Study designs can be divided into two basic types
observational and experimental designs.
In observational studies,
the investigator is simply observing and recording what is happening or has happened; there is no treatment or intervention involved.
In an experimental design
there is an intervention, treatment, advice, or procedure involved and the investigator seeks to record the eff ect of this intervention on the outcome of interest.
Observational studies
1. Case report — this describes a single clinical subject, N = 1.
2. Case series — this describes the characteristics of a group of clinical subjects, N > 1
what they have in common
whether these subjects are different from non-diseased people, i.e. there is comparison.
3. Cross-sectional study — addresses the questions of how widespread a disease is and what risk factors are shared by sufferers. A cross-sectional study may provide prevalence but not incidence data, or causality.
The results of cross-sectional studies are usually analysed using Chi square.
4. Case control study — a sample of diseased individuals is compared with disease-free people by examining them retrospectively for the presence or absence of some hypothesized risk factors.
This type of study can provide some insight into causality, but not incidence or prevalence.
The results of case control studies are usually presented as odds ratio (OR).
5. Cohort study — samples of disease-free individuals with a given risk factor and those without the risk factor are followed forward over time for the appearance or not of a disease. Because of its temporal sequence it can indicate causality and incidence but not prevalence.
The results of cohort studies are usually presented as relative risks (RR) and absolute risks (AR).
Experimental study designs
1. Randomized controlled trial (RCT) — the participants are randomly allocated to the study and control groups, reducing the chance of bias.
2. Double-blind RCT — the blinding of the investigator removes his/her expectations in addition to random allocation of the participants.
3. Crossover study (trial) — a longitudinal study in which subjects receive a sequence of different treatments or exposures,
it allows participants to receive treatment at some stage of the study and not be disadvantaged by exposure to placebo only, for example.
About the Author
