Snap shot

1
Snap shot

• Cross-sectional surveys
• FETP India

2
Competency to be gained from this lecture

• Design the concept of a cross-sectional survey

3
Key areas

• The concept of a survey
• Planning a survey
• Analytical cross-sectional studies

4
Definition of a survey

• Oxford English Dictionary
• The act of viewing/examining / inspecting in
  detail especially for some specific purpose
• Merriam Webster Online Dictionary
• To query (someone) in order to collect data for
  the analysis of some aspect of a group or area
• Abrahmson
• An investigation in which information is
  systematically collected, but in which the
  experimental method is not used

Concept of survey
5
Survey The epidemiological concept

• Observation of a cross-section of a population at
  a single point in time
• Unit of observation and analysis The individual
• Usually conducted to collect information about
  prevalence
• Also known as prevalence studies
• No independent reference groups
• May be repeated
• Surveillance of risk factors for cardio-vascular
  diseases

Concept of survey
6
Examples of research questions that can be
addressed through a survey

• What is the prevalence of hypertension in
  Chennai?
• What is the prevalence and distribution of known
  risk factors for cardio-vascular diseases in
  rural Tamil Nadu?
• How satisfied are patients attending government
  hospitals in Chennai?

Concept of survey
7
Uses of cross-sectional surveys in public health

• Estimate prevalence of disease or their risk
  factors
• Estimate burden
• Measure health status in a defined population
• Plan health care services delivery
• Set priorities for disease control
• Generate hypotheses
• Examine evolving trends
• Before / after surveys
• Iterative cross-sectional surveys

Concept of survey
8
The place of surveys among other study designs

• Observational
• Not interventional
• Cross-sectional in logistic
• The logic maybe cross-sectional or retrospective

Concept of survey
9
Collection of information on prevalence during a
survey

• Disease
• Exposure to potential risk factors
• Practices
• Dietary intake
• Costs and utilization of health care services
• Healthy / unhealthy behaviours
• Physiologic measurements

Concept of survey
10
Collection of information on incidence during a
survey

• The logistic of the survey is always
  cross-sectional
• The logic maybe retrospective to estimate
  retrospective incidence
• Village visit to estimate the retrospective
  incidence of measles
• Retrospective cohort the day following a food
  poisoning

Concept of survey
11
Important considerations in planning
cross-sectional surveys

1. Study objectives
2. Study population
3. Analysis plan
4. Information to collect
5. Data collection methods
6. Sampling methods
7. Sample size
8. Data recording and processing

Preparing a survey
12
1. Potential objectives of a cross-sectional
study

• Descriptive
• Estimate prevalence
• Analytic
• Compare the prevalence of a disease in various
  subgroups, exposed and unexposed
• Compare the prevalence of an exposure in various
  subgroups, affected and unaffected

Preparing a survey
13
2. Populations that may be studied with a survey

• General
• District survey
• National survey
• Specific
• School survey
• Institutional survey
• Populations with specific behaviours (e.g.,
  injection drug users) or characteristics (e.g.,
  diabetic patients)

Preparing a survey
14
3. Analysis plan for a survey

• Define the indicator needed
• Identify the information needed to calculate the
  indicator
• Example Dental caries indicators require
  information on
• Number of permanent teeth decayed
• Number of teeth missing
• Number of teeth filled

Preparing a survey
15
4. Information to collect Operational
definitions

• Need precision to reduce inter-observation
  variability
• Examples of definitions of obesity
• A weight, in under clothes without shoes which
  exceeds by 10 or more of standard weight for
  age, height in a specified population
• Sex and a skin fold thickness of 25mm or more,
  measured with a Harpenden skinfold caliper at
  the back of the right upper arm, midway between
  the tip of the acromial process and tip of the
  olecranon process

Preparing a survey
16
5. Data collection methods during a survey

• Interviews
• Phone interview, direct interview
• Record reviews
• Medical records for nosocomial infection survey
• Structured observations
• E.g., Health care facility surveys to describe
  health care delivery
• Measurements (e.g., WHO STEPWISE approach)
• Anthropometry (e.g., height and weight)
• Biological measurements (e.g., blood tests)

Preparing a survey
17
6. Sampling strategies during a survey

• Simple random sampling
• Sampling frame available
• Study participants selected at random
• Systematic sampling
• Sampling frame organized sequentially
• Selection of every nth individual
• Cluster sampling
• Selection of clusters / communities with a
  probability proportional to population size
• Selection of an equal number of individuals
  within each cluster / community

Preparing a survey
18
Example of simple random sampling
Numbers are selected at random
19
Example of systematic sampling
Every eighth house is selected
20
Example of cluster sampling
Section 2
Section 1
Section 3
Section 5
Preparing a survey
Section 4
21
7. Sample size for a survey

• Use formula / software
• Parameters
• Expected frequency
• Prevision
• Confidence level
• Need to double sample size if comparisons
  required
• Before / after
• Exposed / unexposed (analytical survey)

Preparing a survey
22
8. Data recording /processing

• Establish the structure of the database
• Unique level
• Multiple levels
• Village
• Household
• Individual
• Set up relational link between databases if
  required

Preparing a survey
23
Example of survey results
Anemia and use of iron/ folic acid (IFA) tablets
among pregnant women, Dhenkanal district,
Orissa, India, 2004
Number Total Percentage
Hb lt 11 g/dl 285 456 63
IFA coverage according to health workers 100 days 404 456 89
IFA coverage according to health workers lt 100 days 52 456 11
IFA consumption according to women 90 days 382 456 84
IFA consumption according to women lt 90 days 74 456 16
Preparing a survey
24
Advantages of cross-sectional surveys

• Fairly quick
• Easy to perform
• Less expensive
• Adapted to chronic / indolent diseases

Preparing a survey
25
Limitations of cross-sectional surveys

• Limited capacity to document causality (exposure
  and outcome measured at the same time)
• Not useful to study disease etiology
• Not suitable for the study of rare / short
  diseases
• Not adapted to severe / acute diseases
• NEYMAN BIASE
• Not adapted to incidence measurement

Preparing a survey
26
Presentation of the data of an analytical
cross-sectional study in a 2 x 2 table
Ill Non-ill Total Exposed a b ab Non-exposed c
d cd Total ac bd abcd
Analytical surveys
27
Limitations of causal inference in analytical
cross-sectional studies

• Prevalent cases
• Exposure and outcome examined at the same time

Analytical surveys
28
Measuring association in analytical
cross-sectional surveys

• Prevalence ratio
• Prevalence among exposed / prevalence among
  unexposed
• Formula equivalent to risk ratio
• Concept different
• No incidence
• Only prevalence

29
Take home messages

• Surveys are a snap shot that can look back or
  compare to generate hypotheses
• Surveys require careful preparation and detailed
  protocol writing
• Analytical cross-sectional surveys require (1)
  double sample size and (2) caution in
  interpretation