Design of an algorithm to identify persons with mental illness in a police administrative database

Kathleen Hartford, Lisa Heslop, Larry Stitt, Jeffrey S Hoch

Research output: Contribution to journalArticle

30 Scopus citations

Abstract

North American police maintain a database to track events and information related to their involvement with the public that contain a series of electronic caution/dependency flags attached to an individual's name for internal communication. To identify persons with mental illness in a police administrative database, an algorithm was developed that was composed of (a) caution/dependency flags, (b) addresses, and (c) key search words indicative of mental illness. Based on the level of confidence of the algorithm, persons with mental illness (PMI) were then assigned to one of three categories: Definite, Probable and Possible PMI. Results for 2000 include the sociodemographic characteristics of PMI and non-PMI in the database. The mean number of contacts, types of interactions, re-involvement with a year, charges and dispositions are described. The algorithm provides a cheap, quick method to identify PMI for North American police. It enables police to monitor the effectiveness of pre-arrest diversion programs and allows researchers to analyze questions of criminalization and mental illness.

Original languageEnglish (US)
Pages (from-to)1-11
Number of pages11
JournalInternational Journal of Law and Psychiatry
Volume28
Issue number1
DOIs
StatePublished - Jan 2005
Externally publishedYes

    Fingerprint

Keywords

  • Mental illness
  • Police

ASJC Scopus subject areas

  • Pathology and Forensic Medicine
  • Applied Psychology
  • Social Psychology
  • Law
  • Psychiatry and Mental health

Cite this