Development and Validation of a Risk Score for Diabetes Screening in Oman

Area of Science:

• Public Health
• Epidemiology
• Diabetes Research

Background:

• Type 2 diabetes mellitus (T2DM) poses a significant public health challenge globally and in Oman.
• Non-invasive and accessible screening tools are crucial for early detection and prevention of T2DM.
• Existing diabetes risk scores may not perform optimally in diverse populations.

Purpose of the Study:

• To develop and validate a diabetes risk score model tailored for the general Omani population.
• To create a self-administered, non-invasive screening tool for prevalent T2DM.
• To assess the performance of the developed score against existing models.

Main Methods:

• Utilized the 2008 World Health Survey (WHS) data from Oman (n=2720) for model development.
• Employed multivariable logistic regression to identify significant risk factors (e.g., age, BMI, hypertension).
• Validated the model using an independent Omani cohort (Sur Survey 2006, n=1355) via ROC curve analysis.

Main Results:

• A robust diabetes risk score model was developed, incorporating eight variables: age, sex, education, marital status, residence, hypertension, smoking, and BMI.
• Internal validation (WHS 2008) showed an AUC of 0.80 with optimal cutoff ≥ 15 (sensitivity 71.1%, specificity 74.4%).
• External validation (Sur 2006) yielded an AUC of 0.74 with optimal cutoff ≥ 13 (sensitivity 54.0%, specificity 79.0%). Existing scores performed poorly in the Omani population.

Conclusions:

• The developed diabetes risk score is an easy-to-use, self-administered tool for screening prevalent T2DM in Oman.
• The score identifies individuals at risk and raises awareness about diabetes risk factors.
• Findings can inform policymakers in establishing targeted diabetes prevention programs.