Publication: Validation and effectiveness of a new screening tool for pulmonary tb detection among diabetic patients in Kelantan
Loading...
Date
2024-03
Authors
Ismail, Muhammad Ikhwan
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Introduction: Tuberculosis (TB) is a dangerous infectious disease worldwide. Diabetes mellitus (DM) increases the risk of infection up to four times compared to non-DM patients. With the increase in DM patients, the rate of TB infection is increasing worldwide. Until now, there is no specific TB screening tool for DM patients causing the lack of Pulmonary TB detection among DM patients.
Objective: To determine the factors that contribute to Pulmonary TB disease among DM patients, formulate a TB screening scoring tool and confirm the effectiveness of the Tuberculosis Diabetes Mellitus Predictive Tool (TBDM-PT) in Kelantan.
Methodology: This study consists of three phases. The case-control study for the Phase 1 involved 270 DM patients either suffering from TB (n=90) and non-TB patients (n=180). DM disease status, age 18 years and above and in the health clinic appointment were included in the analysis. Patient data was taken from the DM medical record, TBIS 10-1A form and National Diabetes Registry (NDR) website. Simple and multiple logistic regression was used to analyze the data. During Phase 2, the risk score was formulated based on multiple logistic regression and the area under the curve (AUC) method. The content validation study involved nine experts in TB management. The values of the factor content validation index (F-CVI) and the average index to the content validation index at the scale level (S-CVI/Ave) were obtained. Twenty health workers in health clinics were selected for face validation studies and face validation index (FVI) values were obtained.
A pilot study involving 34 DM patients who had undergone TB screening at a health clinic in the Kota Bharu district three months prior, was conducted to determine the diagnostic performance for TBDM-PT such as accuracy, sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV).
In a quasi-experimental study for phase 3, Kota Bharu was selected as the intervention group, using TBDM-PT as a screening tool while Pasir Mas was the control group using existing screening methods. The patient was followed up for four months until the patient completed the Pulmonary TB diagnosis test as the primary outcome while the secondary outcome was the latent TB infection (LTBI). Incidence rates and risk ratios (RR) were used in comparative analysis to demonstrate the effectiveness of screening tools
Result: The majority of patients were female (56.7%) and aged 60 years and above (60.7%). Seven factors (items) were identified to be closely related by statistical analysis to the risk of TB disease; male (Adj OR: (95% CI); 2.78 (1.17-6.63)), Body mass index (BMI) <23kg/m2 (Adj OR: (95% CI); 7.22 (2.74-19.04)), duration of DM ≤ 10 years (Adj OR: (95% CI); 2.90 (1.13-7.39)), no BCG scar (Adj OR: (95% CI);19.76 (2.98-130.9)), HbA1C level in 6 months ≥8% (Adj OR: (95% CI);19.76 8.31 (3.28-21.05)), history of cough over two weeks (Adj OR: (95% CI); 31.08 (11.64-82.98)) and history of fever over two weeks (Adj OR : (95% CI); 6.25 (1.15-33.99)). TB contact history was found to be clinically closely related to the risk of TB disease through consultation with a TB specialist and was included as a screening item as a high risk in TBDM-PT.
The F-CVI and S-CVI/Ave values reached the accepted adequate screening tool validation level. The pilot study showed the accuracy was 58.8% with a sensitivity 83%, specificity 54%, positive predictive value (PPV) 28% and negative predictive value (NPV) 94%. A total of 227 DM patients underwent an experimental study, the risk ratio (RR) for the detection of TB disease was 3.29; 95%CI (1.12,9.69), while detection of latent TB (LTBI) involving 223 DM patients found the risk ratio (RR) was 3.67; 95%CI (1.01,12.76)
Conclusion: TBDM-PT is a newly developed and validated Pulmonary TB risk scoring tool. It effectively increases the detection rate three- to fourfold of Pulmonary TB disease and latent TB infection in the DM patient population. The recommendation is that this study needs to be improved with cost-effective studies before being used in health clinics.
Description
Keywords
Tuberculosis , diabetes mellitus