Joel Howell, MD, PhD
What is the accuracy of instruments for detecting incapacity for medical treatment decision-making in patients who do not have significant comorbid psychiatric conditions?
Included studies evaluated instruments that could be used in offices or at bedside to assess the capacity of patients to make medical treatment decisions. Studies that evaluated capacity for consent to research or only included patients with significant psychiatric conditions were excluded. Outcomes were sensitivity, specificity, and likelihood ratios (LRs).
MEDLINE and EMBASE/Excerpta Medica (to Apr 2011) and reference lists were searched for English-language, high-quality (≥ level 3 on the Levels of Evidence for the Rational Clinical Examination), prospective studies. 9 studies (n = 829, mean age range 54 to 86 y, 42% to 92% women) evaluating 9 instruments met the selection criteria and compared capacity assessment instruments with a gold standard (competency panel or physicians with expertise in capacity assessment); these are reported here. 3 studies assessed medical inpatients (n = 288), 2 nursing home residents (n = 146), 1 ambulatory patients (n = 100), 1 psychiatric and medical inpatients (n = 41), 1 competency clinic patients (n = 14), and 1 general population (n = 240).
The test characteristics are in the Table. Prevalence of incapacity ranged from 6% in elderly controls to 73% in elderly patients with Alzheimer disease. 4 of 9 instruments had +LRs > 5 and –LRs < 0.25: Hopkins Competency Assessment Test, Aid to Capacity Evaluation (ACE), Understanding Treatment Disclosure, and the Fazel Questionnaire (Table).
Several instruments are accurate for evaluating capacity for medical treatment decision-making in patients without severe psychiatric conditions.
*ACE = Aid to Capacity Evaluation; CCT = Cognitive Competency Test; CQ-M = Cognitive Questionnaire—Medicine; HCAI = Hopemont Capacity Assessment Interview; HCAT = Hopkins Competency Assessment Test; SCI = Specific Capacity Instrument; UTD = Understanding Treatment Disclosure; other abbreviations defined in Glossary. Gold standard = assessment by forensic or expert psychiatrist, or competency panel.
†1 study reported in each row. Levels of Evidence for the Rational Clinical Examination (level 1 = highest) ≥ 2 unless stated otherwise.
‡UTD and HCAI were evaluated in the same study.
§Rational Clinical Examination level of evidence = 3.
Howell J. Review: Several instruments are accurate for evaluating patient capacity for medical treatment decision-making. Ann Intern Med. ;155:JC5–12. doi: 10.7326/0003-4819-155-10-201111150-02012
Download citation file:
Published: Ann Intern Med. 2011;155(10):JC5-12.
Dementia, Geriatric Medicine, Hospital Medicine, Neurology.
Results provided by:
Copyright © 2018 American College of Physicians. All Rights Reserved.
Print ISSN: 0003-4819 | Online ISSN: 1539-3704
Conditions of Use