Screening for Type 2 Diabetes

1
Screening for Type 2 Diabetes

• Charlotte Glümer
• STAR-course
• Epidemiology and Biostatistics
• Aurangabad 2005

2
What is screening

• Screening for a disease is defined as an
  examination of asymptomatic persons to classify
  them as likely or unlikely to have the disease.
• Persons who appear likely to have the disease are
  further investigated for a final diagnosis.
• Those who are found to have the disease are then
  treated.
• The goal of screening is to reduce morbidity and
  mortality from the disease by early treatment and
  control in the detected individuals.

Morrison 1985
3
Types of screening

• Mass screening
• Selective screening
• Opportunistic case-finding

4
Criteria for Screening (Wilson and Jungner, WHO,
Geneva 1968, Publ. H. Papers 34)

• Important health problem
• Natural history should be known
• An effective and accepted treatment should be
  available
• Facilities for diagnosis and treatment should be
  available
• Diagnosis in latent stage should be possible and
  important
• A reliable and harmless screening test should be
  available
• The test should be acceptable for the population
• The cost of screening should be balanced against
  the health
• care system
• Screening should be a continuous process, not a
  one-time event

5
Obesity Trends Among U.S. Adults BRFSS, 1991,
1995 and 2000
(BMI ? 30, or 30 lbs overweight for 54
person)
No Data lt10 10-14
15-19 ?20
Source Mokdad A H, et al. JAMA 199928216,
200128610.
6
Obesity Trends Among U.S. AdultsBRFSS, 2001
Source Mokdad A H, et al. J Am Med Assoc
199928216, 200128610.
7
Diabetes and Gestational Diabetes Trends Among
Adults in the U.S., BRFSS 1990, 1995 and 2001
Mokdad AH, Ford ES, Bowman BA, et al. Prevalence
of obesity, diabetes, and other obesity-related
health risk factors, 2001. JAMA 2003 Jan
1289(1).
8
The Prevalence in India

age
Ramachandran, Diabetologia 2001, 1094-1101
9
CVD and Risk Factors
CVD death rate and number of risk factors
(cholesterol, blood pressure, smoking), age
adjusted (per 10.000 person years)
Numbers of risk factors
MRFIT, Diabetes Care 1993 2 434-444
10
Questions to be answered

• Is previously undiagnosed diabetes a problem?
• Does it have any implications to be undiagnosed?
• Can we find/detect individuals with previously
  undiagnosed diabetes?
• Should we find them?

11
Questions to be answered

• Is previously undiagnosed diabetes a problem?
• Does it have any implications to be undiagnosed?
• Can we find/detect individuals with previously
  undiagnosed diabetes?
• Should we find them?

12
Diabetes prevalence in Inter99


Glümer C.Prevalences of Diabetes and Impaired
Glucose Regulation in a Danish Population Diabetes
Care 2003
13
2 minutes
14
Questions to be answered

• Is previously undiagnosed diabetes a problem in
  Denmark? YES
• Does it have any implications to be undiagnosed?
• Can we find/detect individuals with previously
  undiagnosed diabetes?
• Should we find them?

15
Undiagnosed Diabetes is not a benign condition
DECODE Study, Lancet 1999
16
Risk factors in individuals with diabetes

Inter99
17
SDM vs. NDM

• 195 SDM
• Hypertension m70, f81
• Dyslipidaemia
• m23 f18
• Mean HbA1c 6.3
• MI 13
• IHD40

• NDM (60)
• Hypertension m47, f70
• Dyslipidaemia
• m23 f21
• Mean HbA1c 8.5
• MI 3
• IHD24

Spijkerman Journal of Internal Medicine 2004
256429-436
18
Questions to be answered

• Is previously undiagnosed diabetes a problem in
  Denmark? YES
• Does it have any implications to be undiagnosed?
  YES
• Can we find/detect individuals with previously
  undiagnosed diabetes?
• Should we find them?

19
Screening

• Screening for a disease is defined as an
  examination of asymptomatic persons to classify
  them as likely or unlikely to have the disease
  (Morrison 1985).
• The diagnosis is based on an OGTT
• Time consuming, inconvenient for patients and
  health care system

20
Screening tests

• Risk scores
• Measurements hyperglycaemia
• Blood glucose measurements
• random capillary blood glucose
• fasting plasma glucose
• HbA1c

21
Risk Scores

• A Risk Score
• Is a useful method to identify people at
  increased risk of having T2DM
• Reduces the number of people requiring testing
  for undiagnosed diabetes
• Achieves acceptable sensitivity, specificity,
  predictive values for detecting undiagnosed T2DM
• Can be used in a stepwise screening strategy
• Probably only used in Caucasians

22
Risk Scores
23
Performance of a Risk Score
24
Screening Tests

• Risk scores
• Measurements hyperglycaemia
• Blood glucose measurements
• random capillary blood glucose
• fasting plasma glucose
• HbA1c

25
Random Blood Glucose

• Pros
• Not fasting
• Can be taken the whole day
• If 11.1 mmol/l ? diagnostic
• Cons
• Gender specific cut-offs
• Dependent on post prandial time

26
Fasting Plasma Glucose

• Pros
• Is the most optimal singular test according to
  performance, sensitivity 79, specificity 88
  and 13 subsequent testing
• Relatively simple
• Part of the diagnostic test
• Cons
• Fasting

27
HbA1c

• Pros
• Not fasting
• Easy and acceptable
• Cons
• Price
• Lack of standardisation? comparison problematic
• Distribution in the population

28
Glucose Tolerance vs. HbA1c
Literature Sensitivity 65-95 Specificity
67-98
12.1
0.06
29
Why Stepwise Screening?
ICER Incremental Cost Effectiveness Rate

• Stepwise screening is cheaper than mass screening
• It is important to lower the expense of the first
  step

Glümer et al Diabetic Medicine 2004 874-880
30
2 minutes
31
Questions to be answered

• Is previously undiagnosed diabetes a problem in
  Denmark? YES
• Does it have any implications to be undiagnosed?
  YES
• Can we find/detect individuals with previously
  undiagnosed diabetes? YES
• Should we find them?

32
Three key questions

• Effective and accepted treatment for screen
  detected type 2 diabetes?
• Is screening accepted in the population?
• Economical and organizational implications of
  screening?

33
Intervention studies in Type 2 DM
Effect of different interventions on clinical (c)
or surrogate (s) endpoints in Type 2 diabetic
patients
post-hoc analyses in type 2 diabetic patients
from the original cohort
34
Is treatment effective?

• Clinically diagnosed versus screen detected
• Steno-Type-2 T2DM and micro alb, 6 years of
  duration
• 4S, HPS, HOT pts had complications
• UKPDS Clin diagnosed i.e. duration 5-12y.
• No RCT in screen detected individuals

35
Is treatment effective?

• Are the results applicable to screen
  detected individuals
• Very low prevalence of retinopathy, nephropathy
• retinopathy 3-7
• microalbuminuria 6-17
• How should we treat them ?

36
Treating known diabetic patients or screening for
new cases?

• Treatment
• HbA1c too high
• lack of facilities for treatment
• lack of funding for treatment
• lack of funding for complication screening and
  treatment

• Screening
• With these problems should we ask for more?
• Would we dilute funding

37
Three key questions

• Effective and accepted treatment for screen
  detected type 2 diabetes?
• Is screening accepted in the population?
• Economical and organizational implications of
  screening?

38
Is Screening Accepted In the Population?

• Population based screening
• No
• High risk screening strategies
• Hoorn screening study
• 11.000 invited (50-75 yr)

Total population
80
Questionnaire
87
FBG
89
OGTT
Diabetes Care 20021784-1789
39
Personal costs

• Fear, anxiety, stress
• How big is the problem
• How frequent
• How to prevent it
• Systematic research needed

40
Potential Harms

• Harm and anxiety
• 1253 outpatients (aged 45-64 yr), without
  diabetes were screened for diabetes
• Measure of health related quality of life by
  Questionnaires using SF36, HRQol-score
• Baseline score and 1 yr after screening
• Did not find any find any differences in the
  score between the individuals with DM compared to
  individuals without diabetes

Edelman Diabetes Care 20021022-1026
41
Quality of life in SDM vs. NDM

• SDM (116)
• T2DM symptom score 0.24
• Mental Component Score 53.5
• General well being 27.5
• No difference over time

• NDM(49)
• T2DM symptom score 0.56
• Mental Component Score 48.5
• General well being 24.8
• improved within 12 months

Adriaanse et al. Diabetic medicine 2004 1075-1081
42
Three key questions

• Effective and accepted treatment for screen
  detected type 2 diabetes?
• Is screening accepted in the population?
• Economical and organizational implications of
  screening?

43
Economic burden of screening

• Is not examined
• A simulation model from U.S have shown that
  screening for diabetes may be beneficial for
  development of microvascular diseases. Risk
  reductions used are from RCTs on known diabetics
  ? not applicable to screen detected
• No economic models on the cost-effectiveness on
  CVD

44
Economic consequences of polypharmacy

• Current annual cost 0.5 billion d.kr
• Implementation of systematic polypharmacy
  additional annual cost of 1.7 billion d.kr
• Implementation of systematic polypharmacy in
  screen detected individuals causes an annual cost
  of 3-4 billion d.kr

45
2 minutes
Shall we recommend screening?
46
Shall we implement screening ?

• Mass- Populationbased screening
• High Risk screening Stepwise
• Case- finding

47
(No Transcript)