
==== Front
Int Health
Int Health
inthealth
International Health
1876-3413
1876-3405
Oxford University Press

37971028
10.1093/inthealth/ihad104
ihad104
Original Article
AcademicSubjects/MED00390
Influence of the new dispersible fixed-dose combination anti-Tuberculosis drug on treatment adherence among children with Tuberculosis in Osun State, Nigeria
Chijioke-Akaniro Obioma National Tuberculosis, Leprosy and Buruli Ulcer Control Programme, Nigeria

https://orcid.org/0000-0001-6253-1059
Akinyemi Patrick A National Postgraduate Medical College of Nigeria, Ijanikin Lagos State, Nigeria

Asuke Sunday Bingham University Karu, Nasarwa state, Nigeria

Anyaike Chukwuma National Tuberculosis, Leprosy and Buruli Ulcer Control Programme, Nigeria

Uwaezuoke Ndubuisi A Department of Paediatrics, University of Nigeria Teaching Hospital Ituku-Ozalla Enugu state, Nigeria

Ochuko Urhioke National Tuberculosis, Leprosy and Buruli Ulcer Control Programme, Nigeria

Ubochioma Emperor National Tuberculosis, Leprosy and Buruli Ulcer Control Programme, Nigeria

Omoniyi Amos WHO Country Office Nigeria

Merle Corinne S Special Programme for Research & Training In Tropical Diseases (TDR), World Health, Organization, Geneva Switzerland

Daniel Soji Olabisi Onabanjo University, Ago-Iwoye, Ogun State, Nigeria

Corresponding author: Tel: +2349092307226; E-mail: ocakaniro@gmail.com
9 2024
16 11 2023
16 11 2023
16 5 534543
23 5 2023
31 7 2023
24 10 2023
© The Author(s) 2023. Published by Oxford University Press on behalf of Royal Society of Tropical Medicine and Hygiene.
2023
https://creativecommons.org/licenses/by-nc/4.0/ This is an Open Access article distributed under the terms of the Creative Commons Attribution-NonCommercial License (https://creativecommons.org/licenses/by-nc/4.0/), which permits non-commercial re-use, distribution, and reproduction in any medium, provided the original work is properly cited. For commercial re-use, please contact journals.permissions@oup.com

Abstract

Background

The dispersible fixed-dose combination drug has been recommended as the mainstay of treatment for TB in children. However, more needs to be known about its effect on treatment. This study aimed to assess the effectiveness of the formulation on treatment adherence among children with TB.

Methods

A historical cohort design was used to assess and compare adherences of old loose non-dispersible and new dispersible fixed-dose anti-TB drugs, using a convergent parallel mixed-method approach for data collection. Determinants of treatment adherence were assessed using binary logistic regression.

Results

The proportion of children with good treatment adherence was higher in the new dispersible formulation group (82 [64.6%]) relative to the proportion among the loose non-dispersible formulation group (29 [23.4%]). Reports of forgetfulness, travelling and pill burden were significantly higher among those with poor adherence in the loose non-dispersible formulation group. Significant predictors of treatment adherence were acceptability (adjusted OR [AOR]=4.1, p=0.013, 95% CI 1.342 to 12.756), travelling from treatment areas (AOR=8.9, p=0.002, 95% CI 2.211 to 35.771) and forgetfulness (AOR=74.0, p<0.001, 95% CI 23.319 to 234.725).

Conclusions

The determinants of treatment adherence are multifactorial. In addition to ensuring universal access to the drug, flexible referral in case of travelling and ensuring treatment partners' participation to minimise forgetfulness to take pills, are essential.

adherence
dispersible fixed-dose combination
loose non-dispersible formulation
TB
==== Body
pmcIntroduction

Adherence to anti-TB treatment is essential for the successful management of TB among children living with the infection. The goal of TB treatment is to achieve a cure in the case of pulmonary TB and treatment completion in the case of extrapulmonary TB, in which treatment adherence is essential. Non-adherence to treatment has been linked to negative treatment outcomes like treatment failure, relapse, development of drug-resistant TB, worsening medical condition and, consequently, death.1 Despite the importance of the concept of adherence in the management of TB, there is no universally accepted definition or measure of adherence in the programme. Adherence to TB treatment is simply viewed as the extent to which a patient takes medications as prescribed by the healthcare service providers.2 A study conducted in Ethiopia was more specific in its operational definition of adherence to anti-TB treatment. Adherence was defined as the usage of at least 95% of the medications as prescribed within a reference period of 30 d.3 Apart from the use of pill counts, the use of the Tuberculosis Treatment Adherence Scale is another means of assessing adherence to anti-TB medication.4 It is a self-reporting tool appropriate for retrospective study, especially in a health system with poor documentation. The tool is capable of assessing treatment adherence over the entire treatment period.

The appropriate method for the assessment of adherence is contextual. This could involve direct assessment like Directly Observed Therapy (DOT), biomarker assay, pill counts and other proxy measures like patient self-report and the use of attendance visits. In Nigeria, a modified DOT short course (DOTs) is practised in the management of drug-sensitive TB, which could be regarded as a community-based DOT; intake of drugs and keeping records are supervised by an appointed relative or friends. This thus makes pill counting in the face of poor unreliable documentation of drug intake a less desirable and tedious proxy measure of adherence to TB treatment. Self-report of missed doses is a simple and inexpensive measure of non-adherence to TB treatment. Self-reporting has been shown to have almost equal accuracy with other expensive and tedious methods like the use of 99 DOTS (a cell phone-based technology) and higher accuracy than pill estimates.5 Also, a systematic review of studies that assessed the relationship between adherence assessment outcomes using self-reporting methods and monitoring devices established moderate to high correlations between the two broad measures of adherence to TB treatments.6

Adherence to TB treatment, being a long-term treatment, is affected by multiple factors that are interrelated. The factors that have been identified in related studies could be classified into individual factors, namely, drug factors, health system factors and other environmental factors like a lack of social support.7–10 Drug formulation, dosages and occurrence of side effects were among the identified factors influencing the adherence of children to anti-TB medication.8,11 Commonly identified individual factors include a lack of belief in the efficacy of the drug, age group of the child and gender of the child.8,12 Younger age group, female gender and living in urban areas were significant predictors of good adherence to TB treatment.10 Other identified individual factors comprise perception of the disease and its treatment (perceived benefit, barrier, threat and self-efficacy), caregivers’ poor knowledge of TB and its treatment, forgetfulness and being busy with daily activities (and school activities by the child).10,13–15 Health system factors have been shown to also play a significant role in adherence to TB treatment. Relationships between patients or caregivers and healthcare providers, access to healthcare facilities and the availability of drugs were the common health system factors affecting treatment adherence.10,13 Good communication between healthcare workers and the patients or caregivers, coupled with the caregiver's concern for the child's health, have been shown to be a strong motivation for adherence.10,15

Consequent to the observed challenges militating against adherence to anti-TB treatment among children, especially the drug formulation factor, the WHO introduced the use of a child-friendly dispersible anti-TB drug in 201516 and this was introduced to the Nigerian Health System in 2018. There is a dearth of studies that assess the effect of the new drug formulation on treatment adherence relative to the pre-existing non-dispersible fixed-dose medications since its introduction. This study, therefore, assessed and compared the adherence to TB treatment among patients on the new child-friendly fixed dose and the non-dispersible fixed-dose formulation. The findings would form the basis for future public health interventions in TB and other disease control programmes that require prolonged medication among children.

Materials and Methods

Description of study area

The research was conducted in Osun State. It is a landlocked state located in the Southwest region of Nigeria, 7°30′N 4°30′E.17,18 It has a total population of 3 423535 million (National Population Figures of 2006) with a 2023 population projection figure of 5 123 586 based on an annual growth rate of 2.4%.19 Osun state is mainly an agrarian state in the tropical rainforest of Nigeria. The state has 30 local government areas (LGAs) that have been classified into 15 urban LGAs and 15 rural LGAs.

Each LGA has 10–11 wards with at least one model primary health centre (PHC) per ward to ensure access to healthcare. All the PHCs have facilities for vaccination, including BCG, and provision of primary care for locally endemic diseases like diarrheal diseases. Osun State has 428 DOTs sites that provide treatment services for children with drug-sensitive TB. The total number of cases of TB in Osun State in 2022 was 22 799, out of which 904 were paediatric cases, according to the Monitoring and Evaluation Unit of the National Tuberculosis, Leprosy, and Buruli Ulcer Control Programme. The TB treatment success rate among children in Osun State increased by 2% from 2020 to 2022; 89% in 2020 and 91% in 2022. Treatment failure rates were also observed to have declined from 1% to 0% while TB-specific death rates declined from 5% to 3% within the same period. Osun State introduced fixed-dose combination (FDC) dispersible anti-TB medications for children in 2020. All the healthcare facilities providing DOTs services currently use the new child-friendly dispersible fixed-dose anti-TB drugs. However, some facilities still use both the new dispersible and old non-dispersible drugs concurrently. The state also has referral centres for drug-resistant TB and other complications resulting from TB infection.

Study design

The study was conducted using a historical cohort study design, using quantitative and qualitative data collection techniques. The retrospective section of the study involved interviewing children who used non-dispersible loose anti-TB drugs from 2018 to 2020 as well as their caregivers. The children who took child-friendly dispersible fixed-dose anti-TB drugs during 2020–2022 were also interviewed to determine their acceptability of anti-TB drug formulation. The study population was children aged 0–14 y with drug-sensitive pulmonary TB in Osun State.

Sample size and sampling technique

The sample size (N) was calculated to obtain an absolute precision of ±5% using the sample size formula for comparison between two proportions.20 Because the study population was relatively small or finite (i.e. <10 000), the finite population correction for proportions was used by dividing the sample size N by 1+[N–1/10 000].20 After correcting for an anticipated non-response rate of 10%, the minimum sample size was 123 for each group. However, 127 participants were recruited into the FDC group while 124 participants were recruited into the non-dispersible loose treatment group. The sample size was calculated based on the proportion of children who developed hepatotoxic-related side effects to child-friendly dispersible fixed-dose anti-TB drugs (9.6%) and non-dispersible loose anti-TB drugs (16.4%) in a study conducted among children with drug-sensitive pulmonary TB.21 This study was a subset of a larger study that involved the assessment of side effects of old loose non-dispersible and FDC dispersible anti-TB formulations, as well as acceptability and adherence to the formulations. The participants for the quantitative arm of the study were selected using a purposive sampling technique based on preset criteria: children aged 0–14 y with drug-sensitive pulmonary TB who used either of the two anti-TB drug formulations under study. Caregivers of eligible children were contacted and invited via telephone calls using the contact details documented at DOTs sites while their children were on treatment. Caregivers that honoured the invitations were enrolled into the study. The study sample for the qualitative aspect was selected using a convenient sampling technique. Four focused group discussions (FGDs), eight key informant interviews (KIIs) and 25 in-depth interviews (IDIs) were conducted. The interviewees in the FGDs and IDIs were mothers whose children had pulmonary TB and received dispersible and non-dispersible anti-TB medications, while the interviewees in the KIIs were healthcare workers and teachers providing support for pupils with the infection.

Data collection

Data were collected during December 2022–January 2023. The data were collected by six trained research assistants with a minimum qualification of Bachelor in Social Sciences. Quantitative data were collected using an interviewer-administered questionnaire, administered electronically using an android application, mWater Survey, by research assistants trained for the study. The first section of the questionnaire assessed the characteristics of the children and their households, while the second section comprised questions that assessed the acceptability of anti-TB drugs. Acceptability was assessed using questions that asked for the perceptions of caregivers about the characteristics of the drug formulation administered: frequency of dosage, appearance, taste, dissolubility and pill burden. The third section comprised six questions that measured treatment adherence. The questions were adapted from the eight-item Morisky Medication Adherence Scale that has been used to explore the extent of adherence to treatment in some chronic diseases like HIV and diabetes mellitus.22 The adapted questions were tested for reliability, leading to the exclusion of two items with a higher Cronbach's alpha value if the items were deleted. The Cronbach alpha value of the six-item questionnaire used was 0.822. The last two sections assessed the level of knowledge of the caregivers about TB and the support given while receiving treatment.

The qualitative aspect adopted KIIs, IDIs and FGD guides to explore the determinants of adherence to the treatment of TB. The guides explored the perception of participants about the disease, treatments and their experience with anti-TB drugs. Likewise, common factors that enhanced treatment adherence were also investigated by the guides. The interviews were conducted by experienced trained research assistants who are fluent in the native language of the study area, Yoruba. The interview guides were translated into Yoruba to encourage the active participation of those caregivers with a limited understanding of English. The interviews were audio-recorded while salient points were documented by notetakers. The audio records were transcribed, while the interviews conducted in Yoruba were translated into English by a linguist.

Data analysis plan

Quantitative data were analysed using SPSS (https://portal.mwater.co/#/) version 27 for Windows while the qualitative data were analysed using NVIVO 10.

Qualitative data analysis

The data were coded using a hybrid approach consisting of a combination of deductive and inductive coding methods. The qualitative data were analysed using thematic analysis. Codes were initially generated based on the research question of the study and identified key constructs, while new codes identified in the course of the analysis were also considered and coded. The data collected were reviewed to identify related excerpts then were coded. The codes were grouped into themes and the findings were triangulated and used to strengthen the findings from the quantitative data.

Quantitative data analysis

Univariate analysis: categorical data like treatment adherence and reasons for non-adherence were summarised using frequencies and proportions, while numerical variables like age of the children were summarised using median and IQR due to the skewness of the data. Multiple response analysis was used to assess the reasons for non-adherence. Non-adherence was assessed based on the responses to six questions. A child was considered non-adherent if the response was ‘Yes’ to any of the questions. The knowledge of caregivers about TB was assessed by composite scoring of seven questions, where the correct response attracted one mark while the wrong response attracted zero marks. The total scores were ranked into tertiles. Those with scores in the first tertile were classified as having a poor knowledge of TB, while those with scores in the second tertile were classified as having an average knowledge of the disease. Those with scores in the third tertile were classified as having a good knowledge of the disease.

Bivariate analysis: factors reported to have contributed to non-adherence were compared between the two study groups using χ2. Associations between sociodemographic characteristics of the children and their caregivers were compared with the level of adherence using χ2, while the age of children was compared using the Mann–Whitney U-test.

Multivariate analysis was conducted to assess determinants of treatment adherence using binary logistic regression analysis. A p-value of <0.05 was considered statistically significant.

Results

The median age of the children that took the new dispersible fixed-dose anti-TB drugs was 6.0 (4.0–10.0) y, while the median age of the children that took the old single-dose non-dispersible drugs was 10.0 (8.0–13.0) y (p<0.001). There was a statistically significant difference between the educational status of children in the two drug formulation groups (p<0.001). There were no out-of-school children among those children that took the old non-dispersible formulation, while 20 (15.7%) were out of school among those children that took the new FDC dispersible drug. The proportion of caregivers/mothers with a good knowledge of TB was statistically significantly higher among caregivers with children that used old non-dispersible formulation (54 [56.8%]) relative to the proportion with a good knowledge among those that used the new dispersible formulation (41 [43.2%]) (p=0.023). There were no statistically significant associations between drug formulation groups and other participant characteristics like religion, level of education of parents/caregivers and wealth index (Table 1).

Table 1. Comparison of participants’ characteristics across the study groups

Variable	New dispersible drug, n (%)	Old single doses non-dispersible drugs, n (%)	Statistics	
Age of the child (years)	6.0 (4.0–10.0)	10.0 (8.0–13.0)	U=3853.5	
			p<0.001	
Gender				
 Male	62 (48.8)	68 (54.8)	X2=0.9106	
 Female	65 (51.2)	56 (45.2)	p=0.3399	
Religion				
 Christianity	51 (40.2)	62 (50.0)	LR=4.069	
 Islam	76 (59.8)	61 (49.2)	p=0.131	
 Other	0 (0.0)	1 (0.8)		
Mother’s occupation				
 Artisan	31 (24.4)	31 (25.0)		
 Civil servant	14 (11.0)	6 (4.8)	LR=17.868	
 Farming	25 (19.7)	44 (35.5)	p=0.003	
 Housewife	6 (4.7)	2 (1.6)		
 Trading	46 (36.2)	41 (33.1)		
 Other	5 (3.9)	0 (0.0)		
Child’s educational status				
 In school	107 (84.3)	124 (100.0)	X2=21.218	
 Out of school	20 (15.7)	0 (0.0)	p<0.001	
 Mother’s level of education				
 No formal education	16 (12.6)	21 (16.9)		
 Primary	44 (34.6)	42 (33.9)		
 Secondary	37 (29.1)	42 (33.9)	X2=3.486	
 Diploma	21 (16.5)	13 (10.5)	p=0.480	
 University degree	9 (7.1)	5 (4.8)		
Father’s level of education				
 No formal education	12 (9.4)	23 (18.5)		
 Primary	30 (23.6)	33 (26.6)	X2=7.368	
 Secondary	52 (40.9)	47 (37.9)	p=0.118	
 Diploma	17 (13.4)	8 (6.5)		
 University degree	16 (12.6)	13 (10.5)		
Household’s wealth index				
 First quintile	19 (15.0)	30 (24.2)		
 Second quintile	24 (18.9)	27 (21.8)		
 Third quintile	29 (22.8)	22 (17.7)	X2=4.931	
 Fourth quintile	26 (20.5)	24 (19.4)	p=0.294	
 Fifth quintile	29 (22.8)	21 (16.9)		
Knowledge of TB				
 Poor	27 (21.3)	12 (9.7)	X2=7.522	
 Average	59 (46.5)	58 (46.8)	p=0.023	
 Good	41 (32.3)	54 (43.5)		
Abbreviations: U, Mann-Withney U Test; LR, Likelihood Ratio; X2, Chi-Square.

Figures 1 and 2 show the proportion of children with good adherence to TB treatment and the proportion with good adherence disaggregated based on the drug formulation used. The overall prevalence of good adherence to TB treatment across the two groups was 111 (44.2%). A higher proportion of children (64.6%) on the new FDC dispersible had good adherence to treatment compared with their counterparts on the old drug formulation (23.4%) and this was statistically significant (p<0.001).

Figure 1. TB treatment adherence.

Figure 2. Treatment adherence across the two study groups.

The submissions from the qualitative data align with the findings from the quantitative data on the effectiveness of the FDC dispersible formulation on treatment adherence:

[M]y child likes the taste, so he reminds [me] in most cases by saying ‘Where is my morning vitamin C?’ because he could not differentiate both (grandmother of a child on FDC dispersible formulation, IDI participant).

The timing of the administration also seemed to be helpful in ensuring good adherence, as presented in the qualitative findings. The convergence of the findings was predicated on the timing of the administration of the medication by the interviewee. As stated by some of the caregivers, because the medication is administered first thing in the morning, the drug administration is therefore saved from other daily activities that could serve as distractions:

I didn't miss it, every morning when they wake up, that's what I give them before they eat

(39-y-old mother with two children on loose non-dispersible formulation, IDI participant).

It's very easy for us to give them, immediately you wake up, you take water and give the drugs. I do that every day before the commencement of cooking and household chore[s] so that I don't get carried away with the rush to meet up with their school bus (33-y-old mother of a child on FDC dispersible formulation).

There was a statistically significant association between the age of a child and adherence to TB treatment. The median age of children with good adherence to treatment (7.0 [4.0–10.0] y) was statistically significantly lower than the median age of children with poor adherence (9.5 [7.0–12.0] y) (p<0.001). There was no statistically significant association between adherence to treatment and other children and caregivers’ sociodemographic characteristics, like the level of education of mothers or relevant caregivers, household wealth index, religion or gender of the child. The age of a child was a significant factor in treatment adherence and was supported by findings from the qualitative data:

Administering the drug to the younger one was easier because I can easily hold him and make him take the drug. Giving the older one could be challenging and I sometimes gave up on drug administration when the father or our male neighbour that could hold him are not around (39-y-old mother with two children on loose non-dispersible formulation, IDI participant).

Table 2 compares variables that suggest non-adherence between the two study groups. A statistically significantly higher proportion of caregivers that used old loose non-dispersible drugs (88 [71.0%]) reported forgetfulness as the cause of non-adherence relative to the proportion among the caregivers that used the new dispersible drugs (33 [26.0%]) (p<0.001). More children that took loose non-dispersible drugs experienced side effects of the drugs and felt worse; the experience of side effects necessitated cutting back on the medication in 23 (18.5%) compared with five children (3.9%) using new FDC dispersible drugs (p<0.001). More than one-fifth, 28 of the children (22.6%) had their treatment duration extended for >3 wk because of non-adherence compared with seven (5.5%) with a similar extension among children who took the new dispersible anti-TB medication (p<0.001). A higher proportion of caregivers that used the old non-dispersible formulation (39 [31.5%]) forgot their children's medications while travelling compared with the proportion of caregivers that used the FDC dispersible formulation (7 [5.5%]) (p<0.001). The proportion of caregivers that discontinued medication before the recommended duration of treatment because the children felt better was significantly higher among those that used old non-dispersible formulation (16 [12.9%]) compared with those caregivers that used the FDC dispersible formulation (2 [1.6%]) (p<0.001).

Table 2. Comparison of variables used for adherence assessment across the two study groups

	Study group		
Variable	New dispersible drug, n (%)	Old non-dispersible drug, n (%)	Statistics	
Do you sometimes forget to take your anti-TB medication?				
 Yes	33 (26.0)	88 (71.0)	X2=50.848	
 No	94 (74.0)	36 (29.0)	p<0.001	
Did you cut back on your medication because you felt worse or experienced side effects of drugs?				
 Yes	5 (3.9)	23 (18.5)	X2=13.515	
 No	122 (96.1)	101 (81.5)	p<0.001	
Did you ever forget to take your pills along when you travelled?				
 Yes	7 (5.5)	39 (31.5)	X2=28.205	
 No	120 (94.5)	85 (68.5)	p<0.001	
Was your treatment extended by more than 3 weeks?				
 Yes	7 (5.5)	28 (22.6)	X2=15.233	
 No	120 (94.5)	96 (77.4)	p<0.001	
Did you stop the medication at any time when the illness was under control?				
 Yes	2 (1.6)	16 (12.9)	X2=12.095	
 No	125 (98.4)	108 (87.1)	p<0.001	
Did you experience difficulties in adhering to the treatment plan?				
 Yes	20 (15.7)	76 (61.3)	X2=55.097	
 No	107 (84.3)	48 (38.7)	p<0.001	

Table 3 compares commonly reported reasons for non-adherence to treatment across the two study groups. The proportion of caregivers that attributed their non-adherence to simply forgetting the medication was significantly higher among the loose non-dispersible formulation group (87 [91.6%]) compared with 27 (60.0%) among the new FDC dispersible formulation group (p<0.001). Travelling away from the treatment area was also reported more frequently as the reason for non-adherence among the loose non-dispersible formulation group (46 [48.4%]) compared with 10 (22.7%) with a similar reason among the new FDC dispersible formulation group (p=0.004). Pill burden as a reason for non-adherence was statistically significantly higher among the loose non-dispersible formulation group (25 [26.3%]) compared with the new FDC dispersible formulation group (p=0.020).

Table 3. Comparison of factors that affect adherence to TB treatment among patients with poor adherence to treatment across the two study groups

	Study group		
Variable	New dispersible drug, n (%)	Old non-dispersible drug, n (%)	Statistics	
Simply forgot				
 No	18 (40.0)	8 (8.4)	X2=20.136	
 Yes	27 (60.0)	87 (91.6)	p<0.001	
Felt good				
 No	41 (93.2)	79 (83.2)	X2=2.561	
 Yes	3 (6.8)	16 (16.8)	p=0.110	
Travel away from home				
 No	34 (77.3)	49 (51.6)	X2=8.252	
 Yes	10 (22.7)	46 (48.4)	p=0.004	
Too many pills to take				
 No	40 (90.9)	70 (73.7)	X2=5.404	
 Yes	4 (9.1)	25 (26.3)	p=0.020	
Side effects of medication				
 No	43 (95.6)	85 (89.5)	X2=1.441	
 Yes	2 (4.4)	10 (10.5)	p=0.230	
Felt sick while on treatment				
 No	44 (97.8)	87 (91.6)	X2=1.951	
 Yes	1 (2.2)	8 (8.4)	p=0.163	
The caregiver was not around to pick up drugs				
 No	43 (95.6)	80 (84.2)	X2=3.684	
 Yes	2 (4.4)	15 (15.8)	p=0.055	
Busy with school activities				
 No	43 (95.6)	88 (92.6)	X2=0.434	
 Yes	2 (4.4)	7 (7.4)	p=0.510	

The findings from the qualitative study further affirmed the influence of travelling out of the treatment area on treatment adherence:

My daughter could not use her medication when I travelled to a religious camp. I didn't know where to get the drug and could not ask people around me because I don't want to be isolated. The two weeks she missed was added by the nurse, so she used the drug for more than six months (27-y-old mother of a child that took the FDC dispersible drug, FGD participant).

When my husband was transferred to this town, we waited for some weeks before we were directed to this health centre by a nurse in my neighbourhood. You know it’s not everybody that should know your challenges (27-y-old mother with a child on FDC dispersible formulation, IDI participant).

I got a refill before I travelled but the challenge I had was giving my child many pills in the presence of my in-laws. Sometimes I used the drugs when it is convenient, despite the irregular time of taking the medication, I still ensure he used the medication every day (45-y-old mother with a child on the loose non-dispersible anti-TB drug).

Table 4 shows binary logistic regression analysis of associations between participants' characteristics and treatment adherence. Acceptability of drug formulation was a significant predictor of treatment adherence. Participants classified as having good acceptability were four times more likely to adhere to treatment compared with those with poor drug acceptability (adjusted OR [AOR]=4.1, p=0.013, 95% CI 1.342 to 12.756). Travelling away from the place of residence (treatment area) was also a significant factor affecting treatment adherence. Participants that did not travel from the primary area of treatment throughout the treatment period were more likely to adhere to treatment than those with a history of travelling (AOR=8.9, p=0.002, 95% CI 2.211 to 35.771). Simply forgetting to take the medication plays a significant role in treatment adherence. Participants with no report of simply forgetting the medication were 74 times more likely to adhere to treatment (AOR=74.0, p<0.001, 95% CI 23.319 to 234.725).

Table 4. Binary logistic regression analysis of associations between selected variables and adherence

Variable	OR	p	95% CI	
Study group				
 Old non-dispersible formulation	Ref			
 New dispersible formulation	0.6	0.454	0.187–2.115	
 Age	0.043	0.500	0.922–1.181	
Acceptability				
 Poor	Ref			
 Good	4.1	0.013	1.342–12.756	
Travelling away from the treatment area				
 No	8.9			
 Yes	Ref	0.002	2.211–35.771	
Usually forgot to take the pills				
 No	74.0			
 Yes	Ref	<0.001	23.319–234.725	

Discussion

The difference in the educational status of children in the two groups could be due to the significant difference in their age ranges. Because the number of out-of-school children recorded were in the group that used the new FDC anti-TB drugs with a relatively younger age group, it could be partly due to the proportion of the children that were of preschool age. The proportion of mothers with a good knowledge of TB and TB treatment was higher in the group that used old loose non-dispersible anti-TB drugs relative to the proportion of caregivers that used the new dispersible formulation. This could be due to variations in the mode, sources and frequency of health education as the majority of participants in the two study groups were on treatments at different periods.

Overall, about four out of 10 participants had good adherence to TB treatment. This is relatively low compared with the prevalence of good adherence to TB treatment in studies conducted in other regions of Nigeria.23,24 However, the previous studies were conducted among adults with TB, while the target population of the current study was children aged 0–14 y. Because it is adults that served as caregivers for the children with TB, the variation in the adherence rate is probably due to the difference in value placed on individual health and the health of others. It could also be due to variations in the value placed on health and health-seeking behaviour across the study area because the studies were conducted in varying sociocultural environments within Nigeria. A similar study conducted among children in Uganda that focused on discontinuation of treatment as one of the indicators of non-adherence found that only about one out of 11 children discontinued treatment. The non-adherence among this Uganda population may be higher with consideration of other indicators of treatment adherence.25 A higher adherence was recorded among children in Ethiopia that is probably due to differences in the method of assessment of adherence and varying health system factors; the treatment adherence was restricted to the number of pills used out of the prescribed doses within 30 d from the study.8

Treatment adherence in this study was significantly higher among those who used FDC dispersible anti-TB drugs compared with children who used loose non-dispersible formulation. This could be partly due to numerous drug factors that may affect adherence to treatment: formulation, taste, frequency of dosage and pill burden, as well as dissolubility in water for children.26 The finding is similar to that of a systematic review that assessed and compared the FDCs and single-drug formulations in the treatment of pulmonary TB, in which adherence was observed to be higher among patients on FDC formulation.27 However, the systematic review was based on studies conducted among adults. Other similar studies and systematic reviews on drug formulations and medication adherence were in support of the finding that medication adherence is better among patients on FDC formulations.28,29 However, a randomised control trial of FDC vs loose drug regimen among adults with TB showed that there was no significant difference in the participants’ adherence to treatment in the two study groups.30 This could be due to some other factors, such as health workers’ monitoring, health education and perception of the disease, which have been shown to significantly influence treatment adherence.10,15

Acceptability of drug formulation was a significant predictor of adherence to TB treatment. This could be due to various properties assessed to derive acceptability, which are drug factors that could enhance adherence, comprising the taste, solubility in water, frequency of intake and formulation. (The acceptability of FDC dispersible anti-TB medication is discussed in another publication). The travelling of patients and/or caregivers away from the treatment area was also a predictor of treatment adherence. This can be attributed to the structure of the TB treatment system, which restricts access to the medications; the medications are only available at the dedicated clinic and are rarely available over the counter. This, therefore, limits the access of patients to the drugs (despite them being free) while away from the treatment area without a referral (transfer out).

Simply forgetting to take the drugs was the most common reason for non-adherence among caregivers. This could be because most children with TB do not have treatment supporters other than their parents, who have the responsibility of administering the medication. It could also be attributed to an improvement in health status because the child takes the medication, thus reducing the caregiver’s perceived threat of the disease. Forgetting to take the medication was also a major cause of non-adherence to treatment in previous studies.31–33

Conclusion

There was significantly better treatment adherence among children that used FDC dispersible tablets compared with the loose non-dispersible formulation group. Therefore, it may be expedient to increase the coverage of the dispersible FDC across all clinics in the study area. In addition to the improvement in the drug formulation and ensuring its ease of access, continuous health education on the importance of treatment adherence helped caregivers in ensuring that their children adhered to the treatment. Thus, the use of the FDC formulation coupled with strengthened health education and consistent follow-up by healthcare workers will further improve the adherence of children to anti-TB treatment.

Limitations of the study

The majority of participants in the two study groups were on treatments at different periods because the new dispersible anti-TB drug was introduced into the Nigerian Health System in 2018. The treatment adherence could have been affected by varying factors particular to the treatment periods. However, findings from those who are still using old non-dispersible drugs due to a shortage of the new formulation in some areas of study were not different from those who used it before the introduction of the new formulation in 2018. The study was also prone to recall bias because it entailed recalling past events when the caregiver's child was on treatment. This applied to both study groups because only children who had completed their treatments were enrolled in the study. The questions were structured to be practical, which eased the recall ability of caregivers.

Authors’ contributions

OC, APA and SD designed the study. All the authors were involved in study implementation. OC, APA and SD conducted data analysis and interpretation. OC and APA drafted the manuscript. All the authors critically revised the manuscript. All the authors read and approved the final version of the manuscript. SD mentored other authors.

Acknowledgements

We acknowledge the logistic support and provision of access to children with TB by the Osun State Tuberculosis and Leprosy Control Programme Office.

Funding

None.

Competing interests

CSM is a staff member of the World Health Organization; the author alone is responsible for the views expressed in this publication and they do not necessarily represent the decisions, policy or views of the WHO.

Ethical approval

Ethical approval was obtained from the State Ministry of Health of Osun State and the National Tuberculosis, Leprosy, and Buruli Ulcer Control Programme. Informed consent was obtained from the caregivers of the children with TB who were involved in the study.

Data availability

The dataset for this study is not publicly available but it will be made available upon reasonable request to the corresponding author.
==== Refs
References

1 Alipanah  N, Jarlsberg  L, Miller  C  et al.  Adherence interventions and outcomes of tuberculosis treatment: A systematic review and meta-analysis of trials and observational studies. PLoS Med. 2018;15 (7 ):e1002595.29969463
2 Valencia  S, León  M, Losada  I  et al.  How do we measure adherence to anti-tuberculosis treatment?  Expert Rev Anti Infect Ther. 2017;15 (2 ):157–65.27910715
3 Tesfahuneygn  G, Medhin  G, Legesse  M. Adherence to anti-tuberculosis treatment and treatment outcomes among tuberculosis patients in Alamata District, northeast Ethiopia. BMC Res Notes. 2015;8 :503.26420164
4 Yin  X, Tu  X, Tong  Y  et al.  Development and validation of a tuberculosis medication adherence scale. PLoS One. 2012;7 (12 ):e50328.23251363
5 Subbaraman  R, Thomas  BE, Kumar  JV  et al.  Measuring tuberculosis medication adherence: A comparison of multiple approaches in relation to urine isoniazid metabolite testing within a cohort study in India. Open Forum Infect Dis. 2021;8 (11 ):ofab532.35559123
6 Monnette  A, Zhang  Y, Shao  H  et al.  Concordance of adherence measurement using self-reported adherence questionnaires and medication monitoring devices: An updated review. Pharmacoeconomics. 2017;36 :17–27.
7 Muture  BN, Keraka  MN, Kimuu  PK  et al.  Factors associated with default from treatment among tuberculosis patients in nairobi province, Kenya: A case control study. BMC Public Health. 2011;11 (1 ):696.21906291
8 Fekadu  G, Bekele  F, Bekele  K  et al.  Adherence to anti-tuberculosis treatment among pediatric patients at Nekemte Specialized Hospital, Western Ethiopia. Patient Prefer Adherence. 2020;14 :1259–65.32801656
9 Gebreweld  FH, Kifle  MM, Gebremicheal  FE  et al.  Factors influencing adherence to tuberculosis treatment in Asmara, Eritrea: A qualitative study. J Health Popul Nutr. 2018;37 :1–9.29304840
10 Nakiranda  R . Factors associated with adherence to anti-tuberculosis treatment among children attending TB clinics in  Mulago Hospital, Nsambya Home Care and Kawempe (KCCA Health Centre). 2014, http://makir.mak.ac.ug/handle/10570/4562?show=full
11 Panuganti  S, Bhavani  K, Vijayasree  L  et al.  A cross sectional study on factors influencing non-adherence to treatment among paediatric tuberculosis patients in field practice area of Osmania Medical College, Hyderabad. Int J Comm Med Public Health. 2022;9 (6 ):2479–82.
12 OʼDonnell  MR, Wolf  A, Werner  L  et al.  Adherence in the treatment of patients with extensively drug-resistant tuberculosis and HIV in South Africa: A prospective cohort study. J Acquir Immune Defic Syndr. 2014;67 (1 ):22–9.24872138
13 Mekonnen  HS, Azagew  AW. Non-adherence to anti-tuberculosis treatment, reasons and associated factors among TB patients attending at Gondar town health centers, Northwest Ethiopia. BMC Res Notes. 2018;11 (1 ):691.30285907
14 Azizi  N, Karimy  M, Naseri  V. Determinants of adherence to tuberculosis treatment in Iranian patients: Application of health belief model. J Infection Dev Countries. 2018;12 :706–11.
15 Gebremariam  RB, Wolde  M, Beyene  A. Determinants of adherence to anti-TB treatment and associated factors among adult TB patients in Gondar city administration, Northwest, Ethiopia: Based on health belief model perspective. J Health, Popul Nutr. 2021;40 (1 ):49.34838120
16 World Health Organization . WHO75 - 2015 - child-friendly Formulations of Anti-TB Medicines. 2015. Available at https://photos.hq.who.int/galleries/1075 [accessed March 11, 2023].
17 Nigeria Governors’ Forum . Public Finance Database. 2023. Retrieved 4/11/2023, from https://www.publicfinance.ngf.org.ng/nigeria-state-budget/osun-state-public-finance-database-2/.
18 Osun State Government . Osun State-Geography. 2023. Retrieved 5/11/2023, from https://www.osunstate.gov.ng/about/geography/.
19 World Bank . Population growth (annual %) - Nigeria. 2021. Available at https://data.worldbank.org/indicator/SP.POP.GROW?locations=NG [accessed March 11, 2023].
20 Katz  DL, Elmore  JG, Wild  DM  et al.  Jekel's Epidemiology, Biostatistics, Preventive Medicine, and Public Health. 4th edn. Philadelphia, PA: Elsevier Saunder; 2014.
21 Pakhlavonova  A, Aksenova  V, Klevno  N  et al.  Acceptability of a new child-friendly dispersible tablet for drug-sensitive pulmonary tuberculosis in children. Eur Respir J. 2020;56 (suppl 64 ):1564.
22 Sakthong  P, Chabunthom  R, Sakulbumrungsil  R. Psychometric properties of the Thai version of the 8-item Morisky Medication Adherence Scale in patients with type 2 diabetes. Ann Pharmacother. 2009;43 :950–7.19366872
23 Iweama  CN, Agbaje  OS, Umoke  PCI  et al.  Nonadherence to tuberculosis treatment and associated factors among patients using directly observed treatment short-course in north-west Nigeria: A cross-sectional study. Sage Open Med. 2021;9 :2050312121989497.33614034
24 Ubajaka  CF, Azuike  EC, Ugoji  JO  et al.  Adherence to drug medications amongst tuberculosis patients in a tertiary health institution in South East Nigeria. Int J Clin Med. 2015;6 (06 ):399–406.
25 Kibirige  L, Izudi  J, Okoboi  S. Discontinuation of tuberculosis treatment among children in the Kampala Capital City Authority health facilities: A mixed-methods study. BMC Infect Dis. 2021;21 (1 ):511.34074268
26 Gugssa Boru  C, Shimels  T, Bilal  AI. Factors contributing to non-adherence with treatment among TB patients in Sodo Woreda, Gurage Zone, Southern Ethiopia: A qualitative study. J Infect Public Health. 2017;10 (5 ):527–33.28189508
27 Gallardo  CR, Rigau Comas  D, Valderrama Rodríguez  A  et al.  Fixed-dose combinations of drugs versus single-drug formulations for treating pulmonary tuberculosis. Cochrane Database Syst Rev. 2016;2016 (5 ):Cd009913.27186634
28 Wei  Q, Zhou  J, Li  H  et al.  Medication adherence with fixed-dose versus free-equivalent combination therapies: Systematic review and meta-analysis. Front Pharmacol. 2023;14 :14.
29 Böhm  AK, Schneider  U, Aberle  J  et al.  Regimen simplification and medication adherence: Fixed-dose versus loose-dose combination therapy for type 2 diabetes. PLoS One. 2021;16 (5 ):e0250993.33945556
30 Aseffa  A, Chukwu  J, Vahedi  M  et al.  Efficacy and safety of ‘fixed dose’ versus ‘loose’ Drug regimens for treatment of pulmonary tuberculosis in two high TB-burden African countries: A randomized controlled trial. PLoS One. 2016;11 :e0157434.27322164
31 Adane  AA, Alene  KA, Koye  DN  et al.  Non-adherence to anti-tuberculosis treatment and determinant factors among patients with tuberculosis in Northwest Ethiopia. PLoS One. 2013;8 (11 ):e78791.24244364
32 Zegeye  A, Dessie  G, Wagnew  F  et al.  Prevalence and determinants of anti-tuberculosis treatment non-adherence in Ethiopia: A systematic review and meta-analysis. PLoS One. 2019;14 (1 ):e0210422.30629684
33 Neze , nega  ZS, Perimal-Lewis  L, Maeder  AJ. Factors influencing patient adherence to tuberculosis treatment in Ethiopia: A literature review. Int J Environ Res Public Health. 2020;17 (15 ):5626–41.32759876
