Non-adherence and its predictors among pediatric household contacts on tuberculosis preventive therapy in the 3HP era: A multicenter observational study

Abstract

Background:

Despite prophylactic pharmacotherapy being ranked as the best intervention yet, poor treatment adherence tempers tuberculosis preventive therapy (TPT) efficacy.

Objectives:

This study aimed to investigate TPT non-adherence among household contacts and its determinants following the rollout of the 3HP regimen in northwest Ethiopia.

Design:

A facility-based multicenter retrospective study was conducted in selected hospitals of northwest Ethiopia from October 17 to February 21, 2024.

Methods:

Target populations were TPT-eligible household contacts (⩽15 years old) based on the national tuberculosis treatment guideline. Univariate and multivariate binary logistic regression model was fitted to identify potential predictors. p-Value < 0.05 was employed to adjudicate the significance level.

Results:

Among 494 household contacts (HHCs) enrolled in this study, 27.94% were non-adherent. Age (5–10 years: adjusted odds ratio (AOR) (95% CI): 1.92 (1.05–3.19); 10–15 years: AOR (95% CI): 2.04 (1.01–4.12)), parental status (AOR (95% CI): 1.87 (1.02–5.24)), comorbidity (AOR (95% CI): 2.92 (1.08–5.69)), poor nutritional status (AOR (95% CI): 1.30 (1.06–2.43)), adverse drug reactions (AOR (95% CI): 4.01 (1.03–8.20)), adherence support (AOR (95% CI): 1.56 (1.08–6.69)) and TPT regimen type (AOR (95% CI): 4.23 (1.85–6.53)) predicts non-adherence.

Conclusion:

Altogether, this study revealed a high level of non-adherence among HHCs in northwest Ethiopia, falling short of the national 2025/26 TPT targets. TPT should be accompanied by firm counseling and routine supervision to improve adherence.

Plain language summary

A study on tuberculosis preventive therapy non-adherence among pediatric household contacts in the 3HP era

Tuberculosis (TB) is a serious disease. To stop it from developing in people who have been exposed, a prophylactic therapy is applied. While this medicine is the best available, it only works if people take it as directed. This study looked at the prevalence of TPT non-adherence and its predictors among children and teenagers in northwest Ethiopia. Over one-quarter (27.94%) of the children and teens did not adhere to their preventive treatment. In brief, this study discovered that the following groups had a higher likelihood of not following their treatment plan: Older children and teens (ages 5-15), those who had lost one or both parents, had other comorbid health conditions, with poor nutritional status, experienced adverse drug reactions from the medicine. The specific type of treatment regimen also played a role. This study concludes that the rate of household contact children adherence to TB prevention treatment in northwest Ethiopia is too low to meet national health goals. To mitigate this, clinicians should prefer to initiate 3HP-based TPT with firm counselling to them and their families, and ongoing support and vigilant for ADRs.

Keywords

adherence 3HP latent tuberculosis infection tuberculosis preventive therapy

Introduction

Tuberculosis (TB) has been recognized as a silent epidemic in children and adolescents recently.¹ There were 10.8 million new TB cases in 2023, of which 12% were children and adolescents (0–14 years).² Based on a study by Villavicencio and his colleges approximately 6·5 million children and adolescents (<20 years) died globally in 2021, of which about 3.14% were TB-related.³

A regional analysis study based on Global Burden of Disease 2021 revealed that Eastern Sub-Saharan Africa had the highest age-standardized incidence rate of TB.⁴ The overall prevalence of TB among children and adolescents (0–14 years old) in Ethiopia was 9.35%.^5,6

Pediatric patients (younger than 15 years) are at high risk of contracting TB after close contact with an adult household member diagnosed with pulmonary TB (PTB).^7–9 The global prevalence of latent TB infection (LTBI) in the general population, according to a systematic review and meta-analysis study by Cohen et al., was 21.2% to 24.8% globally.¹⁰ In addition, an analytical study based on GBD-2019 revealed that the global burden of LTBI was 23.67%.¹¹ In India, the prevalence of LTBI among pediatric HHCs (<15 years) was 15.4%. In Ethiopia, a study by Smith et al. involving adult diabetic and pre-diabetic HHCs in Addis Ababa revealed that 71% were diagnosed with LTBI.¹² In addition, a study in southern Ethiopia revealed that the epidemiology of active TB among adult household contacts of patients with smear-positive PTB was 6.9% (95% CI 4%–10%).¹³ Latent TB infections are the major source of TB disease.^14,15

As per the sustainable development goal 3.3.2, established to eradicate infectious diseases such as TB by 2030, preventing it arising from the large reservoir is pivotal.^16,17 However, TB preventive therapy (TPT) is often neglected.^14,16 Globally, TPT coverage among eligible HHCs, including both children and adults, remained low at 21% in 2023, far below the WHO target of 90% by 2027, indicating a substantial unmet need.² Furthermore, fewer than one in five were successfully completed TPT.¹⁴

According to Ethiopian guidelines for clinical and programmatic management of TB, TPT is prioritized for children and adolescents who are HHCs of bacteriologically confirmed PTB cases.¹⁸ In addition, adults and adolescent patients living with HIV who do not report any of the symptoms of current cough, fever, weight loss, or night sweats are unlikely to have active TB and should receive TPT. Despite a significant advancement in its approach in 2018, TPT coverage was 22% among ⩽15 years old HHCs of PTB in Ethiopia.^19–21

Non-adherence to TPT among HHCs in Ethiopia is a healthcare concern.^22,23 A key strategy used to increase treatment adherence in Ethiopia is initiating TPT with a treatment supporter, in accordance with WHO guidelines.²⁴ Ethiopia plans to achieve an 80% of TPT completion by 2025/26.¹⁹

Common barriers to TPT adherence include age (2–15 years),²⁵ female gender, low education and income levels, poor patient acceptance, limited drug access, stigma, undernutrition,²⁶ inadequate awareness of caregivers,^27,28 adverse drug reactions,²⁹ and regimen types.³⁰ The World Health Organization (WHO) heralds a new era in TB prevention policy by endorsing the 3HP regimen.³¹ Ethiopia was an early adopter of it, since 2018,³² and rapidly scaled up its implementation by 2020, aiming to improve adherence with a shorter and safer regimen.

Following 3HP, as far as the authors’ knowledge is concerned, no study exists on the initiation and non-adherence rate of TPT in Ethiopia. Consequently, this study evaluates ongoing programs, sheds light on the development of evidence-based policies, and targeted interventional strategies to improve public health outcomes in settings with high TB burden.

Methodology

Study design, setting and period

A multicenter retrospective study was conducted through medical record reviews. It was done in the TB clinics of five public teaching University hospitals of northwest Ethiopia selected purposively (Figure 1). These hospitals serve more than 12 million people in aggregate. They were selected purposively due to their high patient flow and located in urban settings characterized by a dense population.

Figure 1.

Study area map employing ArcGIS Pro version 3.4 (Esri; Redlands, California, USA).

Study population

Target population was children and adolescents (<15 years of age) in contact with households of bacteriologically confirmed pulmonary-drug-susceptible TB patients (index cases) and eligible for TPT at those selected hospitals during August 2023 to July 2024. Children and adolescents who were referred to other health facilities and had incomplete medical records were excluded. Five public teaching University hospitals providing TB diagnostic and treatment services in northwest Ethiopia were selected for this study. In those TB clinics’ TB registration book, 878 drug-susceptible PTB index cases were identified. From them, 654 HHC target populations were screened (Figure 2).

Figure 2.

Involvement procedure of study participants at TB clinics of teaching University hospitals in Amhara region, Northwest Ethiopia (N = 878).

Study variables

Non-adherence rate to TPT among HHCs was the principal dependent variable. Explanatory variables include sociodemographic, healthcare system, clinical, and treatment-related variables.

Operational definitions

Adolescent: a person aged 10–15 years, according to this study.

Children: a person aged <10 years.

Family size: the number of persons counted as members of an individual’s household; when a family have ⩽4 persons, we classify it as a low family size.^33,34

Household contact: a person who shared the same enclosed living space as the index person for one or more nights or for frequent or extended daytime periods during the 3 months before the start of anti-TB treatment.¹⁸

Index case with TB: is the initially identified adult patient with new or recurrent TB in a specific household. An index person is the one on whom a contact investigation is centered, but is not necessarily the source.³¹

Non-adherence/Non-completion of course of TPT: According to the Ethiopian TB prevention and control guideline, a patient would be labeled as “non-adherent” if the patient had taken less than 80% of the recommended dose within 120% of the planned TPT duration, or less than 90% of the recommended dose within 133% of the planned TPT duration.¹⁸ The following table guides the TPT adherence criteria in Ethiopia (Table 1). Treatment was classified as late if a patient had taken the prescribed doses after the scheduled TPT period.

Full TPT completion was defined as completion of 100% of prescribed doses with a designed treatment schedule.

Lost to follow-up: TPT interrupted by person for eight consecutive weeks or more for 6H, four consecutive weeks or more for 3HP and 3 months isoniazid and rifampicin (3HR).¹⁸ Lost to follow-up patients were labeled as non-adherent if they fulfilled the above definition of non-adherence unless otherwise.

Undernutrition: When the body does not have an adequate amount of one or more nutrients, reflected in biochemical tests like hemoglobin level for anemia, in anthropometric indicators such as stunting (low height-for-age) or wasting (low weight-for-height) and/or weight-for-age (underweight).³³

Table 1.

Assessment of TPT adherence status of pediatric HHCs in Amhara comprehensive specialized hospitals, northwest Ethiopia, based on the national TB treatment and control guideline, 2023/24.

Regimen	Total duration (months)	Expected No of doses (100%)	Extended time for treatment Completion (treatment duration plus 33% additional time)	Remark on completion criteria for TPT regimens in Ethiopia
3HP	3	12	120 days (4 months)	Completion of the 12 doses within 4 months period
3RH	3	84	120 days (4 months)	Completion of the 84 doses within 4 months period
6H	6	168	240 days (8 months)	Completion of the 168 doses within 8 months period

6H, 6 months isoniazid; HHC, household contact; 3HP, 3 months isoniazid and rifapentine; 3RH, 3 months rifampicin and isoniazid; TPT, tuberculosis preventive therapy.

Data management

Data collection and quality assurance

To meet our objectives, secondary data were extracted from patient medical records between 17 October and 21 February 2024. Data were gathered by utilizing a standardized data abstraction tool adapted from the Ethiopian Federal Ministry of Health TB clinic intake and follow-up form and from previous literature.^20,35-37 Initially, it was pretested in a pilot study at Woreta primary hospital TB clinic on 17 sample medical records. In light of this study, the tool was refined accordingly for clarity and comprehensiveness. Fifteen trained health professionals: five physicians as supervisors, five pharmacists, and five nurses who were experienced in working in TB clinics were selected as data collectors. A 1-day training was given to both supervisors and data collectors. The data retrieval process was closely supervised by strict supervisors, and the overall quality of the data collection was also monitored by the principal and co-investigators. The collected data were checked for completeness and consistency.

Statistical analysis

The data were entered into EpiData version 4.7 and exported to Stata version 17 software (STATA corp., College Station, TX, USA) for analysis. The study participants were characterized using descriptive statistics using measures such as frequency, percentage, and mean value with standard deviation were all included in this analysis.

At last, adherence status to TPT was dichotomized into adherent or nonadherent. The bivariable logistic regression model was fitted for each potential predictor variable. Moreover, those variables having a p-value ⩽0.25 in bivariate analysis were fitted into a multivariable logistic regression model to control the impact of confounders. Adjusted odds ratios with 95% confidence intervals and p-values were used to measure the strength of association and to identify statistically significant determinants. Factors were deemed significantly associated with the outcome variable if their p-values were <0.05 at 95% confidence interval. The Shapiro-Wilk normality test revealed that our data were not normally distributed. The model’s fitness was verified using the Hosmer-Lemeshow goodness-of-fit test and it revealed that the model was adequately fit (p-value = 0.07). The variance inflation factor (VIF) was used to diagnose the presence of multicollinearity among predictor variables. There is no collinearity among the independent variables in the model, as seen by the mean VIF of 2.33. The discriminative ability of this model was assessed employing c-statistics, revealing that 81% of children and adolescents were correctly classified into adhering versus non-adhering (c statistic = 0.81).

Furthermore, upon performing an in-depth expert review, some factors might be included in the multivariate analysis even though their p-value was above the cut-point value in bivariate analysis based on existing knowledge and evidence.

Results

Tuberculosis preventive therapy pathway of care

Upon a retrospective clinical audit on the TB clinic contact investigation registration book, a total of 2250 HHCs were claimed from 502 index cases. The mean age of HHCs was 31.9 years (standard deviation 5.31). Among the total HHCs identified, 654 were declared eligible for TPT. Of them, 493 were asymptomatic, whereas 161 were symptomatic. Five hundred forty one (82.72%) of them were successfully initiated TPT (Figure 3).

Figure 3.

LTBI care pathway: screening and TPT enrollment among pediatric HHCs in Amhara comprehensive specialized hospitals, northwest Ethiopia, 2023/24 (n = 494).

Healthcare system and sociodemographic-related features of participants in non-adherence study

Among 541 HHCs initiating TPT, only 494 have been enrolled in the non-adherence study due to incomplete medical records and referral to other health institutions. The majority of the study participants were females (59.31%), 5–10 years old (48.78%), and had attained only a primary school level of education (65.38%). Furthermore, 29 (5.87%) children and adolescents in the study population were identified as orphanhood. A little over half (51.21%) of the caregivers do not have formal education (Table 2).

Table 2.

Sociodemographic profile of pediatric household contacts on TPT with their caregivers and healthcare system-related features in Amhara comprehensive specialized hospitals, northwest Ethiopia, 2023/24 (N = 494).

Sociodemographic variable	Total	%
Age (years)
<5	81	16.4
5–10	241	48.78
10–15	172	34.82
Gender
Female	293	59.31
Male	201	40.69
Domicile
Rural	218	44.13
Urban	276	55.87
Educational status of HHCs
No formal education	53	10.73
Kindergarten	54	10.93
Primary	323	65.38
Secondary	64	12.96
Educational status of caregiver
No formal education	253	51.21
Primary	174	35.22
Secondary	49	9.92
Tertiary	18	3.64
Ethnicity
Amhara	321	64.98
Tigrie	97	19.64
Oromo	44	8.91
Others^a	32	6.47
Family size
⩾4	198	40.08
<4	296	59.92
Parental status
Both alive	284	57.49
Father alive	83	16.80
Mother alive	98	19.84
Both died	29	5.87
Relationship between child and caregiver
Parent	361	73.08
Relative	77	15.59
Guardian	56	11.34
Distance of the hospital from client
<10 km	223	45.14
>10 km	271	54.86
Chronic healthcare services integration with other services
Yes	385	77.94
No	109	22.06
Qualification of prescribing healthcare professional
Internist	196	39.68
General practitioner	298	60.32
Programmatic adherence support by healthcare professionals
Yes	103	20.85
No	391	79.15
Drug stock-out
Yes	136	27.53
No	358	72.47
Health insurance
CBHI	219	44.33
Out of pocket	275	55.67

Guragie, Afar.

CBHI, community-based health insurance; TPT, tuberculosis preventive therapy; HHC, household contact.

Clinical profile of the study population

Almost a quarter of children and adolescents did not have a history of Bacillus Calmette-Guérin (BCG) vaccination. More than a quarter 181 (36.64%) of them had poor nutritional status. Comorbidities, primarily central nervous system disorders led by attention deficit hyperactivity disorder (ADHD), were observed in 4.25% (n = 21) of contacts. Adverse drug reaction (ADR) occurrence rate in these patients was high, 26.52% (n = 131), where flu-like cutaneous reactions were the predominant (Table 3).

Table 3.

Clinical characteristics of pediatric HHCs on TPT in Amhara comprehensive specialized hospitals, northwest Ethiopia, 2023/24 (N = 494).

Clinical variable	Total	%
HIV status
Negative	494	100
BCG vaccination profile
Yes	351	71.05
ADR profile
Yes	131	26.52
Comorbidity status
No comorbidity	428	86.64
Type 1 diabetes mellitus	13	2.64
Asthma	12	2.43
CNS disorders^a	21	4.25
Cardio-renal disorders	13	2.63
Others^b	7	1.41
Medications for comorbidities
Insulin	13	2.63
Salbutamol	12	2.43
Methylphenidate	7	1.42
Amphetamines	14	2.83
Valproic acid	3	0.61
Phenobarbitone	6	1.21
Ethosuximide	1	0.20
Enalapril	9	1.82
Furosemide	12	2.43
Corticosteroids	14	2.83
Spironolactone	12	2.43
Digoxin	2	0.40
Benzathine penicillin G	8	1.62
Use of complementary and alternative medicines
Yes	255	51.62
No	239	48.38
Nutritional status
Undernutrition	181	36.64
Non-malnourished	313	63.36

Attention deficit hyperactivity disorder, Autism Spectrum Disorder, and Epilepsy. ^bGIT disorders and Cancer.

ADR, adverse drug reaction; BCG, Bacillus Calmette-Guérin; HHC, household contacts; TPT, tuberculosis preventive therapy.

Adherence status to TB preventive treatment

Non-adherence to TPT among children and adolescents was 27.94% (138/494). The non-adherence rate was higher among the 3RH regimen, 39.13% (9/23), than among HHCs who were on 3HP, 27.39% (129/471). Among children and adolescents on TPT, 343 (69.43%) achieved full adherence by completing all doses. HHCs on the 3HP regimen has higher rate of completing TPT fully (341/471 = 72.4%) as compared to HHCs on the 3RH regimen (2/23 = 8.7%; Figure 3).

Regarding treatment retention rate per each dose of TPT, the maximum number of participants who lost their treatment was during tenth week of therapy (n = 83; Figure 4).

Figure 4.

Weekly treatment retention rate among pediatric household contacts by dose in Amhara comprehensive specialized hospitals, northwest Ethiopia, 2023/24 (N = 494).

Adverse drug reaction profile of household contacts

Six hundred seventy-three ADRs were reported from 131 HHCs. The causality, severity, and preventability data of ADRs were not found. ADR occurrences in this study have been classified into six main types. Cutaneous ADRs were hypersensitivity reactions revealing angioedema, urticaria, rash, pruritus, and potentially life-threatening anaphylaxis during treatment. Flu-like symptoms encompass fever, chills, weakness, fatigue, muscle pain, aches, syncope, tachycardia (>100 bpm), palpitations, hypotension, flushing, dizziness, conjunctivitis (red eyes), or excessive sweating. Gastrointestinal ADRs include vomiting (with or without nausea) or abdominal pain. Respiratory ADRs include bronchoconstriction, cough, dyspnea, or chest pain, where presumptive TB was ruled out. Among 673 ADRs, 41.75% (281), 40.56% (273), 6.24% (42), 4.61% (31), 3.57% (24), and 3.27% (22) were flu-like symptoms, cutaneous, gastrointestinal, neurological, hepatotoxic, and respiratory ADRs, respectively. Seven participants discontinued TPT due to flu-like symptoms (n = 2), peripheral neuropathy (n = 1), and hepatotoxicity (n = 4). Among individuals who developed hepatotoxicity, 83.3% (10 of 12) were non-adherent to TPT (Figure 5).

Figure 5.

Impact of adverse drug reactions on TPT adherence among pediatrics in Amhara comprehensive specialized hospitals, northwest Ethiopia, 2023/24 (N = 494).

Factors associated with non-adherence

In the multivariate analysis, factors with a p-value ⩽ 0.25 in the univariate analysis were qualified for and included. In addition, gender was incorporated into the multivariate analysis despite its p-value exceeding the cutoff, as previous studies identified it as a significant determinant factor of non-adherence to TPT.^29,38,39 Explanatory variables such as age, relationship of children with caregiver, parental status, drug stock-out, commorbidity, ADR experience, regimen type, lack of adherence support, and poor nutritional status were significantly associated with non-adherence, and the odds were in the predicted direction. The association between regimen type and ADR with non-adherence was clear-cut. Patients who were on 3RH (AOR (95% CI): 4.23 (1.85–6.53)) and experienced ADR (AOR (95% CI): 4.01 (1.03–8.20)) were four times more likely to have failed to adhere to their medication compared to their counterparts. Among all types of ADRs, patients who revealed neurological reactions (AOR (95% CI): 6.1 (1.23–8.20)) and hepatotoxicity (AOR (95% CI): 7.2 (1.26–9.04)) were at a higher odds of being non-adherent to TPT (Table 4).

Table 4.

Multiple logistic regression analysis of factors associated with non-adherence to TPT among pediatrics in Amhara comprehensive specialized hospitals, northwest Ethiopia, 2023/24 (N = 494).

Variables	Covariates category	Univariate analysis			Multivariate analysis
Variables	Covariates category	OR	95% CI	p-Value	OR	95% CI	p-Value
Gender (Ref-Male)	Female	1.18	(0.83–2.72)	0.26	1.29	(0.53–4.11)	0.37
Age (Ref-<5)	5–10	2.38	(1.11–3.59)	0.02	1.92	(1.05–3.19)	0.03
	10–15	2.31	(1.09–4.71)	0.001	2.04	(1.01–4.12)	0.009
Family size (Ref-⩽4)	>4	1.67	(0.63–2.19)	0.31
Domicile (Ref- Urban)	Rural	1.38	(0.85–2.74)	0.28
Educational status of children (Ref- Secondary school)	No formal education	1.19	(0.64–5.35)	0.31
	Kindergarten	2.04	(0.95–6.19)	0.29
	Primary school	1.37	(0.77–3.22)	0.42
Distance (Ref-<10 km)	>10 km	1.79	(0.13–3.07)	0.3
Relationship (Ref-Parent)	Relatives	1.23	(0.51–3.08)	0.21	1.02	(0.60–2.08)	0.54
	Guardian/Neighbors	2.02	(1.07–4.17)	<0.001	1.92	(1.03–3.61)	0.01
Parental status (Ref-Both alive)	Father/mother alive	0.38	(0.03–1.51)	0.12	0.52	(0.03–1.78)	0.99
	Both died	2.17	(1.08–4.19)	0.002	1.87	(1.02–5.24)	0.01
Educational status of caregiver (Ref- Tertiary)	No formal education	1.30	(0.88–3.10)	0.44
	Primary	1.21	(0.97–2.72)	0.39
	Secondary	1.74	(0.99–4.82)	0.55
Prescribers’ qualification (Ref- Internist)	GP	1.29	(0.04–3.19)	0.89
Service integration (Ref- Yes)	No	1.71	(1.12–6.69)	0.53
Comorbidity (Ref- No)	Yes	2.37	(1.12–6.69)	0.005	2.92	(1.08–5.69)	0.03
Drug stock-out (Ref-No)	Yes	1.52	(1.02–2.17)	<0.001	1.34	(1.02–2.06)	0.02
Use of CAM (Ref-No)	Yes	2.90	(0.82–5.29)	0.81
Nutritional status (Ref-Nourished)	Malnourished	1.41	(1.02–2.19)	<0.001	1.30	(1.06–2.43)	0.001
ADR (Ref-No)	Yes	5.53	(1.63–7.08)	<0.001	4.01	(1.03–8.20)	0.003
Flu-like symptoms (Ref-No)	Yes	1.60	(1.2–3.10)	0.02	1.41	(1.02–2.40)	0.04
Cutaneous reactions (Ref-No)	Yes	3.47	(1.4–5.06)	<0.001	2.50	(1.28–4.01)	<0.01
Gastrointestinal (Ref-No)	Yes	1.90	(1.02–3.19)	0.02	1.36	(1.05–4.07)	0.03
Neurological (Ref-No)	Yes	7.90	(1.3–10.20)	<0.001	6.1	(1.23–8.20)	<0.001
Hepatotoxicity (Ref-No)	Yes	8.7	(1.03–11.08)	<0.01	7.2	(1.26–9.04)	<0.01
Respiratory (Ref-No)	Yes	1.52	(1.11–4.03)	0.02	1.30	(1.06–2.01)	0.04
Regimen type (Ref-3HP)	3RH	5.37	(1.08–3.76)	<0.001	4.23	(1.85–6.53)	0.01
Adherence support (Ref-Yes)	No	1.83	(1.12–6.69)	0.02	1.56	(1.08–6.69)	0.03

ADR, adverse drug reaction; aOR, adjusted odds ratio; CM, complementary medicine; ETB, Ethiopian birr; GP, general practitioner; LTBI, latent TB infection; OR, odds ratio; Ref, reference category; TPT, TB preventive therapy.

Discussion

Several preventive measures are employed, yet preventive pharmacotherapy ranked as the best intervention.^40,35 However, non-adherence tempers TPT program efficacy.⁴¹ In this study, among TPT eligibles, 82.72% initiated TPT. Despite failing to attain the national strategic plan by 2025/2026 (90%),¹⁹ it corroborates an earlier study conducted in Ethiopia^20,42-44 and Sub-Saharan Africa, such as Rwanda and Eswatini.^45,46

Among the participants in the study, 27.94% were non-adherent, indicating a failure to meet the strategic goal.¹⁹ However, this study’s finding was in accordance with earlier studies in Ethiopia.^23,36 This finding also fairly aligns with an earlier study in Mozambique.⁴⁷

Herein, multiple variables predict non-adherence. Compared to HHCs under five, those older than five had a roughly twofold-increased odds of developing non-adherence to TPT. The results of this investigation mirror those of previous studies in Ethiopia,²⁵ and Cambodia.⁴² This may be due to the fact that school-aged children are more prone to non-adherence due to a lack of firm adherence support from parents/caregivers.⁴⁸ In the contrary, a cross-sectional study in China reported that age failed to reach statistical significance threshold.³⁰ This discrepancy might be due to the involvement of adolescents and adults (14–52 years old), differences in TPT regimen type and study area.

This study revealed that being orphanhood increases the risk of non-adherence, possibly due to a lack of strong social support.^43,44 Furthermore, child contacts whose caregivers were guardians/neighbors were twice as likely to have failed to adhere as compared to those cared for by biological relatives. In the Ethiopian context, TPT was conducted as a self-administered therapy. Therefore, biologic-relative adherence supporters achieves higher rate of adherence than their counterparts.

Chronic comorbid illnesses predict TPT non-adherence in agreement with earlier studies.^49,50 This may be potentially due to the medication regimen complexity associated with the treatment of comorbidities. Poor nutritional status was also found to be a pertinent risk factor in agreement with previous studies.^26,51 On the contrary, earlier studies involving patients with LTBI reported that nutritional status failed to reach a statistically significant level.^30,52 This may be due to differences in study design and patients’ age group.

HHCs on the 3RH regimen were more likely to have failed to adhere than those who were on the 3HP regimen. This study finding mirrors an earlier study in Ethiopia.²⁵ In addition, a study in Cambodia reported that individuals who initiated 3RH or 6H were more likely to not complete the treatment compared to those on 3HP.⁴² This may be due to the medication regimen complexity and treatment burden with 3RH and 6H regimens.

Children experiencing ADRs were found to have a higher odds of treatment non-adherence, consistent with other studies.⁵³ Among all types of ADRs, patients who revealed neurological reactions and hepatotoxicity were at a higher odds of being non-adherent to TPT in line with the earlier studies.⁵⁴ This may be due to the development of a negative attitude toward the medications as they experienced TPT-related ADRs.⁵⁵ In contrast, a study in a similar setting revealed that ADR experience was not associated with non-adherence.⁵⁶ This might possibly be due to a target population discrepancy. Finally, drug stock-out was also predict TPT non-adherence. A study in rural Delhi revealed that the primary reason for non-adherence to TPT among HHCs was the unavailability of medicine.⁵⁷

Strength and limitation of the study

Our study has several limitations. In the first place, purposive hospital selection and exclusion of participants due to referral and incomplete medical records may affect the study’s integrity. However, this is considered a low risk due to the relatively small size of the excluded group, and their characteristics were less likely to be systematically different. Despite this study trying to incorporate multiple potential predictors, it might be affected by unmeasured variables due to its retrospective nature. Although there are methodological limitations, following the adoption of the 3HP-based regimen in Ethiopia this study is the first of its kind.

Conclusion and recommendations

Non-adherence among HHCs to TPT was alarmingly high in the study area, indicating a failure to achieve the strategic plan set by Ethiopia for 2025/2026. It remains the program’s concern despite the introduction of rifamycins-based 3HP-regimen into the Ethiopian TPT algorithm.

To mitigate TPT non-adherence in the study area, the government should set focused policies on curbing risk factors. This includes access to healthcare institutions and medications, improving nutritional status, and clinicians should prefer to initiate 3HP-based TPT accompanied with firm counseling, supervision, and vigilance for ADRs. Altogether, special emphasis should be given to those who were school-aged children, orphans, comorbid, malnourished, and ADR experienced, considering them as vulnerable groups.

Supplemental Material

sj-docx-1-tai-10.1177_20499361261426969 – Supplemental material for Non-adherence and its predictors among pediatric household contacts on tuberculosis preventive therapy in the 3HP era: A multicenter observational study

Supplemental material, sj-docx-1-tai-10.1177_20499361261426969 for Non-adherence and its predictors among pediatric household contacts on tuberculosis preventive therapy in the 3HP era: A multicenter observational study by Woretaw Sisay Zewdu, Muluken Adela Alemu, Tilaye Arega Moges, Sisay Sitotaw Anberbr, Desalegn Addis Mussie, Abel Temeche Kassaw, Demewoz Kefale, Tigabu Eskeziya Zerihun, Fisseha Nigussie Dagnew and Samuel Berihun Dagnew in Therapeutic Advances in Infectious Disease

Footnotes

Acknowledgements

The authors would like to thank the data collectors, the hospitals, and the respective TB clinic staffs.

Declarations

ORCID iDs

Woretaw Sisay Zewdu

Tilaye Arega Moges

Sisay Sitotaw Anberbr

Desalegn Addis Mussie

Abel Temeche Kassaw

Tigabu Eskeziya Zerihun

Fisseha Nigussie Dagnew

Samuel Berihun Dagnew

Supplemental material

Supplemental material for this article is available online.

References

International Union Against Tuberculosis and Lung Disease (The Union). Silent epidemic: A call to action against child tuberculosis. Geneva, Switzerland, 2025.

World Health Organization. Global tuberculosis report 2024. Geneva, Switzerland: World Health Organization, 2024.

Villavicencio

Perin

Eilerts-Spinelli

, et al. Global, regional, and national causes of death in children and adolescents younger than 20 years: an open data portal with estimates for 2000–21. Lancet Global Health 2024; 12(1): e16–e17.

Guo

, et al. Tuberculosis incidence, deaths and disability-adjusted life years in children and adolescence, 1990-2021: results from the Global Burden of Disease Study 2021. PLoS ONE 2025; 20(3): 1–19.

Alene

Python

Weiss

DJ.

, et al. Mapping tuberculosis prevalence in Ethiopia using geospatial meta-analysis. Int J Epidemiol 2023; 52(4): 1124–1136.

Dememew

Deribew

Zegeye

, et al. A high burden of infectious tuberculosis cases among older children and young adolescents of the female gender in Ethiopia. Trop Med Infect Dis 2025; 10(3): 1–11.

Salazar-Austin

Dowdy

Chaisson

, et al. Seventy years of tuberculosis prevention: efficacy, effectiveness, toxicity, durability, and duration. Am J Epidemiol 2019; 188 (12): 2078–2085.

Seid

Alemu

Diriba

, et al. Active tuberculosis in household contacts of bacteriologically confirmed pulmonary tuberculosis patients: a multicenter study finding the ‘Missed One’ in Central Ethiopia. PLoS ONE 2025; 20(2): 1–17.

Swathy

Krishnan

Geetha

Prevalence of latent TB infection among childhood contacts of adults with sputum positive pulmonary TB. Clin Epidemiol Global Health 2024; 28(101612): 1–4.

10.

Cohen

Mathiasen

Schön

, et al. The global prevalence of latent tuberculosis: a systematic review and meta-analysis. Eur Respiratory J 2019; 54 (3): 1–14.

11.

Ding

Guo

, et al. Prevalence trends of latent tuberculosis infection at the global, regional, and country levels from 1990–2019. Int J Infect Diss 2022; 122: 46–62.

12.

Smith

AGC

Kempker

Wassie

, et al. The impact of diabetes and prediabetes on prevalence of mycobacterium tuberculosis infection among household contacts of active tuberculosis cases in Ethiopia. Open Forum Infect Dis 2022; 9(7): 1–8.

13.

Daniel

Tadesse

Tamirat

, et al. Epidemiology of active tuberculosis among adult household contacts of patients with smear-positive pulmonary tuberculosis in Kembata-Tembaro zone, southern Ethiopia: a cross-sectional study. BMJ Public Health 2025; 3(e001531): 1–6.

14.

Matteelli

Lovatti

Sforza

, et al. Programmatic management of tuberculosis preventive therapy: past, present, future. Int J Infect Dis 2023; 130: S43–S46.

15.

Chekesa

Gumi

Chanyalew

, et al. Prevalence of latent tuberculosis infection and associated risk factors in prison in East Wollega Zone of western Ethiopia. PLoS ONE 2020; 15(5): 1–11.

16.

Churchyard

Swindells

Controlling latent TB tuberculosis infection in high-burden countries: a neglected strategy to end TB. PLoS Medicine 2019; 16(4): 5–8.

17.

United Nations Development Programme. Sustainable development goal 3. New York: Department of Economic and Social Affairs, 2017.

18.

Federal Democratic Republic of Ethiopia Ministry of Health (FDREMOH). Guidelines for clinical and programmatic management of TB, TB/HIV, DR-TB and leprosy in Ethiopia, 7th ed. Addis Ababa, Ethiopia: FMOH, 2021.

19.

Federal democratic republic of Ethiopia ministry of health. Tuberculosis-Leprosy national sterategic plan final draft 2021-2026. Addis Ababa: FMoH, 2020.

20.

Kebede

Fekadu

Jerene

Ethiopia’s experience on scaling up latent TB infection management for people living with HIV and under-five child household contacts of index TB patients. J Clin Tuberculosis Mycobacterial Dis 2018; 10: 29–31.

21.

Salazar-Austin

Bergman

Mulder

, et al. Improving access to tuberculosis preventive treatment for children in Ethiopia: designing a home-based contact management intervention for the CHIP-TB trial through formative research. BMC Health Services Res 2024; 24(1043): 1–14.

22.

Garie

Yassin

Cuevas

LE.

Lack of adherence to isoniazid chemoprophylaxis in children in contact with adults with tuberculosis in Southern Ethiopia. PLoS ONE 2011; 6 (11): 13–17.

23.

Salazar-austin

Cohn

Nonyane

BAS

, et al. Effectiveness of a home-based approach to child contact investigation and TPT management by community health workers in ethiopia: a pragmatic cluster-randomized trial. Clin Infect Dis 2025; ciaf203: 1–8.

24.

World Helath Organization. WHO operational handbook on tuberculosis. Module 1: prevention-tuberculosis preventive treatment. 6th ed. Geneva, Switzerland: World Health Organization, 2024.

25.

Seid

Alemu

Diriba

, et al. Routine tuberculosis contact investigation yield and preventive treatment cascade in central Ethiopia. Heliyon 2024; 10 (10): e30942.

26.

Okwara

Oyore

Were

FN.

, et al. Correlates of isoniazid preventive therapy failure in child household contacts with infectious tuberculosis in high burden settings in Nairobi, Kenya - a cohort study. BMC Infect Dis 2017; 17(623): 1–11.

27.

Chiang

SS.

Senador

Altamirano

, et al. Adolescent, caregiver and provider perspectives on tuberculosis treatment adherence: a qualitative study from. BMJ Open 2023; 13: 1–12.

28.

Leddy

AM.

Jaganath

Triasih

, et al. Social determinants of adherence to treatment for tuberculosis infection and disease among children, adolescents, and young adults: a narrative review. J Pediatric Infect Dis Soc 2022; 11(S3): S79–S84.

29.

Rahman

Hossain

Banu

, et al. Uptake and completion of tuberculosis preventive treatment using 12-dose, weekly isoniazid–rifapentine regimen in Bangladesh: a community-based implementation study. MDPI: Trop Med Infect Dis 2024; 9(4).

30.

Huang

Yuan

, et al. Effects of preventive therapy for latent tuberculosis infection and factors associated with treatment abandonment: a cross-sectional study. J Thoracic Dis 2018; 10(7): 4377–4386.

31.

WHO. WHO consolidated guidelines on tuberculosis: Module 1: Prevention-tuberculosis preventive treatment. 6th ed. Geneva, Swizerland: World Health Organization, 2024.

32.

FDRE Ministry of Health. National guidelines for TB, DR-TB and leprosy in Ethiopia. 6th ed. Addis Ababa, Ethiopia: FDRE Ministry of Health, 2018.

33.

Hailu

Jara

Bogale

EK.

Nutritional status and its associated factors among infants and young children aged 6-23 months in Addis Ababa, Central Ethiopia, 2021: a cross-sectional study. J Nutr Sci 2024; 13: e43.

34.

Central Statistical Agency (CSA), ICF. Ethiopia demographic and health survey 2016. CSA, Addis Ababa, Ethiopia, 2016.

35.

Liyew

Gilmour

Clements

ACA

, et al. Comparative effectiveness of interventions for preventing tuberculosis: systematic review and network meta-analysis of interventional studies. eClinicalMedicine 2023; 64: 102209.

36.

Seid

Tsedalu

Ayele

Adherence to isoniazid preventive therapy among under-five children in contact with adult bacteriologically confirmed pulmonary tuberculosis patients: a mixed-method study. Int J Microbiol 2020; 2020: 1–8.

37.

Abelti

Dememew

Gebreyohannes

, et al. Community-based tuberculosis preventive treatment among child and adolescent household contacts in Ethiopia. Trop Med Infect Dis 2025; 10(102): 1–15.

38.

Sandoval

Mtetwa

Devezin

, et al. Community-based tuberculosis contact management: caregiver experience and factors promoting adherence to preventive therapy. PLOS Global Public Health 2023; 3(7): 1–14.

39.

Plans-Rubió

Godoy

Toledo

, et al. Factors associated with non-adherence to tuberculosis preventive treatment among adult contacts of pulmonary tuberculosis cases with latent tuberculosis infection in Catalonia, Spain, in 2019−2021. Trop Med Infect Dis 2024; 9(3): 1–13.

40.

Teixeira

LAA

Santos

Correia

, et al. Long-term protective effect of tuberculosis preventive therapy in a medium/high tuberculosis incidence setting. Clin Infect Dis 2024; 78(5): 1321–1327.

41.

Ihesie

Chukwuogo

Eneogu

, et al. Acceptance and completion rates of 3-month isoniazid-rifampicin (3HR) tuberculosis preventive treatment (TPT) among contacts of bacteriologically confirmed TB patients—patients’ and healthcare workers’ perspectives. Trop Med Infect Dis 2024; 9(301): 1–14.

42.

Khun

KE.

Factors associated with incomplete tuberculosis preventive treatment: a retrospective analysis of six-years programmatic data in Cambodia. Sci Rep 2024; 14(18458): 1–7.

43.

Liu

, et al. The influence of parents on medication adherence of their children in China: a cross-sectional online investigation based on health belief model. Front Public Health 2022; 10(845032), 1–13.

44.

Amuge

Ndekezi

Mugerwa

, et al. Facilitators and barriers to initiating and completing tuberculosis preventive treatment among children and adolescents living with HIV in Uganda: a qualitative study of adolescents, caretakers and health workers. AIDS Res Therapy 2024; 21(59), 1–12.

45.

Birungi

FM.

Graham

SM.

Uwimana

, et al. Adherence to isoniazid preventive therapy among child contacts in Rwanda: a mixed-methods study. PLoS ONE 2019; 14 (2): 1–16.

46.

Kay

Sandoval

Mtetwa

, et al. Vikela Ekhaya: a novel, community-based, tuberculosis contact management program in a high burden setting. Clin Infect Diss 2022; 74(9): 1631–1638.

47.

I de Manuel-Rimbau

Nguenha

Mambuque

, et al. Initiation and adherence to isoniazid preventive therapy in children under 5 years of age in Manhiça, Southern Mozambique. J Trop Pediatrics 2023; 69(1): 1–4.

48.

Jerene

Assefa

Tesfaye

, et al. Effectiveness of women-led community interventions in improving tuberculosis preventive treatment in children: results from a comparative, before-after study in Ethiopia. BMJ Open 2022; 12(7), 1–9.

49.

World Health Organization. WHO Operational handbook on tuberculosis: Module 6: Tuberculosis and comorbidities. 3rd ed. Geneva, Switzerland: World Health Organization, 2024.

50.

Sentís

Vasconcelos

Machado

, et al. Failure to complete treatment for latent tuberculosis infection in Portugal, 2013–2017: geographic-, sociodemographic-, and medical-associated factors. Eur J Clin Microbiol Infect Dis 2020; 39(4): 647–656.

51.

Islam

Das

Kumar

DD.

Nutritional status and adherence to anti-tubercular treatment among tuberculosis patients in a community development block of Eastern India. Indian J Tuberculosis 2024; 71: S178–S183.

52.

Stærke

Martinsen

Jensen

, et al. A cohort study of the long-term outcome of latent tuberculosis infection among socially marginalized people in a low-incidence country. Int J Infect Dis 2022; 124: S56–S62.

53.

Omar

Mohamoud

Adam

, et al. Assessment of non-adherence to anti-TB drugs and associated factors among patients attending TB treatment centers during COVID-19 pandemic in Mogadishu, Somalia: a cross-sectional study. Infect Drug Res 2024; 17: 3879–3890.

54.

Ozaltun

SC.

An evaluation of medication adherence in new tuberculosis cases in Ankara: a prospective cohort study. Healthcare 2024; 12(2353), 1–17.

55.

Felisia

Triasih

Nababan

BWY

, et al. High tuberculosis preventive treatment uptake and completion rates using a person-centered approach among tuberculosis household contact in Yogyakarta. Trop Med Infect Dis 2023; 8(520): 1–10.

56.

Sharma

Bakshi

Basu

, et al. Implementation of tuberculosis preventive therapy with INH-Rifapentine (3HP) for latent tuberculosis infection management in household tuberculosis contacts in India: a prospective study. Trop Med Int Health 2023; 28 (12): 890–900.

57.

Kumar

Yadav

Lukhmana

Community-based assessment of tuberculosis preventive therapy among household contacts living with index pulmonary TB patient in rural Delhi. Indian J Tuberc 2025; 72(4): 494–499.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.10 MB