Acceptability of multidrug-resistant tuberculosis preventive treatment: a retrospective survey

Extract

Rifampicin- or multidrug-resistant tuberculosis (MDR-TB) is infectious, difficult to treat, and causes high rates of morbidity and mortality [1]. Therefore, preventing MDR-TB and reducing the risk of secondary MDR-TB cases is highly desirable from both clinical and public health perspectives.

Shareable abstract

We retrospectively surveyed patient experiences and acceptability of a 6-month levofloxacin multidrug-resistant tuberculosis preventive treatment regimen. While the regimen was broadly acceptable, most respondents would not take it again. https://bit.ly/3I3O4fl

To the Editor:

Rifampicin- or multidrug-resistant tuberculosis (MDR-TB) is infectious, difficult to treat, and causes high rates of morbidity and mortality [1]. Therefore, preventing MDR-TB and reducing the risk of secondary MDR-TB cases is highly desirable from both clinical and public health perspectives.

The World Health Organization (WHO) has recently issued a strong recommendation for a 6-month levofloxacin regimen for MDR-TB preventive treatment (MDR-TPT) among exposed contacts. However, these guidelines identified a persistent knowledge gap regarding the preferences of affected communities for different regimens [2, 3]. Understanding patient perspectives is crucial to successful MDR-TPT scale-up and effective prevention [4], since the strategy requires otherwise asymptomatic individuals to accept a prolonged course of drug treatment. This should also guide the design of better patient-centred care that facilitates MDR-TPT completion [5]. Therefore, we aimed to explore preferences and experiences among participants of a levofloxacin MDR-TPT trial [6].

A retrospective survey was nested within the double-blind, multi-site randomised controlled VQUIN trial [6]. VQUIN compared the efficacy of a 6-month regimen of daily oral levofloxacin versus indistinguishable placebo for MDR-TB prevention in household contacts of people with bacteriologically-confirmed MDR-TB (“index cases”) treated by the Vietnam National TB Programme. Household members living with an index case were invited to be screened for TB infection, unless they were HIV-positive or malnourished; individuals with positive tuberculin skin result (≥10 mm induration) were randomised. Levofloxacin daily dosing ranged from 10–15 mg·kg⁻¹ in adults and from 15–20 mg·kg⁻¹ for children, with a maximum dose in both groups of 750 mg. Per protocol treatment completion was defined as taking at least 80% of doses within 270 days of starting therapy. Treatment adherence was assessed every four weeks at prescription refill via pill count. Ultimately, 2041 household contacts (1981 adults ≥15 years, 60 children <15 years) from 10 urban and rural provinces in Vietnam were enrolled between March 2016 and August 2019 and followed for 30 months.

To evaluate participant perspectives of their MDR-TPT, a subset of household contacts were contacted by phone in September and October 2023, after all participants had completed follow-up. Trial participants were eligible if they i) had taken at least one dose of levofloxacin, ii) consented to further follow-up contact, and iii) had up-to-date contact information. A short survey regarding levofloxacin's palatability and acceptability was completed (and can be viewed at https://osf.io/4rdjh/?view_only=02310f2c3f254d89aafd3472389cbffa), based upon a questionnaire used in a concurrent MDR-TPT trial [7].

Following an initial period of pilot evaluation, we contacted all participants who were enrolled in 2019, the final year of VQUIN enrolment. Verbal consent was obtained from all participants before administering the short survey. We performed descriptive analyses, with proportions compared using two-sided binary tests of proportions and categorical variables compared using chi-squared tests.

Among 292 eligible individuals, 251 (86.0%) household contacts were contacted, 11 of whom declined to consent, for a response rate of 82.2% (240 out of 292). The median time between trial end and survey completion was 1.89 years (interquartile range (IQR):1.74–2.20). Although we contacted a slightly higher proportion of participants in the placebo arm (128 out of 240, 53.3%) versus the levofloxacin arm (112 out of 240, 46.7%), this difference was not statistically significant (p=0.171).

Most (158 of 240, 65.8%) survey participants were women, with median age at enrolment of 35 years (IQR 21–46). Respondents were mainly from the southern (133 of 240, 55.4%) and northern (85 of 240, 35.4%) regions, with a minority from the central region (22 of 240, 9.2%). No participants reported previous MDR-TPT use, although 12 (5.0%) reported a history of TB. Household contacts were mostly close relatives of the index case – children (65 of 240, 27.1%); spouses (59 of 240, 24.7%); parents (31 of 240, 12.9%); siblings (28 of 240, 11.7%). Most (136 of 240, 56.7%) reported sleeping in the same building as the index case, with many sharing a room or bed (64 of 240, 26.6%). Treatment adherence was high: 84.6% (203 of 240) completed MDR-TPT as per protocol.

Most participants (180 of 240, 75.0%) who received MDR-TPT were “neutral” regarding its taste (neither “liked” nor “disliked” it), though 46 (19.1%) could not recall. A majority (160 of 240, 66.7%) found the tablet size acceptable (“ideal”, “fine”, “neutral”), though 39 (16.3%) participants said tablets were too large. Importantly, preferences regarding tablet taste (p=0.116) and size (p=0.681) did not differ by treatment arm. When asked about treatment frequency, only a small proportion (29, 12.1%) of participants thought daily dosing was “too often”, while a large majority (174 of 240, 72.5%) considered dosing acceptable (“fine”, “neutral”); preferences did not differ by treatment arm (p=0.224). However, a sizable minority found the 6-month treatment duration “too long” (67 of 240, 27.9%), with preferences not differing between arms (p=0.242). When examining responses to the acceptability survey by treatment adherence, no significant differences were noted (table 1).

TABLE 1.

Acceptability of 6-month levofloxacin regimen or placebo for the prevention of multi-drug resistant tuberculosis among 240 household contacts of people with multi-drug resistant tuberculosis in Vietnam

Acceptability question	Response	Participants completing treated per protocol (n=203)	Participants not completing treatment per protocol (n=37)	p-value
How do you feel about the process of taking the medication doses?	Easy/Very easy	58 (29%)	7 (19%)	0.2133
	Neutral	113 (56%)	21 (57%)
	Difficult/Very difficult	12 (6%)	5 (14%)
	Cannot recall	20 (10%)	4 (11%)
How did you feel about the quantity of medication you had to take each day?	Liked/Liked very much	8 (4%)	0 (0%)	0.2133
	Neutral	156 (77%)	29 (78%)
	Disliked/Disliked very much	7 (3%)	2 (5%)
	Cannot recall	32 (16%)	6 (16%)
How did you feel about the frequency of taking medications?	Fine/Ideal frequency	63 (31%)	10 (27%)	0.2381
	Neutral	85 (42%)	17 (46%)
	Too frequent/Much too frequent	24 (11%)	5 (14%)
	Cannot recall	31 (15%)	5 (14%)
How did you feel about the duration of treatment?	Fine/Ideal duration	33 (16%)	1 (3%)	0.2133
	Neutral	83 (41%)	20 (54%)
	Much too long/Too long	57 (28%)	10 (27%)
	Cannot recall	30 (15%)	6 (8%)

Results stratified by per protocol treatment completion.

A minority of participants stated they would take this regimen again (33 of 112 (29.5%) in the active arm, 32 of 128 (25.0%) in the placebo arm; p=0.528). Given the high number who were unprepared to re-take MDR-TPT, no difference (p=0.8628) could be demonstrated between those who experienced an adverse event of any grade (23 out of 81, 28.4%) versus none (42 out of 159, 26.4%); or a serious adverse event (grade ≥3) (5 of 22, 22.3%) compared to absent or lower grade events (60 of 218, 27.5%; p=0.818); or among older (≥55 years) (9 of 29, 31.0%) versus younger participants (56 of 211, 26.5%; p=0.774).

Slightly higher proportions of participants in each arm responded “yes” when asked whether they would recommend the treatment to other people in their situation (46 of 112 (41.1%) in treatment versus 49 of 128 (38.3%) in placebo arms, p=0.758). There were no significant differences in recommending this treatment based on presence of adverse event (p=0.663), serious adverse event (p=0.580), or older age (p=1.0).

In summary, no significant differences were observed in experiences and acceptability of a 6-month regimen of MDR-TB TPT by treatment arm or adherence. Interestingly, less than one third of participants said they would take the treatment again, although the finding that approximately 40% would recommend MDR-TPT to others suggests some appreciation of its protective value. Despite these findings, per protocol treatment adherence was high. This may be partially attributable to trial conditions (e.g., monthly pill counts, regular clinical exams, phone calls between medication refills) and social desirability factors (e.g., avoiding appearing “difficult” to study doctors). Further qualitative work would allow deeper understanding of the drivers behind the low willingness to repeat MDR-TPT. Survey findings may provide some indication, including that almost 30% of participants found the regimen to be “too long”.

The most commonly reported response to each question in the survey was “neutral”, which indicates that while study participants did not feel overwhelmingly favourably towards the regimen, neither did they feel especially negative. This may imply that preventative therapy was broadly acceptable, as the experience does not stand out in participants’ memories. However, as this was a retrospective survey, poor recall was certainly a factor in respondents’ answers and may have contributed to bias towards the null; consequently, results should be interpreted cautiously.

Despite its limitations, our study addresses an important knowledge gap regarding MDR-TPT acceptability in a high-prevalence setting, although generalisability is uncertain. Our findings differ from patient preferences research performed within a concurrent levofloxacin MDR-TPT trial, TB-CHAMP, done in South Africa [7]. TB-CHAMP conducted in-depth qualitative work, including interviews and participatory research activities, to understand children's and adolescents’ experiences of this regimen. There, children found levofloxacin to be broadly acceptable, as young people adapted to the medicine's taste over time. Key barriers for parents/guardians in helping their children complete six months of treatment were identified [8]. Other studies examining fluoroquinolone-based MDR-TPT regimens among children have found it to be tolerable and safe, with high rates of completion observed [9, 10]. In these cohorts [9, 10], rates of adverse events, particularly serious adverse events, were substantially lower than in VQUIN [6], which may contribute to the observed differences in MDR-TPT acceptability between age groups. Identifying barriers that explain the low rate of recommendation specifically among adults taking MDR-TPT themselves would be important for scaling up MDR-TB preventive therapy.

In future, studies of MDR-TPT acceptability should be embedded within trials to complement adverse event findings. This would minimise the issue of poor recall and allow survey responses greater validity and interpretability. Understanding patient experiences and preferences is important to guide the scale-up of new regimens by facilitating adherence, augmenting the public health benefit of preventive therapy.