Clinical and lifestyle characteristics associated with older age among tuberculosis patients with diabetes in Bangladesh: An analytical cross-sectional study

Abstract

Objective

Studies on the clinical and lifestyle characteristics of tuberculosis patients with diabetes mellitus in Bangladesh remain limited. This study compared the demographic, clinical, and lifestyle characteristics between older and younger tuberculosis patients with diabetes mellitus.

Methods

This multicenter analytical cross-sectional study was conducted from January to December 2024 in 18 centers of the Diabetic Association of Bangladesh. Adult patients with a history of completed tuberculosis treatment and diabetes diagnosed >2 weeks after initiation of tuberculosis treatment were considered eligible. Among 393 eligible participants, all 80 patients aged ≥50 years were included. For balanced comparison, 80 participants aged <50 years were selected from the remaining eligible pool. Data were collected through face-to-face interviews using a pretested semistructured questionnaire and checklist. Categorical variables were compared using the chi-square test or Fisher’s exact test. Multivariable binary logistic regression was used to identify characteristics independently associated with older age. Written informed consent was obtained, and ethical issues were maintained.

Results

Male patients were predominant in both older (80.0%) and younger (93.8%) tuberculosis patients with diabetes mellitus. Pulmonary tuberculosis was more common in younger patients (97.5%), while family history of diabetes (46.3%) and comorbidity (47.5%) were more common in older patients with diabetes mellitus. Moreover, occupation, body mass index category, physical inactivity, family history of diabetes, comorbidity, and type of tuberculosis were found to be significant (p < 0.05) between older and younger patients. In multivariable analysis, family history of diabetes (adjusted odds ratio = 3.873, p = 0.001) and pulmonary tuberculosis classification (adjusted odds ratio = 12.820, p = 0.003) remained independently associated with older age.

Conclusion

Family history of diabetes and type of tuberculosis were found to be independently associated with older tuberculosis patients with diabetes mellitus. These findings may help inform age-sensitive screening, referral, and follow-up strategies for integrated tuberculosis–diabetes mellitus care; however, larger and more representative studies are warranted.

Keywords

Tuberculosis older adults diabetes mellitus tuberculosis–diabetes mellitus comorbidity Bangladesh

Introduction

Tuberculosis (TB) remains a major global public health concern and is a leading cause of death from infectious diseases. In 2024, the World Health Organization (WHO) estimated that 10.7 million people developed TB, with 1.23 million deaths worldwide.¹ In Bangladesh, the estimated TB incidence was 221 per 100,000 population, placing the country among the eight nations accounting for two-thirds of the global TB burden.¹ Key risk factors for TB include malnutrition, poverty, smoking, alcohol use, and comorbidities such as human immunodeficiency virus infection and diabetes mellitus (DM).²

DM is also a growing global health challenge, with its prevalence expected to rise sharply in the coming years. In Bangladesh, the prevalence of DM—driven by rapid urbanization, population aging, sedentary behavior, and rising obesity rates—reached 14.2% in 2021.^3,4 Although DM complications contribute significantly to the disease burden, preventive strategies can help reduce the impact.

TB and DM often coexist and interact bidirectionally. DM increases susceptibility to TB, while TB-related physiological stress and treatment may contribute to or worsen existing DM. This dual burden strains healthcare systems, especially in low- and middle-income countries such as Bangladesh, where both conditions are widespread, particularly among the older population.

In particular, older individuals are vulnerable due to weakened immunity or multiple comorbidities. TB–DM comorbidity is usually associated with poorer treatment outcomes, higher mortality, and longer recovery periods.^5,6 Previous studies have also demonstrated that these treatment outcomes may vary across different age groups.^7,8 Understanding these differences is important for planning an age-sensitive TB–DM care model.

Although several studies have examined the prevalence of DM among patients with TB, relatively few studies have explored how demographic, clinical, and lifestyle characteristics vary between older and younger TB patients with DM, particularly in low- and middle-income countries such as Bangladesh. Understanding these differences may help improve early detection of DM among TB patients, support risk stratification, and inform age-sensitive screening and follow-up strategies in TB–DM comorbidity management. In this study, older age was defined as age ≥50 years in line with TB literature^9,10 and programmatic practice in which TB-related vulnerability and comorbidity burden often increase from this age onward.

The present study, therefore, compared demographic, clinical, and lifestyle characteristics between older (≥50 years) and younger (<50 years) TB–DM patients at selected BADAS centers in Bangladesh. It also explored the characteristics associated with older TB–DM patients.

Methods

Study design and settings

This analytical cross-sectional study was conducted from January to December 2024 in centers affiliated with the Diabetic Association of Bangladesh (BADAS). BADAS operates a nationwide network of DM care facilities that provide outpatient and inpatient services and maintain clinical records of patients with DM and related comorbidities.

Eighteen BADAS-affiliated centers were selected for this study to ensure geographic representation. Two centers were selected from each of the eight administrative divisions of Bangladesh, and given the larger population size and higher patient load in the Dhaka Division, two additional centers from this division were included.

These centers routinely provide DM care and maintain clinical records of patients with TB–DM comorbidity, making them suitable sites for identifying eligible study participants. The multicenter approach allowed inclusion of patients from diverse geographic and healthcare settings across Bangladesh.

Study population

The study population comprised patients aged ≥18 years who attended the selected BADAS centers with a history of completed TB treatment and DM diagnosed more than 2 weeks after initiation of TB treatment. This timing was used in accordance with national TB–DM comorbidity management guidance to reduce misclassification related to transient stress hyperglycemia.¹¹

Study participants were further categorized into two groups based on age:

older TB–DM patient: age ≥50 years

younger TB–DM patient: age <50 years

In this study, the older TB–DM patients were treated as the reference category to identify demographic, clinical, and lifestyle characteristics that distinguished older TB–DM patients from younger TB–DM patients.

Sample size and sampling technique

Eighteen BADAS-affiliated centers were selected for this study: two centers from each of the country’s eight administrative divisions. Considering the large population in the Dhaka Division, two additional centers were included in the study. From these centers, a total of 489 TB–DM patients were identified from center records during the study period. After excluding patients with incomplete information (n = 6) and those who declined participation (n = 90), the final cohort comprised 393 eligible participants. The participant selection procedure is presented in Figure 1.

Figure 1.

Selection process of TB–DM patients and classification into older and younger age groups.

Among these, all 80 eligible TB–DM patients aged ≥50 years were included. From the remaining eligible participants aged <50 years, 80 were selected from the same centers to create a balanced comparison group. This approach was used to facilitate group-wise analytical comparison rather than estimate age-specific prevalence in the source population.

Data collection methods and instruments

Data were collected through face-to-face interviews and reviewing medical records. Background characteristics and disease-related information were collected using a pretested, semi-structured questionnaire. Information on background characteristics and some disease-related variables was collected from eligible participants. Variables related to background characteristics included the following: (a) sex (male/female); (b) religion (Islam/Sonaton/others); (c) marital status (married/unmarried); (d) educational level (illiterate/primary/secondary/higher secondary and above); (e) occupation (service/self-employed or business/homemaker/unemployed or retired); (f) type of family (nuclear/joint); (g) body mass index (BMI) (underweight/normal/overweight/obese); and (h) smoking habit (current smoker/not-current smoker). Disease-related variables included the following: (a) family history of DM (yes/no); (b) comorbidity (yes/no); and (c) type of TB (pulmonary TB/extra-pulmonary TB). After obtaining written consent from all selected patients, data were collected through face-to-face interviews. Other relevant information was collected by reviewing patient records.

Data management

Data were checked and verified at both field and central levels to ensure quality. Data were kept safely under the principal investigator’s control. All data were thoroughly checked to verify their relevance and consistency. Incomplete and missing data were excluded. Data were coded, categorized, cleaned, and entered into SPSS software (version 25.0). All participant information was de-identified before analysis, and no personally identifiable information was retained in the manuscript.

Statistical analysis

Data analysis was performed using SPSS. Categorical variables were presented as frequencies and percentages. Comparisons between older TB–DM patients and younger TB–DM patients were made using the chi-square test or Fisher’s exact test, as appropriate. Variables that were clinically relevant and demonstrated an association in bivariate analysis were entered into the multivariable binary logistic regression model to identify characteristics independently associated with the older TB–DM patient group. Collinearity among included variables was assessed before final model fitting. Adjusted odds ratios (AORs) with 95% confidence intervals (CIs) were reported. A p-value <0.05 was considered statistically significant. All tests were two-sided.

Ethical approval and informed consent statements

Ethical clearance was taken from the ethical review committee of the BADAS, Dhaka, Bangladesh (BADAS-ERC/EC/24/525). The confidentiality and anonymity of the data were strictly maintained. The study was conducted in accordance with the ethical principles of the Declaration of Helsinki. Written informed consent was obtained from all participants before the interview. Participants had the right to withdraw from the study at any time during data collection.

Reporting guideline

This study was reported in accordance with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines for observational studies.¹²

Results

The study compared demographic, lifestyle, and clinical characteristics between older and younger TB–DM patients. Occupation demonstrated a statistical significance between older and younger TB–DM patients (p = 0.005). Among other characteristics, physical inactivity (p = 0.044) and BMI category (p = 0.011) exhibited significant differences between older and younger TB–DM patients. Although the mean BMI was quite similar between these patients, it was 20.198 ± 3.628 in older TB–DM patients and 20.246 ± 3.171 in younger TB–DM patients. However, no significant differences were found in other background characteristics, such as sex, marital status, educational level, and smoking habits (Table 1).

Table 1.

Comparison between older and younger TB–DM patients based on background variables.

Attributes	Older TB–DM patientsf (%)	Younger TB–DM patientsf (%)	p-value(χ² test)
Sex
Male	64 (80.0)	67 (83.8)	0.682
Female	16 (20.0)	13 (16.2)	0.682
Marital status
Married	67 (83.8)	61 (76.2)	0.323
Unmarried	13 (16.2)	19 (23.8)	0.323
Educational level
Illiterate	10 (12.5)	21 (26.3)	0.087
Primary	6 (7.5)	3 (3.8)
Secondary	8 (10.0)	11 (13.8)
Higher secondary and above	56 (70.0)	45 (56.3)
Occupation
Service	6 (7.5)	7 (8.8)	0.005
Self-employed/Business	11 (13.8)	20 (25.0)
Homemaker	14 (17.5)	26 (32.5)
Unemployed/Retired	49 (61.3)	27 (33.8)
Smoking habit
Current smoker	41 (51.2)	34 (42.5)	0.342
Non-current smoker	39 (48.8)	46 (57.5)	0.342
BMI^*
Underweight	29 (36.3)	18 (22.5)	0.011
Normal	32 (40.0)	51 (63.7)
Overweight/Obese	19 (23.8)	11 (13.8)
Physical activity
Yes	10 (12.5)	21 (26.3)	0.044
No	70 (87.5)	59 (73.8)	0.044

TB: tuberculosis; DM: diabetes mellitus; f: frequency; BMI: body mass index (weight in kg/height in meter²–Asian Scale).

Among the study participants, family history of DM was more commonly reported in older TB–DM patients (46.3%) than in younger TB–DM patients (21.3%); significant differences were observed between older and younger TB–DM patients (p = 0.001). In addition, significant differences were observed in the presence of comorbid disease conditions between older and younger TB–DM patients (p = 0.022). Moreover, the type of TB or disease classification also demonstrated statistically significant differences between older and younger TB–DM patients (p < 0.001) (Table 2).

Table 2.

Comparison between older and younger TB–DM patients based on disease-related variables.

Attributes	Older TB–DM patientsf (%)	Younger TB–DM patientsf (%)	p-value(χ² test)
Family history of diabetes
Yes	37 (46.3)	17 (21.3)	0.001
No	43 (53.8)	63 (78.8)	0.001
Comorbidity
Yes	38 (47.5)	23 (28.7)	0.022
No	42 (52.5)	57 (71.3)	0.022
Type of TB
Pulmonary TB	62 (77.5)	78 (97.5)	<0.001
Extra-pulmonary TB	18 (22.5)	2 (2.5)	<0.001

TB: tuberculosis; DM: diabetes mellitus; f: frequency.

In the regression analyses, several characteristics were associated with older TB–DM patients. In the univariable analysis, comorbidity was significantly associated with older TB–DM patients (odds ratio = 2.242, p = 0.015). In the multivariable logistic regression, family history of DM and type of TB remained independently associated with older TB–DM patients. These findings suggest that both clinical and familial factors play an important role in distinguishing older TB–DM patients from younger TB–DM patients (Table 3).

Table 3.

Binary logistic regression analysis of some selected attributes of older TB–DM patients.

Attributes		Univariable model			Multivariable model
Attributes		COR	95% CI	Sig	AOR	95% CI	Sig
Occupation	Service	Reference			Reference
	Unemployed/Retired	2.117	0.646–6.941	0.216	2.598	0.636–10.616	0.184
	Self-employed/Business	0.642	0.172–2.391	0.509	0.537	0.113–2.549	0.434
	Homemaker	0.628	0.177–2.235	0.473	0.631	0.137–2.907	0.554
BMI	Underweight	Reference			Reference
	Overweight/Obese	1.072	0.416–2.765	0.885	1.008	0.309–3.282	0.990
	Normal	0.389	0.187–0.813	0.012	0.521	0.219–1.240	0.140
Physical activity	Yes	Reference			Reference
Physical activity	No	0.401	0.175–0.919	0.031	2.568	0.883–7.469	0.083
Family history of diabetes	No	Reference			Reference
Family history of diabetes	Yes	3.189	1.595–6.376	0.001	3.873	1.713–8.759	0.001
Comorbidity	No	Reference			Reference
Comorbidity	Yes	2.242	1.167–4.310	0.015	1.130	0.494–2.587	0.772
Type of TB	Extra-pulmonary	Reference			Reference
Type of TB	Pulmonary	11.323	2.530–50.666	0.002	12.820	2.432–67.585	0.003

COR: crude odds ratio, CI: confidence interval, AOR: adjusted odds ratio, Sig: significance; TB: tuberculosis; DM: diabetes mellitus.

Discussion

This multicenter analytical cross-sectional study compared older TB–DM patients and younger TB–DM patients in Bangladesh. The analysis revealed that family history of DM and type of TB were independently associated with belonging to the older TB–DM patient group after adjustment for other measured variables. Several other characteristics, including occupation, BMI category, physical inactivity, and comorbidity, were significant between older and younger TB–DM patients in the bivariate analysis but were not significant in the adjusted model.

Family history of DM was one of the strongest characteristics associated with the older TB–DM patients. A substantially higher proportion of older TB–DM patients reported a positive family history, and this remained significant in both the univariable and multivariable models. The result is similar to that of a study from Tanzania.¹³ This finding may reflect the cumulative role of familial susceptibility together with long-term metabolic vulnerability over the life course.

The type of TB was also independently associated with older TB–DM patients in this study. Pulmonary TB was more common in the younger TB–DM patients, while extra-pulmonary TB was relatively more frequent in the older TB–DM patients. The adjusted model nonetheless demonstrated a strong association between TB classification and the older TB–DM patient. As this was a cross-sectional study, the results do not imply that pulmonary TB causes DM or that age is a biological outcome. Rather, it indicates that the type of TB differed between older and younger TB–DM patients. Previous studies have also reported an association between pulmonary TB and DM among patients with TB.^14–16

In the bivariate analysis, occupation, BMI category, and physical inactivity differed significantly between older TB–DM patients and younger TB–DM patient; however, these associations were not significant after adjustment. These findings may still be important from a descriptive and programmatic perspective, as they suggest that older TB–DM patients may differ from younger TB–DM patients in social and lifestyle characteristics that could influence care needs, counseling, and follow-up planning.

In this study, a larger proportion of older TB–DM patients were unemployed or retired. Although occupation was not independently associated with the older TB–DM patients in the adjusted model, this pattern may still reflect broader socio-demographic differences between the older and younger TB–DM patients. A study from Nigeria found no consistent evidence that occupation was associated with DM risk among patients with TB.¹⁷ In the present study, the observed difference may partly relate to reduced workforce participation and more sedentary daily routines among older TB–DM patients, which have been associated with DM-related metabolic risk.¹⁸

Although the mean BMI of older and younger TB–DM patients did not vary significantly, the BMI category demonstrated a significant association in bivariate analysis. This pattern may reflect differences in the nutritional and metabolic profiles between older and younger TB–DM patients. Earlier studies have also reported an association between BMI category and TB–DM patients.^16,19 However, in this study, BMI category was not independently associated with older TB–DM patients after multivariable adjustment.

Physical inactivity also varied significantly between older and younger TB–DM patients in the bivariate analysis. Previous evidence has demonstrated that lower physical activity is associated with DM, particularly among middle-aged and older adults.²⁰ In the present study, this difference may reflect the combined effects of age, illness burden, frailty, and reduced mobility among older TB–DM patients. Nevertheless, as the association did not remain significant in the adjusted model, this finding should be interpreted as a group difference rather than an independent distinguishing factor.

Comorbidity was more common among older TB–DM patients and was significantly associated in the univariable analysis; however, it did not remain significant in the adjusted model. This finding is still clinically relevant, as multiple chronic conditions tend to accumulate with age and may complicate both TB and DM care. Previous studies also suggested that the presence of multiple chronic conditions is more common in older patients and may contribute to worsening TB–DM health conditions and overall health status.^7,21 In the Bangladesh context, this may have practical implications for integrated care and follow-up in older TB–DM patients.

Overall, the findings suggest that the older and younger TB–DM patients are not entirely similar in their clinical and familial profiles. Family history of DM and type of TB were the two characteristics most strongly associated with belonging to the older TB–DM patient group. Recognizing these differences may support more age-sensitive approaches to screening, counseling, referral, and continuity of care in joint TB–DM services.

Limitations

This study has certain limitations. As this was an analytic cross-sectional study, causation cannot be established; only association is demonstrated. Additionally, because this study was center-based, its findings may not generalize to community-dwelling TB–DM patients. Moreover, because it is a center-based study, it may be subject to selection bias, particularly since the younger comparison group was selected from the available eligible pool rather than through random sampling. The sample size was modest, and the operational age threshold of 50 years may limit comparability with studies that use higher age thresholds for older age. Finally, several potentially relevant variables, including alcohol intake, dietary factors, medication history, and glycemic control indicators such as glycated hemoglobin, were not available for adjustment; thus, residual confounding cannot be excluded.

Conclusion

This study identified key demographic and clinical characteristics associated with belonging to the older TB–DM patient, particularly family history of DM and type of TB. These findings may help inform age-sensitive screening, referral, and continuity-of-care strategies for joint TB–DM management in Bangladesh. Larger and more representative studies are warranted to confirm these observations.

Footnotes

Acknowledgement

The authors thank BADAS, Dhaka, Bangladesh, and all the participants of the study for their support and contribution. AI was used solely for language and editing purposes.

Author contribution

Mohammad Afsarul Habib: Conceptualization (lead); writing–original draft (lead); formal analysis (lead), and project administration (lead).

Syeda Sumaiya Efa: Conceptualization (lead); methodology (lead); writing–original draft (lead); formal analysis (lead); writing–review and editing (equal), and visualization (lead).

Kaniz Afrin: Methodology and formal analysis (equal) and software (lead).

Nasreen Islam: Conceptualization (supporting) and writing–original draft (supporting).

Mohammad Delwar Hossain: Review and editing (equal).

Consent for publication

Not applicable

Data availability statement

The datasets used and analyzed during the current study are available from the corresponding author on reasonable request.

Declaration of AI usage

All ideas, interpretations, and analyses are the original work of the authors. AI was used solely as a linguistic and editorial aid, and its contributions were reviewed and approved by the authors.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Ethical approval and informed consent statements

Ethical clearance was obtained from the ethical review committee of BADAS, Dhaka, Bangladesh (BADAS-ERC/EC/24/525). Confidentiality and anonymity of data were maintained strictly. The study was conducted in accordance with the ethical principles of the Declaration of Helsinki. Informed written consent was obtained from all participants before the interview. Participants had the right to withdraw from the study at any time during data collection.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

ORCID iD

Mohammad Afsarul Habib

References

World Health Organization. Global tuberculosis report 2025. Geneva: World Health Organization, https://www.who.int/teams/global-tuberculosis-programme/tb-reports/global-tuberculosis-report-2025 (2025, accessed 11 March 2026).

Bisht

Dahiya

Ghosh

, et al. The cause-effect relation of tuberculosis on incidence of diabetes mellitus. Front Cell Infect Microbiol 2023; 13: 1134036.

International Diabetes Federation. IDF Diabetes Atlas Belgium 2021. https://diabetesatlas.org/ (2021, accessed 6 January 2025).

Chowdhury

MAB

Islam

Rahman

, et al. Diabetes among adults in Bangladesh: changes in prevalence and risk factors between two cross-sectional surveys. BMJ Open 2022; 12: e055044.

Sharma

Goel

Sharma

, et al. Prevalence of diabetes mellitus and its predictors among tuberculosis patients currently on treatment. Indian J Community Med 2018; 43: 302–306.

Abdullah Al Mamun

Sohael

Prevalence of diabetes mellitus among tuberculosis patients in Bangladesh. Diabetes Complications 2022; 6: 1–8.

Habib

Afrin

Efa

, et al. Effects of diabetes mellitus on retreatment of tuberculosis: a multi-centered case-control study from Bangladesh. J Clin Tuberc Other Mycobact Dis 2024; 36: 100450.

Rehman

Khattak

Mushtaq

, et al. The impact of diabetes mellitus on the emergence of multi-drug resistant tuberculosis and treatment failure in TB-diabetes comorbid patients: a systematic review and meta-analysis. Front Public Health 2023; 11: 1244450.

Negin

Abimbola

Marais

BJ.

Tuberculosis among older adults–time to take notice. Int J Infect Dis 2015; 32: 135–137.

10.

Coffman

Chanda-Kapata

Marais

, et al. Tuberculosis among older adults in Zambia: burden and characteristics among a neglected group. BMC Public Health 2017; 17: 804.

11.

Hossain

Ahmed

Rahim

, et al. Bangladesh national guidelines on the management of tuberculosis and diabetes mellitus co-morbidity (summary). Indian J Endocrinol Metab 2016; 20: 853–857.

12.

Von Elm

Altman

Egger

; STROBE Initiativeet al. Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. BMJ 2007; 335: 806–808.

13.

Mabula

Kazinyingi

Chavala

, et al. Prevalence and risk factors for diabetes mellitus among tuberculosis patients in Moshi Municipal Council, Kilimanjaro, Tanzania. East Afr Health Res J 2021; 5: 69–74.

14.

Delgado-Sánchez

García-García

Castellanos-Joya

, et al. Association of pulmonary tuberculosis and diabetes in Mexico: analysis of the national tuberculosis registry 2000–2012. PloS One 2015; 10: e0129312.

15.

Lee

Kang

, et al. Prevalence and impact of diabetes mellitus among patients with active pulmonary tuberculosis in South Korea. Lung 2017; 195: 209–215.

16.

Hossain

Rahim

Islam

, et al. TB prevalence among patients with diabetes in Bangladesh. Int J Tuberc Lung Dis 2022; 26: 119–125.

17.

Ekeke

Ukwaja

Chukwu

, et al. Screening for diabetes mellitus among tuberculosis patients in Southern Nigeria: a multi-centre implementation study under programme settings. Sci Rep 2017; 7: 44205.

18.

Hamilton

Zderic

TW.

Sedentary behavior as a mediator of type 2 diabetes. Med Sport Sci 2014; 60: 11–26.

19.

Sembiah

Nagar

Gour

, et al. Diabetes in tuberculosis patients: an emerging public health concern and the determinants and impact on treatment outcome. J Family Community Med 2020; 27: 91–96.

20.

Tian

Shilko

, et al. The relationship between physical activity and diabetes in middle-aged and elderly people. Medicine (Baltimore) 2023; 102: e32796.

21.

Stanikzai

Zafar

Akbari

, et al. Prevalence and predictors of depression symptoms among Afghan hypertensive patients: secondary analysis of a multi-center cross-sectional study. Integr Blood Press Control 2024; 17: 41–50.