A Spotters Guide Til Study Designs

EBN notebook
A spotter’s guide to study designs

When searching for evidence to answer our clinical questions, studies are experimental or observational. The list is not
the ability to rapidly recognise different types of studies is exhaustive but covers most basics designs.
helpful for finding the one that best answers the question. Our first distinction is whether the study is analytic or non-
‘‘Levels of evidence’’ tables make suggestions for which design is analytic. A non-analytic theoretical study tries to explain and
best for which type of question. For instance, you would develop our understanding of processes that occur for partici-
naturally consider a randomised controlled trial as the most pants. Theoretical studies are qualitative and usually are based
appropriate study design for intervention decisions. But for on grounded theory. A non-analytic descriptive study does not try
potential harms of interventions, we may need case–control to quantify the relation between factors but tries to give us a
studies. And for aetiology, we often need to use cohort studies: picture of what is happening in a population (eg, prevalence,
you wouldn’t randomise someone to cigarette smoking to see if incidence, or experience of a group). Descriptive studies include
they did worse—that would also be unethical. But you would case reports, case-series, qualitative studies, and surveys (cross-
want investigators to follow up cigarette smokers and non- sectional studies), which measure the frequency of several
smokers for a long time, just as Richard Doll did.1 factors, and hence the size of the problem. They may sometimes
This short Notebook is a brief guide to the different study also include analytic work (comparing factors—see below).
types and their advantages and disadvantages. In trying to An analytic study attempts to quantify the relation between 2
understand why investigators chose a particular study type, factors—that is, the effect of an intervention (I) or exposure (E) on
several factors need to be taken into account. The first thing to an outcome (O). To quantify the effect, we need to know the rate
recognise is that both clinical questions and study designs have of outcomes in a comparison (C) group as well as the intervention
similar components (as we’d expect from PICO): or exposed group. Whether the researcher actively changes a
c a defined population (P) from which groups of participants factor or imposes an intervention determines whether the study is
are studied considered to be experimental (active involvement of researcher)
c outcomes (O) that are measured OR or observational (passive involvement of researcher).
c a situation (S) that is described
In experimental studies, the researcher manipulates the
PO is sufficient for questions about frequency, such as the exposure—that is, he or she allocates participants to the
prevalence of hepatitis C in specific groups. But for experimental intervention or exposure group. Experimental studies, or
and analytic observational studies, we need 2 extra elements: randomised controlled trials (RCTs), are similar to experiments
c interventions (I) or exposures (E) that are applied to
in other areas of science. That is, participants are allocated to
different groups of participants >2 groups to receive an intervention or exposure and are then
followed up under carefully controlled conditions. Such
A SIMPLE CLASSIFICATION controlled trials, particularly if randomised and blinded, have
The figure shows the tree of possible designs, branching into the potential to control for most of the biases that can occur in
subgroups of study designs by whether the studies are scientific studies, but whether bias is actually controlled
theoretical, descriptive, or analytic and by whether the analytic depends on the quality of the study design and implementation.
Simplified classification of different types of studies

PO, Population, Outome; PS, Population, Situation; PICO, Patient, Intervention, Comparison, Outcome; PECO, Population, Exposure, Comparison,
Outcome. Q1, Q2, and Q3 refer to the 3 questions outlined under ‘‘Spotting the study design.’’
EBN July 2009 Vol 12 No 3 71

EBN notebook
In analytic observational studies, the researcher simply Q2. If analytic, was the intervention randomly allocated?
measures the exposure or treatments of the groups. Analytic Yes ) randomised controlled trial
observational studies include case–control studies, cohort No ) observational study
studies, and some population (cross-sectional) studies. These
For observational studies, the main types will then depend on
studies all include matched groups of participants and assess
the timing of the measurement of outcome, so our third
associations between exposures and outcomes.
question is
Observational studies investigate and record exposures (such
as interventions or risk factors) and observe outcomes (such as Q3. When were the outcomes determined?
disease) as they occur. Such studies may be purely descriptive or (a) Some time after the exposure or intervention ) cohort
more analytical. study (‘‘prospective study’’).
We should finally note that studies can incorporate several (b) At the same time as the exposure or intervention )
design elements. For example, the control group of a randomised cross-sectional study or survey.
trial may also be used as a cohort study, and the baseline measures (c) Before the exposure was determined ) case–control
of a cohort study may be used as a cross-sectional study. study (‘‘retrospective study’’ based on recall of the
exposure).
SPOTTING THE STUDY DESIGN
The type of study can generally be worked out by looking at 3 Paul Glasziou MBBS, Carl Heneghan, MB ChB
issues (as per the tree of possible designs shown in the figure): Centre for Evidence-Based Medicine; University of Oxford, Oxford, UK
Q1. Was the aim of the study to develop theoretical understanding (PS
questions)—qualitative—or to simply describe a population (PS or A modified version of this Notebook appears in Evidence-Based Medicine.
PO questions)—qualitative or descriptive—or to quantify the 1. Doll R, Peto R, Boreham J, Sutherland I. Mortality in relation to smoking: 50 years’
relation between factors (PICO questions)—analytic. observations on male British doctors. BMJ 2004;328:1519.
72 EBN July 2009 Vol 12 No 3

A Spotters Guide Til Study Designs

Uploaded by

Copyright:

Available Formats

A Spotters Guide Til Study Designs

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

A Spotters Guide Til Study Designs

Uploaded by

Copyright:

Available Formats

EBN notebook

A spotter’s guide to study designs

Simplified classification of different types of studies

EBN July 2009 Vol 12 No 3 71

72 EBN July 2009 Vol 12 No 3

You might also like

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.