The health-related quality of life of drug-resistant tuberculosis patients receiving treatment in Botswana Authors: Kebayaone P Gare, Makamu Sebakeng, Mooketsi Molefi Published: 2024 ABSTRACT AbstractBackgroundDrug-resistant tuberculosis (DR-TB) poses a significant global health challenge and requires treatment with potentially toxic second-line anti-TB drugs. Assessing the health-related quality of life (HRQoL) of DR-TB patients is crucial, given the impact of disease and treatment on their well-being. This study aimed to evaluate HRQoL among DR-TB patients undergoing treatment in Botswana and identify predictors of variability during the treatment period.MethodsA cross-sectional study involving almost all the eligible clients in the DR-TB treatment database was conducted at four of the six DR-TB sites in Botswana. The SF-36v2-based questionnaire was administered to all patients receiving treatment between March 2022 and June 2023. Data analysis was performed with QualityMetric Inc., LLC PRO CoRE scoring software and Stata 13.1 for the HRQoL scoring and regression analyses, respectively. A score ≤ 47 on the norm-based scoring (NBS) indicated poor HRQoL. Shorter, all-oral DR-TB regimens were introduced since 2018 but Botswana had not yet fully implemented those in the years 2022/2023. Patients had to go on treatment for 18–24 months during the time of the study.ResultsSeventy-two of the 81 eligible participants were enrolled. Participants on treatment for 13–24 months exhibited better HRQoL scores (53.3 ± 8.4) than those in the initial (0–12 months, 46.9 ± 10.8) and latter phases of treatment (> 24 months, 44.3 ± 10.1) for the Physical Component Summary (PCS) even though it was not statistically significant (p = 0.0996). The mental component summary (MCS) scores were 41.6 ± 11.3, 51.6 ± 7.8, and 41.3 ± 10.7 for 0–12, 13–24, and > 24 months, respectively, with significant differences observed for the MCS (p = 0.0097). Multivariate analysis identified renal impairment as a predictor of PCS variability, while alcohol consumption, prior TB treatment, and lung cavity on chest X-ray imaging predicted MCS variability.ConclusionDR-TB patients in Botswana demonstrated comparable or improved HRQoL (> 47 NBS) in their second year (13–24 months) of treatment, contrasting with poorer HRQoL scores in the initial and final years for both the PCS and MCS. The findings underscore the necessity for tailored psychosocial support, advocated for its integration into the Botswana National TB Program as a pilot initiative before widespread implementation.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12879-024-10258-7.Keywords: Health-related quality of life, HRQoL, Drug-resistant tuberculosis, DR-TB, SF-36v2, Botswana, Predictors Background Drug-resistant tuberculosis (DR-TB) is caused by a Mycobacterium tuberculosis (MTB) strain that is resistant to at least one of the two most potent TB drugs (isoniazid or rifampin) [1, 2]. DR-TB - a term that defines five categories (of DR-TB) - is a chronic, devastating disease that requires treatment with potentially toxic and less efficacious second-line anti-TB drugs [2, 3]. From the Global TB Report 2023, the World Health Organization (WHO) uses five categories to classify cases of DR-TB: isoniazid-resistant TB (Hr-TB); rifampicin-resistant TB (RR-TB); multi-drug resistant TB (MDR-TB); extensively drug-resistant TB (XDR-TB); and pre-XDR-TB. Pre-XDR-TB is TB that is resistant to rifampicin and any fluoroquinolone (a class of second-line anti-TB drug) [3]. XDR-TB is TB that is resistant to rifampicin, plus any fluoroquinolone, plus at least one of either bedaquiline or linezolid [3]. TB has a substantial adverse impact on patients’ quality of life (QoL) [4], and DR-TB has been described by Isaakidis as “the worst of the worst illnesses”, and its treatment is “worse than the disease itself” [2, 5]. The participants of a qualitative study conducted in India described the intensive phase of DR-TB treatment as worse than the disease itself [6]. HRQoL is defined as “the extent to which patient’s subjective perception of physical, mental and social well-being are affected on a day-to-day basis by a disease and its treatment(s)” [2, 5]. The need to measure HRQoL has become important due to the broadening concept of measuring health status beyond conventional indicators, such as mortality and morbidity [2, 7]. Globally DR-TB treatment success was approximately 63% as of 2023, and it was reported to be 71% for the WHO African region, with high rates of death and loss to follow-up [3]. In 2023, DR-TB treatment success rate for Botswana had been ranging above 70% in the previous five years with a 77% success rate reported in 2020 [8]. In Botswana, increasing trends in resistance to first-line anti-TB drugs have been observed among new patients [9], a situation that might lead to an increase in new DR-TB patients. Botswana has reported an average of 100 patients on second-line TB drugs since the establishment of the first DR-TB centres in 2006 [8]. Adverse events (AEs) are common during treatment for DR-TB, and little is known about the HRQoL of patients receiving treatment for DR-TB or the effect of AEs on HRQoL [4]. The treatment for DR-TB takes much longer, is more complex, and is often more toxic than the treatment used for DS-TB [4, 6]. AEs range from those that temporarily reduce QoL to those that cause long-term disability and those that are potentially life threatening [6]. In Botswana, treatment duration ranged from 18 to 24 months and sometimes 18–24 months from the time of culture conversion [10]. In rare instances, near treatment completion, sputum cultures could revert to positive, prolonging treatment beyond 24 months. Botswana used all-oral DR-TB regimens. However, the country has since transitioned to the shorter, all-oral regimens recommended by the WHO. One may anticipate a high prevalence of residual impairment in the HRQoL of these patients, given the lingering and destructive nature of DR-TB and lengthy therapy with toxic drugs for 2 years [2, 6]. The purpose of this study was to evaluate the HRQoL among DR-TB patients who were receiving treatment in Botswana and to identify predictors of variability during the treatment period, thereby providing data concerning the effect of DR-TB treatment on the HRQoL of patients. Methods Study design and settings This was a cross-sectional study conducted at four of the six healthcare facilities in Botswana, which offered DR-TB treatment during the study period. These facilities were Princess Marina Hospital (PMH) in Gaborone; Nyangabgwe Referral Hospital (NRH) in Francistown; Sekgoma Memorial Hospital (SMH) in Serowe; and Ghanzi Primary Hospital (GPH) in Ghanzi. Patients who were confirmed to have DR-TB were treated for TB either at the DR-TB site or at a primary healthcare facility (clinic or health post). All patients visited the DR-TB treatment site every month for routine investigations, consultation and review. Study population The study was conducted on patients who were receiving DR-TB treatment anytime from March 1st, 2022, to June 30th, 2023. Inclusion criteria were those who: (i) had been on DR-TB treatment for at least 4 weeks, (ii) were 18 years of age or older, (iii) comprehended English/ Setswana, and (iv) were followed up at or being hospitalized in the DR-TB site. Excluded were those who were on DR-TB treatment but were mentally disabled or critically ill in such a way that they won’t be able to participate in the data collection process. The resistance patterns for DR-TB patients in this population have been defined under Table 2 footnotes. Table 2Clinical characteristics of the participants (n = 72)Variablen (%) Registration Category New DR-TB69 (95.8)Treatment after loss to follow-up2 (2.8)Other: MDR contact1 (1.4) Resistance Pattern RR-TB1 (1.4)MDR-suspect*4 (5.6)MDR-TB66 (91.7)Pre XDR-TB1 (1.4) History of Injectable use No72 (100.0) Number of months since Current treatment start 0–1253 (73.6)13–2414 (19.4)More than 245 (6.9) Treated for TB before Yes38 (52.8) Drug-susceptible37 (51.4) Drug-resistant1 (1.4)Length of previous treatment 6 months35 (48.6) 7–12 months3 (4.2)No34 (47.2)Length of sickness prior to DR-TB diagnosis (months)0–662 (86.1)7–121 (1.4)12–186 (8.3)More than 183 (4.2) HIV Status HIV negative32 (44.4)HIV positive on ART38 (52.8)HIV status unknown2 (2.8) Comorbidities None61 (84.7)Diabetes2 (2.8)Hypertension8 (11.1)Asthma1 (1.4) Adverse event within the last 4 weeks Yes47 (65.3)No25 (34.7) Current clinical outcome Hearing loss Yes3 (4.2)No64 (88.9)Some5 (6.9) Anaemia Yes4 (5.6)No68 (94.4) Renal impairment Yes4 (5.6)No68 (94.4) Liver Impairment Yes5 (6.9)No67 (93.1) Lung Cavity Yes14 (19.4)No54 (75.0)X-ray findings not available4 (5.6)MDR-TB- Multidrug-resistant tuberculosis, DR-TB- Drug-resistant tuberculosis, RR TB- Rifampicin-resistant tuberculosis, XDR- Extensively drug-resistant tuberculosis, X-ray- Chest radiography film, ART- Antiretroviral therapy. MDR-suspect*- the resistance patterns of these patients were undetermined, Adverse event (AE) within the last 4 weeks – establishment of whether an AE occurred between the last healthcare facility visit and the visit at the time of the study interview since DR-TB patients receiving treatment came for review every month at the healthcare facility Clinical characteristics of the participants (n = 72) Number of months since Current treatment start MDR-TB- Multidrug-resistant tuberculosis, DR-TB- Drug-resistant tuberculosis, RR TB- Rifampicin-resistant tuberculosis, XDR- Extensively drug-resistant tuberculosis, X-ray- Chest radiography film, ART- Antiretroviral therapy. MDR-suspect*- the resistance patterns of these patients were undetermined, Adverse event (AE) within the last 4 weeks – establishment of whether an AE occurred between the last healthcare facility visit and the visit at the time of the study interview since DR-TB patients receiving treatment came for review every month at the healthcare facility Sampling and sample size determination According to unpublished data from Ministry of Health at time of the study, the total number of patients who were receiving DR-TB treatment nationwide was more than 100. The sample size was initially calculated manually using the formula: sample size= [Z1-α/22 SD2]/d2. Statulator [11], an online calculator, was also used to compute the sample size with the confidence interval (CI), standard deviation (SD) and precision or margin of error (d) set as follows: CI = 95%, Z = 1.96; SD= +/-10; and d = 2.5. Since no similar local studies were found, the authors decided these statistics from literature review and the guidance of a reference article by Charan J et al. [12]. The initial calculated sample size was 65 participants and was adjusted for a nonresponse rate of 10% to 72 participants. However, a decision to perform exhaustive sampling was reached when it was found that the number of patients at the selected sites was less than expected as per preliminary information from unpublished data. At the end of the data collection period, 72 participants from all the four research sites had successfully completed the questionnaire, while a total of 9 eligible participants did not participate in the study. This is summarised by Fig. 1 below. The enrollment of study participants at the four DR-TB sites. *Other reasons- participant felt that all questions were the same, ultimately requesting that responses be excluded from the analysis Piloting and data Collection Data collection was carried out from March 22nd, 2022 to June 30th, 2023. This process started with piloting the data collection tool for two weeks with five participants, from PMH DR-TB clinic in Gaborone, whose data were not included in the actual study data analysis. Through the pilot results, anticipated challenges were addressed and necessary adjustments were made in order to make data collection effective. During the study period, eligible participants were asked to self-administer the SF-36v2-based questionnaire at any point in time during their treatment as long as they had been in treatment for a minimum of 4 weeks. Participants provided informed consent prior to data collection. The trained research assistants (RAs), or at times the researcher, administered the questionnaire face-to-face to participants who were unable to write. After answering the questionnaire, it was checked if the participants omitted any questions, and if they did, it was established why; then, they were encouraged to complete it. Patient records were also reviewed for all participants, to capture variables that could not be provided by the participants. Data quality control The RAs were trained in the data collection process, and the completeness of the questionnaire was checked before the closure of each interview. All entries were crosschecked after they were entered into a Microsoft Excel spreadsheet. The researcher regularly supervised the RAs by visiting the research sites and through virtual calls. The health-related quality of life assessment questionnaire and its scoring Despite the abundance of other tools currently available to measure HRQoL in TB patients, the Short Form-36 (SF-36) health survey remains the most valid option [13–18] and has been validated in African populations [19]. Physical and psychometric evaluations of the SF-36v2 health survey provide evidence that it is a reliable and valid tool for detecting differences among groups defined by socio-demographic status and social factors [20]. The SF-36v2 tool has been utilized by some TB studies and has shown acceptable reliability and validity [2]. After thorough review of the literature and discussion with study supervisors, the SF-36v2 tool was selected for the study, adopted as it is and included in the study questionnaire. Two independent native Setswana speakers who also speak English translated the SF-36v2 tool from English to Setswana, and another similar pair back-translated it. The translations were then reconciled. Since no similar studies have been published in Botswana, the Setswana version of the SF-36v2 was made available to participants alongside the English version (as the primary tool). The SF-36v2 has eight scales that gauge eight domains of HRQoL: physical functioning (PF, 10 items), role physical (RP, 4 items), bodily pain (BP, 2 items), general health (GH, 5 items), vitality (VT, 4 items), social functioning (SF, 2 items), role-emotional (RE, 3 items), and mental health (MH, 5 items). The eight health domains are further summarized into two parts: the physical component summary (PCS) and the mental component summary (MCS). The standard norm-based scores (NBSs) of eight health domains and component summary measures were obtained using standard scoring algorithms (USA weights). Standard scoring algorithms and standard component summary measures were used based on the recommendations of developers (QualityMetric) and other similar studies from sub-Saharan Africa (SSA). The QualityMetric scoring software retroactive licence The QualityMetric Incorporated, LLC, USA, granted this study a retroactive licence for QualityMetric (QM) PRO CoRE scoring software and the SF-36v2 United Kingdom (English) Standard tool. The licence was granted after a request to QualityMetric Incorporated, LLC. Recommendations that were made by the QualityMetric Form Review and Science Team were implemented before scoring the HRQoL data with the QM PRO CoRE scoring software. The recommendations that were made were to harmonize the SF-36v2 UK (English) tool with the SF-36v2 tool adapted to the questionnaire for data collection since there were few differences in terms of the arrangement of the items. Data processing and analysis The questionnaire data were entered into a Microsoft Excel spreadsheet, which was password protected, and variables were ordered as they appeared on the study questionnaire. Missing data estimation (MDE) was applied for question 11a of the SF-36v2 tool by treating the data in that column as missing completely at random (MCAR) to provide valid estimates of the available data. This was because the wording in the SF-36v2 used in the interviews was slightly different from that in the United Kingdom (English) SF-36v2 provided by QualityMetric Incorporated, LLC. The questions read “I seem to get a little sick easier than other people” in the former and “I seem to get ill more easily than other people” in the latter. For this reason, QualityMetric LLC recommended the use of MDE for question 11a. To execute MDE for the 11a column, three datasets were created, and regression was run to determine whether there were statistically significant differences between the datasets. For the QM PRO CoRE software, Microsoft Excel spreadsheet columns were formatted to match the requirements of the software. The spreadsheet was then imported into the software to score the data to obtain the HRQoL data. Further analysis of the data was performed using Stata 13.1. Categorical data were summarized as frequencies (expressed as percentages). The treatment experiences of the participants and the groups were categorized based on the number of months they were on treatment (i.e., between 0 and 12, 13–24 or more than 24 months). The three groups and their mean scores were compared by one-way ANOVA. To determine the predictors of variability in HRQoL scores, regression analysis was performed. This analysis started with bivariate regression analysis to recruit variables with p values < 0.25 and those of clinical significance into the parsimonious model for multivariate regression analysis. For all the study variables and analyses, the confidence intervals were set at 95%, and a p value < 0.05 was considered to indicate statistical significance. Results Participants’ sociodemographic characteristics The sociodemographic characteristics of the participants are displayed in Table 1. Seventy-two (72) participants participated in the study. This was the optimal sample size from the beginning, even though sampling ended up being exhaustive because COVID-19 affected the TB notification rate. The mean age of the participants was 42.1 (± 14.6) years, and 51 (70.8%) of the participants were male. Twenty-four (33.3%) of the participants were aged 35–44 years, and they were the most common age group. Thirty-six (50.0%) of the participants resided in rural areas, and 55 (76.4%) of the participants were single (of marital status). Six (8.3%) participants were smokers, and 12 (16.7%) of the participants drank alcohol. Sixty-eight (94.4%) of the participants were Botswana citizens, and thirty-eight (52.8%) of the participants had attained secondary education. Forty-four (61.1%) of the participants were unemployed, with 43 (59.7%) receiving no monthly income. Table 1Sociodemographic characteristics of the participants (n = 72)Variablen (%) Gender Male51 (70.8)Female21 (29.2) Age (years) Mean (± SD), min- max42.1(± 14.6), 19–8818–2410 (13.9)25–3413 (18.1)35–4424 (33.3)45–5411 (15.3)55–648 (11.1)> 656 (8.3) Residence Rural36 (50.0)Semi-urban27 (37.5)Urban9 (12.5) Marital status Married17 (23.6)Single55 (76.4) Smoking Yes6 (8.3)No66 (91.7) Drinking alcohol Yes12 (16.7)No60 (83.3) Nationality Motswana68 (94.4)Other nationality4 (5.6) Education Level No formal education12 (16.7) Primary11 (15.3) Secondary38 (52.8) Tertiary11 (15.3) Employment Status Yes21 (29.2) No44 (61.1) Other*7 (9.7) Monthly income (BWP) & None43 (59.7) Less than 200011 (15.3) 2001–500011 (15.3) 5001–10,0004 (5.6) More than 10,0003 (4.2)&Exchange rate: 1 USD = 13.36 BWP (average for year 2023)USD- United States Dollar, BWP- Botswana PulaEmployment Status Other*: Total 7 (Self-employed- 5; Pensioner- 1; Unspecified- 1) Sociodemographic characteristics of the participants (n = 72) &Exchange rate: 1 USD = 13.36 BWP (average for year 2023) USD- United States Dollar, BWP- Botswana Pula Employment Status Other*: Total 7 (Self-employed- 5; Pensioner- 1; Unspecified- 1) Participants’ clinical characteristics The clinical characteristics of the participants are displayed in Table 2. Almost all (95.8%) participants were registered as having new DR-TB. MDR-TB constituted the majority (91.7%) of the resistance patterns. All (100.0%) participants had no history of TB injectable medication. Fifty-three (73.6%) of the participants were in the first year of treatment. Approximately half (52.8%) of the participants were treated for TB prior to the current treatment; among them, 51.4% were treated for DS-TB, while 1.4% was treated for DR-TB. Thirty-five (48.6%) of the participants’ previous TB treatment lasted for 6 months. Sixty-two (86.1%) of the participants were sick for a range of 0–6 months before being diagnosed with DR-TB. Approximately half (52.8%) of the participants were HIV positive on ART, and 84.7% of all the participants had no comorbidities. Forty-seven (65.3%) of the participants had experienced an adverse event within the last four weeks. Three (4.2%) participants had hearing loss, 5 (6.9%) had some hearing loss, 4 (5.6%) had anaemia, 4 (5.6%) had renal impairment, 5 (6.9%) had liver impairment, and 14 (19.4%) had lung cavity on chest radiography at the beginning of treatment. Resistance patterns (DR-TB categories) defined RR-TB (R only) – defined as resistance to rifampicin detected using genotypic or phenotypic methods with or without resistance to other first-line anti-TB drugs. MDR-TB (R and I) – defined as a form of TB disease caused by a strain of M. tuberculosis complex that is resistant to rifampicin and isoniazid. Pre XDR-TB – defined as TB that is resistant to rifampicin and any fluoroquinolone (a class of second-line anti-TB drug). MDR-suspect – defined as a Category II TB patient who is smear positive at the end of the fourth month of treatment or later. The HRQoL scores for different treatment groups of patients are displayed in Table 3 among the eight health domains (PF, RP, BP, GH, VT, SF, RE & MH) of the HRQoL, which are explained as scales. These are based on norm-based scoring (NBS) for the SF-36v2 standard. Table 3Health-Related Quality of Life (HRQoL) SF-36v2 scores using norm-based Scoring (NBS) for different treatment groups of patients (n = 72)ScaleTreatment →Mean (± SD)0–12 months13–24 monthsMore than 24 monthsPF45.4 (± 12.9)52.9 (± 9.1)45.7 (± 13.8)RP41.9 (± 12.5)50.5 (± 9.1)34.7 (± 9.1)BP42.5 (± 12.3)52.4 (± 10.7)39.2 (± 7.6)GH50.5 (± 10.5)55.6 (± 8.1)52.8 (± 7.0)VT50.9 (± 10.5)59.6 (± 7.9)50.2 (± 17.5)SF39.3 (± 13.0)47.0 (± 12.2)37.3 (± 12.3)RE38.2 (± 13.5)49.5 (± 9.8)35.3 (± 7.4)MH43.1 (± 10.8)52.9 (± 8.8)44.1 (± 13.4)PF- Physical functioning, RP- Role physical, BP- Bodily pain, GH- General health, VT- Vitality, SF- Social functioning, RE- Role emotional, MH- Mental health Health-Related Quality of Life (HRQoL) SF-36v2 scores using norm-based Scoring (NBS) for different treatment groups of patients (n = 72) PF- Physical functioning, RP- Role physical, BP- Bodily pain, GH- General health, VT- Vitality, SF- Social functioning, RE- Role emotional, MH- Mental health Table 4 shows the physical component summary (PCS) and mental component summary (MCS) scores. The data are presented as the means and standard deviations (SDs). Table 4Component summary scores for different treatment groups of patients (n = 72)ScaleTreatment→Mean (± SD)0–12 months13–24 monthsMore than 24 monthsPCS46.9 (± 10.8)53.2 (± 8.4)44.3 (± 10.1)MCS41.6 (± 11.3)51.6 (± 7.8)41.3 (± 10.7)SD- standard deviation, PCS- physical component summary, MCS- mental component summary Component summary scores for different treatment groups of patients (n = 72) SD- standard deviation, PCS- physical component summary, MCS- mental component summary Table 5 shows the results of one-way ANOVA comparing the mean scores for the three treatment groups. The degrees of freedom were taken as 2 because there were three groups. The F-statistic F(2,69) = 2.39, p = 0.0996 for PCS and F(2,69) = 4.96, p = 0.0097 for MCS, showing statistical significance for the latter. Table 5Mean variability of HRQoL scores among the treatment groups (n = 72)Treatment GroupMean ScoreDFF-statisticp-value PCS 0–12 months46.9 (± 10.8)22.390.099613–24 months53.2 (± 8.4)> 24 months44.3 (± 10.1) MCS 0–12 months41.6 (± 11.3)24.960.0097*13–24 months51.6 (± 7.8)> 24 months41.3 (± 10.7)DF, degrees of freedom; PCS, physical component summary; MCS, mental component summary; *p < 0.05Treatment groups: 0–12 months, 13–24 months, and > 24 months Mean variability of HRQoL scores among the treatment groups (n = 72) DF, degrees of freedom; PCS, physical component summary; MCS, mental component summary; *p < 0.05 Treatment groups: 0–12 months, 13–24 months, and > 24 months Multivariate linear regression analysis revealed that renal impairment (coefficient: -11.57, 95% CI: -21.81, -1.33) was statistically significant for PCS, as shown in Table 6. Drinking alcohol (coefficient: 7.75, 95% CI: 1.23, 14.27), being treated for TB before (coefficient: -6.46, 95% CI: -11.63, -1.28) and having a lung cavity on chest X-ray film (coefficient: 5.14, 95% CI: -0.27, 10.01) were statistically significant for MCS, as shown in Table 7. Table 6Multivariate Regression to determine predictors of variability among physical component Summary (PCS) HRQoL scores of participants (n = 72)Predictor variableCoefficient95% CIp valueGender3.09-2.48, 8.660.271Age-0.07-0.28, 0.140.529Drinking Alcohol4.80-1.45, 11.050.129Nationality5.82-4.26, 15.900.253Educational Level1.42-1.81, 4.650.383Resistance Pattern-1.46-6.64, 3.730.576Treated for TB Before4.09-0.59, 8.770.086Length of Sickness-1.99-4.90, 0.930.179Adverse Event-2.27-7.09, 2.550.349Hearing Loss-2.64-8.44, 3.150.365Anaemia8.42-1.33, 18.160.089Renal Impairment-11.57-21.81, -1.330.027*PCS- Physical component summary, MCS- Mental component summary, CI- Confidence interval, Number of months Rx- Number of months since the start of current treatment, Length of sickness- Length of sickness prior to DR-TB diagnosis, Adverse event- Adverse event within the last 4 weeks, CXR- Chest X-ray film, *p < 0.05. R-squared for regression model = 0.305 Multivariate Regression to determine predictors of variability among physical component Summary (PCS) HRQoL scores of participants (n = 72) PCS- Physical component summary, MCS- Mental component summary, CI- Confidence interval, Number of months Rx- Number of months since the start of current treatment, Length of sickness- Length of sickness prior to DR-TB diagnosis, Adverse event- Adverse event within the last 4 weeks, CXR- Chest X-ray film, *p < 0.05. R-squared for regression model = 0.305 Table 7Multivariate Regression to determine predictors of variability among Mental Component Summary (MCS) HRQoL scores of participants (n = 72)Predictor variableCoefficient95% CIp valueAge-0.20-0.42, 0.030.089Residence2.49-1.53, 6.500.220Marital Status2.60-4.18, 9.380.446Drinking Alcohol7.751.23, 14.270.021*Educational Level-2.39-5.86, 1.080.174Employment status2.29-0.57, 5.140.115Number of Months Rx3.81-0.33, 7.950.071Treated for TB Before-6.46-11.63, -1.280.015*Comorbidities1.44-4.22, 7.100.613Adverse Event-3.36-8.19, 1.460.168Hearing Loss-1.94-6.97, 3.090.442Renal Impairment-9.62-20.12, 0.880.072Lung Cavity on CXR5.140.27, 10.010.039*PCS- Physical component summary, MCS- Mental component summary, CI- Confidence interval, Number of months Rx- Number of months since the start of current treatment, Length of sickness- Length of sickness prior to DR-TB diagnosis, Adverse event- Adverse event within the last 4 weeks, CXR- Chest X-ray film, *p < 0.05. R-squared for regression model = 0.348 Multivariate Regression to determine predictors of variability among Mental Component Summary (MCS) HRQoL scores of participants (n = 72) PCS- Physical component summary, MCS- Mental component summary, CI- Confidence interval, Number of months Rx- Number of months since the start of current treatment, Length of sickness- Length of sickness prior to DR-TB diagnosis, Adverse event- Adverse event within the last 4 weeks, CXR- Chest X-ray film, *p < 0.05. R-squared for regression model = 0.348 Discussion We aimed to assess and estimate the HRQoL of DR-TB patients receiving treatment in Botswana, and we interviewed majority (almost all) of the DR-TB patients in the country. The WHO recommendations for shorter, all-oral regimens for DR-TB were made since 2018 [3] but Botswana had not yet fully implemented those in the years 2022/2023. DR-TB patients had to go for duration of 18–24 months while still awaiting the implementation of the new regimens. One retrospective cohort study examined severe adverse events (AEs) during second-line TB treatment in the context of high HIV co-infection in South Africa [21]. The study reported on the occurrence of AEs during a standardized DR-TB treatment regimen at two outpatients, decentralized, public-sector sites in Johannesburg, South Africa [21]. Overall, 578 reports of AEs were captured for 204 (35.3%) patients, and 110 (19.0%) patients had at least one severe AE reported. Patients with at least one AE experienced a median of 3 (IQR: 2–4) AEs per patient [21]. In contrast, two-thirds (65.3%) of the patients in our study experienced AEs compared to one-third (35.3%) of the patients in the Johannesburg, South Africa study. Our study also revealed that 47 (65.3%) of the patients had experienced between one and six AEs within the past 4 weeks to the time of the interview. These AEs included the following: nausea, vomiting, diarrhea, dizziness, severe headache, seizures, numbness, rash/itching skin and others, which were specified by the participants. The patterns of the intervals in our study showed that those with a treatment interval of 0–12 months had < 47 NBSs in most of the domains (PF = 45.4, RP = 41.9, BP = 42.5, GH = 50.5, VT = 50.9, SF = 39.3, RE = 38.2 and MH = 43.1), those with a treatment interval of 13–24 months had better scores (PF = 52.9, RP = 50.5, BP = 52.4, GH = 55.6, VT = 59.6, SF = 47.0, RE = 49.5 and MH = 52.9), and those with a > 24-month interval had decreased NBS points (PF = 45.7, RP = 34.7, BP = 39.2, GH = 52.8, VT = 50.2, SF = 37.3, RE = 35.3 and MH = 44.1). These findings are comparable to those of a prospective cohort study performed in Yemen to evaluate HRQoL during and after treatment and to identify the risk factors that are predictive of HRQoL score differences. At the beginning of treatment, the mean scores for all health domains were < 47 NBS points (PF = 40.7, RP = 16.1, BP = 21.6, GH = 28.3, VT = 14.55, SF = 25.9, RE = 13.7, and MH = 14.7) [22]. At the completion of treatment, all eight health domains increased compared to those at the beginning of treatment (PF = 59.3, RP = 31.1, BP = 40.9, GH = 48.5, VT = 30.5, SF = 46.6, RE = 26.6 & MH = 27.7), but a follow-up duration of 1 year after completing treatment showed decreased NBS points in all domains (PF = 51.5, RP = 30.6, BP = 39.1, GH = 47.8, VT = 30.2, SF = 43.7, RE = 26.4 & MH = 27.2) [22]. A common finding for both studies was that the scores were low at the beginning of treatment and increased towards the end of treatment, after which they decreased as the time in or after treatment progressed. Anecdotally, it can happen that towards treatment completion, one of the sputum cultures turns positive hence prolonging treatment for that patient beyond 24 months. It may seem like treatment is not working in such cases leading to lower HRQoL scores. One cross-sectional study among adult patients with laboratory-confirmed rifampicin-resistant TB receiving DR-TB treatment at a public-sector outpatient DR-TB clinic in Johannesburg, South Africa, between 02/2015 and 01/2018 showed that DR-TB had a substantial impact on patients’ QoL but that AEs during the early months of treatment may be responsible for reducing HRQoL even further [23]. Our study had similar findings because one-way ANOVA showed that patients who were receiving treatment for 13–24 months performed better than those who were receiving treatment for 0–12 months and those who were receiving treatment for > 24 months according to the NBS scores for both the PCS and MCS, however p-value (0.0996) was not statistically significant for the former. Those in the early months (0–12 months) of treatment had reduced HRQoL. A study in Eritrea had an objective to evaluate the HRQoL between RR/MDR-TB in comparison to DS-TB patients. It was a cross-sectional study conducted in RR/MDR-TB and DS-TB patients under treatment [24]. Data collected using the World Health Organization Quality of Life (WHOQOL-BREF) questionnaire were analyzed. RR/MDR-TB patients had statistically lower mean scores in all domains than DS-TB patients [24]. HRQOL was impaired in both groups, but RR/MDR-TB patients had a worse HRQoL [24]. Even though our study was not comparing DR-TB versus DS-TB, DR-TB patients in all groups in our study had relatively low to fair scores, which were around 47 NBS overall. However, there was statistically significant difference for MCS scores in our study. Another cross-sectional study was conducted over a period of 6 months at the DR-TB centre, of a tertiary care centre in the eastern Uttar Pradesh, India [25]. It aimed to find the impact of MDR-TB on QoL domains during the initiation of the MDR-TB treatment regimen [25]. Patients with age > 18 years diagnosed with MDR-TB were included in the study [25]. The WHOQOL-BREF questionnaire was used to assess the HRQoL of patients and data were analyzed using SPSS. The mean value of different domains of WHO QOL-BREF was low in MDR-TB patients, with social relation domain being the most affected [25]. Even though our study was using the SF-36v2 tool, participants showed lower or marginal (≤ 47 NBS) social functioning (SF) for all the studied groups. This reflects that SF was negatively affected despite it being studied at different time points as opposed to the Indian study (which studied social relation). This points out to that DR-TB affects social functioning or social relation negatively. One more study conducted in Nanjing, China investigated the current status of the QOL of DR-TB patients and analyzed the influencing factors [26]. The Chinese version of the WHOQOL-BREF questionnaire was used to investigate the QOL levels of patients with DR-TB, and a multiple linear regression model was used to analyze the QOL influencing factors [26]. The findings showed that in Nanjing, China, patients with younger age, higher education level, living in urban areas, high family per-capita monthly income, no adverse drug reactions, no comorbidities, and having caregivers have better quality of life [26]. The study concluded that future interventions to improve the QoL of patients with DR-TB could be tailored to a specific factor [26]. In relation to our study context, improvement of HRQoL of DR-TB patients could be tailored to those predictors, which were shown to be statistically significant in multivariate regression modeling. Limitations and strengths This study was not without limitations. We acknowledge that the study was prone to recall bias since participants had to recall some of the information from the past when responding to questions. Furthermore, 9 (12.5%) patients from the 4 sites we accessed did not participate, bringing the total of eligible patients to 81. Had we had more resources, adding 2 of the remaining DR-TB treatment sites could have led to the identification of more cases. This may or may not have led to different results. Moreover, since the study was cross-sectional in nature, the causal relationship between an explanatory variable and an outcome variable could not be established for all. Despite these limitations, our study has the noteworthy strength that, to our knowledge, it is the first HRQoL study focusing on DR-TB patients in Botswana to comprehensively assess their situation. The data were collected through exhaustive sampling, making them more representative despite the relatively small population of DR-TB patients in Botswana for the study period. Conclusion DR-TB patients on treatment in Botswana demonstrated comparable or improved HRQoL (≥ 47 NBS) in their second year (13–24 months) of treatment, contrasting with poorer HRQoL scores in the initial and final years for both PCS and MCS, with no statistical significance for PCS. The findings underscore the necessity for tailored psychosocial support and other forms of support, advocating for their integration into the Botswana National TB Programme (BNTP) as a pilot initiative before widespread implementation. Therefore, the BNTP should implement an intervention programme to enhance HRQoL at the early months and towards the end of treatment to avoid any further negative consequences of DR-TB and its treatment in patients during those periods. We recommend a longitudinal study in the future in order to establish any cause-effect relationships that may exist in HRQoL, as we are hopeful that this current study would have set ground for other studies locally. Electronic supplementary material Below is the link to the electronic supplementary material. Supplementary Material 1 Acknowledgements We would like to thank the participants, staff and management from all the DR-TB sites for their support during the study period. Similarly, we would like to thank the University of Botswana Faculty of Medicine and the Ministry of Health, Botswana. Our gratitude also goes to the following people for their role in data collection: Dineo Lekgari, Donna Tshegetsang, Bonani Modikwa, Wananani Lackson and Oteng Orelathile. Abbreviations Analysis of variance Drug-resistant tuberculosis Drug-susceptible tuberculosis Health-related quality of life Mental Component Summary Norm-based Scoring Physical Component Summary Patient-related Outcomes Quality of Life 36v2-Short Form 36-item version 2 tool Author contributions Conceptualization, KPG, MM, and MS; Methodology, KPG, MM, and MS; Software, KPG and MM; Formal Analysis, KPG, and MM; Investigation, KPG, MS, and MM; Data Curation, KPG, MM; Writing – Original Draft Preparation, KPG; Writing – Review & Editing, KPG, MS, and MM; Visualization, KPG, MS and MM; Supervision, MS, and MM; Project Administration, KPG; Funding Acquisition, KPG. Funding This research received funding from the Office of Research and Development (ORD), University of Botswana. Data availability The datasets for the current study are not publicly available in order to protect the privacy of participants but are available from the corresponding author upon reasonable request from credible individuals and institutions. Declarations Ethical approval and consent to participate All ethical clearance procedures were followed before the start of the study. Ethical approval was obtained from the University of Botswana Institutional Review Board (IRB) under reference number UBR/RES/IRB/BIO/GRAD/158 and the Ministry of Health Review Board under permit number HPDME: 13/18/1 of 01 December 2021, and permission to conduct the study on sites was granted by the PMH, NRH, SMH and GPH IRBs. All respondents were requested to sign a consent form and were informed that participation in the study is voluntary. All participants who successfully completed the study questionnaire were included in the analysis. All participants provided written informed consent before study participation. Clinical trial number: not applicable. Consent for publication Not applicable. Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Supplementary Materials Supplementary Material 1 Data Availability Statement The datasets for the current study are not publicly available in order to protect the privacy of participants but are available from the corresponding author upon reasonable request from credible individuals and institutions.