Bringing morphometrics into the Fetal Alcohol Report: Statistical Language for the Forensic Neurologist or Psychiatrist

• Published in: Journal of psychiatry & law 2010-12, Vol.38 (4), p.449-473
• Main Authors: Bookstein, Fred L., Kowell, Arthur P.
• Format: Article
• Language: English
• Subjects: Behavior; Family medical history
• ISSN: 0093-1853; 2163-1794
• DOI: 10.1177/009318531003800405

The components of a diagnosis in the fetal alcohol spectrum consider, among other domains, evidence of prenatal alcohol exposure, facial features, neurological or neuropsychological deficits, and neuroanatomical form. Most of these assessments are based on traditional sorts of quantifications (e.g., length measures, IQ scores, quantity of alcohol drunk by the mother), but for one of the domains, the morphometric assessment of neuroanatomical form as abnormal in a way suggesting fetal alcohol damage, the associated quantifications are relatively recent. For purposes of a forensic report, this morphometric component of the diagnostic judgment would seem to go best in the form of one classic statistical quantity, the likelihood ratio (odds ratio) of two hypotheses as they both attempt to explain the form of a particular structure within the defendant's brain (in our approach, the midcurve of the corpus callosum). On one hypothesis, the subject is presumed to be from a normal population; on the other, he is from a group all characterized by diagnosable fetal alcohol damage. Our article explains in simple terms how the forensic physician might wish to couch that component of the FASD diagnostic system for the purpose of aiding those in the courtroom who have the task of determining guilt or culpability.