A spotter’s guide to study designs
When searching for evidence to answer our clinical questions, the ability to rapidly recognise different types of studies is helpful for finding the one that best answers the question. “Levels of evidence” tables make suggestions for which design is best for which type of question. For instance, you would naturally consider a randomised controlled trial as the most appropriate study design for intervention decisions. But for potential harms of interventions, we may need case–control studies. And for aetiology, we often need to use cohort studies: you wouldn’t randomise someone to cigarette smoking to see if they did worse—that would also be unethical. But you would want investigators to follow up cigarette smokers and non-smokers for a long time, just as Richard Doll did.1
This short Notebook is a brief guide to the different study types and their advantages and disadvantages. In trying to understand why investigators chose a particular study type, several factors need to be taken into account. The first thing to recognise is that both clinical questions and study designs have similar components (as we’d expect from PICO):
a defined population (P) from which groups of participants are studied
outcomes (O) that are measured OR
a situation (S) that is described
PO is sufficient for questions about frequency, such as the prevalence of hepatitis C in specific groups. But for experimental and analytic observational studies, we need 2 extra elements:
interventions (I) or exposures (E) that are applied to different groups of participants
A SIMPLE CLASSIFICATION
The figure shows the tree of possible designs, branching into subgroups of study designs by whether the studies are theoretical, descriptive, or analytic and by whether the analytic studies are experimental or observational. The list is not exhaustive but covers most basics designs.
Our first distinction is whether the study is analytic or non-analytic. A non-analytic theoretical study tries to explain and develop our understanding of processes that occur for participants. Theoretical studies are qualitative and usually are based on grounded theory. A non-analytic descriptive study does not try to quantify the relation between factors but tries to give us a picture of what is happening in a population (eg, prevalence, incidence, or experience of a group). Descriptive studies include case reports, case-series, qualitative studies, and surveys (cross-sectional studies), which measure the frequency of several factors, and hence the size of the problem. They may sometimes also include analytic work (comparing factors—see below).
An analytic study attempts to quantify the relation between 2 factors—that is, the effect of an intervention (I) or exposure (E) on an outcome (O). To quantify the effect, we need to know the rate of outcomes in a comparison (C) group as well as the intervention or exposed group. Whether the researcher actively changes a factor or imposes an intervention determines whether the study is considered to be experimental (active involvement of researcher) or observational (passive involvement of researcher).
In experimental studies, the researcher manipulates the exposure—that is, he or she allocates participants to the intervention or exposure group. Experimental studies, or randomised controlled trials (RCTs), are similar to experiments in other areas of science. That is, participants are allocated to ⩾2 groups to receive an intervention or exposure and are then followed up under carefully controlled conditions. Such controlled trials, particularly if randomised and blinded, have the potential to control for most of the biases that can occur in scientific studies, but whether bias is actually controlled depends on the quality of the study design and implementation.
In analytic observational studies, the researcher simply measures the exposure or treatments of the groups. Analytic observational studies include case–control studies, cohort studies, and some population (cross-sectional) studies. These studies all include matched groups of participants and assess associations between exposures and outcomes.
Observational studies investigate and record exposures (such as interventions or risk factors) and observe outcomes (such as disease) as they occur. Such studies may be purely descriptive or more analytical.
We should finally note that studies can incorporate several design elements. For example, the control group of a randomised trial may also be used as a cohort study, and the baseline measures of a cohort study may be used as a cross-sectional study.
SPOTTING THE STUDY DESIGN
The type of study can generally be worked out by looking at 3 issues (as per the tree of possible designs shown in the figure):
Q1. Was the aim of the study to develop theoretical understanding (PS questions)—qualitative—or to simply describe a population (PS or PO questions)—qualitative or descriptive—or to quantify the relation between factors (PICO questions)—analytic.
Q2. If analytic, was the intervention randomly allocated? Yes ⇒ randomised controlled trial No ⇒ observational study
For observational studies, the main types will then depend on the timing of the measurement of outcome, so our third question is
Q3. When were the outcomes determined?
(a) Some time after the exposure or intervention ⇒ cohort study (“prospective study”).
(b) At the same time as the exposure or intervention ⇒ cross-sectional study or survey.
(c) Before the exposure was determined ⇒ case–control study (“retrospective study” based on recall of the exposure).
A modified version of this Notebook appears in Evidence-Based Medicine.