Drug resistance pulmonary tuberculosis among prisoners in Ethiopia: A systematic review and meta-analysis

Highlights

• •This systematic review and meta-analysis study has shown DR-TB in Ethiopian prisoners.
• •Multidrug resistance tuberculosis was detected at 3.0% in this study.
• •The most frequent drug resistance observed was rifampin resistance (4.0%).
• •Any drug resistance of M. tuberculosis isolates was 5.0%.

Abstract

Background

Tuberculosis (TB) is a severe public health issue in prison inmates in Ethiopia since there is no routine screening for TB during prison admission. Also, drug-resistant tuberculosis (DR-TB) is a significant public health problem. Prisons are the most important permissive environments for TB transmission. However, less attention has been given to this segment of the population. Ethiopia’s condition is worse because of poor living circumstances and inefficient health care in the prisons.

Objective

The study determined the pooled prevalence of DR-TB among prisoners in Ethiopia.

Methods

A systematic search was conducted to retrieve records from databases such as PubMed/MEDLINE, ScienceDirect, Cochrane Library, and Google Scholar. The search did not entail a lower time limit, and articles published up until January 2024 were considered. This study was conducted in accordance with the PRISMA guidelines. The data were extracted using a standardized data extraction format. Meta-analysis was computed using STATA version 16 software. Heterogeneity was assessed by the I^2 and publication bias through a funnel plot. The random-effects meta-analysis model was computed to estimate the pooled prevalence of pulmonary tuberculosis (PTB) and DR-TB among prisoners.

Results

Out of 338 records, six cross-sectional studies with 3277 study participants were included in this systematic review and meta-analysis. Of the 3277 study participants included in this study, 5.2% (169) were confirmed positive for PTB. Among 169 PTB cases the pooled prevalence of any DR-TB was 5.0% (95% CI: 2–9%), isoniazid (INH) resistance was 3.0% (95% CI: 0–6%), rifampin (RIF) resistance was 4.0% (95% CI: 0–8%), Multidrug-resistant tuberculosis (MDR-TB) was 3.0% (95% CI 0–6%).

Conclusion

This systematic review and meta-analysis study has shown DR-TB in Ethiopian prisoners. These findings suggest the need for attention in prisons to the control of DR-TB in prisoners in Ethiopia.

1. Introduction

Tuberculosis (TB) is a communicable disease caused by Mycobacterium tuberculosis, and it is the second leading cause of death among infectious diseases, next to COVID-19 [1]. According to the World Health Organization (WHO) 2023 Global Tuberculosis Report, there were 10.6 million cases, and 1.3 million were children. People living with HIV accounted for 6.3% of the total. TB caused an estimated 1.30 million deaths, including 167,000 people with HIV [2]. In Ethiopia, the 2024 WHO report estimated an incidence rate of 146 cases per 100,000 population in 2023 [3].

Multidrug-resistant tuberculosis (MDR-TB) is still a major public health concern in many countries. MDR-TB represents another important clinical challenge and refers to TB that is resistant to at least isoniazid (INH) and rifampin (RIF) [4], [5]. Drug-resistant TB (DR-TB) is a man-made problem, largely the consequence of human error as a result of poor supply chain management, substandard quality of anti-TB drugs, and inadequate or improper treatment practices [6]. Inadequate infection control practices have also been identified as a major contributing factor to the spread of DR-TB [7], [8]. People living with HIV are at a significantly higher risk of developing MDR-TB, which is associated with increased mortality and markedly reduced survival time. The co-occurrence of HIV and MDR-TB presents a particularly fatal combination, with both infections contributing synergistically to poor treatment outcomes [1]. Despite the considerable public health implications, the precise relationship between HIV infection and MDR-TB remains inadequately understood. Moreover, a systematic review has reported an elevated risk of primary (transmission-associated) MDR-TB among HIV-positive individuals, highlighting the need for further investigation into the dynamics of co-infection [9]. According to the 2022 WHO report, the burden of DR-TB increased between 2020 and 2021, with 450,000 new cases of rifampicin-resistant tuberculosis (RR-TB) [9]. The 2024 report indicated that MDR/RR-TB was found in 3.2% of new TB patients and 16% of previously treated cases globally. In Ethiopia, MDR-TB was found in 1.12% of newly diagnosed cases and 11.83% of patients previously treated with TB[3].

In Ethiopia, the National Tuberculosis, Leprosy and other Lung Diseases Prevention and Control Program (NTLLDP) leads the programmatic management of DR-TB. The country has adopted WHO-recommended short and longer all-oral regimens, guided by regularly updated national treatment protocols. DR-TB services are delivered through treatment initiation and follow-up centers, with additional community-based support provided by health extension workers. Despite progress in expanding access to rapid diagnostics, treatment, and patient support systems, the program still faces challenges, including delays in diagnosis, loss to follow-up, and emerging resistance to newer drugs.

Prison populations are particularly vulnerable to TB and DR-TB due to overcrowding, poor ventilation, inadequate nutrition, and limited healthcare access [10], [11]. In Ethiopia, few studies have been conducted on the prevalence of DR among prisoners. These studies reported different findings and provided local information on DR-TB [10], [12], [13], [14], [15], [16]. Prison settings have often been identified as important but neglected reservoirs for DR-TB [17]. However, no comprehensive review has been conducted to assess the burden of DR-TB among prisoners in Ethiopia. This systematic review and meta-analysis aim to pool the reported results of the existing studies and provide scientifically consistent results on DR-TB among Ethiopian prisoners. The findings of our study can provide useful insights for health policymakers to devise appropriate interventions to reduce the subsequent complications of the disease.

2. Materials and methods

2.1. Protocol and registration

The protocol for this systematic review and meta-analysis was reviewed and registered by the University of York's PROSPERO (International Prospective Register of Systematic Reviews) with registration number CRD42024498070.

2.2. Search strategy

This systematic review and meta-analysis study was conducted following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) standards [18]. We conducted systematic searches of electronic databases until June 2025, including MEDLINE (PubMed), ScienceDirect, Cochrane Library, Google Scholar, and grey literature sources search for articles published in English without limiting the publication year. Articles that reported the prevalence of DR-TB among prisoners in Ethiopia were included in the analysis. The search was conducted by two independent investigators (GD and AA). Any uncertainty or disagreement about eligibility was resolved through discussion. Searching was employed using the following keywords: “prevalence”, “magnitude”, “proportion”, “burden”, “tuberculosis”, “drug resistance”, “prisoners”, and “Ethiopia”, both in MeSH and free text. For Google Scholar, we systematically screened the first 256 results, were included a review.

2.3. Study selection and data extraction

Two authors (GD and AA) independently screened articles based on predefined criteria. The inconsistencies between the two authors were resolved by the third author (BD) through discussion. Finally, the selected articles were included in the final analysis. The data were extracted by two independent authors (BD and GD) using a 2016 Microsoft Excel Spreadsheet. Inconsistencies in data recording among the two authors were handled through discussion. The publication year, study design, study area, region, study time, total number of prisoners, number of samples, number of positive cases, laboratory method, and each drug resistance profile were extracted from the articles.

2.4. Inclusion criteria

For our review, the following criteria were used to determine whether studies were qualified for inclusion: describing the prevalence of DR-TB among prisoners in Ethiopia.

2.5. Exclusion criteria

Articles that contained epidemiological data only on TB cases, case studies, author responses, editorials, commentaries, expert opinions, and general reviews were excluded from the final analysis.

2.6. Outcome of interest

The main outcome of interest was the prevalence of DR-TB calculated by dividing the number of DR-TB cases by the total number of TB-positive cases.

2.7. Study quality assessment

The Joanna Briggs Institute (JBI) critical appraisal tools for cross-sectional studies were used to assess the study quality of each study. The study quality assessment scale was employed to assess the quality of the included studies by two authors (GD and AA) independently. The disagreement between the two authors was resolved by consensus. In the case of a persistent disagreement, a third author was consulted. The checklist comprises eight indicators, each scored equally. The total score was converted to a percentage scale, with overall study quality categorized as follows: high quality (>80%), moderate quality (60–80%), and low quality (<60%).

2.8. Statistical analysis

Stata version 16 (Stata Corp LP, College Station, TX) was used for analysis. The random-effect model was used, and a 95% confidence interval was calculated to pool the estimated DR-TB. The I² statistics were used to measure study heterogeneity. An I² value ≥ 50% was considered for the presence of heterogeneity. Potential publication bias was assessed using a funnel plot.

3. Results

3.1. Summary of literature search

A total of 627 studies were retrieved from three databases (Science Direct: 72 studies, PubMed: 56 studies, Cochrane Library: 240, Google Scholar: 256 studies, and Additional studies identified through other sources: 3). Due to duplication, 382 studies were excluded. Next, 228 articles were excluded based on their title and abstracts. Seventeen articles that passed the next stage were assessed through a full-text review. After the full-text review, eleven articles were excluded because there was no DR-TB data. Finally, six articles on the prevalence of drug-resistant TB were used for final analysis after a full-text article assessment (meta-analysis). The PRISMA flowchart summarizes the complete selection procedure (Fig. 1).

Fig. 1.

Demonstrates the PRISMA flow diagram, which depicts the systematic review and meta-analysis study screening and selection process.

3.2. Summary of study and population characteristics

To summarize the study characteristics, a total of 40,670 prison inmates were included in six published articles. Among the total prison population, 3277 (8.1%) were screened for pulmonary tuberculosis (PTB) based on clinical or diagnostic criteria. Of those screened, 169 individuals (5.2%) were confirmed to have bacteriologically positive PTB, as diagnosed using methods such as Xpert MTB/RIF assay, MGIT liquid culture, and line probe assays. All 169 bacteriologically confirmed cases underwent drug susceptibility testing (DST) to assess resistance to first-line anti-TB drugs. Cross-sectional studies were included in the analysis. Data was collected in the period between 2013 and 2021, with publication years ranging from 2016 to 2023. The lowest sample size of the included studies was 259 [16], and the maximum sample size was 1334 [12]. The studies were reported from four regions and one administrative city in Ethiopia. Specifically, of the total of 6 articles reviewed, two studies [10], [14] were from the Amhara region, one study [13] was from the SNNP region, one study [16] was from the Oromia region, one study [15] was from the Oromia, SNNPRS and Harari region, and the remaining one study [12] was from Addis Ababa. Regarding DST methods, two studies [12], [13] used the Xpert MTB/RIF assay and MGIT liquid culture, two studies [10], [14] used the Xpert MTB/RIF assay, and one study [16] used the Xpert MTB/RIF assay and Line probe assay, and one study used the MGIT liquid culture [15]. (Table 1).

Table 1.

General characteristics of studies describing the drug resistance TB profile.

First Author [ref.]	Publication Year	Study Design	Study area (Region)	Study time	Type sample	Total prisoners	Number of Sample	Number of Positive case	Diagnostics Method	Drug resistance
										Any DR	STM	INH	RIF	EMB	PZA	MDR-TB
Ali et al.	2016	cross-sectional	Oromia, SNNPRS and Harari	2013	PTB	15,495	765	71	MGIT liquid culture	4	4	2	2	2		2
Beza et al.	2017	cross-sectional	Amhara	2016	PTB	2700	265	9	Xpert MTB/RIF assay	1			1
Gebrecherkos et al.	2016	cross-sectional	Amhara	2015	PTB	3900	282	15	Xpert MTB/RIF assay	1			1
Hordofa et al.	2023	cross-sectional	Oromia	2021	PTB	2704	259	14	Xpert MTB/RIF assay and Line probe assay	1		1	0
Merid et al.	2018	cross-sectional	SNNPRS	2015	PTB	2068	372	38	Xpert MTB/RIF assay and MGIT liquid culture	1	1	1	1	1		1
Sahle et al.	2019	cross-sectional	Addis Ababa	2014–2016	PTB	13,803	1334	22	Xpert MTB/RIF assay and MGIT liquid culture	5	1	2	4		2	2

3.3. Pulmonary tuberculosis drug resistance profile

3.3.1. Any anti-TB drug resistance

Any drug resistance was reported in 6 studies, with a weighted pooled prevalence of 5.0% (95% CI: 2–9%) among all 169 bacteriologically confirmed M. tuberculosis cases that tested first-line drug susceptibility testing (DST), resulting in 100% DST coverage for these isolates. No clinically diagnosed (non-bacteriologically confirmed) TB cases were included in the pooled analysis of drug resistance. Three of the included studies reported the highest proportion of any drug resistance, ranging from 11.0% to 23.0% [10], [12]. The highest prevalence of any anti-TB drug resistance was reported by Sahle et al. with 23% in Addis Ababa [12], while the lowest proportion was reported by Merid et al. with 3.0% in the SNNPRS region [13] (Fig. 2).

Fig. 2.

Forest plot of the meta-analysis on any drug resistance TB cases.

3.3.2. Isoniazid resistance

In our review, we have assessed the prevalence of INH resistance across all confirmed positive TB cases. Isoniazid resistance was reported in four studies. Individual studies have found prevalence rates ranging from 3.0% to 9.0% [12], [13], [15]. Overall, the weighted pooled prevalence of INH is estimated at 3.0% (95% CI: 0–6%) among a total of 169 M. tuberculosis isolates. Geographically, the highest proportion of INH resistance was recorded in two studies with 7.0% and 9.0% reported in South West Ethiopia, Oromia, and central Ethiopia, and Addis Ababa [12], [16] (Fig. 3).

Fig. 3.

Forest plot of the meta-analysis on INH resistance TB cases.

3.3.3. Rifampicin resistance

In this meta-analysis, data on a total of 169 patients with bacteriologically confirmed pulmonary tuberculosis (PTB)were pooled from five studies included in the review. Of the total of 169 patients, 8 had RIF resistance. The weighted pooled prevalence of any RIF resistance was 4.0% (95% CI: 0–8). Geographically, the prevalence of RIF resistance varied. Two studies reported the highest prevalence of RIF resistance with 11.0% and 18.0% from Northwest Ethiopia and Central Ethiopia, Addis Ababa [10], [12] (Fig. 4).

Fig. 4.

Forest plot of the meta-analysis on RIF resistance TB cases.

3.3.4. Multi-drug-resistant tuberculosis

Overall, the prevalence of MDR-TB ranged from 3.0 to 9.0%, with a pooled prevalence of 3% (95% CI 0–6%). The proportion of MDR-TB among all TB cases varies from place to place. Among the reviewed research, one article reported a high prevalence of MDR-TB, a rate of 9.0% reported in central Ethiopia, Addis Ababa [12] (Fig. 5).

Fig. 5.

Forest plot of the meta-analysis on MDR-TB cases.

3.3.5. Quality assessment results of included studies

The quality of the included studies was evaluated using the Joanna Briggs Institute (JBI) critical appraisal tools for cross-sectional studies. The total score ranged from zero to eight, with higher scores indicating better methodological quality. Among the six studies assessed, five received scores above 80%, indicating high quality, while one article received a 75% score. Based on these results, the overall risk of bias was judged to be low; all included studies were of high methodological quality.

3.3.6. Publication bias and heterogeneity

There was no heterogeneity between the studies for any anti-TB drug resistance, INH resistance, RIF resistance, and MDR-TB (I² = 5.66%, P = 0.38; I² = 0.00%, P = 0.72; I² = 24.01%, P = 0.27; I² = 0.00%, P = 0.6), respectively. A funnel plot was used to determine the presence of publication bias. Visual inspection of the funnel plot suggested relative symmetry; however, due to the small number of included studies (n = 6), this method has limited power to detect publication bias. (Fig. 6).

Fig. 6.

A funnel plot showing publication bias in research studies of the drug-resistant.

4. Discussion

The current meta-analysis identified the burden of DR-TB among prison inmates in Ethiopia. The overall pooled prevalence of MDR-TB and INH resistance was 3.0% and 3.0%, respectively. Four percent of confirmed TB cases were resistant to RIF. In addition, the weighted pooled prevalence of resistance to any first-line anti-TB drug was 5.0%, indicating a considerable level of drug resistance within correctional settings. These findings highlight the need for strengthened routine drug susceptibility testing, early detection strategies, and targeted TB control interventions in prisons to prevent further transmission and amplification of resistant strains.

In this review, we estimated the pooled prevalence of any anti-TB drug resistance to be 5.0%. Our finding is comparable to a study conducted among the prison population in southern Brazil, which reported a prevalence of 4.5% [19]. In contrast, the pooled prevalence of any drug resistance was significantly lower than that reported in the general population of Ethiopia, where a pooled prevalence of 14.25% was found [20]. A possible explanation for the difference between our findings and previous reports could be the number of articles reviewed. The previous study included more articles compared to our study.

Overall, the pooled prevalence of INH resistance was 3.0% in the current study. This finding is higher than a study conducted among the prison population in southern Brazil, which reported a prevalence of 1.5% [19]. However, it is slightly lower than the surveillance of drug-resistant tuberculosis based on reference laboratory data in Ethiopia, with a prevalence was 5.8% [21]. Similarly, the pooled prevalence of any INH resistance was higher in a meta-analysis study 15.62% [20] compared to the current study. Our findings reported low INH resistance compared to the other studies [20], [21]. This might be due to the limited number of published articles, accounting for few data sources in the country on the DR-TB among prisoners in Ethiopia. Another reason could be that diagnostic methods, such as GeneXpert, do not detect INH. Additionally, these differences from place to place could be due to differences in patient selection.

In the present review, the pooled prevalence of RR-TB was 4.0%. The pooled prevalence of the present review was relatively lower than the findings reported previously from the surveillance of drug-resistant tuberculosis based on reference laboratory data in Ethiopia, in which the prevalence of RR-TB was 6.3% [21]. Additionally, a notably higher rate of RR-TB (8.5%) has been reported among prisoners in the Democratic Republic of the Congo [22]. This difference might be due to study design differences. Moreover, the surveillance report estimated the prevalence of RR-TB in new TB cases and previously treated cases separately. In our study, we could not estimate the prevalence of RR-TB in new TB cases and retreated TB cases separately due to reported variations in the published papers.

In this review, the pooled prevalence of MDR-TB was estimated to be 3.0%. The pooled prevalence of the present review was relatively in agreement with the global MDR-TB prevalence data in the WHO TB report in 2023, which was estimated at 3.3% among new cases [1]. A similar prevalence was also observed in a study conducted in Mali, which reported a rate of 4.7% [23]. However, the prevalence of MDR-TB resistance in the current review was lower than the result (43.14%) reported in former Soviet countries [24], and a similar previous review reported from a different country also showed a comparable figure (54.8%) [17]. In contrast, a systematic review and meta-analysis focusing on MDR-TB in prisons reported a much lower pooled prevalence of 0.48% [25]. The difference in the rates is related to the difference in the patients included (newly and previously treated) as well as the difference in the study period. The WHO data [1] are more recent, and most of the included studies in this review were old studies.

The study has a few limitations. First, the number of articles that were reviewed was very few due to the lack of published articles in the country. Second, we were unable to obtain individual-level data such as treatment history (new vs. previously treated cases), disease severity, and comorbidities, including HIV status was not consistently reported across studies, limiting the ability to conduct subgroup analyses. Third, there is no differentiation in the resistance between different age groups. However, the study is important in documenting overall resistance to programmatic control and is a starting point to further describe the resistance over time and in different prisoners.

5. Conclusion

In conclusion, this systematic review and meta-analysis summarized the prevalence of DR-TB among Ethiopian prisoners. While the analysis confirmed the presence of DR-TB, the limited number of included studies restricted the ability to identify independent predictors, thereby constraining the strength of recommendations for targeted interventions. The rate of INH resistance and MDR TB was relatively low compared to RIF resistance. Nevertheless, the observed levels of drug resistance underscore the need to strengthen strategies for the early detection and programmatic management of drug-susceptible and drug-resistant TB in prisoners, as well as sustained monitoring of anti-TB drug resistance in the country. We recommend that a national anti-TB drug resistance survey be carried out on all prisoners in Ethiopia to determine the actual prevalence of DR-TB.

Ethical considerations

Since this study is based on previously published articles, ethical approval is not applicable.

Funding

The authors received no specific funding for this systematic review and meta-analysis.

CRediT authorship contribution statement

Getu Diriba: Writing – review & editing, Writing – original draft, Validation, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. Ayinalem Alemu: Writing – review & editing, Validation, Supervision, Software, Methodology, Formal analysis. Abyot Meaza: Writing – review & editing, Visualization, Validation, Methodology. Muluwork Getahun: Writing – review & editing, Methodology, Data curation. Getachew Seid: Writing – review & editing, Validation, Formal analysis, Data curation. Bazezew Yenew: Writing – review & editing, Validation, Methodology, Data curation. Hilina Mollalign: Writing – review & editing, Validation, Supervision. Biniyam Dagne: Writing – review & editing, Validation, Methodology, Investigation. Shewki Moga: Writing – review & editing, Validation, Resources. Gemechu Tadesse: Writing – review & editing, Validation, Methodology, Investigation, Data curation, Conceptualization.

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.

Appendix A. Supplementary data

The following are the Supplementary data to this article: