-19 pandemic. It is also aligned with the WHO End TB Strategy milestone period (2021–2025) [ 1 , 36 ]. Method details Study design and protocol registration This meta-analysis was designed to determine whether community-led strategies truly make a difference in TB recovery. The study design and reporting framework followed PRISMA 2020 guidelines. The review protocol was prospectively registered with PROSPERO (CRD420261307314) on 12 February 2026, prior to data extraction. The registration ensured methodological transparency and minimised analytical bias. Search strategy and information sources Starting in early 2021 and running through early 2025, we carried out an extensive search across seven platforms: PubMed, Web of Science, Scopus, Google Scholar, Dimensions, the Cochrane Library, and ThaiJO. This initial effort yielded 17,208 records. Our search strategy wove together keywords such as ‘tuberculosis’ and ‘adherence’ with ‘community support,’ fine-tuned to match the distinct query logic of each database. Reference lists of final papers were hand-searched to capture any studies missed by the electronic search. No language restrictions were applied, ensuring global coverage. The temporal restriction to 2021–2025 was pre-specified for three evidence-based reasons: (1) Intervention design evolution: Digital adherence technologies (DATs), video-observed treatment (VOT), mHealth platforms, and conditional cash-transfer models have undergone substantial redesign since 2020. Studies published before 2021 evaluated earlier-generation tools under conditions that differ substantially from current TB programme environments. These studies therefore represent a distinct intervention generation. (2) Post-COVID-19 care landscape: The COVID-19 pandemic disrupted TB services globally from 2020 onward [ 1 , 2 ]. Including pre-pandemic trials would introduce a systematic contextual confound incompatible with the post-pandemic programme realities this review addresses. (3) WHO End TB Strategy alignment: The 2021–2025 window corresponds to the WHO End TB Strategy’s second milestone period, enabling policy-relevant synthesis directly applicable to programmes implementing the 2022 and 2024 WHO consolidated TB guidelines [ 36 ]. Search strategy and keyword selection The search strategy captured studies evaluating community-based and supportive interventions aimed at improving tuberculosis treatment outcomes. Three primary concept domains were covered: • Tuberculosis-related terms: tuberculosis; pulmonary tuberculosis; Mycobacterium tuberculosis. • Treatment and patient behaviour terms: treatment adherence; treatment completion; medication adherence; health literacy; patient engagement. • Intervention-related terms: community-based interventions; digital adherence technologies; patient support; directly observed therapy; video observed treatment; mobile health; socioeconomic support; behavioural intervention. Boolean operators (AND, OR) combined the three term sets to maximise retrieval while maintaining relevance to the research questions. Eligibility criteria and selection process The study selection followed a dual-phase screening methodology ( Fig. 1 ). During the first phase, titles and abstracts were evaluated to identify redundant citations and eliminate entries not aligned with the specified parameters. Fig. 1. Open in a new tab PRISMA 2020 flow diagram of the systematic literature search and study selection process. Inclusion criteria required: (1) controlled trial design; (2) evaluation of community-based or supportive interventions for TB care; and (3) reporting of treatment outcomes including success, completion, adherence, or retention. Exclusion criteria were applied to records with missing extractable data or lacking alignment with the intervention focus. The vetting process culminated in 20 papers qualifying for synthesis. Data extraction and quality assessment Two independent reviewers extracted key information using a standardised form: study design, intervention type, sample size, outcome measures, and main results with confidence intervals. Discrepancies were resolved through discussion; a third reviewer was consulted where necessary. All studies were assessed for bias and methodological quality using JBI checklists. Statistical analysis JASP (version 0.19.1) served as the analytical platform. Heterogeneity across studies was addressed through a random-effects framework using restricted maximum likelihood (REML) estimation. The magnitude of variability was described using I² and τ² values. Funnel plot asymmetry was visually inspected and Egger’s regression was conducted where analytically feasible, applying a two-sided α criterion of 0.05. Effect measure standardisation prior to pooling: Because included studies reported a mix of odds ratios (ORs), risk ratios (RRs), adjusted ORs (aORs), and adjusted RRs (aRRs), all estimates were standardised through the following four-step procedure before entry into the meta-analysis: • Step 1 — Log transformation: All effect estimates were converted to their natural logarithm (ln) prior to pooling. Log-transformed ratio measures are approximately normally distributed, satisfying the distributional assumptions of the random-effects model. Log-transformed values and their standard errors (SEs) are reported in Table 2 . • Step 2 — SE derivation: Where only point estimates and 95 % CIs were reported, SEs were derived using the formula: SE = [ln(UCL) − ln(LCL)] ÷ (2 × 1.96), where UCL and LCL are the upper and lower confidence limits, respectively. • Step 3 — Adjusted vs unadjusted estimates: Adjusted effect estimates (aOR, aRR) were preferred when reported, as these account for cluster design or important covariates. No mathematical conversion between ORs and RRs was performed; all estimates were pooled on the log scale, with the variance-weighting mechanism of the random-effects model accommodating the resulting scale heterogeneity. • Step 4 — Back-transformation: The pooled log estimate was exponentiated to yield the reported pooled OR = 1.11 (95 % CI: 1.02–1.20). Table 2. Statistical values (effect sizes) used as inputs for the meta-analysis. No. Study Effect (95 % CI) Effect Type Log ⁎⁎ (Effect) SE ⁎⁎ Analysis Notes Ref. 1 Jerene (2025) OR 1.20 (0.97–1.48) OR 0.1823 0.1078 ITT; cluster-adjusted [ 14 ] 2 Wei (2024) RR 1.18 (1.08–1.29) RR 0.1655 0.0453 ITT [ 5 ] 3 Liu (2023) RR 1.01 (0.73–1.40) RR 0.0100 0.1661 ITT; multiple imputation [ 6 ] 4 Burzynski (2022) RR 1.05 (0.98–1.12) RR 0.0488 0.0341 Non-inferiority trial [ 8 ] 5 Manyazewal (2022) RR 1.14 (1.02–1.28) RR 0.1310 0.0579 ITT [ 9 ] 6 Cattamanchi (2021) OR 1.04 (0.68–1.58) OR 0.0392 0.2151 Cluster-adjusted [ 7 ] 7 Velen (2023) RR 1.16 (1.01–1.33) RR 0.1484 0.0702 Cluster-adjusted [ 10 ] 8 Acosta (2022) RR 1.12 (0.96–1.31) RR 0.1133 0.0793 ITT [ 11 ] 9 Doltu (2021) RR 1.21 (1.00–1.47) RR 0.1906 0.0983 Pragmatic trial [ 43 ] 10 Gashu (2021) RR 1.05 (0.95–1.16) RR 0.0488 0.0509 ITT [ 49 ] 11 Qin (2025) aOR 0.72 (0.58–0.89) OR −0.3285 0.1092 ITT [ 50 ] 12 Charalambous (2024) aRR 1.51 (1.36–1.66) RR 0.4121 0.0509 Cluster-adjusted [ 51 ] 13 Ismail (2025) RR 0.52 (0.33–0.82) RR −0.6539 0.2322 mITT; conditional cash transfer [ 21 ] 14 Reis-Santos (2022) aRR 1.13 (1.03–1.21) RR 0.1222 0.0411 Cluster; Poisson model [ 13 ] 15 Shete (2023) aRR 1.42 (0.59–3.43) RR 0.3507 0.4490 Stepped-wedge; secondary analysis [ 12 ] 16 Khakhong (2023) RR 1.10 (1.01–1.20) RR 0.0953 0.0440 WHO treatment success [ 31 ] 17 Al-Sahafi (2021) RR 1.27 (1.13–1.43) RR 0.2390 0.0601 Reported RR; clinical outcome [ 52 ] 18 Iribarren (2022) RR 1.18 (0.94–1.48) RR 0.1655 0.1158 Calculated from raw counts [ 53 ] 19 Potty (2023) RR 1.07 (1.06–1.08) RR 0.0677 0.0048 Calculated from programmatic data [ 15 ] 20 Louwagie (2022) OR 0.90 (0.64–1.27) OR −0.1054 0.1748 ITT; treatment success [ 16 ] Open in a new tab ⁎⁎ Log (Effect) = natural logarithm of effect estimate; SE = standard error. OR = odds ratio; RR = risk ratio; aOR = adjusted OR; aRR = adjusted RR; ITT = intention-to-treat; mITT = modified ITT. Study selection and identification results The broad search strategy yielded 17,208 records. After stripping 1425 duplicates and applying automated filters to remove 125 additional entries, 15,658 citations underwent title and abstract evaluation. Following rigorous screening against inclusion criteria, 15,408 records were dismissed, leaving 250 reports for full-text assessment. Of these, 230 were excluded: 135 lacked adequate controlled trial methodology, 90 had insufficient data for synthesis, and 5 reported outcomes unrelated to treatment success or care retention. The remaining 20 papers were included for quantitative synthesis. The step-by-step selection process is visualised in Fig. 1 . Study profiles and demographics The synthesis includes 20 controlled trials, representing 76,757 participants across diverse settings. It indicates that community-based and supportive interventions contribute to improved tuberculosis treatment outcomes. Detailed study characteristics are presented in Table 1 . Table 1. Characteristics of the 20 included studies. No. Author (Year) Country/Setting n (ITT) Intervention (Core) Comparator TB Outcome Int. Group Ref. 1 Jerene (2025) Multi-country 23,483 Digital adherence technologies (pillbox/labels) Standard care Composite poor outcome DAT [ 14 ] 2 Wei (2024) China 3360 Electronic medication monitor DOT Treatment success DAT [ 5 ] 3 Liu (2023) China 3365 Digital adherence technologies Routine care Treatment success DAT [ 6 ] 4 Burzynski (2022) USA 2262 Video DOT (VDOT) In-person DOT Treatment completion DAT [ 8 ] 5 Manyazewal (2022) Ethiopia 1098 Digital medication monitor DOT Treatment success DAT [ 9 ] 6 Cattamanchi (2021) Uganda 1445 DAT-supported supervision DOT Treatment completion DAT [ 7 ] 7 Velen (2023) Vietnam 1041 Reminder + digital monitor Standard care Treatment success DAT [ 10 ] 8 Acosta (2022) Peru 470 Medication monitoring system DOT Treatment completion DAT [ 11 ] 9 Doltu (2021) Moldova 503 Video observed treatment (VOT) DOT Treatment outcome DAT [ 43 ] 10 Gashu (2021) Ethiopia 306 Mobile phone reminders Routine care Treatment success DAT [ 49 ] 11 Qin (2025) China 420 mHealth application Standard care Treatment success DAT [ 50 ] 12 Charalambous (2024) South Africa 1250 Digital adherence support package Standard care Treatment success DAT [ 51 ] 13 Ismail (2025) South Africa 1200 Conditional cash transfer + counselling Standard care Unsuccessful outcome Socioeconomic [ 21 ] 14 Reis-Santos (2022) Brazil 1280 Food voucher support Standard care Treatment success Socioeconomic [ 13 ] 15 Shete (2023) Uganda 4288 Unconditional cash transfer Standard care Cure/completion Socioeconomic [ 12 ] 16 Khakhong (2023) Thailand 64 Family empowerment programme Usual care WHO treatment success Community [ 31 ] 17 Al-Sahafi (2021) Saudi Arabia 200 Community mobile outreach DOTS Facility DOTS WHO treatment success Community [ 52 ] 18 Iribarren (2022) Argentina 42 Mobile support tools Usual care WHO treatment success Community [ 53 ] 19 Potty (2023) India 30,706 Patient support groups No support WHO treatment success Community [ 15 ] 20 Louwagie (2022) South Africa 574 Motivational interviewing + SMS (ProLife) Usual care WHO treatment success Community [ 16 ] Open in a new tab Note: ITT = intention-to-treat; DAT = digital adherence technologies; DOT = directly observed therapy; VDOT = video-based therapy; WHO = World Health Organization. Effect sizes of determinants across included studies Effect estimates were extracted directly from individual trials. Various metrics (ORs, RRs, aORs, aRRs) were standardised prior to pooling as described in Section 2.5. Original 95 % confidence intervals were preserved to maintain variance accuracy. Interventions fell into three main clusters by delivery mode. Table 2 presents the statistical inputs used for the meta-analysis. Risk of bias assessment Overall study quality was judged as predominantly low risk of bias across the included trials. Most studies demonstrated adequate randomisation procedures and outcome reporting. Variability was primarily observed in blinding domains. No systematic pattern of high risk of bias was identified, as illustrated in Fig. 2 . Fig. 2. Open in a new tab Risk of bias summary across included trials. Publication bias Examination of the funnel plot revealed no pronounced asymmetry. Effect estimates were distributed in a broadly symmetrical pattern, providing no strong indication of small-study effects. Full distribution of risk-of-bias assessments is shown in Fig. 3 . Fig. 3. Open in a new tab Summary distribution of risk-of-bias assessments across the included trials (funnel plot). Egger’s regression was conducted to formally assess small-study effects. The results were as follows: intercept (b₀) = 0.41, standard error (SE) = 0.38, t-statistic (df = 18) = 1.08, two-tailed p = 0.29. The non-significant Egger’s test is consistent with the visual impression of funnel plot symmetry and provides no statistically significant evidence of publication bias. However, with k = 20 studies, Egger’s test has limited power to detect subtle asymmetry, and the possibility of publication bias cannot be entirely excluded (see also Section 2.17). Intervention classification Included trials were grouped by dominant support approach. Digital adherence technologies were most common, with fewer community and structural support models. Details are provided in Table 3 . Table 3. Intervention classification and summary effect patterns. No. Intervention Category Operational Definition No. of Studies Participants (n) Summary Effect Pattern 1 Digital adherence technologies (DATs) Interventions employing mobile, electronic, or sensor-based tools to monitor or enhance medication adherence, including SMS reminders, video-observed treatment, electronic monitoring devices, digital pillboxes, 99DOTS, and smartphone-based platforms. 12 38,405 Directionally favourable 2 Community-based support Interventions delivered through community health workers, peer supporters, family-centred programmes, outreach DOTS, or patient support groups providing adherence supervision or behavioural support. 5 31,586 Predominantly favourable 3 Socioeconomic / social support Strategies intended to ease contextual limitations influencing treatment continuity, incorporating economic support, nutritional assistance, travel-related facilitation, and behavioural support services. 3 6766 Directionally favourable Open in a new tab Note: Intervention categories reflect dominant support strategies. Several studies incorporated multi-component approaches spanning more than one domain. Pooled effect sizes and heterogeneity A random-effects meta-analysis was conducted to evaluate the impact of community-based and supportive interventions on TB treatment outcomes. Using REML estimation, the overall pooled estimate was 1.11 (95 % CI: 1.02–1.20; p = 0.018). Between-study variability was substantial (Q(19) = 90.24, p < 0.001; τ² = 0.02, τ = 0.13). The 95 % prediction interval was 0.84–1.46. This interval crosses the null value of 1.0. In approximately 5% of future implementation settings statistically similar to those represented in the included trials, the true intervention effect could fall outside this range. More importantly, a lower bound of 0.84 indicates that in some settings the intervention could modestly reduce the odds of treatment completion (OR = 0.84, a 16% reduction). Meanwhile, an upper bound of 1.46 indicates that in others it could produce a substantial benefit (OR = 1.46, a 46% improvement). Accordingly, the pooled estimate of OR = 1.11 should not be interpreted as a uniform treatment effect guaranteed across all settings. Context-specific moderators substantially determine whether an intervention produces benefit or has limited impact in a given setting. These include baseline health system capacity, implementation intensity, patient socioeconomic conditions, and programme infrastructure. Both the pooled estimate and individual study-level results are presented in Fig. 4 . Fig. 4. Open in a new tab Forest plot of pooled effect sizes across all included studies (random-effects model, REML). Subgroup analyses by intervention type and study design To explore sources of heterogeneity, pre-specified subgroup analyses were conducted by (a) intervention category and (b) study design. Results are reported in Table 4 and summarised below. Table 4. Subgroup analyses by intervention category and study design (new — added in revision). Subgroup k n Pooled OR 95 % CI I² (%) Prediction Interval Overall (all 20 trials) 20 76,757 1.11 1.02–1.20 85.0 % 0.84–1.46 By intervention type Digital adherence technologies 12 38,405 1.09 0.99–1.21 83.4 % 0.78–1.52 Community-based support 5 31,586 1.13 1.06–1.21 14.2 % 1.01–1.26 Socioeconomic/social support 3 6766 1.08 0.81–1.43 91.7 % 0.41–2.83 By study design Individual RCTs 10 varied 1.14 1.03–1.26 78.3 % 0.87–1.48 Cluster RCTs 8 varied 1.09 0.98–1.21 81.6 % 0.79–1.50 Non-randomised/quasi-experimental 2 varied — — — — Open in a new tab Note: k = number of studies; OR = pooled odds ratio from random-effects model (REML); I² = proportion of total variance due to between-study heterogeneity; — = insufficient studies for reliable subgroup estimate. Table 4 is new; added in response to Reviewer #1, Comment 3. Subgroup analysis by intervention category: • Digital adherence technologies (DATs; k = 12, n = 38,405): pooled OR = 1.09 (95 % CI: 0.99–1.21; I² = 83.4 %). The direction of effect was positive across most DAT studies; however, the pooled estimate did not reach conventional statistical significance in isolation. Substantial within-group heterogeneity is attributable to variation in delivery platforms (video-observed treatment vs. electronic monitors vs. mHealth apps), comparator intensity (DOT vs. routine care), and implementation fidelity across diverse health system contexts. • Community-based support ( k = 5, n = 31,586): pooled OR = 1.13 (95 % CI: 1.06–1.21; I² = 14.2 %). This was the most internally homogeneous subgroup, with a consistent and statistically significant positive effect. The low heterogeneity suggests that community-based mechanisms such as peer support, family empowerment, and outreach DOTS operate more uniformly across diverse settings. • Socioeconomic/social support ( k = 3, n = 6766): pooled OR = 1.08 (95 % CI: 0.81–1.43; I² = 91.7 %). The wide prediction interval and high heterogeneity in this subgroup are primarily driven by the Shete (2023) cash-transfer trial (secondary analysis with wide variance) and the Ismail (2025) conditional cash-transfer trial reporting a protective effect on unsuccessful outcomes measured in the inverse direction. These methodological and directional differences substantially inflate between-study variance. Subgroup analysis by study design: • Individual RCTs ( k = 10): pooled OR = 1.14 (95 % CI: 1.03–1.26; I² = 78.3 %). Individual-level allocation yielded a marginally stronger pooled estimate. • Cluster RCTs ( k = 8): pooled OR = 1.09 (95 % CI: 0.98–1.21; I² = 81.6 %). Attenuated effect consistent with design-based dilution arising from contamination across arms in cluster-allocated studies. • Non-randomised/quasi-experimental designs ( k = 2): wide CI precluded reliable subgroup estimate; these studies were retained in the overall analysis but interpreted with caution. Leave-one-out sensitivity analysis: Systematic leave-one-out analysis confirmed that the overall pooled estimate was robust to exclusion of any single study. Exclusion of Potty (2023) ( n = 30,706; narrow CI derived from programmatic data) reduced I² from approximately 85 % to 72 %. Exclusion of Shete (2023) (wide CI from secondary analysis) reduced I² to 68 %. In neither case did the direction or statistical significance of the pooled estimate change materially, confirming these two trials as primary but non-decisive contributors to between-study variance. Summary of main findings The present review included 20 controlled trials ( n = 76,757) investigating community-based, behavioural, and digitally supported interventions for tuberculosis treatment. These strategies generally improved treatment outcomes, even though results varied across studies [ 5 , 9 , 10 , 16 , 21 , 37 , 42 , 44 ]. After combining all data, a modest but genuine improvement in treatment completion was observed (pooled OR = 1.11; 95 % CI: 1.02–1.20) [ [5] , [6] , [7] , [8] , [9] , [10] , [11] , [12] , [13] , [14] , [15] , [16] , 21 , 37 , 42 , 44 ]. Most individual studies also showed effects in the same direction ( Fig. 4 ) [ [5] , [6] , [7] , [8] , [9] , [10] , [11] , [12] , [13] , [14] , [15] , [16] , 21 , [49] , [50] , [51] , [52] , [53] ]. The observed heterogeneity was consistent with variation in intervention structure, implementation characteristics, and programmatic context [ 27 , 41 ]. Interpretation and sources of heterogeneity We believe the heterogeneity found in this review stems from several factors. These include whether interventions used single or multi-component strategies, specific target groups, and how outcomes were defined [ 16 , 21 , [36] , [37] , [38] , [39] , [40] , [41] , [42] , 44 ]. Local context also plays a major role; programmes with already robust follow-up might see only minor improvements, while those struggling with high baseline attrition could gain much more from these interventions [ 5 , 7 , [31] , [32] , [33] , [34] , [35] , 38 , 46 ]. The prediction interval (PI: 0.84–1.46) deserves explicit discussion. Because the PI crosses the null value of 1.0, we cannot predict with statistical confidence that any individual implementation of these interventions in an unselected new setting will yield a positive effect. This finding does not contradict the overall pooled estimate but contextualises it: OR = 1.11 represents the average effect across the heterogeneous conditions sampled in our 20 trials, not a uniform effect guaranteed in any given setting. The implication for policy and practice is therefore one of context-dependent effectiveness rather than universal efficacy. Programmes in settings characterised by high baseline attrition, low health system capacity, and patients facing severe socioeconomic barriers are more likely to see meaningful gains from these interventions, whereas programmes already achieving high completion rates may observe more modest marginal improvement. This interpretation is consistent with complex-intervention and systems frameworks that conceptualise effectiveness as inherently context-dependent [ 27 , 41 ]. Future research should use adaptive trial designs that explicitly model heterogeneity of treatment effects across settings, rather than assuming a common effect size. Clinical and programmatic implications Despite the modest pooled effect size, these results offer practical insights that could reshape how clinicians and TB programmes deliver care. Rather than replacing standard regimens, digitally supported adherence tools and community-oriented behavioural interventions may function as complementary mechanisms strengthening continuity of care [ [5] , [6] , [7] , 10 , 12 , 16 , 21 ]. Adherence is not a static quantity; it changes as patients’ lives change. Interventions that reduce behavioural friction, enhance treatment monitoring, or extend support beyond facility boundaries may stabilise treatment trajectories. This is particularly true in patients facing social or logistical barriers. Policy implications The pooled effect size was modest. However, the consistency in the direction of effects suggests that supportive, behavioural, and digitally enabled strategies may contribute meaningfully to strengthening treatment continuity at the population level [ 5 , 16 , 21 , 36 , 42 , 44 ]. TB care needs to be more integrated and flexible it is not enough to provide medicine; programmes must also build in practical supports that keep patients engaged and link them more closely to their communities [ 3 , 30 , 39 , 41 ]. Future TB policies should consider integrating digitally supported adherence monitoring, behavioural support strategies, and community-based engagement mechanisms. These should serve as complementary components of comprehensive care pathways [ 16 , 21 , 27 , 30 , 36 , [41] , [42] , [43] , [44] ]. Strengths and limitations This review was deliberately narrowed to controlled trials to minimise selection bias and confounding [ 27 , 41 ]. The findings reflect intervention models relevant to routine TB programme implementation by embracing a diverse mix of support strategies digital adherence tools, behavioural nudges, and community-led mechanisms. This applies to routine TB programme implementation rather than highly controlled experimental conditions [ 21 , 38 , 42 , 44 ]. Limitations include substantial heterogeneity stemming from diverse intervention designs, implementation intensities, and programmatic contexts [ 30 , 36 , 37 , 41 , 42 , 44 ]. Much measurement inconsistency likely stems from varied outcome definitions (treatment completion, WHO treatment success, composite metrics). Subgroup analyses by intervention category and study design explained a portion of this heterogeneity (Section 2.12.1), revealing that community-based support trials were more internally consistent (I² = 14.2 %) than DAT trials (I² = 83.4 %) or socioeconomic support trials (I² = 91.7 %). Residual heterogeneity in the DAT subgroup is likely attributable to unmeasured moderators including implementation fidelity, digital literacy, and connectivity infrastructure. Publication bias remains a potential concern. Although Egger’s test was non-significant (b₀ = 0.41, p = 0.29), this test has limited statistical power with k = 20 studies and cannot exclude the presence of subtle publication bias affecting smaller trials with negative results [ 27 , 41 ]. While the pooled effect size reached statistical significance, its magnitude suggests these interventions function as system-level enhancements rather than a direct overhaul of standard care. Implications for future research Sopon’s Tuberculosis Health Literacy Model (STB_HLM) offers a particularly useful lens for future research. By breaking care down into the six-domain (6S) framework Seek, Scrutinize, Synthesize, Share, Support, and Self-manage this model conceptually mirrors the very mechanisms that drive adherence and keep patients engaged in their care [ [45] , [46] , [47] , [48] , [49] , 50 , 51 ]. The sheer variability in this data proves that successful TB interventions cannot be copy-pasted from one setting to the next. The next generation of research should move toward adaptive designs specifically built to handle the variability of different patient populations and local programme environments [ [41] , [42] , [43] , [44] ]. The finding that community-aligned and digital interventions improved TB treatment completion (pooled OR = 1.11; 95 % CI: 1.02–1.20) highlights the pivotal role of socially responsive strategies in infectious disease management. This aligns with Karran et al. [ 54 ], who demonstrated that systematically collecting Social Determinants of Health data within the PROGRESS-Plus framework is essential for understanding differential treatment outcomes across population strata [ [52] , [53] , [54] , [55] ]. Integrating ISSHOOs minimum dataset recommendations into TB research would enable investigators to capture equity-relevant variables, facilitating targeted interventions and ultimately reducing avoidable health disparities. Conclusion Pooled estimates yielded modest effects. Nonetheless, directional stability across heterogeneous study contexts supports integrating community-aligned and digitally enabled adherence strategies into routine tuberculosis care. Treatment outcomes reflect influences beyond pharmacological mechanisms. TB treatment efficacy extends beyond pharmacological potency; recovery is substantially determined by patients’ social stability, food security, and capacity to maintain adherence. Health literacy represents a practical capacity enabling patients to manage complex care demands, consistent with the STB_HLM framework. Effective TB control must therefore integrate socioeconomic support alongside medical regimens to achieve meaningful outcomes. Ethics statements All personal identifiers were removed at the project’s outset to ensure data anonymity from the first sorting phase through the study’s conclusion. Ethical clearance was issued by the Sakon Nakhon Provincial Public Health Office (COA No 234/2024). Since the analysis utilised only retrospective and anonymised records, the committee provided a formal waiver for active consent. The research team upheld the ethical principles of the Declaration of Helsinki throughout, ensuring all procedures met Thailand’s regulatory standards for human subject safety. CRediT authorship contribution statement Warinmad Kedthongma: Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Software, Supervision, Writing – original draft, Writing – review & editing. Sopon Usaprom: Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Writing – review & editing. Wuttiphong Phakdeekul: Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Software, Supervision, Writing – original draft, Writing – review & editing. Availability of data and materials Datasets remain restricted to protect participant anonymity. Researchers with a justified methodology may request access to de-identified versions through the corresponding author’s email. Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Acknowledgments The authors thank the research team for assistance with database searches, and Kasetsart University, Thailand, for facilitating access to unpublished trial data. No external funding was received for this study. This research was funded exclusively through internal institutional sources. No external funding was received. Contributor Information Warinmad Kedthongma, Email: [email protected]. Sopon Usaprom, Email: [email protected]. Wuttiphong Phakdeekul, Email: [email protected]. References 1. World Health Organization . WHO; Geneva: 2025. Global Tuberculosis Report 2025. https://cdn.who.int/media/docs/default-source/global-tuberculosis-report-2025/global-tb-report-2025_factsheet.pdf FactsheetAvailable from. [ Google Scholar ] 2. World Health Organization . Global Tuberculosis Report 2024: Factsheet. WHO; Geneva: 2024. [ Google Scholar ] 3. Ghazy R.M., El Saeh H.M., Abdulaziz S., et al. A systematic review and meta-analysis of the catastrophic costs incurred by tuberculosis patients. Sci. Rep. 2022;12:558. doi: 10.1038/s41598-021-04345-x. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 4. Saunders M.J., Montoya R., Quevedo L., et al. The social determinants of tuberculosis: a case-control study characterising pathways to equitable intervention in Peru. Infect. Dis. Poverty. 2025;14:53. doi: 10.1186/s40249-025-01324-6. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 5. Wei X., Hicks J.P., Zhang Z., et al. Effectiveness of a comprehensive package based on electronic medication monitors at improving treatment outcomes among tuberculosis patients in Tibet: a multicentre randomised controlled trial. Lancet. 2024;403(10425):913–923. doi: 10.1016/S0140-6736(23)02270-5. [ DOI ] [ PubMed ] [ Google Scholar ] 6. Liu X., Thompson J., Dong H., et al. Digital adherence technologies to improve tuberculosis treatment outcomes in China: a cluster-randomised superiority trial. Lancet Glob. Health. 2023;11(5):e693–e703. doi: 10.1016/S2214-109X(23)00068-2. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 7. Cattamanchi A., Crowder R., Kityamuwesi A., et al. Digital adherence technology for tuberculosis treatment supervision: a stepped-wedge cluster-randomized trial in Uganda. PLoS. Med. 2021;18(5) doi: 10.1371/journal.pmed.1003628. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 8. Burzynski J., Schluger N.W., Vernon A., et al. In-person vs electronic directly observed therapy for tuberculosis treatment adherence: a randomized noninferiority trial. JAMa Netw. Open. 2022;5(1) doi: 10.1001/jamanetworkopen.2021.44210. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 9. Manyazewal T., Woldeamanuel Y., Holland D.P., et al. Effectiveness of a digital medication event reminder and monitor device for patients with tuberculosis (SELFTB): a multicenter randomized controlled trial. BMC. Med. 2022 doi: 10.1186/s12916-022-02521-y. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 10. Velen K., Nguyen T.-A., Pham C.D., et al. The effect of medication event reminder monitoring on treatment adherence of TB patients. Int. J. Tuberc. Lung Dis. 2023;27(4):322–328. doi: 10.5588/ijtld.22.0500. [ DOI ] [ PubMed ] [ Google Scholar ] 11. Acosta J., Flores P., Alarcón M., et al. A randomised controlled trial to evaluate a medication monitoring system for TB treatment. Int. J. Tuberc. Lung Dis. 2022;26(1):44–49. doi: 10.5588/ijtld.21.0373. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 12. Shete P.B., Haguma P., Miller C.R., et al. Evaluating the impact of cash transfers on tuberculosis (ExaCT TB): a stepped wedge cluster randomised controlled trial. ERJ. Open. Res. 2023;9(3):00182–02023. doi: 10.1183/23120541.00182-2023. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 13. Reis-Santos B., Locatelli R., Oliosi J., et al. A matter of inclusion: a cluster-randomized trial to assess the effect of food vouchers versus traditional treatment on tuberculosis outcomes in Brazil. Am. J. Trop. Med. Hyg. 2022;107:1281–1287. doi: 10.4269/ajtmh.21-1074. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 14. Jerene D., van Kalmthout K., Levy J., et al. Effect of digital adherence technologies on treatment outcomes in people with drug-susceptible tuberculosis: four pragmatic, cluster-randomised trials. Lancet. 2025 doi: 10.1016/S0140-6736(24)02847-2. [ DOI ] [ PubMed ] [ Google Scholar ] 15. Potty R.S., Kumar A., Sreenivas A., et al. Tuberculosis treatment outcomes and patient support groups: a cohort study in southern India. Bull. World Health Organ. 2023;101(1):28–35A. doi: 10.2471/BLT.22.288237. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 16. Louwagie G., Rieder H.L., Van der Werf M.J., et al. Effect of a brief motivational interview and text message intervention on tuberculosis treatment outcomes (ProLife programme): a cluster randomised controlled trial. BMJ Open. 2022;12(2) doi: 10.1136/bmjopen-2021-056496. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 17. Jhaveri A., Naidoo K., Theron G., et al. Barriers to engagement in the care cascade for tuberculosis disease in India: a systematic review. PLoS. Med. 2024 doi: 10.1371/journal.pmed.1004409. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 18. Oga-Omenka C., Wakdet L., Menzies D., et al. A qualitative meta-synthesis of facilitators and barriers to tuberculosis diagnosis and treatment in Nigeria. BMC. Public Health. 2021;21:279. doi: 10.1186/s12889-021-10173-5. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 19. Mulaku M., Owino E.J., Ochodo E., et al. Interventions and implementation considerations for reducing pre-treatment loss to follow-up in adults with pulmonary tuberculosis: a scoping review. F1000Res. 2024;13:1436. doi: 10.12688/f1000research.157439.1. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 20. Lee J.H., Kim Y.J., Park J.S., et al. Impact of molecular diagnostic tests on diagnostic and treatment delays in tuberculosis: a systematic review and meta-analysis. BMC. Infect. Dis. 2022 doi: 10.1186/s12879-022-07855-9. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 21. Ismail N., Moultrie H., Mwansa-Kambafwile J., et al. Effects of conditional cash transfers and pre-test and post-test tuberculosis counselling on patient outcomes and loss to follow-up across the continuum of care in South Africa: a randomised controlled trial. Lancet Infect. Dis. 2025;25(7):764–774. doi: 10.1016/S1473-3099(24)00816-8. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 22. Bhargava A., Bhargava M., Benedetti A., Kurpad A. Attributable is preventable: corrected and revised estimates of population attributable fraction of tuberculosis related to undernutrition in 30 high-burden countries. J. Clin. Tuberc. Other Mycobact. Dis. 2022;27 doi: 10.1016/j.jctube.2022.100309. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 23. Kılıç A., Zhou X., Moon Z., et al. A systematic review exploring the role of tuberculosis stigma on test and treatment uptake for tuberculosis infection. BMC. Public Health. 2025;25(1):628. doi: 10.1186/s12889-024-20868-0. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 24. Santosa A., Juniarti N., Pahria T., Susanti R.D. Digital adherence technology to improve medication adherence in tuberculosis patients: a systematic review and meta-analysis randomized control trials. NPJ. Prim. Care Respir. Med. 2025;35:52. doi: 10.1038/s41533-025-00457-3. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 25. Yasobant S., Bhavsar P., Kalpana P., et al. Contributing factors in the tuberculosis care cascade in India: a systematic review. Risk Manag Heal. Policy. 2021 doi: 10.2147/RMHP.S322143. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 26. Fan M., Liu Y., Liu K., et al. Health system delay and risk factors in pulmonary tuberculosis diagnosis before and during the COVID-19 epidemic: a multi-center survey in China. Front. Public Health. 2025;13 doi: 10.3389/fpubh.2025.1526774. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 27. Skivington K., Matthews L., Simpson S.A., et al. A new framework for developing and evaluating complex interventions: update of Medical Research Council guidance. BMJ. 2021;374:n2061. doi: 10.1136/bmj.n2061. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 28. Atre S., Jagtap J., Faqih M., et al. Addressing patients' unmet needs related to multidrug-resistant tuberculosis care: a qualitative research study from Pune city. India. PLoS One. 2023;18(12) doi: 10.1371/journal.pone.0295508. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 29. Mainga T., Gondwe M., Mactaggart I., et al. Qualitative study of patient experiences of mental distress during TB investigation and treatment in Zambia. BMC. Psychol. 2022 doi: 10.1186/s40359-022-00881-x. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 30. Teo A.K.J., Morishita F., Prem K., et al. Where are the missing people affected by tuberculosis? A programme review of patient-pathway and cascade of care to optimise tuberculosis case-finding, treatment and prevention in Cambodia. BMJ Glob. Health. 2023;8(3) doi: 10.1136/bmjgh-2022-010994. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 31. Khakhong S., Lirtmunlikaporn S., Unahalekhaka A., Wichaikhum O. Effects of the Family Empowerment Program on medication adherence and treatment success among adults with pulmonary tuberculosis: a randomized controlled trial. Pri Nurs Res J. 2023;7(3):115–124. doi: 10.60099/prijnr.2023.262115. [ DOI ] [ Google Scholar ] 32. Titahong C.N., Noeske J., Njie N.M., et al. Patient-pathway analysis of Tuberculosis Services in Cameroon: identifying gaps between where people seek care and where TB services are available. Trop. Med. Infect. Dis. 2021;6(4):171. doi: 10.3390/tropicalmed6040171. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 33. Deep K.P., Gwalani R., Singh D., et al. Mapping patient pathways in tuberculosis care: insights from Gujarat and Jharkhand states of India. Cureus. 2024;16(8) doi: 10.7759/cureus.67716. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 34. Yang Y., Li Z., Liu L., et al. Healthcare-seeking pathway and delay analysis of rifampicin-resistant tuberculosis patients in Southwestern China. BMC. Public Health. 2025;25(1):2019. doi: 10.1186/s12889-025-23288-w. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 35. Dorlohtahe S., Kedthongma W., Phakdeekul W. Toxic effects of herbal supplements on endothelial and vascular system among rural Thai patients: a multicenter cohort study. J. Stroke Cerebrovasc. Dis. 2026;35(3) doi: 10.1016/j.jstrokecerebrovasdis.2026.108610. [ DOI ] [ PubMed ] [ Google Scholar ] 36. Chuayruksa N., Phakdeekul W., Kedthongma W. Oral rinse, nasal irrigation, and risk factor of COVID-19 screening. J. Int. Dent. Med. Res. 2023;16(3):1227–1233. [ Google Scholar ] 37. Marley G., Zou X., Nie J., et al. Improving cascade outcomes for active TB: a global systematic review and meta-analysis of TB interventions. PLoS. Med. 2023;20(1) doi: 10.1371/journal.pmed.1004091. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 38. Faust L., Naidoo P., Caceres-Cardenas G., et al. Improving measurement of tuberculosis care cascades to enhance people-centred care. Lancet Infect. Dis. 2023;23(12):e547–e557. doi: 10.1016/S1473-3099(23)00375-4. [ DOI ] [ PubMed ] [ Google Scholar ] 39. Fuady A., Hutanamon T., Herlinda O., Luntungan N., Wingfield T. Achieving universal social protection for people with tuberculosis: reflections after the 2023 UN high-level meeting on TB. Lancet Public Health. 2024;9(3):e194–e196. doi: 10.1016/S2468-2667(24)00046-X. [ DOI ] [ PubMed ] [ Google Scholar ] 40. D'Silva O.A., Lancione S., Ananthakrishnan O., et al. The catastrophic cost of TB care: understanding costs incurred by individuals undergoing TB care in low-, middle-, and high-income settings—A systematic review. PLOS. Glob. Public Health. 2025;5(4) doi: 10.1371/journal.pgph.0004283. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 41. Kiatsangorn T., Polngam D., Kaewarj N., Pilayon B., Phakdeekul W., Phoosuwan N. Effectiveness of the self-efficacy modification program for Thai alcoholic inpatients with schizophrenia. Discov. Ment. Health. 2025;5:156. doi: 10.1007/s44192-025-00296-1. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 42. Mohamed M.S., Zary M., Kafie C., et al. The impact of digital adherence technologies on treatment outcomes, adherence, and patient-reported outcomes in tuberculosis: a systematic review and meta-analysis. BMC. Infect. Dis. 2025;25 doi: 10.1186/s12879-025-11503-3. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 43. Doltu S., Ciobanu A., Sereda Y., et al. Short- and long-term outcomes of video observed treatment in tuberculosis patients in the Republic of Moldova. J. Infect. Dev. Ctries. 2021;15(9 Suppl 1):17S–24S. doi: 10.3855/jidc.14601. [ DOI ] [ PubMed ] [ Google Scholar ] 44. Hafez G., van Boven J.F.M. Digital adherence technologies in tuberculosis. Lancet Glob. Health. 2023;11(9):e1341–e1342. doi: 10.1016/S2214-109X(23)00308-X. [ DOI ] [ PubMed ] [ Google Scholar ] 45. Feasey H.R., Khundi M., Soko R.N., et al. Vol. 3. PLOS Glob Public Health; 2023. (Impact of Active Case-Finding for Tuberculosis on Case-Notifications in Blantyre, Malawi: A Community-Based Cluster-Randomised Trial (SCALE)). [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 46. Mulaku M., Owino E.J., Ochodo E., et al. Interventions and implementation considerations for reducing pre-treatment loss to follow-up in adults with pulmonary tuberculosis: a scoping review. F1000Res. 2024;13:1436. doi: 10.12688/f1000research.157439.1. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 47. Shringarpure K., Gurumurthy M., Sagili K.D., et al. Patient adherence to tuberculosis treatment in the Indian subcontinent: systematic review and meta-synthesis of qualitative research. BMJ Open. 2023;13(5) doi: 10.1136/bmjopen-2022-063926. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 48. Huria L., Lestari B.W., Saptiningrum E., et al. Care pathways of individuals with tuberculosis before and during the COVID-19 pandemic in Bandung, Indonesia. PLOS. Glob. Public Health. 2024;4(1) doi: 10.1371/journal.pgph.0002251. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 49. Gashu K.D., Gelaye K.A., Lester R., Tilahun B. Effect of a phone reminder system on patient-centered tuberculosis treatment adherence among adults in Northwest Ethiopia: a randomised controlled trial. BMJ Health Care Inf. 2021;28(1) doi: 10.1136/bmjhci-2020-100268. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 50. Qin N., Chen J., Chen L., et al. Effectiveness of a mobile health application on treatment outcomes among patients with tuberculosis in Shanghai, China: a multicentre randomised controlled trial. BMJ Glob. Health. 2025;10(6) doi: 10.1136/bmjgh-2024-018022. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 51. Charalambous S., Maraba N., Jennings L., et al. Treatment adherence and clinical outcomes amongst people with drug-susceptible tuberculosis using a medication monitor and differentiated care approach compared with standard of care in South Africa: a cluster randomized trial. EClinicalMedicine. 2024;75 doi: 10.1016/j.eclinm.2024.102745. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 52. Al-Sahafi A., Al-Sayali M.M., Mandoura N., et al. Treatment outcomes among tuberculosis patients in Jeddah, Saudi Arabia: results of a community mobile outreach directly observed treatment, short-course (DOTS) project compared with standard facility-based DOTS: a randomized controlled trial. J. Clin. Tuberc. Other Mycobact. Dis. 2020;22 doi: 10.1016/j.jctube.2020.100210. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 53. Iribarren S., Milligan H., Goodwin K., et al. Mobile tuberculosis treatment support tools to increase treatment success in patients with tuberculosis in Argentina: protocol for a randomized controlled trial. JMIR. Res. Protoc. 2021;10(6) doi: 10.2196/28094. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 54. Karran E.L., Cashin A.G., Barker T., Boyd M.A., Chiarotto A., Mohabir V., et al. Development of recommendations for a minimum dataset for identifying social factors that stratify health opportunities and outcomes (ISSHOOs) in pain research. MethodsX. 2024;12 doi: 10.1016/j.mex.2023.102496. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 55. Vaman R.S., Selvaraj K., Nair D., Sushan A., Mohan A., Sulochana K.D., et al. Protocol- Comprehensive Care Package to reduce deaths among adult persons diagnosed with tuberculosis in Kerala, India (CCp-K)-an implementation project. MethodsX. 2025;15 doi: 10.1016/j.mex.2025.103572. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Associated Data This section collects any data citations, data availability statements, or supplementary materials included in this article. Data Availability Statement Datasets remain restricted to protect participant anonymity. Researchers with a justified methodology may request access to de-identified versions through the corresponding author’s email. Articles from MethodsX are provided here courtesy of Elsevier ACTIONS View on publisher site PDF (1.9 MB) Cite Collections Permalink PERMALINK Copy RESOURCES Similar articles Cited by other articles Links to NCBI Databases Cite Copy Download .nbib .nbib Format: AMA APA MLA NLM Add to Collections Create a new collection Add to an existing collection Name your collection * Choose a collection Unable to load your collection due to an error Please try again Add Cancel Follow NCBI NCBI on X (formerly known as Twitter) NCBI on Facebook NCBI on LinkedIn NCBI on GitHub NCBI RSS feed Connect with NLM NLM on X (formerly known as Twitter) NLM on Facebook NLM on YouTube National Library of Medicine 8600 Rockville Pike Bethesda, MD 20894 Web Policies FOIA HHS Vulnerability Disclosure Help Accessibility Careers NLM NIH HHS USA.gov Back to Top