A Dataset on Demographic and Lifestyle Factors for Prognosticating Chronic Kidney Disease Progression in Diabetic Patients

Published: 2 July 2025| Version 2 | DOI: 10.17632/hjkzgbxgv5.2
Contributors:
,
,
,
,
,

Description

This dataset, collected from BIRDEM General Hospital, Dhaka, presents a comprehensive compilation of demographic and lifestyle factors for predicting the development of Chronic Kidney Disease (CKD) in diabetic patients. It includes health records from 400 individuals, 185 diagnosed with CKD and 215 without, all of whom have been living with diabetes for over a decade. The data was carefully collected, cleaned, and authenticated under the direct supervision of Dr. Rafi Nazrul, a nephrologist at BIRDEM. Ethical approval was obtained from the Ethical Review Committee (ERC) of the Diabetic Association of Bangladesh (BADAS), and a structured questionnaire was used to gather relevant demographic and non-invasive lifestyle data. The feature set was curated through an iterative process, combining prior research findings with expert input from diabetologists and nephrologists to ensure clinical validity and applicability. The dataset includes variables such as gender, job type, family history of diabetes, age, BMI, comorbidities (e.g., hypertension, heart disease), and lifestyle habits (diet adherence, physical activity, smoking, water intake), along with detailed dietary intake to estimate daily calorie consumption. These easily accessible, non-invasive factors recorded longitudinally through patients’ diabetes record books, which will allow researchers to explore and model the trajectory of CKD development over a 10-year span, offering early insights into risk prediction and disease progression, particularly in resource-constrained healthcare settings.

Files

Institutions

  • Ahsanullah University of Science and Technology

Categories

Machine Learning

Licence