Comparing the performance of FINDRISC, CANRISK, and a Saudi-specific dysglycemia risk score in the Kingdom of Saudi Arabia Público

Guo, Angela Wenxia (2016)

Permanent URL: https://etd.library.emory.edu/concern/etds/5999n4223?locale=es
Published

Abstract

Objective: To assess the performance of three diabetes risk scores: a rapid Saudi-specific dysglycemia risk score, the Finnish Diabetes Risk Score (FINDRISC), and the Canadian Diabetes Risk Assessment (CANRISK) as a screening tool for dysglycemia, using data from the Saudi Household Interview Survey, a recent nationally representative cross-sectional study in Saudi Arabia.

Research Design and Methods: Data from 4,461 Saudi adults aged 15 years or older who participated in the Saudi Health Interview Survey and completed an HbA1c lab test were included in the analysis. Anthropologic measurements, socio-demographic information, lifestyle information, and past medical history were collected through household surveys. Participants were referred to a local clinic to complete biochemical measurements, including HbA1c. We assessed the performance of a Saudi-specific risk score, the FINDRISC, and CANRISK tools by calculating sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV). We also estimated the kappa statistic for agreement for each risk score to assess the frequency of the risk scores predicting dysglycemia when true dysglycemia was present, and no dysglycemia when true dysglycemia was not present.

Results: Overall, CANRISK was the most sensitive in identifying those with dysglycemia (CANRISK: 70.57% [70 .54, 70.59] vs. Saudi-specific score: 53.58% [51.36, 55.80] vs. FINDRISC: 39.46% [39.43, 39.49]), but the least specific in detecting those without dysglycemia (CANRISK: 55.87% [55.85, 55.89] vs. Saudi-specific score: 75.54 [73.86, 77.21] vs. FINDRISC: 82.95 [82.83, 82.86]). The PPV was similar among the three risk scores, ranging from 55.2%-63.9%. The NPV was also similar among the three risk scores, ranging from 64.0%-71.2%. The Saudi-specific score had the highest kappa statistic (Saudi-specific score: 0.296, se=0.014; CANRISK: 0.260, se=0.015; FINDRISC=0.233, se=0.014) and demonstrated the best balance in correctly detecting individuals with and without dysglycemia. All three risk scores performed better in identifying individuals with diabetes, compared to identifying individuals with dysglycemia. All three risk scores did not perform as well when detecting undiagnosed dysglycemia and undiagnosed diabetes.

Conclusion : Based on our analysis, the rapid Saudi-specific dysglycemia risk score performs as well as the FINDRISC and CANRISK.

Table of Contents

CHAPTER 1: INTRODUCTION ......................................................................... 1 Introduction: .................................................................................................. 1
Natural History of Type 2 Diabetes Mellitus .......................................................... 1
Type 2 Diabetes in Saudi Arabia ......................................................................... 3
Screening for Type 2 Diabetes ........................................................................... 5
Risk Scores .................................................................................................... 6
Problem Statement .......................................................................................... 9
Purpose Statement: ......................................................................................... 9
Research Questions: ........................................................................................ 10
Significance: .................................................................................................. 10
Definition of terms: ......................................................................................... 11
References: ................................................................................................... 12
CHAPTER 2: COMPREHENSIVE REVIEW OF THE LITERATURE ........................... 16
Measure for validating risk score performance ..................................................... 17
Known risk scores ........................................................................................... 19
References: .................................................................................................... 26 CHAPTER 3: MANUSCRIPT ................................................................................. 29
ABSTRACT ...................................................................................................... 29
INTRODUCTION: .............................................................................................. 30 METHODS ....................................................................................................... 34
Data Source ............................................................................................. 34
Study Design and Data Collection ................................................................. 34
Study Variables ......................................................................................... 35
Data Cleaning ........................................................................................... 37
Known diabetes risk calculators and associated variables ................................. 38
Data Analysis ........................................................................................... 39
RESULTS ........................................................................................................ 40
DISCUSSION: ................................................................................................. 44
Implications and Future Directions ............................................................... 47
REFERENCES: .................................................................................................. 48
CHAPTER 4: CONCLUSIONS AND RECOMMENDATIONS ..................................... 54
TABLES AND FIGURES: ................................................................................... 56
TABLE 1. Criteria for risk scores for screening for dysglycemia using the Saudi Health Interview Survey (SHIS) ................................ 56
TABLE 2. Characteristicsa of SHIS participants that completed HbA1c testing for blood glucose compared to all participants ............... 59
TABLE 3. Performancea of risk scores in identifying dysglycemia among Saudi adults ...................................................................... 60
TABLE 4. Performancea of risk scores in identifying dysglycemia among Saudi adults without diagnosed dysglycemia (diabetes or prediabetes)... 61
TABLE 5. Mean scoresa by gender and diabetes status for each risk score ........... 62
TABLE 6. Performancea of risk scores in identifying dysglycemia among Saudi adults by gender ....................................................... 63
TABLE 7. Performancea of risk scores in identifying diabetes among Saudi adults using dysglycemia risk scores ...................................... 64
TABLE 8. Performancea of risk scores in identifying diabetes among Saudi adults without diagnosed diabetes using dysglycemia risk scores ......... 65

About this Master's Thesis

Rights statement
  • Permission granted by the author to include this thesis or dissertation in this repository. All rights reserved by the author. Please contact the author for information regarding the reproduction and use of this thesis or dissertation.
School
Department
Degree
Submission
Language
  • English
Research field
Palabra Clave
Committee Chair / Thesis Advisor
Committee Members
Partnering Agencies
Última modificación

Primary PDF

Supplemental Files