Categorical versus dimensional approaches to diagnosis: methodological challenges

J Psychiatr Res. 2004 Jan;38(1):17-25. doi: 10.1016/s0022-3956(03)00097-9.


The arguments pitting categorical versus dimensional approaches to psychiatric diagnosis have been long ongoing with little sign of imminent resolution. We argue that categorical and dimensional approaches are fundamentally equivalent, but that one or other approach is more appropriate depending on the clinical circumstances and research questions being addressed. This paper aims to demonstrate (a) how these two approaches necessarily interdigitate, (b) to clarify the conditions under which one should utilize one approach over the other, and (c) to alert psychiatric clinicians and researchers to issues in the methodology literature that might facilitate their considerations. Using an example from the Infant Health and Development Program (IHDP), we illustrate the importance of using dimensional approaches for hypothesis testing, identify the problems with power and with interpretation that arise from employing a categorical approach, and underscore the importance of identifying the appropriate cutpoints when a categorical approach is necessitated. We argue that failure to utilize the correct approach under the appropriate circumstances can result in impaired clinical and research decision-making.

Publication types

  • Comparative Study

MeSH terms

  • Adult
  • Child
  • Child, Preschool
  • Diagnosis, Differential
  • Diagnostic and Statistical Manual of Mental Disorders*
  • Early Intervention, Educational
  • Humans
  • Infant
  • Infant, Low Birth Weight
  • Infant, Newborn
  • Infant, Premature, Diseases / diagnosis
  • Infant, Premature, Diseases / therapy
  • Intellectual Disability / classification
  • Intellectual Disability / diagnosis*
  • Intellectual Disability / therapy
  • Intelligence
  • Longitudinal Studies
  • Mental Disorders / classification
  • Mental Disorders / diagnosis*
  • Multicenter Studies as Topic / statistics & numerical data
  • Outcome Assessment, Health Care / statistics & numerical data
  • Psychological Tests / statistics & numerical data*
  • Psychometrics / statistics & numerical data
  • ROC Curve
  • Randomized Controlled Trials as Topic / statistics & numerical data
  • Reproducibility of Results