• Users Online: 42
  • Print this page
  • Email this page

Table of Contents
Year : 2022  |  Volume : 1  |  Issue : 3  |  Page : 187-191

Factor associated with treatment noncompletion during implementation of isoniazid-preventive therapy in ten pilot sites of Lome, Togo

Departement of Service de Pneumo-Phtisiologie, Chu Sylvanus Olympio, Lomé, Togo

Date of Submission20-Jun-2022
Date of Decision29-Jun-2022
Date of Acceptance08-Aug-2022
Date of Web Publication18-Sep-2022

Correspondence Address:
Dr. Doevi Mawuena Biaou
100 Route de Moval, 90400 Trevenans
Login to access the Email id

Source of Support: None, Conflict of Interest: None

DOI: 10.4103/jpdtsm.jpdtsm_48_22

Rights and Permissions

BACKGROUND: Tuberculosis (TB) represents the leading cause of death among people living with HIV (PLHIV). Several studies have proven the effectiveness of isoniazid-preventive treatment (IPT) in reducing the incidence and mortality of TB. The study aims to identify the factors associated with noncompletion of the treatment during the pilot phase of IPT implementation.
METHODS: This was a retrospective cohort study using data routinely collected in ten PLHIV care centers in Lomé, Togo, conducted between June 1, 2019, and March 31, 2020. All confirmed PLHIV newly enrolled in care who gave consent were included. IPT tolerance and signs of TB were checked at each appointment. To identify the factors associated with noncompletion of treatment, a logistic regression model was developed.
RESULTS: Sixty-five percent of the 301 patients were female. The mean age was 39.8 ± 13.2 years. Respectively, 43.8% and 41.9% of the patients were at clinical stages 1 and 2. The mean completion rate was 42.9% with extremes depending on the center ranging from 9.5% to 90.0%. The main causes of noncompletion were isoniazid stockouts (45.9%), lost to follow-up (37.8%), and noncompliance (12.8%). Multivariate analysis identified stockouts, public type of center, and poor general condition at the inclusion as risk factors for noncompletion.
CONCLUSIONS: Good results obtained by some centers prove that the implementation of the intervention is indeed possible, as long as good management of drug stocks and better follow-up of patients are ensured.

Keywords: HIV, isoniazid-preventive treatment, Togo, tuberculosis

How to cite this article:
Biaou DM, Stephane Adambounou TA, Ogounde NA, Tchedeke Efio MM, Adjoh KS. Factor associated with treatment noncompletion during implementation of isoniazid-preventive therapy in ten pilot sites of Lome, Togo. J Prev Diagn Treat Strategies Med 2022;1:187-91

How to cite this URL:
Biaou DM, Stephane Adambounou TA, Ogounde NA, Tchedeke Efio MM, Adjoh KS. Factor associated with treatment noncompletion during implementation of isoniazid-preventive therapy in ten pilot sites of Lome, Togo. J Prev Diagn Treat Strategies Med [serial online] 2022 [cited 2022 Dec 7];1:187-91. Available from: http://www.jpdtsm.com/text.asp?2022/1/3/187/356293

  Introduction Top

HIV infection is the most potent immunosuppressive risk factor for developing active tuberculosis (TB) disease,[1] and TB represents the leading cause of death among people living with HIV (PLHIV), accounting for one-third of all death causes.[2]

Antiretroviral therapy (ART) alone can help reduce the risk of developing active TB by 65%. However, PLHIV are still at significantly increased risk of TB even with high CD4+ counts.[3] Several studies have demonstrated the effectiveness of isoniazid-preventive therapy (IPT) in reducing TB incidence and mortality, irrespective of ART.[4],[5],[6] Although recommended since the late 1990s by the WHO and United Nations Program on HIV/AIDS as part of a complete HIV/AIDS care strategy, its implementation has been severely hampered by several obstacles, including a lack of a definitive safe strategy to exclude active TB and restrictions on the use of isoniazid due to concerns about the development of drug resistance.[6],[7]

Togo National Program against HIV/AIDS and sexually transmitted infections Programme National de Lutte contre le Sida et les Infections Sexuellement Transmissibles du Togo (PNLS-IST) is implementing IPT for PLHIV. The main objective of this study was to determine the factors associated with noncompletion of IPT during the pilot phase initiated before national scale-up.


Verbal approval was taken from the patients and/or their parents before IPT started and follow-up since this was not fully implemented in national politics. The PNLS-IST Ethics Committee approved the study protocol and enabled access to patient data. In addition, the forms have been made anonymous.

  Methods Top

Study design

This was a retrospective cohort study using routinely collected program data.

Study setting

Of the 20 clinics initially involved in the pilot study of IPT implementation in Lomé, Togo, half were not finally included (seven had not exploitable recorded data and three had enrolled less than ten patients). Three of the included clinics were administrated by nongovernmental organizations implicated in the HIV/AIDS response.

Enrollment and patient follow-up

According to an established protocol, every newly enrolled PLHIV in ART care was screened with the WHO's clinical algorithm to exclude active TB. Patients who present any of the four symptoms (cough, fever, night sweating, and weight loss) were considered as presumptive of TB, and sputum was collected for smear microscopy and Xpert® MTB/RIF (Cepheid, Sunnyvale, CA, USA) test. Those with negative smear test or with none of the above symptoms were considered as not having active TB. Patients with no active TB were enrolled and given isoniazid (at a dose of 5 mg/kg body weigh) after adherence counseling. Follow-up was carried out on a monthly basis. A chart was provided detailing the dates of the next appointments. At each visit, active TB risk, IPT adverse effects, and therapeutic compliance were assessed, and isoniazid was supplied to the patient. Treatment was considered complete when the patient finished the 6 months of isoniazid and returned to the appointment at the end of the program.


We opted for an exhaustive sampling, including all PLHIV newly enrolled in ART care who started treatment between June 1 and October 31, 2019, in the ten pilot sites who gave consent. This period was chosen to include only those patients who would normally have completed their follow-up 3 months before the start of data collection. Cases were excluded when data, especially on outcome, were too incomplete.

Data collection and statistical analysis

Data collection was made by three interns (7th-year medical students) between August and November 2020. Demographic and clinical data were extracted from the ART care booklets. IPT data were abstracted from an enrollment registry and follow-up chart designed for this purpose in the study. The data collection was made into a standardized electronic form deployed with KoboToolbox.

Data cleaning and statistical analysis were made with R 4.0.2 (R Core Team, Vienna, Austria). Categorical variables were presented as counts and proportions and the quantitative variables as means with standard deviations. A Chi-square or Fisher's exact test was used to compare proportions. To identify factors associated with noncompletion of treatment, a logistic regression model was developed. Following the univariate regression, variables with a sufficient association (P < 0.2) with noncompletion of treatment were included in an initial model for multivariate analysis. A stepwise downward selection process using Akaike Information Criterion was used to improve the model.

  Results Top

According to a status report from the centers, of the 440 patients who started IPT during the period, we retrieved records from 308 and excluded seven cases for incomplete data. This left 301 patients (68.4%) in the final sample.

The mean age was 39.8 ± 13.2 years, with 65.1% of patients being female. The follow-up was handled in a public center for 72.8% of patients. At enrollment, 92.7% of PLHIV had a performance status of 0 or 1. Of the 115 patients with CD4+ T-cell counts at enrollment, 35.7% had <200 cells/μL and 22.6% had more than 500 cells/μL. All the included patients were under ART. [Table 1] presents demographic and clinical characteristics relative to HIV infection at enrollment.
Table 1: Demographic and characteristics relative to HIV infection at enrollment

Click here to view

Isoniazid-preventive therapy initiation, follow-up, and outcome

At the clinical screening before isoniazid administration, 6.3% of the patients presented with at least one sign suggestive of TB. The most frequent sign was cough (5.7%) followed by fever (1.7%) and weight loss (0.6%). These two signs were associated in two cases. During subsequent visits, TB was suspected in 3.8% of PLHIV and confirmed after sputum examination in one patient only.

Adverse drug reactions attributable to isoniazid were reported in 5.2% of cases (9 of 174 valid records).

The completion rate was 42.9%. The main reasons for noncompletion were stockouts (45.9%) and lost to follow-up (37.8%), as shown in [Table 2]. Drug stockouts occurred in three public centers where 54.5% of enrolled PLHIV were being followed. After excluding these centers, the completion rate was 73.7%. The patient presented none of the signs included in the clinical algorithm. TB was suspected because of persistent asthenia. Bascilloscopy returned negative. Diagnosis was made by a positive sputum polymerase chain reaction. Mycobacterium tuberculosis isolated was rifampicin sensitive. Another patient discontinued treatment because of poorly tolerated side effects, but the nature of the side effects was not reported.
Table 2: Isoniazid preventive treatment outcome and reasons for noncompletion

Click here to view

Risk factors associated with noncompletion of isoniazid-prevention therapy

The associated risk factors in the final model after excluding less powerful and confounding variables were existence of drug stockout, the type of administration of the center: public, and general condition at initiation performance status ≥2 [Table 3].
Table 3: Risk factors associated with treatment noncompletion

Click here to view

  Discussion Top

Our study encountered difficulties inherent to its retrospective nature, especially missing data. Despite these shortcomings, the multicenter design of this study provides an assessment of IPT implementation in different settings and contexts.

The completion rate in our study is lower than the 80.6% reported by Takarinda et al.[8] when isoniazid was introduced in patients already on treatment, or the 91.5% reported in children living with HIV in Kenya,[9] but higher than the completion rate obtained in Cameroon (44.3%) and Namibia (45.7%).[10],[11] The much better results of Masini et al.[9] in Kenya could be explained by the exclusion of children with a history of nonadherence to ART and the existence of a system of active search for those lost to follow-up.

In the Takarinda cohort, the reasons for noncompletion were “lost to follow-up” in 60% and uninformed in 26.9% of patients. Discontinuations due to side effects and stockouts were marginal (7.2 and 4.5%, respectively). Abandoning treatment and “lost to follow-up” were the main causes of noncompletion also reported by Simo et al.[10] The high proportion of discontinuations due to stockouts found in our cohort is related to the conduct adopted in the three centers where these stockouts occurred: definitive interruption of all patients on treatment at the time of the stockout and no restarting after isoniazid had been resupplied.

In other settings, Zimbabwe, for example, stockouts had only a marginal effect on treatment completion, probably because treatment was resumed in patients once the anti-TB drug became available again. The risk factors found were rather the existence of an already ongoing ART, the delivery of 2 months of isoniazid at the first visit and the irregularity of the patient during the previous year of follow-up.[8]

However, other organizational factors not explored here could also explain unsatisfactory outcomes. In a Namibian study by Roscoe et al., although 72.2% of health workers surveyed described their clinical performance as very good, many gaps were found in their training, their understanding and practice of initiation timing IPT, and even the screening algorithm. Health workers often reject responsibility for difficulties on patients and downplaying challenges such as staff shortages and medication stockouts.[11]

Since isoniazid stocks are managed by the National TB Program, better collaboration with the NTP-IST would allow better coordination at the central level. Some authors propose that, at the local level, sufficient stocks should be kept for the entire course of treatment at the time of initiation and that the centers should be able to request “de-stocking” through simplified procedures.[3]

Even after the initial training, regular updatings and supervision of prescribers are essential for the proper execution of the established protocol.[3] Patient acceptability and adherence can be improved by prescribers' delivery of positive messages, patient education, thorough motivational counseling, and treatment support.[3]

  Conclusions Top

Good results obtained by some centers prove that the implementation of the intervention is indeed possible, as long as good management of drug stocks and better follow-up of patients are ensured and treatment is initiated at an early disease stage.

Study limitations

The limitations of this study are related to its retrospective nature, with the difficulty of missing data. The quality of agent training and active search mechanisms of the loss to follow-up patients has not been evaluated.

Financial support and sponsorship


Conflicts of interest

There are no conflicts of interest.

  References Top

Mirsaeidi M, Sadikot RT. Patients at high risk of tuberculosis recurrence. Int J Mycobacteriol 2018;7:1-6.  Back to cited text no. 1
[PUBMED]  [Full text]  
Joint United Nations Programme on HIV/AIDS-UNAIDS. UNAIDS Data 2020. Geneva: UNAIDS; 2020. p. 436.  Back to cited text no. 2
Pathmanathan I, Ahmedov S, Pevzner E, Anyalechi G, Modi S, Kirking H, et al. TB preventive therapy for people living with HIV: Key considerations for scale-up in resource-limited settings. Int J Tuberc Lung Dis 2018;22:596-605.  Back to cited text no. 3
Badje A, Moh R, Gabillard D, Guéhi C, Kabran M, Ntakpé JB, et al. Effect of isoniazid preventive therapy on risk of death in west African, HIV-infected adults with high CD4 cell counts: Long-term follow-up of the Temprano ANRS 12136 trial. Lancet Glob Health 2017;5:e1080-9.  Back to cited text no. 4
Moh R, Badjé A, N'takpé JB, Kouamé GM, Gabillard D, Ouassa T, et al. Screening for active tuberculosis before isoniazid preventive therapy among HIV-infected West African adults. Int J Tuberc Lung Dis 2017;21:1237-44.  Back to cited text no. 5
Jena L, Harinath BC. Efficacy and safety of isoniazid preventive therapy in light of increasing multi-drug resistance in tuberculosis. Int J Mycobacteriol 2015;4:354-5.  Back to cited text no. 6
  [Full text]  
Grace SG. Barriers to the implementation of isoniazid preventive therapy for tuberculosis in children in endemic settings: A review. J Paediatr Child Health 2019;55:278-84.  Back to cited text no. 7
Takarinda KC, Choto RC, Harries AD, Mutasa-Apollo T, Chakanyuka-Musanhu C. Routine implementation of isoniazid preventive therapy in HIV-infected patients in seven pilot sites in Zimbabwe. Public Health Action 2017;7:55-60.  Back to cited text no. 8
Masini EO, Sitienei J, Weyeinga H. Outcomes of isoniazid prophylaxis among HIV-infected children attending routine HIV care in Kenya. Public Health Action 2013;3:204-8.  Back to cited text no. 9
Simo L, Etoundi Mballa GA, Pefura Yone EW, Kuaban C. Rev Mal Respir 2015;32:A217.  Back to cited text no. 10
Roscoe C, Lockhart C, de Klerk M, Baughman A, Agolory S, Gawanab M, et al. Evaluation of the uptake of tuberculosis preventative therapy for people living with HIV in Namibia: A multiple methods analysis. BMC Public Health 2020;20:1838.  Back to cited text no. 11


  [Table 1], [Table 2], [Table 3]


    Similar in PUBMED
   Search Pubmed for
   Search in Google Scholar for
 Related articles
    Access Statistics
    Email Alert *
    Add to My List *
* Registration required (free)  

  In this article
Article Tables

 Article Access Statistics
    PDF Downloaded25    
    Comments [Add]    

Recommend this journal