Basic statistics for clinicians: 4. Correlation and regression.

G Guyatt; S Walter; H Shannon; D Cook; R Jaeschke; N Heddle

Home/Archive/PubMed

PubMed · 7859197

Basic statistics for clinicians: 4. Correlation and regression.

G Guyatt·S Walter·H Shannon·D Cook·R Jaeschke·N Heddle

View original source ↗Export metadata ↓

Abstract

Correlation and regression help us to understand the relation between variables and to predict patients' status in regard to a particular variable of interest. Correlation examines the strength of the relation between two variables, neither of which is considered the variable one is trying to predict (the target variable). Regression analysis examines the ability of one or more factors, called independent variables, to predict a patient's status in regard to the target or dependent variable. Independent and dependent variables may be continuous (taking a wide range of values) or binary (dichotomous, yielding yes-or-no results). Regression models can be used to construct clinical prediction rules that help to guide clinical decisions. In considering regression and correlation, clinicians should pay more attention to the magnitude of the correlation or the predictive power of the regression than to whether the relation is statistically significant.

Explore related subjects

Data Interpretation, Statistical Female Humans Logistic Models Male Odds Ratio Regression Analysis Statistics as Topic Journal Article Research Support, Non-U.S. Gov't

Keep this discovery

Explore connections, maps & timelines

BibTeX RIS

G Guyatt, S Walter, H Shannon, D Cook, R Jaeschke, N Heddle. 1995-02-15. Basic statistics for clinicians: 4. Correlation and regression.. https://pubmed.ncbi.nlm.nih.gov/7859197/

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related citations

PubMed1996

Use of a stochastic model for repeated binary assessment.

We discuss a stochastic model appropriate for binary data in clinical studies where assessments are made at various nominal times during a treatment phase. The model is then applied to data on headache relief, nausea and photophobia/phonophobia in a migraine study. The transition rates and probabilities during the initial 240 minutes after treatment administration are derived using the method of maximum likelihood. The results are then compared with analysis at each nominal time point. Stochastic modelling is considered more appropriate for the analysis of repeated binary assessments than analysis at each nominal time since each patient's assessments are modelled simultaneously.

Data Interpretation, Statistical↗

PubMed1996

Fitting multi-parameter distributions to sister chromatid exchange data.

Bender et al. (1992) presented the number of sister chromatid exchanges (SCE) in 50 peripheral blood lymphocytes from each of 393 normal human subjects. In that study of 19650 cells, the number of SCE per cell ranged from 0 to 32. We examine the resulting frequency distributions, and show how they may be fitted, by the method of moments, to discrete Pearson-type and Johnson translation system distributions.

Data Interpretation, Statistical↗

PubMed1996

Forty years of randomised trials in the New Zealand Medical Journal.

AIM: To identify all randomised trials published in the New Zealand Medical Journal, to document the basic characteristics of these trials and to count the number that were detectable on medline. METHODS: All issues of the New Zealand Medical Journal between 1943 and 1995 were systematically hand-searched. All trials identified were characterised and compared against the trials identified using an optimal medline search strategy. RESULTS: The handsearch identified 152 randomised controlled trials, the first published in 1955. Half the trials recruited less than 34 participants and more than 90% were of pharmaceutical interventions. Only 18% of studies reported on the method of randomisation and 13% provided evidence that final analyses were conducted on an intention to treat basis. Fifty one percent of trials employed a placebo control group and 28% involved a crossover design. Since 1966, when Medline became available, 143 trials were published of which 89 (62%) were identified by the Medline search. CONCLUSIONS: Two of the major difficulties that face those preparing systematic reviews were illustrated by this survey. First, important information on design and analysis is often missing from reports of trials. Second, a large proportion of published randomised trials are not identifiable on Medline. Standard formats for reporting the results of trials and inclusion of trials identified by hand-searching on the Cochrane Collaboration's International Register of Randomised Controlled Trials of Health Care will facilitate the future production of reliable systematic reviews.

Data Interpretation, Statistical↗