Introduction: Diabetes screening questionnaires such as FINDRISC need to be evaluated in the target population taking into account their own characteristics.
Material and Methods: A cross-sectional study of diagnostic tests was carried out. We found Original FINDRISC (OF) and Modified FINDRISC (MF) scores with Latin American abdominal obesity cut-offs (male ≥ 94 cm and female ≥ 90 cm) to determine dysglicemia including impaired fasting glucose (IFT), glucose intolerance (GIT), and diabetes mellitus (DM), according to Oral Glucose Tolerance Test in workers of the María Auxiliadora Hospital. The Area under the ROC Curve (AUROC) of the MF and OF was found. The sensitivity, specificity, positive likelihood ratio (PLR) and negative likelihood ratio (NLR) of each MF and OF point were calculated. A PLR> 2 was rated as useful and one greater than 10 as excellent, and an NLR <0.5 was considered useful and <0.1 as excellent.
Results: Of 549 subjects, 77.5% were female, 70% were> 45 years old, 75% had a BMI> 25 and 65.4% abdominal obesity with Latin American criteria. About 15% had some history of hyperglycemia or hypertension and 70% did not exercise or did not eat vegetables daily. Half of them had a family member with diabetes. There was dysglicemia in 17.8%, of which 11.4% had IFT; 3.8% GIT and 2.6% DM. The MF presented greater accuracy than OF (AUROC: 0.71 vs. 0.69, p = 0.007). There were no differences in the curves between women or men.The score 11 obtained a sensitivity and specificity of 78.6% and 53.0% respectively and the score 14 a sensitivity and specificity of 56.1% and 76.7 respectively. A PLR, a value of 14 is rated useful and an excellent 21. On the other hand, one NLR qualifies a score of 11 as useful and 2 as excellent.
Conclusion: MF had greater diagnostic accuracy than OF for dysglycemia. Scores of 14 and 21, confer high and very high risk; on the contrary, scores of 11 and 2, low and very low risk.
M. Yovera: None. L. Damas: None. R.W. Ticse: None.
