Diabetes and Risk Factors How to conduct prevalence studies Dhaka, August 1, 2005

1
Diabetes and Risk FactorsHow to conduct
prevalence studiesDhaka, August 1, 2005

• A. Samad Shera, TI, SI, FRCP
• Honorary President, International Diabetes
  Federation (IDF)
• Director, WHO Collaborating Centre
• Member, WHO International Expert Advisory Panel
  on Diabetes
• Member, European Expert Committee on Diabetes in
  Immigrants to Europe
• National Coordinator for Diabetes Control
• Secretary General, Diabetic Association of
  Pakistan
• Member Syndicate, Liaquat University of Medicine
  and Health Sciences, Hyderabad - Sindh

2
(No Transcript)
3
The most valuable of all talents is that of never
use two words when one will do  
Thomas Jefferson
4
A-PLANNING THE STUDY

• A number of steps are required from
• the moment of decision to undertake
• a study, until the practical preparation
• can commence

5
B-PREPARING THE STUDY

• a)
• b)
• c) Selecting the survey site Choose a central
  survey site , to which a predermined number of
  subjects are invited each day,
• Characteristics of a survey site
• 1- Indoors or Shamians
• 2- Adequate waiting -------
• 3- Simple flow of subjects
• 4- Privacy for anthropometric measurements.
• 5- Quiet environment for BP measurement

6
C- PREPARING FOR THE STUDY

• a) Obtaining Approval
• b) Choosing the team leader
• c) Selecting the study site
• d) The Pre-Study census
• e) Preparing the chosen subjects for the study
• f) Team training
• g) Quality assurance

7
D- CONDUCTING THE STUDY

• a) Study Procedures
• 1) Registration
• 2) Fasting blood sample
• 3) Glucose drink
• 4) Two hour blood sample
• b) Anthropometry
• c) Questionnaire
• d) Blood pressure

Continued
8
Continued
e) Final assessment f) Non response g) Over
response h) Providing general health care I)
Providing feed back j) Transport of specimens
9
E- SPECIFICATION OF DATA TO BE COLLECTED

• a)
• Collect data related to the topics of immediate
  interest
• Resist to collect opportunistic data which
  serve no obvious purpose

10

• 1) Core data Include basic demographic
• information, medical and family history,
• lifestyle factors, key physical parameters
• (BP,GT,Anthropometry and blood lipids)
• 2) Optimal data Addition material which might
  include dietry habits, genetic markers, evidence
  of disease complications etc

11
b) Specification of study procedure c)
Specification of Laboratory procedures d) Choice
of statistical methods e) Selecting the most
appropriate range f) Preparing the survey form
(example of our survey form for National
Survey)
Continued
12
Continued

• g) Selecting the survey sample
• a) Simple random sampling
• b) Multistage sampling
• c) Cluster sampling
• h) Sample size determination
• i) Choosing the team size and
• composition
• j) Preparing a written protocol

13
F- HANDLING THE DATA

• Data Entry
• a) Data entry / editing
• b) Data verification
• Data Analysis
• a) Create variable
• b) Run Analysis
• c) Create Charts

14
G-PREPAING THE SURVEY FORM

• Examiner enters into appropriate box the
  numerical code of the correct response
• Almost all coded responses are to be found at the
  right hand margin of this form
• The form only includes key item and is short
  enough to fit on two sides of a single sheet of
  paper

15
H-SELECTING THE SURVEY SAMPLE

• Simple Random Sampling Subjects normally
  recruited from a list such as recent electoral
  role. The method is labour intensive
• Multistage Sampling The community is first
  divided into subgroups on the basis of ethnicity,
  ----------, socio-economical level, age group or
  sex. Multistage sampling is useful when over
  sampling under representative groups is required
  e.g, a minority population or elderly age groups
• Cluster Sample Simplest method in Procter

16
I- PREPARING THE REPORT

• a) Introduction
• b) Survey methods
• c) Response
• d) Results
• e) Discussion
• f) Conclusions
• g) Tables and figures