A computer algorithm for calculating the adequacy of antidepressant treatment in unipolar and bipolar depression

J Clin Psychiatry. 2003 Jul;64(7):825-33. doi: 10.4088/jcp.v64n0714.


Background: Major depression is often treated with medications in doses that are too low or too short in duration. We published an early version of the Antidepressant Treatment History Form (ATHF) that rates the adequacy of antidepressant treatment. The updated ATHF presented here includes newer medications and a computer algorithm to automate the evaluation of the adequacy of pharmacotherapy or electroconvulsive therapy for depression.

Method: The computer algorithm was written in MS-DOS Q-BASIC and in Visual Basic 5.0. Treatment data from 47 depressed (Structured Clinical Interview for DSM-III-R) patients were scored by the computer algorithm and assigned a number from 0 to 5 for the adequacy of antidepressant treatment. A psychiatrist blinded to the computer ratings manually rated the treatment using the ATHF.

Results: The computer algorithm, based on an updated version of the ATHF, estimates the adequacy of treatment of unipolar and bipolar depression. Computer algorithm results agreed with those generated by a clinician completing the form manually (kappa = 0.88 to 1.00).

Conclusion: The computer algorithm can be used to analyze large databases and may help reduce the morbidity and mortality associated with major depression by improving the assessment of adequacy of pharmacologic treatments for research and quality assurance purposes. The availability of the updated ATHF on the Internet for downloading allows for modifications according to the user's purposes.

Publication types

  • Research Support, Non-U.S. Gov't
  • Research Support, U.S. Gov't, P.H.S.

MeSH terms

  • Algorithms*
  • Antidepressive Agents / therapeutic use*
  • Bipolar Disorder / diagnosis
  • Bipolar Disorder / drug therapy*
  • Depressive Disorder / diagnosis
  • Depressive Disorder / drug therapy*
  • Diagnostic and Statistical Manual of Mental Disorders
  • Dose-Response Relationship, Drug
  • Electronic Data Processing*
  • Humans
  • Severity of Illness Index
  • Treatment Outcome


  • Antidepressive Agents