CoChat

Declaring independence: why we should be cautious.

Open access
Smith, G D, Phillips, A · 1990 · Journal of Epidemiology and Community Health
Cited by 88

The DOI matches its Crossref record; 2 of the 2 references it cites were found in a registry or an index.

Abstract

Declaring independence: why we should be cautious A review article identified 246 different risk factors for coronary heart disease.'Many of these 246 were risk factors asserted to be "independent", that is, their association with coronary heart disease was not due to relationships with other, confounding, factors.Establishing the independence ofeffects is central in analytical epidemiological studies and is seen as an important step in assigning causality to a relationship.But even given the obsession epidemiologists have for multifactorial theories of disease causation2 it seems unlikely that there actually are 246 separate causes of coronary heart disease.What might be going on here?Consider two examples.We are told that poor dental health3 and coffee consumption4 are both independent risk factors for coronary heart disease.In these cases confounding clearly occurs.Smoking and socioeconomic position are strongly related both to dental health and to coronary heart disease risk; similarly smoking, type A behaviour, and obesity are related to coffee consumption.Thus drinking 10 or more cups of coffee per day was associated with a relative risk of myocardial infarction of 4 9.After adjusting for confounding variables this fell to 2-9, which was still significantly greater than 1.In these papers, as in countless more, the authors declare that a factor is independent because a significant association remains after "adjusting" or "controlling" for the confounding factors in multiple logistic regression models.We discuss here the assumptions made in concluding that a risk factor exerts an independent effect on the basis of such statistical results.In particular we are concerned with error and misclassification which occurs when risk factor levels are measured.Consider the situation with a risk factor of interest and a single confounding factor.Imagine that the relative risk for the association between our risk factor and some disease is significantly different from unity, after adjustment for the confounding factor.When concluding that the effect of the risk factor is independent of the confounding factor it is assumed that the fitted model is correct.With a typical logistic

Overview

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

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

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

Integrity signals

Registry

Crossref · 10.1136/jech.44.4.257

Cites

2 of 2 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.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

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

Research works printed but not linked: 11

  1. Hopkins PN, Williams RR. A survey of 246 suggested coronary risk factors. Atherosclerosis 1981; 40: 1-52. · DOI · Paper page
  2. McCormick J. The multifactorial aetiology of coronary heart disease: a dangerous delusion. Perspect Med Biol 1988; 32: 103-8. · DOI
  3. Mattila KJ, Nieminen MS, Valtonen VV, et al. Association between dental health and myocardial infarction. Br Med J 1989; 298: 779-82. · DOI · Paper page
  4. Rosenburg L, Palmer JR, Kelly JP, Kaufman DW, Shapiro S. Coffee drinking and nonfatal myocardial infarction in men under 55 years of age. Am JEpidemiol 1988; 128: 570-8. · DOI · Paper page
  5. Armstrong BG, Whittemore AS, Howe GR. Analysis of case-control data with covariate measurement error: application to diet and colon cancer. Stat Med 1989;8: 1151-63. · DOI · Paper page
  6. Slattery ML, Robison LM, Schuman KL, et al. Cigarette smoking and exposure to passive smoke are risk factors for cervical cancer. JAMA 1989; 261: 1593-8. · DOI · Paper page
  7. Winkelstein W. Smoking and cervical cancer-current status: a review. AmJ Epidemiol 1990; 131: 945-57. · DOI · Paper page
  8. Martin PMD, Hill GB. Cervical cancer in relation to tobacco and alcohol consumption in Lesotho, Southern Africa. Cancer Detect Prev 1984; 7: 109-14.
  9. Haverkos HW, Pinsky PF, Drotman P, Bregman DJ. Disease manifestation among homosexual men with AIDS: a possible role of nitrites in Kaposi's sarcoma. Sexually Transmitted Dis 1985; 12: 203-8. · DOI
  10. Lauritsen J, Wilson H. Death rush: poppers and AIDS. New York: Pagan Press, 1986.
  11. Watney S. Safer sex as community practice. In: Aggleton P, Davies P, Hart G, eds. AIDS: Individual, cultural & policy dimensions. Basingstoke: Falmer, 1990: 19-33.

Cited by

See all 88

Metadata

Authors
Smith, G. D., & Phillips, A.
Venue
Journal of Epidemiology and Community Health
Publisher
BMJ
Published
1990-12
Volume
44
Issue
4
Pages
257-258
ISSN
  • 0141-7681 (ISSN-L)
  • 0143-005X
  • 1470-2738
OpenAlex
W1976668223
Cited by
88
References
2
FWCI
14.04
Type
journal article
Language
en
Open access
bronze

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.