Search PubMed⌕ Search
Home/Archive/PubMed

PubMed · 16267795

[Statistical models and multivariable analysis].

P Ravani·F Malberti
View original source ↗Export metadata ↓

Abstract

Most clinical research can be simplified as an investigation of an input/output relationship. The inputs are called explanatory (independent) variables or predictors and are thought to be related to the outcome, or response (independent) variable. This relationship is usually complicated by other factors related to both the input and the output (presence of confounding) and can vary according to the levels of the other variables (presence of interaction). This input/output relationship is usually described by statistical models that include a fit part and a residual component or difference between the data and the fit. The most popular models are the general linear models, which can be considered the paradigm of all models used in multi-variable analyzes.

Explore related subjects

Biomedical ResearchHumansModels, StatisticalMultivariate AnalysisEnglish AbstractJournal ArticleReview

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

P Ravani, F Malberti. [Statistical models and multivariable analysis].. https://pubmed.ncbi.nlm.nih.gov/16267795/

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

KEEP EXPLORING

Related citations

PubMed2006

Reporting genetic results in research studies: summary and recommendations of an NHLBI working group.

Prospective epidemiologic studies aid in identifying genetic variants associated with diseases, health risks, and physiologic traits. These genetic variants may eventually be measured clinically for purposes of diagnosis, prognosis, and treatment. As evidence of the potential clinical value of such information accrues, research studies face growing pressure to report these results to study participants or their physicians, even before sufficient evidence is available to support widespread screening of asymptomatic persons. There is thus a need to begin to develop consensus on whether and when genetic findings should be reported to participants in research studies. The National Heart, Lung, and Blood Institute (NHLBI) convened a Working Group on Reporting Genetic Results in Research Studies to discuss if, when, and how genetic information should be reported to study participants. The Working Group concluded that genetic test results should be reported to study participants when the associated risk for the disease is significant; the disease has important health implications such as premature death or substantial morbidity or has significant reproductive implications; and proven therapeutic or preventive interventions are available. Finally, the Working Group recommended procedures for reporting genetic research results and encouraged increased efforts to create uniform guidelines for this activity.

Biomedical Research↗
PubMed2006

Learning to lead.

A successful research career requires not only an aptitude for science but also the mastering of other skills including communication, management, and grant writing. A growing number of programs at universities and research institutes aim to teach these crucial skills to graduate students, postdoctoral fellows, and junior faculty.

Biomedical Research↗