Screening, diagnosis and classification of diabetes

1. Screening, diagnosis and classification of diabetes A. Prof Jonathan Shaw Associate Director Baker IDI Melbourne

2. Identifying people at risk of developing type 2 diabetes • Test everyone – mass screening • Test people who have risk factors for diabetes • Undertake large-scale, non-invasive, self completed screening, with a validated tool, followed by blood tests for those at high-risk • FINDRISK • AUSDRISK • AUSDRISK • Self-completed risk score • Developed and validated on Australian data • Calculates 5-yr risk of developing diabetes

4. 1. Your age group? Under 35 years 0 pts 35 – 44 years 2 pts 45 – 54 years 4 pts 55 – 64 years 6 pts 65 years or over 8 pts 2. Your gender? Female 0 pts Male 3 pts

5. 3. Ethnicity/Country of birth: 3a. Are you of Aboriginal, Torres Strait Islander, Pacific Islander or Maori descent? No 0 pts Yes 2 pts 3b. Where were you born? Australia 0 pts Asia 2 pts Mid-East, N Africa 2 pts S Europe 2 pts Other 0 pts

6. 4. Have either of your parents, or any of your brothers or sisters been diagnosed with diabetes (type 1 or type 2)? No 0 pts Yes 3 pts 5. Have you ever been found to have high blood glucose (sugar) (e.g. in a health examination, during an illness, during pregnancy)? No 0 pts Yes 6 pts

7. 6. Are you currently taking medication for high blood pressure? No 0 pts Yes 2 pts 7. Do you currently smoke cigarettes or any other tobacco products on a daily basis? No 0 pts Yes 2 pts

8. 8. How often do you eat vegetables or fruit? Everyday 0 pts Not everyday 1 point 9. On average, would you say you do at least 2.5 hours of physical activity per week (eg 30 minutes a day on 5 or more days a week)? Yes 0 pts No 2 pts

9. 10. Your waist measurement taken below the ribs (usually at the level of the navel)? • For those of Asian or Aboriginal or Torres Strait Islander descent: • Men Women • < 90 cm < 80 cm 0 pts • 90 – 100 cm 80 – 90 cm 4 pts • >100 cm > 90 cm 7 pts • For all others: • Men Women • < 102 cm < 88 cm 0 pts • 102 – 110 cm 88 – 100 cm 4 pts • > 110 cm > 100 cm 7 pts

10. Your risk of developing type 2 diabetes within 5 years: ≤ 5: Low risk Approximately one person in every 100 will develop diabetes. 6-14: Intermediate risk For scores of 6-8, approximately one person in every 50 will develop diabetes. For scores of 9-14, approximately one person in every 20 will develop diabetes. 15 or more: High risk For scores of 15-19, approximately one person in every seven will develop diabetes. For scores of 20 and above, approximately one person in every three will develop diabetes.

11. Diagnostic criteria for diabetes

12. Diagnostic thresholds should be defined by • Their association with clinically meaningful abnormalities • Level above which intervention is effective • Associations with intermediate (metabolic) disturbances • Normal limits of a healthy population • mean + 2SD • 9?th percentile • Bimodal distribution

13. Cut-points for diabetes based on • Identifying a glucose threshold for the presence of complications • Bi-modal distribution of blood glucose

14. Cut-points for diabetes based on • Identifying a glucose threshold for the presence of complications • Bi-modal distribution of blood glucose

15. 0 Relative risk of CVD mortality by 2-hr glucose - DECODE 6 5 4 3 2 1 0 Relative risk >14.5 <3.0 Normal Diabetes IGT 0 2 4 6 8 10 12 14 16 2-hour plasma glucose (mmol/l)

16. Pima 0 Range of FPG thresholds 7.6-12.5 Egypt 7.2-9.9 US - NHANES 6.7+ ADA. D Care 1997;20:1183-97

17. 0 Limitations of associations with complications • Thresholds have been based on micro- not macrovascular disease • Estimates of threshold values are imprecise (variation between populations, wide limits of deciles) • All data are cross-sectional – longitudinal analyses would be likely to give lower cut-points • ‘Diabetic retinopathy’ occurs at non-diabetic glucose levels • Studies need to have more cases of retinopathy, and be able to use more severe levels of retinopathy

18. HbA1C for DIAGNOSIS PRO Stable Time averaged Reproducible Fasting not required CON Standardisation essential Expensive Not freely available Problems with anaemia, haemoglobinopathies Poor QA schemes in many countries Few data available Cutpoint uncertain

19. Prevalence of Sickle cell disorders

20. PLASMA GLUCOSE for DIAGNOSIS PRO DM is disorder of raised glucose Time honoured Much data Allows international comparisons Accurate assay (?) CON Based on cross-sectional data in relatively small numbers of studies Major pre-analytical problems OGTT required – FPG inadequate Often poor QA

21. Recommends using HbA1c as the preferred diagnostic test for diabetes at a cut-point of 6.5% Diabetes Care July 2009

22. Classification of diabetes

23. WHO reports on classification 1965, 1980, 1985, 1998, 2006

24. POSSIBLE 2009 WHO CLASSIFICATION OF DIABETES MELLITUS • Type 1 diabetes (-cell destruction) • Type 2 diabetes • Other specific types • Gestational diabetes mellitus • Undefined ADA, WHO, 1997

25. Classic Onset Type 1 Age of onset Usually under 30 Usually 30 - 60 Usually over 30 years (except for MODY) Present-ation Rapid onset of thirst, polyuria, weight loss Gradual onset of milder symptoms Often asymptomatic; may present with complications or gradual onset of symptoms. 85% obese. Part of Metabolic Syndrome. Ketonuria Usually present May be absent Absent (except with severe stress eg infection, infarction) Anti-GAD antibodies Present in approx 80% at diagnosis C-peptide level Low or absent, but may be low-normal initially (remission phase) Normal-high (ie hyperinsulinemia) Treatment Insulin required urgently to prevent ketoacidosis Insulin required, but not urgently Healthy eating and exercise, may require oral agents, and/or insulin later Classic Onset Type 1 Slow Onset Type 1 (LADA) Type 2 Absent

26. Summary • Screening • Begin with non-invasive score • Blood testing in those with a high score • Diagnosis • HbA1c may become an accepted test • Clinical diagnosis requires confirmation with a 2nd test • Classification • Differentiation between type 1 and 2 increasingly difficult • For most patients, classification, and need for insulin can be determined on simple, clinical criteria

27. Summary • Screening • Begin with non-invasive score • Blood testing in those with a high score • Diagnosis • HbA1c may become an accepted test • Clinical diagnosis requires confirmation with a 2nd test • Classification • Differentiation between type 1 and 2 increasingly difficult • For most patients, classification, and need for insulin can be determined on simple, clinical criteria

28. Summary • Screening • Begin with non-invasive score • Blood testing in those with a high score • Diagnosis • HbA1c may become an accepted test • Clinical diagnosis requires confirmation with a 2nd test • Classification • Differentiation between type 1 and 2 increasingly difficult • For most patients, classification, and need for insulin can be determined on simple, clinical criteria

29. Summary • Screening • Begin with non-invasive score • Blood testing in those with a high score • Diagnosis • HbA1c may become an accepted test • Clinical diagnosis requires confirmation with a 2nd test • Classification • Differentiation between type 1 and 2 increasingly difficult • For most patients, classification, and need for insulin can be determined on simple, clinical criteria

30. Some things are more complicated than screening and diagnosing diabetes!