Do Logistic Regression and Signal Detection Identify Different Subgroups at Risk? Implications for the Design of Tailored Interventions

• Published in: Psychological methods Vol. 6; no. 1; pp. 35 - 48
• Main Authors: Kiernan, Michaela, Kraemer, Helena C, Winkleby, Marilyn A, King, Abby C, Barr Taylor, C
• Format: Journal Article
• Language: English
• Published: Washington, DC American Psychological Association 01-03-2001
• Subjects: Adult; At Risk Populations; Biological and medical sciences; Cross-Sectional Studies; Data Interpretation, Statistical; European Continental Ancestry Group - statistics & numerical data; Female; Follow-Up Studies; Fundamental and applied biological sciences. Psychology; Hispanic Americans - statistics & numerical data; Human; Humans; Logistic Models; Logistic Regression; Male; Middle Aged; Obesity - epidemiology; Psychology. Psychoanalysis. Psychiatry; Psychology. Psychophysiology; Psychometrics. Statistics. Methodology; Risk Assessment - methods; Risk Assessment - statistics & numerical data; Signal Detection, Psychological; Statistical Analysis; Statistical Regression; Statistics. Mathematics; Human; Statistical method; Logistic regression; Risk factor; Subgroup; Identification; Signal detection; Comparative study
• ISSN: 1082-989X; 1939-1463
• DOI: 10.1037/1082-989X.6.1.35

Identifying subgroups of high-risk individuals can lead to the
development of tailored interventions for those subgroups. This study
compared two multivariate statistical methods (logistic regression and
signal detection) and evaluated their ability to identify subgroups at
risk. The methods identified similar risk predictors and had similar
predictive accuracy in exploratory and validation samples.
However, the 2 methods did not classify individuals into the same
subgroups. Within subgroups, logistic regression identified
individuals that were homogeneous in outcome but heterogeneous in risk
predictors. In contrast, signal detection identified individuals
that were homogeneous in both outcome and risk predictors. Because of
the ability to identify homogeneous subgroups, signal detection may be
more useful than logistic regression for designing distinct tailored
interventions for subgroups of high-risk
individuals.