Student or educator? Sign up with your school email and get 6 months of Pro free.
CoChat
Cited by 173

The DOI matches its Crossref record; 1 of the 1 references it cites were found in a registry or an index.

Classification of epidemiological study designs

Pearce, N. · 2012 · International Journal of Epidemiology

Abstract

In this article, I present a simple classification scheme for epidemiological study designs, a topic about which there has been considerable debate over several decades. I will argue that when the individual is the unit of analysis and the disease outcome under study is dichotomous, then epidemiological study designs can best be classified according to two criteria: (i) the type of outcome under study (incidence or prevalence) and (ii) whether there is sampling on the basis of the outcome. This classification system has previously been proposed by Greenland and Morgenstern (1988)1 and Morgenstern and Thomas (1993),2 all of whom followed previous authors3,4 in rejecting directionality (i.e. prospective/retrospective or from exposure to outcome vs from outcome to exposure) as a key feature for distinguishing study designs. Once this two-dimensional classification system has been adopted, then there are only four basic study designs (Table 1):2,5,6 (i) incidence studies; (ii) incidence case–control studies; (iii) prevalence studies; and (iv) prevalence case–control studies (Rothman et al.7 use the terms ‘incident case–control study’ and ‘prevalent case–control study’ where the adjective refers to the incident or prevalent cases2). Four basic study types Four basic study types In this article, I will briefly illustrate these four different study designs for dichotomous outcomes; I then briefly consider the extension of this classification to include studies with continuous exposure or outcome measures and I briefly mention other possible axes of classification. It should first be emphasized that all epidemiological studies are (or should be) based on a particular population (the ‘source population’) followed over a particular period of time (the ‘risk period’). Within this framework, the most fundamental distinction is between studies of disease ‘incidence’ and studies of disease ‘prevalence’. Once this distinction has been drawn, then the different epidemiological study designs differ primarily in the manner in which information is drawn from the source population and risk period.8 Incidence studies ideally measure exposures, confounders and outcome times of all population members. Table 2 shows the findings of a hypothetical incidence study involving 10 000 people who are exposed to a particular risk factor and 10 000 people who are not exposed. When the source population has been formally defined and enumerated (e.g. a group of workers exposed to a particular chemical), then the study may be termed a ‘cohort study’ or ‘follow-up study’ and the former terminology will be used here. Incidence studies also include studies where the source population has been defined but a cohort has not been formally enumerated by the investigator, e.g. ‘descriptive’ studies of national death rates. Furthermore, there is no fundamental distinction between incidence studies based on a broad population (e.g. all workers at a particular factory or all persons living in a particular geographical area) and incidence studies involving sampling on the basis of exposure, since the latter procedure merely redefines the study population (cohort).4 Findings from a hypothetical cohort study of 20 000 persons followed for 10 years Findings from a hypothetical cohort study of 20 000 persons followed for 10 years Three measures of disease occurrence are commonly used in incidence studies.9 Perhaps the most common measure is the person–time ‘incidence rate’; a second measure is the ‘incidence proportion’ (average risk), which is the proportion of study subjects who experience the outcome of interest at any time during the follow-up period. A third possible measure is the ‘incidence odds’, which is the ratio of the number of subjects who experience the outcome to the number of subjects who do not experience the outcome. These three measures of disease occurrence all involve the same numerator: the number of incident cases of disease. They differ in whether their denominators represent person–time at risk, persons at risk or survivors. Corresponding to these three measures of disease occurrence, the three ratio measures of effect used in incidence studies are the ‘rate ratio’, ‘risk ratio’ and ‘odds ratio’. Incidence studies are usually the preferred approach to studying the causes of disease, because they use all of the available information on the source population over the risk period. However, they are often very expensive in terms of time and resources, and the equivalent results may be achieved more efficiently by using an incidence case–control study design. Table 3 shows the data from a hypothetical incidence case–control study of all 2765 incident cases in the full cohort in Table 2 and a random sample of 2765 controls. Such a study would on an average achieve the same findings as the full cohort study (Table 2), but would be considerably more efficient, since it would involve ascertaining the exposure histories of 5530 people (2765 cas

Overview

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1093/ije/dys049

Cites

1 of 1 references verified0 citation errors · 0 not checkable

Statistics

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Cited by

See all 173

Metadata

Authors
Pearce, N.
Venue
International Journal of Epidemiology
Published
2012-04-01
Volume
41
Issue
2
Pages
393-397
ISSN
  • 0300-5771 (ISSN-L)
  • 1464-3685
OpenAlex
W2117932158
Cited by
173
References
1
FWCI
6.89
Type
journal article
Language
en
Open access
closed

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.