Study Designs

Question 1

Observational studies- descriptive

Answer 1

Routine data, case reports, case series, intended to determine the distribution of disease in a population
Can help find the level of disease or risk factors in a given population
Used to develop public health interventions eg the national nutrition survey

Question 2

Observational studies- analytic

Answer 2

Test one or more hypotheses about the relationship between risk factors and diseases
Cross sectional

Ecological (correlational)

Case control

Cohort

Question 3

Experimental

Answer 3

Rct 
Tightly controlled stud environment limit influences of external factors 
Requires random allocation 
Reduces confounding 
Controlled 
Blinded 
Presentation of results- relative risk 

Advantages:
Closest to true experiment
Groups randomly allocated minimise confounding
Investigating team allocated treatment, not the individual choosing what they will have

Disadvantages:
Ethical issues
Exposing people to experiment needs strong evidence for justification
Time consuming, labour intensive

Question 4

Ecological (correlational)

Answer 4

examine frequency with which a study factor (eg. High fat diet) and an outcome (eg heart disease) occur in a geographically defined population.

Adv: quick, inexpensive

Disad: inability to link exposure with disease, inability to control for confounding, very poor evidence of cause and effect

Question 5

Cross sectional

Answer 5

Sample of population is selected and information is obtained at one point in time
Questionnaires and examinations and investigations
Most common output: prevalence

Adv: measures prevalence of exposure and disease in defined population
Usually cheap and inexpensive
Good for exposures which never change eg blood group
Often used for initial exploration of hypothesis
Can investigate a number of risk factors and outcomes at the same time
No loss to follow up

Disadv: 
Greater need for representative data 
Subject to selection bias
Limited evidence of cause and effect 
No info on temporal relationship between exposure and outcome

Question 6

Case control study

Answer 6

Investigators collect data in a group of subjects with the outcome of interest, as well as in a group of subjects without the income of interest
Subjects in both groups are compared, also referred to as a retrospective design. Subjects with the outcome are cases while those without the outcome are controls

Adv: 
Ability to study rare diseases 
Quick and less costly to perform 
Requires fewer subjects 
Best design to use existing historical data pertaining to both outcome and exposures 
Can stuffy different hypothesis 

Disadv:
Information bias is big problem
- accurate memory
- participants may provide incorrect info
- outdated medical record
- some outcomes require more definitive assessment whennudrd as a research outcome compared to diagnostic

Question 7

Cohort study- gold standard

Answer 7

Prospective follow up,
Following outcome free subjects into the future to adcertain outcome
Compare incidence of disease between exposed and unexposed
Preservation of temporal relationship between exposure and outcome- good evidence of cause and effect

retrospective investigation
All disease and exposure status collected before study begins
Designed to compare those with the exposure of interest to those without the exposure of interest
Must exclude potential subjects who already have the outcome
Investigators watch subjects to see who develops the outcome
Incidence rates of outcome in both groups are compared when study is over

Presentation of results
Relative risk= rate of outcome in exposed group/ rate of outcome in unexposed group

Question 8

Cohort study adv and disadv

Answer 8

Adv
Accurately reflects incidence data
Because incidence is necessary to determine risk, the cohort study design is the design hat allows the identification of risk and risk factors for the outcome
Data collected are generally the highest quality of any of the observational designs
Using a cohort study allows investigators to easily introduce an intervention into the design of the opportunity arises

Disadv
Inefficient for rare diseases
Expensive and time consuming, labour intensive
Validity affected by loss to follow up (rule of thumb need >80%)
Methodology must be developed to minimise risk of loss
Sufficient numbers should be enrolled so that the expected number of loss can be accommodated without damaging the study