USE AND VALIDATION OF RISK SCORES IN SCREENIHG FOR TYPE 2 DIABETES MELLITUS IN A RURAL LOCAL GOVERNMENT AREA OF RIVERS STATE

USE AND VALIDATION OF RISK SCORES IN SCREENIHG FOR TYPE 2 DIABETES MELLITUS IN A RURAL LOCAL GOVERNMENT AREA OF RIVERS STATE

Abstract

Type 2 diabetes mellitus accounts for 90% of all cases of diabetes mellitus in sub-Saharan Africa. The situation is aggravated as 50% of cases are un-diagnosed.The aim of this study was to use and validate a diabetic risk score for screening type 2 diabetes mellitus.
This cross-sectional study carried out between March and May 2013 involved 600 respondents aged between 24 and 64years. A multi-stage sampling method was used to select a total of 41 house-holds from each of the 15 selected clusters studied in Ikwerre Local Government Area. The WHO STEPwise approach consisting of interviewer administered questionnaires, anthropometric measurements, blood pressure, and blood sugar estimates was used to assess risk factors for diabetes. Diabetes was diagnosed as plasma glucose of > 11.1 mmol/l.
Age ≥ 45 years (p < 0.05), waist circumference ≥ 90 cm (p < 0.05), BMI ≥ 35 kg/m2 (p< 0.05), systolic blood pressure ≥ 140mmHg (p < 0.05), and diabetes in first degree relatives (p< 0.05) were associated with type 2 diabetes mellitus. Further analysis using logistic regression provided the predictors for type 2 diabetes mellitus. Compared with respondents age < 45 years, the odd of type 2 diabetes mellitus in those age ≥ 45 years was 2.645(95% CI 1.53-4.58). The odd of type 2 diabetes mellitus in those with body mass index ≥ 35kg/m2 compared to those with lower values was (2.500 [1.04 -6.03] ). The odd of type 2 diabetes mellitus in first degree relatives with the condition compared to those with no family history was (2.828 [1.42- 5.62] ), while respondents with waist circumference ≥ 90cm compared to those with < 90cm had an odd of 2.208(1.37 -3.52). These were the only predictors for type 2 diabetes mellitus.
The sensitivity and specificity were 53.3% and 76.8% respectively. The positive and negative predictive values were 29.4% and 90.1% respectively. The likelihood ratios of positive (LR+) and negative test (LR-) were 2.3 and .61 respectively (results obtained at a cut-off point of 7.5).The area under the curve for the study model (.70) was comparable to

that of the American Diabetes Risk Score (.65) and Trinidad Diabetes Questionnaire (.67). The sensitivity and specificity for American diabetes risk score and Trinidad diabetes questionnaire were (41.3%; 79.9%) and (38%; 88.6%) respectively.
This study showed that a diabetes risk score made of age ≥ 45 years, waist circumference ≥ 90 cm, BMI ≥ 35 kg/m2, and diabetes in first degree relatives may be useful in screening for type 2 diabetes mellitus. The risk score derived from this study is recommended for use at primary health care level, while further studies are recommended to validate the risk score as well as derive a national version.

Title Page———i

Certification——–ii

Dedication———iii

Acknowledgement——-iv

Abstract ———vi

Table of Content——–vii

 

Chapter One

1.0 Introduction ——-1

1.1 Statement of Problem——4

1.2 Purpose of the Study——5

1.3 Significance of Study——8

1.4 Limitation——–9

1.5 Scope of Study——-11

 

Chapter Two

2.0 Review of Related Literature —-12

2.6 Summary of Literature Review—- 19

 

Chapter Three

3.0 Research Methodology and Procedure—22

3.1 Population ——–22

3.2 Sample and Sampling Technique—-22

3.3 Validation of the Instrument —-23

3.4 Reliability of the Instrument —–23

3.5 Data Analysis——-23

 

Chapter Four

4.0 Presentation and Discussion of Result—24

4.1 Analysis and interpretaion of Data—25

4.2 Discussion of Results——38

 

Chapter Five

5.0. Summary, Conclusion, and Recommendation  –40

5.1 Summary——–40

5.2 Conclusion——–41

5.3 Recommendation——42

References ———45

Appendix 1——–47

Appendix ———50

USE AND VALIDATION OF RISK SCORES IN SCREENIHG FOR TYPE 2 DIABETES MELLITUS IN A RURAL LOCAL GOVERNMENT AREA OF RIVERS STATE

 

Sharing is caring!