Article Text

Epidemiology and Control of diabetes - tuberculosis comorbidity in Eswatini: protocol for the prospective study of tuberculosis patients on predictive factors, treatment outcomes and patient management practices
  1. Victor Williams1,2,3,
  2. Alinda Vos1,4,
  3. Kennedy Otwombe3,5,
  4. Diederick E Grobbee1,
  5. Kerstin Klipstein-Grobusch1,3
  1. 1Julius Global Health, Julius Center for Health Sciences and Primary Care, University Medical Center Utrecht, Utrecht University, Utrecht, The Netherlands
  2. 2National Tuberculosis Control Program, Manzini, Eswatini
  3. 3Division of Epidemiology and Biostatistics, School of Public Health, Faculty of Health Sciences, University of the Witwatersrand, Johannesburg, South Africa
  4. 4Ezinthsha, Faculty of Health Sciences, University of the Witwatersrand, Johannesburg, South Africa
  5. 5Perinatal HIV Research Unit, Faculty of Health Sciences, University of the Witwatersrand, Johannesburg, South Africa
  1. Correspondence to Dr Victor Williams; victormw55{at}


Introduction Previous studies indicate people with diabetes mellitus (DM) may have varying treatment outcomes when receiving treatment for tuberculosis (TB) and that TB infection or its treatment may predispose them to develop an abnormal blood glucose or type 2 DM. This has implications for Eswatini which is a high TB burden country and with increasing cases of non-communicable diseases including DM. This study will describe the epidemiology of DM-TB comorbidity in a prospective cohort of patients receiving TB treatment and identify best practices for integration of care for non-communicable diseases into TB services in Eswatini.

Methods and analysis This study will employ a mixed-methods approach. Data from a prospective cohort of newly enrolled patients with TB at 12 health facilities from 1 June 2022 to 30 September 2022, and followed up to 30 April 2023, will be used. For the qualitative, key informants who provide TB services at the health facilities will be interviewed. Quantitative data from patients will be analysed descriptively and by tests of association and multivariate modelling. Key informant interviews from healthcare workers will be analysed using content analysis.

Ethics and dissemination This research has been approved by the Eswatini Health and Human Research Review Board and participant confidentiality will be maintained. COVID-19 safety measures to reduce the risk of infection or transmission by researchers and participants have been instituted. Key programmatic findings and how they can impact healthcare delivery and access will be presented to the specific programme in the Eswatini Ministry of Health and other relevant stakeholders.

  • diabetes & endocrinology
  • tuberculosis
  • COVID-19

This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See:

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Strengths and limitations of this study

  • The use of a prospective cohort design allows for the collection of more accurate and complete data.

  • Interview of key informants provides useful background to different factors which impact services access by patients with tuberculosis (TB).

  • Reduced sample size due to a reduced number of new patients with TB and short follow-up period.

  • Information on the incidence of diabetes mellitus (DM) or hyperglycaemia is limited due to a short follow-up period since DM may develop a few months after treatment.

  • Healthcare workers may provide limited information during the in-depth interview due to fear of reprimand by the health authorities.



The global pandemic caused by the novel coronavirus has affected all countries and territories of the world.1 2 High daily cases and mortality have been recorded in the USA, India and Brazil, closely followed by countries in Europe and Asia.3 4 Mortality data indicated a high mortality rate in patients with comorbidities such as diabetes mellitus (DM), cardiovascular and respiratory diseases, kidney and liver diseases, those recovering from transplants and the critically ill.2 4 5 This indicates that non-communicable diseases (NCDs) have a propensity to coexist and complicate other disease conditions, most times, negatively altering the prognosis. Thus, the development of an appropriate context-specific method of managing commonly occurring NCDs is vital in the context of infectious disease.

In the last two decades, tuberculosis (TB) and HIV infection gained attention from global leaders, healthcare workers, researchers and non-governmental organisations. This was due to their impact on the economy of high burden countries, the health of individuals and pressure on the health system. With concerted efforts from different stakeholders, the prevalence of these two diseases has been controlled in high-income countries while some low-income and middle-income countries are gradually achieving epidemic control with stable infrastructures for a sustained response.6 7

While all efforts concentrated on curtailing the impact of HIV/TB with visible results of its reduction globally, the NCDs, diverse with insidious onset, gradually increased and are now the highest cause of mortality globally.8 9 NCDs now account for about 71% of global mortality.10 This can partly be attributed to less-developed structures to combat NCDs with far less funding for NCD programmes compared with TB and HIV, especially in the low-income and middle-income countries which also have the highest incidence and prevalence of infectious diseases with high levels of poverty and social inequality.11 This neglect of NCDs has become evident as the countries with a high burden of infectious diseases now record high mortality from NCDs. This indicates that both conditions (NCDs and infectious diseases) coexist in the community with each disease acting as an enabler for the other.9 11 12 Major NCDs accounting for increased morbidity and mortality globally include cardiovascular diseases (17.9 million deaths—44%), cancers (23%), respiratory diseases (10%) and DM (1.5%).10 13

The coexistence of infectious diseases, particularly TB, with NCDs such as DM and hypertension has long been recognised by researchers with varying concepts on managing these conditions in the midst of dwindling resources for healthcare services.14 15 People with DM have a greater risk of developing TB. This increased risk is possibly due to poor glycaemic control resulting in abnormal metabolism in macrophages and lymphocytes, which impacts the immune function of these cells. This predisposes to new TB infection or reactivation of latent TB in those who were previously infected.12 14 On the contrary, the causes of impaired blood glucose during TB treatment are not clear. Current evidence points to an impaired glucose tolerance during TB treatment which may or may not resolve once treatment is completed.16–18 This may be due to undiagnosed DM, stress response from infection which elevates stress hormones or abnormal functioning of the liver which results in abnormal endocrine function.16 19

Among known diabetics undergoing treatment for TB, there have been concerns of DM delaying sputum conversion leading to a poor outcome. This is yet to be fully confirmed.12 14 A recent study from Ghana shows significantly fewer patients with hyperglycaemia had sputum conversion at 2 months of TB treatment compared with normoglycaemic patients, but not at 6 months.20 Other factors that could impact TB treatment outcomes among people with DM include the non-integration of services that causes non-adherence, psychosocial factors such as stigma and increased economic burden of treatment for the two conditions which are paid for out-of-pocket in most low-income countries.12 15 More recently, the COVID-19 pandemic impacted all service delivery and access to essential care. This was due to disruption in the supply of essential health commodities, widespread infection of healthcare workers with COVID-19 and restriction of movement which limited patients with TB from visiting health facilities. The impact of the pandemic and the different measures adopted to limit COVID-19 infection on access to TB services and treatment outcomes is yet to be quantified.

The Syndemics concept has been used to describe the symbiotic coexistence of diseases with associated inequity in access to health and social services, poverty and malnutrition resulting in increased morbidity and mortality in at-risk populations.15 21 The Syndemics concept originated from high-income countries’ observations that different disease conditions coexist and affect the communities, notably the minority populations and those with low socioeconomic status. Meanwhile, the concept has been extended to describe the comorbid conditions which exist in low-income and middle-income countries, like TB/HIV and NCDs.15 21 With a gradual increase in lifespan in low-income and middle-income countries, the impact of NCDs particularly DM and hypertension, has become obvious. Morbidity and mortality due to TB and HIV have reduced because patients now access life-saving medications and observed morbidity and mortality is due to NCDs.22–24

In Eswatini, literature on DM in the population and DM-TB comorbidity is scarce with easily accessible data being estimates by WHO and International Diabetes Federation (IDF).25 Available studies have centred on HIV-NCD comorbidity and developing effective integration models to address the increasing cases of NCDs among HIV patients.26 A 2020 study on the prevalence of abnormal blood glucose metabolism in adults indicated a 3.9% prevalence of type 2 DM (T2DM) in adults who attended the outpatient department of a tertiary hospital but no data is available on associated comorbidities with TB or HIV.25 Similarly, the IDF estimates the prevalence of DM in Eswatini is 3.6% in people aged 20–79 years while the age-adjusted prevalence for impaired glucose tolerance is 6.9%.27

Significant progress has been made in the Kingdom of Eswatini in the provision of HIV/TB services with HIV incidence in people 15 years and above reducing from 2.5% in 2011 to 1.4% in 2017.28 Similarly, TB incidence reduced from 1069/100 000 in 2009 to 363/100 000 in 2019.6 Despite these achievements in HIV/TB control, more is required to improve the quality of life of her citizens as more present with NCDs, notably cardiovascular diseases, DM and cancers.29 Data from the Eswatini Ministry of Health indicate DM accounted for 12% of outpatient department visits in 2018 and 5.9% of all in-patient mortality.29 Given the burden of TB in Eswatini, an increase in cases of DM due to lifestyle changes, obesity and ageing may limit further successes in TB prevention activities. Further complicating the dilemma is the absence of reliable data on the prevalence of the common NCDs in the general population and diverse population groups, with the most recent reliable data on the burden of NCDs in Eswatini being the WHO STEPwise approach to surveillance (STEPS) Survey, conducted in 2014.30 Therefore, research on DM in people receiving treatment for TB will provide insight into the different factors that may impact DM and TB treatment outcomes and provide direction for effective health services delivery. This will put Eswatini on track towards achieving WHO’s target of reducing by a third, the burden of NCDs by 2025.9 In this research, reference to diabetes means T2DM.

Rationale for the research

  1. There is a lack of reliable data on the burden of DM in Eswatini, both in the general population and different population subgroups. This study at its conclusion will generate reliable information on DM-TB comorbidity and the prevalence of DM and hyperglycaemia in patients receiving treatment for TB.

  2. With the availability of life-saving HIV/TB medications, people are living longer but are now exposed to developing diabetes due to changing lifestyles, ageing and possibly TB infection. This study will identify the risk factors for developing diabetes or hyperglycaemia in people receiving treatment for TB.

  3. There is an absence of evidence on factors hindering effective management of diabetes at healthcare facilities providing TB services in Eswatini. This research will identify these factors and propose context-specific solutions to improve the integration of TB and DM care.

Conceptual framework

The study will be based on the Social-Ecological Model which examines different interactions which determine the health outcome of an individual.31 The different individual, interpersonal, community, organisational and policy/environmental contexts which can influence health outcomes in Eswatini in line with the social-ecological model will be considered. This will be contextualised to ascertain how these affect the services received by people while receiving treatment for TB and how service delivery can be improved.

The conceptual framework presented in figure 1 highlights the possible interactions which determine the health outcome of an individual. The components of the socialecological model have been unpacked to show the different direct and indirect relationships that exist between the components and the health outcome of an individual. Other determinants of the health outcome of an individual such as demand and supply factors are not included here. This is to enable easy interpretation and application of this framework to the context of Eswatini.

Figure 1

Conceptual framework for the study.

Research questions and objectives

Research question

The study research questions with the desired endpoint and required variables are presented in table 1.

Table 1

Descriptive summary of different research questions and study requirements


This research aims to describe the epidemiology, predictive factors and control measures of diabetes in a prospective cohort of patients who will be treated for TB. The objectives are to:

  1. Describe the epidemiology of diabetes-TB comorbidity in a prospective cohort of patients receiving TB treatment in Eswatini from 1 June 2022 to 30 April 2023.

  2. Identify factors that predict the occurrence of diabetes (or hyperglycaemia) in patients receiving TB treatment in Eswatini from 1 June 2022 to 30 April 2023.

  3. Describe the effect of blood glucose on TB treatment outcome in patients receiving treatment for TB in Eswatini from 1 June 2022 to 30 April 2023, and ascertain if diabetes is a precursor of first-line TB drug resistance.

  4. Ascertain if there is a relationship between baseline BMI, HIV status, blood glucose level and TB treatment outcomes in patients treated for TB in Eswatini.

  5. Identify factors that hinder effective DM care among diabetics receiving TB treatment in Eswatini and propose a context-specific approach to address these factors.


Study design

A mixed-methods prospective study design will be used for this study. For the quantitative part, a prospective cohort approach will be used to review data of consecutive newly diagnosed patients with TB enrolled on care and followed up from 1 June 2022 to 30 April 2023. The qualitative part will involve the interview of select clinical healthcare workers who provide direct care to patients with TB. Data from the prospective cohort will address objectives 1, 2, 3 and 4 while the healthcare worker’s interview will address objective 5.


The study will be conducted at 12 health facilities providing TB services in the four regions of Eswatini (Mbabane Government Hospital, The Luke Commission, Phocweni Clinic, TB Centre, Siphofaneni clinic, Mankayane Hospital, AHF Lamvelase Clinic, Raleigh Fitkin Memorial Hospital, AHF Matsapha, Pigg’s Peak Government Hospital, Nhlangano Health Centre, Hlathikulu Hospital TB Clinic). The health facilities are purposively selected because they see more patients with TB at any given time and have medical officers who review patients with TB with standard laboratories and X-ray facilities to aid patient investigations. Only complicated cases are referred to the National TB Referral Hospital. The National TB Referral Hospital is excluded from this study as it was converted to COVID-19 isolation and treatment centre and all patients relocated to one of these selected facilities. Eswatini is a landlocked country in Southern Africa with a population of about 1.2 million.29 It is surrounded by South Africa, except in the North-East which is bordered by Mozambique.

Patients on drug-susceptible TB treatment in Eswatini receive treatment for 6–9 months depending on if they have received the first-line drug before and are followed up monthly till after two consecutive sputum conversions (expected in the second or third month). Monthly follow-up continues after sputum conversion till they complete treatment. Sputum microscopy for acid-fast bacilli (AFB) and culture are reviewed on each follow-up visit. Drug-resistant TB (DRTB) patients are initially admitted until they have two consecutive negative sputum culture tests (sputum conversion) before discharge and monthly follow-up visits for clinical evaluation which includes medication review, sputum AFB and culture results review, and general assessment. The duration of treatment for DRTB is varied and can last 12–24 months depending on the drug regimen and response to treatment.

Patients receive routine laboratory assessments during their treatment including random blood glucose at baseline, twice during treatment and at end of treatment before final discharge. Three follow-up visits postdischarge is advocated but most clients do not keep this appointment except those receiving antiretroviral medications from the same facility.

Study population

From 2015 to 2020, about 19 000 patients received treatment for TB in Eswatini and 6% of these were paediatric patients.32 The male-female ratio ranged from 1.4 to 1.6 within the same period and the TB/HIV coinfection rate in 2020 was 64%.32 Report of other comorbidities, for example, T2DM, or hypertension is not available. For the years 2015–2020, about 98% of all patients have a documented treatment outcome for all types of TB and the treatment success rate in 2019 and 2020 was 89% and 86%, respectively.32 There are different cadres of healthcare workers who provide care for patients with TB but those who will participate in the key informant interview will be nurses and doctors.

Sampling and sample size

A consecutive sampling approach will be used to enrol newly diagnosed patients with TB on the study. Current TB programme data indicate that on average, in a period of 4 months, about 410 patients are enrolled on TB care at the 12 selected health facilities. Therefore, it is estimated that a minimum of 380–430 participants will be enrolled on this study. Using an estimated diabetes prevalence of 3.6% in Eswatini,27 an effect size of 0.05 and an alpha of 5%, the estimated power of this study is 98%. A sample size range of 106–582 has been used in similar studies,18 33–36 therefore, this anticipated sample size will be adequate for the different outcomes.

One doctor and one nurse will be purposively selected and interviewed per health facility until saturation is achieved in each group and no more new information is obtained from the healthcare workers.37 Since the study will be conducted at 12 health facilities, there will be a minimum of 24 study participants for the qualitative study.

Inclusion criteria

New patients aged 18 and above who will initiate treatment for any form of TB at any of the 12 selected health facilities from 1 June 2022 up to 30 September 2022 will be eligible for inclusion irrespective of sex. Patients meeting the above criteria who are able and willing to provide informed consent will be included.

Healthcare workers to be included in the study must have clinical training (doctors or nurses), regularly review patients’ medical records, have worked at the health facility for a minimum of 12 months, and be willing to provide informed consent to participate in the study.

Data collection

Data collection from patients (baseline and during follow-up) and interviews of healthcare workers will be conducted by two trained research staff conversant with TB data. Data from patients will be entered into an electronic form developed using Research Electronic Data Capture (REDCap). Patients’ demographic and clinical information will be extracted at baseline and during follow-up visits (month 2, month 5/end of treatment) till patients are discharged from care and have a treatment outcome documented in their case file per updated WHO guidance38 (online supplemental file 1).

A structured questionnaire developed in REDCap will be administered electronically to healthcare workers to identify their views on DM-TB management and challenges to DM-TB services provision at the health facilities. This questionnaire has three sections that should take approximately 15 min to complete—demographic information, occupational information and patient care-related questions. An interview will complement the electronically administered questionnaire (online supplemental file 2) to obtain healthcare workers' perspectives on service delivery and recommendations for improvement. An interview guide has been developed to assist the interviewers during interviews (online supplemental file 3) to ensure the quality of the interview responses. The interview guide will be reviewed to ensure it is coherent and the questions asked directly operationalise the study research question. Before interviewing the healthcare workers, deductive codes will be discussed with the research team to act as a guide during the interviews and the final codes will include those raised by the participants. The interviewee will be allowed to respond with limited interruptions. Recorders will be used so that the correct information is transcribed once the interview is completed. The anticipated duration of the interview is 30–45 min.

Approach for patient data collection

The TB units at the different health facilities where patients with TB are enrolled and reviewed maintain patients' clinical information which will be available to the study team. Noting that some TB units may not have the facility for testing and recording baseline glucose measurements for patients and during follow-up, point of care Accu-Check Active Glucometer (Roche) (with test strips and lancets) will be placed at the different TB Units for the measurement and documentation of blood glucose at baseline and during the second and fifth month follow-up visits. Based on this, baseline data will be collected at enrolment, while follow-up data for the second-month visits will commence in the third month and that for the fifth-month visits will commence in the sixth month (figure 2). This is consistent with the guidelines for the recording of patient information when receiving treatment for TB in Eswatini (online supplemental file 4). The provision of Accu-Check Active Glucometer (Roche) is to ensure patients receive a blood glucose measurement at each visit and the study does not become an additional burden to the health facility. Study participants will not be required to fast before a blood glucose test as such a strict routine may not be achievable in programmatic conditions. All study participants with abnormal random baseline or follow-up glucose measurements will be referred to a clinician for further evaluation and care. At the end of the study, the glucometers will be donated to the TB unit for continued use with support from the health facility’s laboratory.

Figure 2

Schedule for data collection at baseline and during follow-up.

The nurses in the TB unit will be oriented on how to use the Glucometer as a point of care test by a trained laboratory technician from the health facility’s laboratory using a standardised guide (online supplemental file 5). The Glucometer will initially be calibrated by the health facility’s laboratory technician at baseline, at the end of month 2 and the end of month 4. This is to ensure the quality of the results produced by the Glucometer is standardised and possible calibration errors are identified and rectified. Additional baseline patient sociodemographic information—educational status, marital status, occupation, smoking and alcohol status which is not routinely collected will be obtained for this study.

Staff at the TB clinic will be oriented on the study and one healthcare worker at each of the 12 health facilities will be trained on how to approach and consent new patients into the study. Facilities will be visited monthly with a follow-up call weekly for updates.

Statistical methods and analysis

Baseline characteristics of variables in the patients’ dataset will be presented in a table. Prevalence of DM or impaired glucose will be determined based on the number of patients with DM or impaired glucose at baseline and during the treatment period expressed as a proportion of all the patients treated in the same period. This will be determined overall and by the type of TB disease (drug-sensitive TB, rifampicin-resistant TB, DRTB or extra drug-resistant TB). The occurrence of abnormal glucose during treatment will be determined based on the number of patients who had normal values at baseline but developed abnormal values during treatment or at the end of treatment. This will be determined overall, and by type of TB disease with additional analysis to estimate the mean and median time between TB diagnosis and identification of abnormal measurements. A logistic regression (or mixed effect model for repeated data) will be used to predict the occurrence of DM or hyperglycaemia.

Statistical tests will be significant if p<0.05. Different subanalysis, comparative and sensitivity analyses will be done to identify possible interactions which may exist between the different patient characteristics (eg, age, sex, HIV status) and hyperglycaemia, for example, testing to ascertain if there is an association between timing of culture conversion and blood glucose. The proposed statistical methods for the different research questions are presented in table 1.

Qualitative analysis in form of analysis of identified codes will be done to identify factors that hinder the care of diabetics receiving TB treatment. Recommendations for improvement will be coded and similar codes will be analysed and presented.

Study data entered into REDCap39 will be extracted in Stata format and imported into Stata V.15 (Stata) for analysis. The software NVivo40 will be used for the analysis of transcribed information from healthcare workers' interviews.

Patient and public involvement

Patients and members of the public were not involved in the study design and the development of this protocol. However, TB programme priorities were considered in the design of the study and protocol. Patients during their routine visits will be informed if their blood glucose measurement is within the normal values. Those with abnormal values will be referred for further review and care. Participating health facilities, healthcare workers, the TB programme and relevant stakeholders will be provided with feedback on the outcome of the study with direct recommendations on how to improve access to TB services and integrate NCD care into TB services.

Ethics and dissemination

Ethical considerations

Approval for the study has been obtained from the Eswatini Health and Human Research Review Board (Protocol Reference Number: EHHRRB036/2021).

Participation in the study will be optional. Patients and healthcare workers who will be interviewed will be oriented and provided with a study information sheet (online supplemental files 6 and 7). They will be required to provide informed consent before participating (online supplemental files 6 and 7) and healthcare workers' consent will include consent for the recording of their comments. Researchers administering the interviews will be required to attest they read out the information sheet to the study participants and answered all questions to their satisfaction before commencing the interview.

Data will be deidentified to ensure confidentiality. Each patient record and healthcare worker interviewed will be assigned a unique identification code. This code will assist with retrieving information in case there is missing data during analysis or a follow-up question. Identifiable information will only be available to the principal investigator and the deidentified data will be accessible to the study team for monitoring of data quality. All project data will be stored in a password protected hard drive to ensure data safety. Transcription will be done without linking names to comments to ensure confidentiality. Healthcare workers will be free to stop participating in the interview at any time without providing a reason. The risks to study participants are minimal and measures have been instituted to ensure confidentiality and safety of data and information from the patients and healthcare workers. Researchers who will access patient data and interview healthcare workers will be required to sign a confidentiality and non-disclosure document (online supplemental file 8).

Standard COVID-19 control measures will be adopted during data extraction and a virtual interview option will be available to limit infection and transmission of SARS-COV-2.


Several articles will be generated from this study for presentation either as a poster or oral presentation at local and international conferences and for publication in peer-reviewed journals.

Ethics statements

Patient consent for publication


Supplementary materials


  • Contributors VW conceptualised the study and developed the first draft and collated input for subsequent drafts. AV, DEG, KO and KK-G participated in the development of the study design, revised and edited the first draft. AV and KO revised the statistical plan and KK-G guided on ethical considerations. All authors read and approved the final manuscript.

  • Funding VW is supported by the Global Health PhD Support Program at University Medical Center, Utrecht, The Netherlands.

  • Competing interests None declared.

  • Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.