Chapter 4 - Study Design Overview and Cross Sectional Studies

Question 1

What are the main categories that studies can be divided into?

Answer 1

Observational and Intervention

Question 2

What can observational studies be divided into?

Answer 2

Descriptive or Analytical

Question 3

What are the general aims of observational studies?

Answer 3

• Describing a disease in a population

- Studying the associations between the occurence of disease and possible causes

Question 4

What are intervention studies typically?

Answer 4

Clinical trials that study the effect of new drug or treatment
They involve asking the question: Does removing the characteristic reduce the probability of developing the disease?

Question 5

What is the aim of a descriptive study?

Answer 5

Aim is to describe the prevalence of a disease and how this varies over time from place to place or by characteristics of the individual (such as age, sex or occupational group)

Question 6

What is an important aspect of descriptive studies?

Answer 6

Identification of the population denominator (either from census data or routine data sources)

Question 7

What are descriptive studies usually based on?

Answer 7

Case series or case registers (vital statistics)

Question 8

Why are descriptive studies useful?

Answer 8

• Suggest clues to the causes of disease
• Enable assessment of the burden (in economic and health terms) of disease in different communities
• Distributions and correlations may be studied between time, place and person

Question 9

Are descriptive studies relevant to the study of causes in disease in populations or individuals?

Answer 9

Populations

Question 10

What is the aim of analytical studies?

Answer 10

They examine associations between the presence of diseases in individuals and populations and potential causative factors

Question 11

Can you infer causality in analytical studies?

Answer 11

No

Question 12

What are three type of questions usually answered in analytic studies and what are their corresponding study designs?

Answer 12

• Have the disease - cross sectional studies
• Develop the disease- cohort disease
• Have a characteristic - case control studies

Question 13

Difference of an analytic study from descriptive study

Answer 13

In analytic study:
Information on controls (those without disease) is as detailed as for cases (those with disease). It is then possible to investigate the reasons why certain individuals become sick within a population.

Question 14

Definition of cross sectional studies

Answer 14

A cross sectional study (prevalence study or survey) focuses on everyone in a define population (or sample of pop) and aims to determine their disease and/or exposure status at THAT POINT IN TIME.

It may be descriptive (assessing burdens of disease or exposure in population)
It may attempt to explain the observed pattern of disease by examining its relationship with possible aetiologic factors.

Often cross sectional studies contains elements of both descriptive and analytical studies.

Question 15

Advantages of cross sectional studies

Answer 15

• Continuous (quantitative) disease outcome can be handled rather than yes/no outcomes
• Case definitions independent of standard health systems can be used so all diseases can be recorded
• Disease prevalence measured (burden)
• Multiple causes of disease can be studied
• Multiple consequences of a given exposure can be studied

Question 16

Disadvantages of cross sectional studies

Answer 16

• Not suitable for rare disease
• Not suitable for disease with short duration
• Statistically inefficient for studying a single disease
• May be misleading if exposures change with disease over time
• Can be difficult to tell whether exposure affects duration or incidence

Question 17

Define

Study Sample
Study Population
Target Population

Answer 17

The study sample is the group on which measurements are made. It should be representative of the study population.

The study population is the group from which study sample is drawn and ideally should be typical of target population.

The target population is the wider population to whom the research might apply.

Question 18

Can you infer from the study sample to the study population?

Answer 18

Yes
Provided the sample is representative

To achieve this take a random sample of study population

Question 19

Sample size

Answer 19

Should be large enough to estimate the condition of interest with reasonable precision in an prevalence survey

Question 20

Two components of non response bias

Answer 20

• Proportion of subjects approached did not participate (non response rate)
• How do non respondents differ from respondents?

Question 21

What is standardised methods of data collection?

Answer 21

Repeatable: it gives same answer when the subject is re-examined

Valid: Measures what it purports to measure

Question 22

In an descriptive study, what is the usual measure of disease definition?

Answer 22

Point prevalence

Question 23

Point prevalence (P)

Answer 23

No. people with disease at time of survey/ no. people in survey x 100%

Question 24

How is prevalence calculated for continuous variables?

Answer 24

• Only possible if results split into above and below a cut off
• or mean and median values used

Question 25

How can strength of association between risk factorand outcome be measured in analytical study?

Answer 25

Risk ratio or odds ratio

Question 26

What is the risk ratio/odds ratio

Answer 26

Ratio of risk or odds of exposure to a risk factor in those with the outcome compared to those without

Question 27

Extensions of cross sectional studies

Answer 27

Multicentre studies
Serial prevalence surveys - carried out in two or more points in time
Case control studies-groups of disease and non diseased subjects are selected and compared
Cohort studies-groups or exposed or not exposed followed up over time