THE RATIONAL CLINICIAN’S CORNER

CLINICAL EXAMINATION

Does This Patient Have Medical

Decision-Making Capacity?
Laura L. Sessums, JD, MD Context Evaluation of the capacity of a patient to make medical decisions should
Hanna Zembrzuska, MD occur in the context of specific medical decisions when incapacity is considered.
Jeffrey L. Jackson, MD, MPH Objective To determine the prevalence of incapacity and assessment accuracy in
adult medicine patients without severe mental illnesses.
CLINICAL SCENARIO Data Sources MEDLINE and EMBASE (from their inception through April 2011) and
One of your patients, a 72-year-old bibliographies of retrieved articles.
woman, comes to you for a preopera- Study Selection We included high-quality prospective studies (n = 43) of instru-
tive evaluation for a total hip replace- ments that evaluated medical decision-making capacity for treatment decisions.
ment. Her medical history includes Data Extraction Two authors independently appraised study quality, extracted rel-
early-stage Alzheimer disease. At a re- evant data, and resolved disagreements by consensus.
cent clinic visit, her husband noted his Data Synthesis Incapacity was uncommon in healthy elderly control participants
wife seemed more forgetful, and on ex- (2.8%; 95% confidence interval [CI], 1.7%-3.9%) compared with medicine inpa-
amination, her Mini-Mental State Ex- tients (26%; 95% CI, 18%-35%). Clinicians accurately diagnosed incapacity (posi-
amination (MMSE) score was 21 out of tive likelihood ratio [LR⫹] of 7.9; 95% CI, 2.7-13), although they recognized it in only
30 points. Today, you ask the patient 42% (95% CI, 30%-53%) of affected patients. Although not designed to assess in-
what she understands about the risks capacity, Mini-Mental State Examination (MMSE) scores less than 20 increased the
and benefits of the planned proce- likelihood of incapacity (LR, 6.3; 95% CI, 3.7-11), scores of 20 to 24 had no effect
(LR, 0.87; 95% CI, 0.53-1.2), and scores greater than 24 significantly lowered the like-
dure. She smiles and tells you it will fix
lihood of incapacity (LR, 0.14; 95% CI, 0.06-0.34). Of 9 instruments compared with
her hip. When you give her informa- a gold standard, only 3 are easily performed and have useful test characteristics: the
tion about risks and alternative treat- Aid to Capacity Evaluation (ACE) (LR⫹, 8.5; 95% CI, 3.9-19; negative LR [LR−], 0.21;
ment options, and query about her un- 95% CI, 0.11-0.41), the Hopkins Competency Assessment Test (LR⫹, 54; 95% CI,
derstanding, she continues to smile and 3.5-846; LR−, 0; 95% CI, 0.0-0.52), and the Understanding Treatment Disclosure (LR⫹,
replies, “It’ll be okay.” You wonder 6.0; 95% CI, 2.1-17; LR−, 0.16; 95% CI, 0.06-0.41). The ACE was validated in the
whether she has the capacity to make largest study; it is freely available online and includes a training module.
the decision to proceed with the op- Conclusions Incapacity is common and often not recognized. The MMSE is useful
eration. only at extreme scores. The ACE is the best available instrument to assist physicians in
making assessments of medical decision-making capacity.
WHY IS THE CLINICAL JAMA. 2011;306(4):420-427 www.jama.com
EXAMINATION IMPORTANT?
Patients are assumed to have capacity the case presented herein, or if the pa- dards for whether a patient meets this
to make medical decisions unless tient disagrees with the physician’s rec- last element also vary from state to state
proven otherwise,1 and many clini- ommendation.9 but are generally based on evaluating
cians lack formal training in capacity The criteria for valid consent to medi-
evaluation. The practical conse- cal treatment vary from state to state but Author Affiliations: Section of General Internal Medi-
quence is that clinicians regularly fail are based on common law and have 3 cine, Walter Reed Army Medical Center, Washing-
ton, DC (Drs Sessums and Zembrzuska); and Division
to recognize incapacity1-8 and gener- elements. The patient must (1) be given of General Internal Medicine, Zablocki VA Medical
ally question a patient’s capacity only adequate information regarding the na- Center and the Medical College of Wisconsin, Mil-
when the medical decision to be made waukee (Dr Jackson).
ture and purpose of proposed treat- Corresponding Author: Laura L. Sessums, JD, MD,
is complex with significant risk, as in ments, as well as the risks, benefits, and General Internal Medicine, Walter Reed Army Medi-
cal Center, 6900 Georgia Ave NW, Washington, DC
alternatives to the proposed therapy, in- 20307 ([email protected]).
CME available online at cluding no treatment; (2) be free from The Rational Clinical Examination Section Editors: Da-
www.jamaarchivescme.com coercion; and (3) have medical deci- vid L. Simel, MD, MHS, Durham Veterans Affairs Medi-
and questions on p 444. cal Center and Duke University Medical Center, Dur-
sion-making capacity. 10 The stan- ham, NC; Drummond Rennie, MD, Deputy Editor.

420 JAMA, July 27, 2011—Vol 306, No. 4 ©2011 American Medical Association. All rights reserved.

Downloaded From: https://jamanetwork.com/ Biblioteca Virtual del SSPA by Jose-Maria Dominguez-Roldan on 12/18/2019
 INCAPACITY AND MEDICAL DECISION MAKING

4 abilities.11 Patients must have the abil- tions—as well as experts’—could ben- used for calculating the confidence in-
ity to (1) understand the relevant in- efit from a standardized approach.15 An terval (CI). For tables where any cell
formation about proposed diagnostic ideal clinical tool for evaluating capac- had a zero value, 0.5 was added to each
tests or treatment, (2) appreciate their ity should be brief and reliable and fa- cell for calculating the LR CIs. Preva-
situation (including their underlying cilitate documentation of the 4 capac- lence and reliability data were pooled
values and current medical situation), ity abilities. A number of instruments using a random-effects model, calcu-
(3) use reason to make a decision, and have been developed for assessing ca- lating variance with exact binomial
(4) communicate their choice.11 pacity to make medical decisions, methods. Diagnostic test accuracy for
Commonly, the ability to commu- mostly in psychiatric patients. While the MMSE2 was pooled using a gener-
nicate a choice is the easiest to assess. there are previous reviews assessing ca- alized linear mixed model, fitting a
Assessing a patient’s understanding of pacity evaluation in the psychiatric lit- 2-level mixed logistic regression model,
the proposed medical decision and his erature,16,17 our review asks several new with independent binomial distribu-
or her individual situation often re- questions. What is the prevalence of in- tions for the true positives and true
quires probing. A decision that fol- capacity in commonly encountered negatives conditional on the sensitiv-
lows from the patient’s individual situ- medical populations? How frequently ity and specificity in each study, and a
ation, the patient’s values, and the do clinicians recognize incapacity? bivariate normal model for the logit
information given about the decision Since physicians frequently assess cog- transforms of sensitivity and specific-
shows the patient’s ability to reason ef- nition and may use it as a surrogate for ity between studies.18 We a priori fo-
fectively. This ability references the pro- capacity, how useful are tests of cog- cused on capacity instruments that were
cess of information manipulation rather nition, such as the MMSE, in assess- possible to perform in an office visit,
than the decision itself, as there is no ing patient capacity? What are the test had moderate to strong levels of qual-
absolute “right” or “wrong” decision characteristics of capacity instru- ity, and had robust likelihood ratios. All
against which the patient’s decision ments for medical patients without sig- analyses were performed using Stata
should be judged. In the case pre- nificant psychiatric comorbidities? Our version 11.1 (StataCorp, College Sta-
sented, the patient’s failure to answer goal is to provide a guide for clini- tion, Texas).
specifically your questions about risks cians selecting a valid, reliable, and
and alternative treatments raises con- clinically useful tool for assessing and RESULTS
cern about her capacity and should re- documenting incapacity in their own The search strategy yielded 5568 unique
sult in a formal assessment. patients. articles. Of these, 4040 articles were
Because a patient’s capacity is both grossly inapplicable based on review of
temporal and situational, capacity METHODS title and abstract. An additional 43 ar-
evaluations should occur in the con- We searched MEDLINE and EMBASE ticles were uncovered by hand review
text of the specific health care deci- from their inception to April 2011 for of the bibliographies, resulting in a total
sion that needs to be made.12 Some pa- English-language articles that studied of 1571 full-text articles retrieved. Af-
tients lack capacity for specific periods instruments assessing medical decision- ter full review, we excluded 1518 ar-
of time, such as when critically ill, but making capacity (not just cognitive abil- ticles (eFigure 1), yielding 53 studies
not permanently. Although some ity) for treatment decisions and were that met eligibility criteria.1,3-8,14,15,19-62
people are completely incapacitated, feasible to use in the office or bedside Of these 53 studies, 10 were excluded
many have limited capacity. Those with (see the eMethods, available at http: because they were duplicate reports or
limited capacity may be able to make //www.jama.com). We excluded stud- subgroup analyses from the same pa-
some diagnostic and treatment deci- ies that involved only patients with se- tient data set.* We used data from all
sions (generally less risky decisions) but vere psychiatric illness, such as suicidal available articles in the abstraction pro-
not others. Physicians commonly hold depression or severe psychosis, and cess, although we included each data
patients to higher standards when judg- those assessing capacity to consent to set only once. All studies were prospec-
ing capacity for riskier medical deci- research (eFigure 1). Capacity for treat- tive, involved a total of 3684 partici-
sions.12-14 ment decisions in the mentally ill is a pants, and came from 7 different coun-
Any licensed physician—not just a significant issue but is usually ad- tries (eMethods).
psychiatrist—can make a determina- dressed by psychiatrists. Further, the
tion of incapacity. The gold standard for literature addressing it is both volumi- How Often Do Patients
capacity determination is a clinical ex- nous and, to some extent, specialized, Lack Capacity?
amination by a physician trained to do so it is not addressed here. The prevalence of patients lacking
the examination who has performed an Sensitivity, specificity, and likeli- capacity varied by type of patient
extensive number of capacity evalua- hood ratios (LRs) were calculated from population (n = 25 studies) (TABLE 1
tions. Most physicians do not meet this raw data. When sensitivity or specific-
standard and their capacity evalua- ity was 1.0 or 0, an exact method was *References 15, 24, 30, 31, 33, 34, 42, 44, 45, 61.

©2011 American Medical Association. All rights reserved. JAMA, July 27, 2011—Vol 306, No. 4 421

Downloaded From: https://jamanetwork.com/ Biblioteca Virtual del SSPA by Jose-Maria Dominguez-Roldan on 12/18/2019
 INCAPACITY AND MEDICAL DECISION MAKING

eTable 1). Most of the studies in-

Table 1. Prevalence of Incapacity in Select Populations
cluded formal measures of cognition
Patients With Incapacity
Disease or No. of No. of (n = 33, 79%), most commonly the
Patient Care Setting Studies Patients No. % (95% CI) Standard Deviation MMSE.‡ A total of 9 capacity instru-
Healthy elderly controls 16 1817 51 2.8 (1.7-3.9) 0.005 ments have been studied in compari-
Mild cognitive impairment 1 147 29 20 (14-26) 0.03 son with a gold standard among adult
Glioma patients 1 26 6 23 (6.9-39) 0.08 medicine patients.§ TABLE 2 shows the
Medicine inpatients 8 816 212 26 (18-35) 0.11 studies comparing these instruments
Parkinson disease 4 148 62 42 (23-60) 0.13 with gold standards, and TABLE 3 shows
Nursing home 5 346 152 44 (28-60) 0.08 the characteristics of these as well as ad-
Alzheimer disease 10 1425 770 54 (28-79) 0.13 ditional selected competency tests (with
Learning disabled 4 208 141 68 (41-97) 0.14 types of validity described in eTable 2).
Abbreviation: CI, confidence interval.
The BOX describes the 3 instruments
that are possible to perform in an of-
and eTable 1) and likely reflected the tated found evidence of a strong rela- fice visit, have robust likelihood ra-
severity of the underlying cognitive defi- tionship between capacity scores and tios, have moderate to strong levels of
cits of the respective population. Par- evidence,67 and are based on US or Ca-
cognition. There was also evidence of
ticipants with learning disabilities had nadian law: the Aid to Capacity Evalu-
a trade-off between sensitivity and
the highest prevalence of incapacity at ation (ACE27) (LR⫹, 8.5; 95% CI, 3.9-
specificity in the relationship between
68% (95% CI, 41%-97%) followed by 19; LR−, 0.21; 95% CI, 0.11-0.41), the
the MMSE score and capacity (eFig-
patients with Alzheimer disease at 54% Hopkins Competency Assessment Test
ure 2). MMSE scores less than 20 in-
(95% CI, 28%-79%). Nursing home (HCAT38) (LR⫹, 54; 95% CI, 3.5-846;
creased the likelihood of incapacity
residents lacked capacity 44% of the LR−, 0; 95% CI, 0.0-0.52), and the Un-
(summary LR, 6.3; 95% CI, 3.7-11; 5
time (95% CI, 28%-60%). Incapacity derstanding Treatment Disclosure
studies), and scores less than 16 in-
was common among medicine inpa- (UTD)54 (LR⫹, 6.0; 95% CI, 2.1-17;
creased the likelihood further (sum-
tients (26%; 95% CI, 18%-35%) but un- LR−, 0.16; 95% CI, 0.06-0.41). (Addi-
mary LR, 12; 95% CI: 5.3-27; 5 stud-
usual among healthy elderly controls tional details about the studies of the
ies; I2 = 0.0%). Scores from 20 through
(2.8%; 95% CI, 1.7%-3.9%). While a 9 instruments are shown in eTable 3,
24 had no effect on the likelihood of in-
prevalence estimate of incapacity as- and descriptions of all 9 instruments ap-
capacity (summary LR, 0.87; 95% CI,
sessed using a single, reliable instru- pear in eTable 4.)
0.53-1.2; 5 studies), and scores greater
ment would be most accurate (these 25 The ACE evaluated 100 consecu-
than 24 significantly lowered the like-
studies used different capacity instru- tive medicine inpatients; the HCAT, 41
lihood of incapacity (summary LR, 0.14; inpatient medicine and psychiatric pa-
ments), the summary average of inca- 95% CI, 0.06-0.34; 8 studies).
pacity in some of the groups studied tients; and the UTD, 50 nursing home
with multiple instruments had reason- Physician Recognition of Incapacity
residents. Of these, the ACE was vali-
ably narrow CIs. Those that did not may dated in the largest study, has excel-
Eight studies reported the rate of rec- lent test characteristics, is the only in-
reflect the varying levels of disease se- ognition of incapacity by the patients’
verity in the underlying, heteroge- strument available online, and is the
physicians.1-8 In all 8 studies, the phy- only instrument for which training ma-
neous study populations as well as sician was blind to the results of for-
methodological and instrument differ- terials are provided. Because the UTD
mal capacity assessments. Physicians does not measure all the capacity abili-
ences between studies. recognized that patients were inca- ties, it is less useful. The HCAT has been
pable of medical decisions in 42% (95% compared with a gold standard in only
Relationship Between Cognition
CI, 30%-53%) of patients indepen- a single study38 from a single center. Of
and Incapacity
dently judged to lack capacity. While the other studies using HCAT, 2 used
The majority of studies included an as- physicians routinely missed the diag-
sessment of cognition (n = 35) (eTable it to assess prevalence1,21 and 2 re-
nosis of incapacity (negative LR [LR−], ported on construct validity5,37 but had
1), with most using the MMSE 0.61; 95% CI, 0.48-0.74), they were
(n=23),† a measure of overall cogni- no gold-standard comparison. In this
usually correct when they made the di- single comparison, the cut point se-
tive ability heavily weighted toward ori- agnosis (positive LR [LR⫹], 7.9; 95%
entation, attention, and memory. Stud- lected yielded perfect results, but it has
CI, 2.7-13). not been replicated, and it is unlikely
ies investigating patient characteristics
associated with being judged incapaci- Instruments for Assessing Capacity
‡References 1-5, 7, 23, 27-29, 37-39, 41, 43, 45, 48,
†References 1-5, 7, 23, 27-29, 37-39, 41, 43, 45, 48,
We found 19 different instruments 49, 54-57, 60.
49, 54-57, 60. for assessing capacity (eMethods and §References 3, 23, 27, 28, 38, 49, 54, 56, 57.

422 JAMA, July 27, 2011—Vol 306, No. 4 ©2011 American Medical Association. All rights reserved.

Downloaded From: https://jamanetwork.com/ Biblioteca Virtual del SSPA by Jose-Maria Dominguez-Roldan on 12/18/2019
 INCAPACITY AND MEDICAL DECISION MAKING

that these perfect discriminatory re- range (20-24) that has no effect on the surgery. Given a pretest probability of
sults will be found in other settings. As likelihood of incapacity. You decide to incapacity of 52% and an LR of 8.5
with other standardized instruments, evaluate her capacity for medical deci- based on her ACE results, you calcu-
patients can find the scenario confus- sion making using the ACE. Since this late that there is a 90% chance that she
ing and inapplicable, making it clini- instrument is based on making deci- lacks capacity to make this decision.
cally difficult to use. sions about her actual problem, you ask You discuss the situation with her hus-
the ACE questions based on the deci- band and agree that he will make the
SCENARIO RESOLUTION sion about whether to have the hip re- decision as her surrogate under her pre-
The question of whether a patient has placed. She is clearly able to commu- viously executed health care power of
impaired capacity for medical decision- nicate her choice, and you find that she attorney. You advise him that, be-
making is common, as exemplified by appreciates and understands her medi- cause capacity is decision specific, she
this patient. She has the capacity for de- cal problem (“I have hip pain from ar- might have capacity for future, less risky
ciding to have her hip replaced if she thritis”) and the proposed treatment decisions. He eventually makes the de-
is able to understand the proposed treat- (“They are planning on replacing my cision that her pain is significant enough
ment as well as the risks, benefits, and hip”) and that she can refuse the pro- to merit the risks and decides to pro-
alternative options, and she is able to posed treatment (“It’s up to me to have ceed with the hip replacement. On fol-
make and communicate her choice.68 the surgery or not”). However, she is low-up several months later, she is am-
Because of her tangential response, you unable to answer questions about the bulating well and is happy that she
are concerned that she may not appre- rehabilitation required after hip sur- “decided” to have the operation.
ciate the risks and treatment alterna- gery and the risk of surgical complica-
tives, so a formal capacity evaluation tions or death. You decide she doesn’t CLINICAL BOTTOM LINE
should be performed. appreciate the foreseeable conse- Capacity is a basic requirement for in-
You repeat the patient’s MMSE evalu- quences of accepting or declining the formed consent and is determined based
ation, and she scores 20 out of 30, simi- treatment. on the process of the patient’s deci-
lar to her previous score of 21. She has Weighing her responses to the ACE sion making rather than the final de-
Alzheimer disease, and this popula- and the moderate risk of the hip sur- cision itself. In most US jurisdictions,
tion has a high prevalence of incapac- gery, you decide she lacks capacity to the patient is required to demonstrate
ity (52%). Her MMSE score is in the make the decision about the proposed 4 abilities to have capacity10: ability to

Table 2. Studies Comparing Capacity Instrument With a Gold Standard

Level of
Source Capacity Instrument Gold Standard Evidence a LR⫹ (95% CI) LR− (95% CI)
Janofsky et al,38 1992 Hopkins Competency Forensic psychiatrist 2 54 (3.5-846) 0 (0.0-0.52)
(US) Assessment Test
(HCAT)
Fazel et al,3 1992 (UK) Fazel Questionnaire Expert psychiatrist 3 9.4 (4.6-19) 0.07 (0.02-0.26)
Etchells et al,27 1999 Aid to Capacity Forensic psychiatrist 2 8.5 (3.9-19) 0.21 (0.11-0.41)
(Canada) Evaluation (ACE)
Pruchno et al,54 1995 Understanding Forensic psychiatrist 2 6.0 (2.1-17) 0.16 (0.06-0.41)
(US) Treatment
Disclosure (UTD)
Fassassi et al,28 2009 Fazel Questionnaire Expert psychiatrist 2 4.4 (2.3-8.3) 0.69 (0.56-0.85)
(Switzerland)
Pruchno et al,54 1995 Hopemont Capacity Forensic psychiatrist 2 3.8 (1.5-9.5) 0.38 (0.21-0.68)
(US) Assessment
Interview (HCAI)
Molloy et al,49 1996 Specific Capacity Competency panel 1 2.0 (1.5-2.8) 0.12 (0.04-0.37)
(Canada) Instrument
(score ⬍16)
Billick et al,23 2009 Competency Expert psychiatrist 2 2.0 (0.7-6.1) 0.33 (0.08-1.4)
(US) Questionnaire–Medicine
(CQ-M)
Schmand et al,57 Clinical Vignette Forensic psychiatrist 1 1.7 (1.1-2.4) 0.29 (0.12-0.71)
1999 (the
Netherlands)
Rutman and Cognitive Competency Multidisciplinary 2 1.5 (0.87-2.7) 0 (0-3.0)
Silberfeld,56 1992 Test (CCT) competency
(Canada) panel
Abbreviations: CI, confidence interval; LR⫹, positive likelihood ratio; LR−, negative likelihood ratio.
a The Rational Clinical Examination level-of-evidence score rates the quality of studies of diagnostic tests on a scale from 1 (highest) to 3 (lowest).

©2011 American Medical Association. All rights reserved. JAMA, July 27, 2011—Vol 306, No. 4 423

Downloaded From: https://jamanetwork.com/ Biblioteca Virtual del SSPA by Jose-Maria Dominguez-Roldan on 12/18/2019
 INCAPACITY AND MEDICAL DECISION MAKING

appreciate the nature of one’s situa- to make decisions; this may be particu- a language in which the patient is not
tion and the consequences of one’s larly appropriate for a patient who does proficient. If a physician determines that
choices (an ability frequently im- not appear to appreciate the impor- a patient is incapacitated, he or she
paired in dementia-related illness and tance of the evaluation. The physician should consider whether the patient’s
not measured by the MMSE), ability to doing the capacity evaluation should do capacity can be optimized and capac-
understand the relevant information, so in the context of a specific decision, ity reassessed at a later date.
ability to reason about the risks and so he or she must be fully knowledge- Optimization can include treating re-
benefits of potential options,37,69 and able about the proposed decision (or versible disorders that affect cogni-
ability to communicate a choice. observe the person who is most knowl- tion (eg, drug-induced or metabolic de-
Capacity is influenced by a variety of edgeable explaining it to the patient), lirium), thought processes, or
factors, including situational, psycho- including the potential risks and ben- communication (eg, patient with mod-
social, medical, psychiatric, and neu- efits. The nature of the decision should erately severe Parkinson disease whose
rological factors.37 Accordingly, capac- be fully explained in the simplest pos- ability to communicate could be im-
ity exists on a continuum, can be sible language, using the patient’s own proved with medication adjustment);
evanescent, and can be optimized. Ex- words if possible, and reviewing infor- shortening and simplifying the infor-
perts recommend that the assessment mation as needed. Pseudo incapacity mation given to and asked of the pa-
begin by informing the patient about the occurs when the patient is provided in- tient; and using alternative methods of
purpose of the examination.70 Some ex- formation in a way he or she cannot un- communication. Methods to simplify
perts advise suggesting the patient do derstand. Examples include excess use the information presented include
his or her best to avoid losing the right of medical jargon or communicating in changing both wording and sentence

Table 3. Characteristics of Selected Competency Tests

No. of Time to Construct Criterion
Test Studies Complete, min Reliability a Validity b Validity c Availability
Aid to Capacity Evaluation (ACE) 1 10-20 Overall ␬ = 0.79 Correlation, Yes Free63
discriminate
Hopkins Competency Assessment 5 10 r = 0.96-0.97 Correlation, Yes Free38
Tool (HCAT) discriminate
Understanding Treatment Disclosure 1 ⬍30 ␣ = 0.55-0.85 Correlation, Yes Grisso and
(UTD) discriminate Applebaum64
Ability to Consent Questionnaire 1 ⬍30 Overall ␬ = 0.85 Correlation No Free, available from
(ACQ) author
Assessment of Capacity of Everyday 1 NS NS Correlation, No Free, available from
Decision Making (ACED) discriminate author
Capacity to Consent to Treatment 9 20-25 ␬ = 0.31-0.57 Correlation, No $200, University of
Instrument (CCTI) on the 5 discriminate, Alabama Research
domains factorial Foundation
Cognitive Competency Test (CCT) 1 60-120 NS Correlation Yes Wang et al65
Cognitive Questionnaire (CQ-M) 2 30 NS Discriminate Yes Free22
Decision Making Rating Scale (DMRS) 1 NS NS Correlation No Not available
Fazel Questionnaire 2 30-45 r ⱖ 0.92 Discriminate Yes Free3
Hopemont Capacity Assessment 4 ⬍30 ␬ = 0.93 Correlation, Yes Free,54 available from
Interview (HCAI) discriminate instrument author
MacArthur Competency Assessment 7 20-25 ICC = 0.87-0.99 Correlation, No $87.95, kit from
Test (MacCAT-T) discriminate Professional
Resource Press
Medical Decision Making Capacity 2 NS NS Correlation, No Free4
Instrument (D.CAPCTY) discriminate
Schmand Vignettes 1 NS ␣ = 0.82 Correlation, Yes Free57
discriminate
Specific Capacity Instrument 1 NS NS Face, Yes Not available
correlation
Structured Interview for 2 20 ␬ = 0.14-0.82 Face No Kitamura and
Competency/Incompetency for 12 Kitamura66
Assessment Testing and Ranking questions
Inventory (SICIATRI)
Vellinga Vignettes 1 NS ␬ = 0.64 Face No Free7
Abbreviations: ICC, intraclass coefficient; NS, not stated.
a ␣ is a Cronbach ␣. r is a Pearson (or Spearman) correlation coefficient.
b Levels of validity are described in eTable 2, available at http://www.jama.com.
c Compared with gold standard in medicine patients.

424 JAMA, July 27, 2011—Vol 306, No. 4 ©2011 American Medical Association. All rights reserved.

Downloaded From: https://jamanetwork.com/ Biblioteca Virtual del SSPA by Jose-Maria Dominguez-Roldan on 12/18/2019
 INCAPACITY AND MEDICAL DECISION MAKING

Box. Capacity Instruments That Can Be Performed in an Office Visit and That Have Robust Likelihood Ratios
and Moderate to Strong Levels of Evidence
Aid to Capacity Evaluation27 (1 question has 4 parts), for a total of 10 points. Scores lower
The Aid to Capacity Evaluation (ACE) uses the patient’s own than 3 suggest incapacity. The 8th-grade and then 6th-grade
medical situation and diagnosis or treatment decision. It pro- level essays are used if the patient does not score more than 3
vides training in presenting the information using a 1-para- using the higher-level essay.
graph scenario in which a patient with gangrene has to make a Understanding Treatment Disclosure54
decision about having an amputation. The instrument con- Understanding Treatment Disclosure (UTD) has 3 possible sce-
sists of 8 questions that assess understanding of the problem, narios: schizophrenia, major depression, and ischemic heart dis-
treatment proposed, treatment alternatives, the option to re- ease. In the original study, participants were given the sce-
fuse treatment, possible consequences of the decision, and the nario corresponding with their disease. They are given (oral
effect of an underlying mental disorder on decision. The in- and written) 5 paragraphs that provide information about the
strument includes a scoring manual that provides objective cri- disorder and treatment options. This is followed by 10 ques-
teria for scoring responses. tions that assess understanding of information, with respon-
Hopkins Competency Assessment Test38 dents providing their own paraphrasing of the information pro-
The Hopkins Competency Assessment Test (HCAT) is a 4-para- vided. The 5 paragraphs are presented again, 1 at a time, with
graph essay written at 3 reading levels (6th grade, 8th grade, questions involving the presentation of 4 statements about the
and 13th grade [completed high school]). The essay explains information, 2 of which are the same but in different words and
the nature of informed consent, the patient’s right to make de- 2 of which are different. Respondents indicate which state-
cisions, how certain decisions can impair decision-making abil- ments are the same. Scoring is from 0 to 2 points on each ques-
ity, and the patient’s right to make advance directives. The ex- tion, and the UTD yields 3 subscale scores: uninterrupted dis-
aminer reads aloud while the patient reads the same material, closure, paraphrased recall (UDPR); element disclosure,
starting with the 13th-grade example. The essay is followed by paraphrased recall (ED-PR); and element disclosure, recogni-
6 questions: 4 open ended, 1 true or false, and 1 sentence tion (ED-RC). The UTD has a scoring manual that provides ob-
completion. Answers are scored 1 point for each correct answer jective criteria for scoring responses.

structure, providing visual aids such as pable of medical decision making. Re- patients in the “gray” zone. Patients
photographs and drawings,19,60 and re- ferral to an expert in capacity evalua- with low cognition nearly always lack
ducing the dependence on verbal and tion may be required for confirmation capacity, and conversely those with
memory ability by presenting only small or when there is diagnostic uncer- high cognitive abilities usually retain ca-
amounts of material at a time.71 Be- tainty. pacity. However, these patients are not
cause capacity is evanescent and deci- Unfortunately, physicians often fail ones physicians are likely to mistak-
sion specific, a patient found to have to recognize incapacity. In our pooled enly categorize. These results show that
or lack capacity for one medical deci- analysis, physicians missed the diag- the MMSE could be used as a screen for
sion should be retested when future nosis in 58% of patients judged inca- patients very likely to have or lack ca-
medical decisions arise. pable, although when physicians do di- pacity and can provide support for per-
Problematic therapeutic relation- agnose incapacity, they are usually forming a formal capacity assessment
ships as well as cultural,72 linguistic, and right. Although patients are presumed when a patient’s score is low or in the
educational barriers can preclude reli- to have capacity unless proven other- gray zone. Any patient with a cogni-
able capacity assessment and should re- wise, missing a diagnosis of incapacity tion problem shown by a low MMSE
sult in referral of the patient. Undiag- means any informed consent the pa- score should be given relevant infor-
nosed depression or other psychiatric tient has given for medical treatment is mation about a health care decision and
illness can confound the capacity as- not valid. have his or her capacity for that deci-
sessment, so the examiner must deter- Physicians often rely on measures of sion explicitly addressed rather than the
mine whether the psychiatric illness is cognition, such as the MMSE, to aid in physician relying solely on the MMSE
affecting the patient’s decision mak- their capacity assessment. While cog- score for determining incapacity.
ing (eg, a depressed patient who feels nition has been consistently found to Numerous instruments have been de-
unworthy of medical treatment or a de- be the most important correlate of ca- veloped to improve physicians’ ability
lusional patient worried that parasites pacity, it is not the only criterion. In our to diagnose incapacity, and most have
about which she has delusions render pooled analysis, the MMSE performed undergone rigorous and repeated evalu-
her a nonsurgical candidate). It is im- well among patients with low cogni- ation for construct validity. Unfortu-
portant to note that psychiatric illness tion (MMSE score ⬍20) or high cog- nately, only a few have been judged
alone does not render a patient inca- nition (MMSE score ⬎24), but not for against a gold standard in a medical
©2011 American Medical Association. All rights reserved. JAMA, July 27, 2011—Vol 306, No. 4 425

Downloaded From: https://jamanetwork.com/ Biblioteca Virtual del SSPA by Jose-Maria Dominguez-Roldan on 12/18/2019
 INCAPACITY AND MEDICAL DECISION MAKING

population. The ACE is the only in- Conflict of Interest Disclosures: All authors have com- mer’s disease. J Am Geriatr Soc. 2000;48(8):919-
pleted and submitted the ICMJE Form for Disclosure of 927.
strument evaluated against a gold stan- Potential Conflicts of Interest and none were reported. 16. Dunn LB, Nowrangi MA, Palmer BW, Jeste DV,
dard, with an acceptable Rational Clini- Disclaimer: All of the views expressed in this article Saks ER. Assessing decisional capacity for clinical re-
are those of the authors and should not be construed search or treatment: a review of instruments. Am J
cal Examination level-of-evidence score to reflect, in any way, those of the Department of Psychiatry. 2006;163(8):1323-1334.
and robust test characteristics, that can Army, the Department of Defense, or the Depart- 17. Sturman ED. The capacity to consent to treat-
be performed in less than 30 minutes, ment of Veterans Affairs. ment and research: a review of standardized assess-
Online-Only Material: The eMethods, eTables, ment tools. Clin Psychol Rev. 2005;25(7):954-
is available for free online, and in- eFigures, and eReferences are available at http://www 974.
cludes training materials. Moreover, the .jama.com. 18. Harbord RM, Deeks JJ, Egger M, Whiting P, Sterne
Additional Contributions: We gratefully acknowl- JA. A unification of models for meta-analysis of diag-
ACE is based on the actual decision the edge Mary Hoffa, MD; Tracey Holsinger, MD; Bar- nostic accuracy studies. Biostatistics. 2007;8(2):
patient is facing. Most of the other in- bara Kamholz, MD; and James Tulsky, MD, Duke Uni- 239-251.
versity Medical Center, for their thoughtful advice in 19. Arscott K, Dagnan D, Kroese BS. Assessing the
struments use a clinical vignette, vio- the preparation of the manuscript. No one received ability of people with a learning disability to give in-
lating the tenet that capacity assess- compensation for their contributions. formed consent to treatment. Psychol Med. 1999;
ment is specific rather than generic. 29(6):1367-1375.
20. Baird AD, Solcz SL, Gale-Ross R, Blake TM. Older
Teaching specifically applicable REFERENCES adults and capacity-related assessment: promise and
legal standards, performing capacity 1. Barton CD Jr, Mallik HS, Orr WB, Janofsky JS. Cli-
caution. Exp Aging Res. 2009;35(3):297-316.
21. Bassett SS. Attention: neuropsychological predic-
evaluations using a standardized nicians’ judgement of capacity of nursing home pa- tor of competency in Alzheimer’s disease. J Geriatr Psy-
instrument when appropriate, and tients to give informed consent. Psychiatr Serv. 1996; chiatry Neurol. 1999;12(4):200-205.
47(9):956-960. 22. Billick SB, Della Bella P, Burgert W III. Compe-
testing capacity-centered cognitive 2. Folstein MF, Folstein SE, McHugh PR. “Mini-
tency to consent to hospitalization in the medical
domains (eg, semantic knowledge, mental state”: a practical method for grading the cog-
patient. J Am Acad Psychiatry Law. 1997;25(2):
nitive state of patients for the clinician. J Psychiatr Res.
short-term recall) could increase the 191-196.
1975;12(3):189-198.
23. Billick SB, Perez DR, Garakani A. A clinical study
consistency and accuracy of physi- 3. Fazel S, Hope T, Jacoby R. Assessment of compe-
of competency to consent to hospitalization and treat-
tence to complete advance directives: validation of a
cians’ capacity judgments. 15 Docu- patient centred approach. BMJ. 1999;318(7182):
ment in geriatric inpatients. J Forensic Sci. 2009;
54(4):943-946.
menting each aspect of a capacity 493-497.
24. Dreer LE, Devivo TA, Novack TA, Krzywanski S,
4. Fitten LJ, Waite MS. Impact of medical hospital-
assessment is important, including a ization on treatment decision-making capacity in the
Marson DC. Cognitive predictors of medical decision-
cognitive test score (if performed); making capacity in traumatic brain injury. Rehabil
elderly. Arch Intern Med. 1990;150(8):1717-1721.
5. Royall DR, Cordes J, Polk M. Executive control and Psychol. 2008;53(4):486-497.
capacity assessment questions asked 25. Dymek MP, Marson DC, Harrell LE. Factor struc-
the comprehension of medical information by elderly
and the patient’s responses; persons retirees. Exp Aging Res. 1997;23(4):301-313. ture of capacity to consent to medical treatment in pa-
tients with Alzheimer’s disease, an exploratory study.
present for the assessment; the exam- 6. Tan J, Hope T, Stewart A. Competence to refuse
J Forensic Neuropsychol. 1999;1(1): 27-48.
treatment in anorexia nervosa. Int J Law Psychiatry.
iner’s conclusions about capacity; and, 2003;26(6):697-707. 26. Dymek MP, Atchison P, Harrell L, Marson DC.
if retesting, the stability of the patient’s 7. Vellinga A, Smit JH, Van Leeuwen E, Van Tilburg Competency to consent to medical treatment in cog-
W, Jonker C. Competence to consent to treatment of nitively impaired patients with Parkinson’s disease.
decision making. Standardized forms geriatric patients: judgements of physicians, family Neurology. 2001;56(1):17-24.
can simplify this process and also pro- members and the vignette method. Int J Geriatr 27. Etchells E, Darzins P, Silberfeld M, et al. Assess-
Psychiatry. 2004;19(7):645-654. ment of patient capacity to consent to treatment.
vide a template to appropriately struc- 8. Wong JG, Clare CH, Holland AJ, Watson PC, Gunn J Gen Intern Med. 1999;14(1):27-34.
ture the examination. Although there M. The capacity of people with a “mental disability” 28. Fassassi S, Bianchi Y, Stiefel F, Waeber G. Assess-
to make a health care decision. Psychol Med. 2000; ment of the capacity to consent to treatment in pa-
are a number of candidate instruments tients admitted to acute medical wards. BMC Med
30(2):295-306.
for evaluating capacity in patients 9. Jourdan JB, Glickman L. Reasons for requests for Ethics. 2009;10:15. doi:10.1168/1472-6939-10-15.
without severe psychiatric disorders, evaluation of competency in a municipal general 29. Fitten LJ, Lusky R, Hamann C. Assessing treat-
hospital. Psychosomatics. 1991;32(4):413-416. ment decision-making capacity in elderly nursing home
and most are based on similar con- 10. Appelbaum PS, Grisso T. The MacArthur Treat- residents. J Am Geriatr Soc. 1990;38(10):1097-
structs, virtually all have been vali- ment Competence Study: I, Mental illness and com- 1104.
petence to consent to treatment. Law Hum Behav. 30. Griffith HR, Dymek MP, Atchison P, Harrell L,
dated against a gold standard in only 1 1995;19(2):105-126. Marson DC. Medical decision-making in neurodegen-
study. More studies are needed, but in 11. Appelbaum PS, Grisso T. Assessing patients’ ca- erative disease: mild AD and PD with cognitive
pacities to consent to treatment. N Engl J Med. 1988; impairment. Neurology. 2005;65(3):483-485.
the interim, we recommend using the 31. Grisso T, Appelbaum PS. Mentally ill and
319(25):1635-1638.
ACE to assist physicians in capacity 12. Abram MB, Ballantine HT, Dunlop GR, et al. Mak- non–mentally-ill patients’ abilities to understand
determination for medicine patients. ing Health Care Decisions. Washington, DC: Presi- informed consent disclosures for medication:
dent’s Commission for the Study of Ethical Problems preliminary data. Law Hum Behav. 1991;15(4):
Author Contributions: Dr Sessums had full access to in Medicine and Biomedical and Behavioral Re- 377-388.
all of the data in the study and takes responsibility for search; 1982. 32. Grisso T, Appelbaum PS. Comparison of stan-
the integrity of the data and the accuracy of the data 13. Kim SYH, Caine ED, Swan JG, Appelbaum PS. Do dards for assessing patients’ capacities to make treat-
analysis. clinicians follow a risk-sensitive model of capacity- ment decisions. Am J Psychiatry. 1995;152(7):
Study concept and design: Sessums, Zembrzuska. determination? an experimental video survey. 1033-1037.
Acquisition of data: Sessums, Zembrzuska, Jackson. Psychosomatics. 2006;47(4):325-329. 33. Grisso T, Appelbaum PS, Mulvey EP, Fletcher K.
Analysis and interpretation of data: Sessums, Jackson. 14. Etchells E, Katz MR, Shuchman M, et al. Accu- The MacArthur Treatment Competence Study: II, Mea-
Drafting of the manuscript: Sessums, Zembrzuska, racy of clinical impressions and Mini-Mental State Exam sures of abilities related to competence to consent to
Jackson. scores for assessing capacity to consent to major medi- treatment. Law Hum Behav. 1995;19(2):127-148.
Critical revision of the manuscript for important in- cal treatment: comparison with criterion-standard psy- 34. Grisso T, Appelbaum PS. The MacArthur Treat-
tellectual content: Sessums, Jackson. chiatric assessments. Psychosomatics. 1997;38 ment Competence Study: III, Abilities of patients to
Statistical analysis: Jackson. (3):239-245. consent to psychiatric and medical treatments. Law
Administrative, technical, or material support: 15. Earnst KS, Marson DC, Harrell LE. Cognitive Hum Behav. 1995;19(2):149-174.
Zembrzuska, Jackson. models of physicians’ legal standard and personal 35. Grisso T, Appelbaum PS, Hill-Fotouhi C. The
Study supervision: Sessums. judgments of competency in patients with Alzhei- MacCAT-T: a clinical tool to assess patients’ capaci-

426 JAMA, July 27, 2011—Vol 306, No. 4 ©2011 American Medical Association. All rights reserved.

Downloaded From: https://jamanetwork.com/ Biblioteca Virtual del SSPA by Jose-Maria Dominguez-Roldan on 12/18/2019
 INCAPACITY AND MEDICAL DECISION MAKING

ties to make treatment decisions. Psychiatr Serv. 1997; A, Harrell LE. Consistency of physician judgments of Medical decision-making capacity in patients with ma-
48(11):1415-1419. capacity to consent in mild Alzheimer’s disease. J Am lignant glioma. Neurology. 2009;73(24):2086-
36. Gurrera RJ, Karel MJ, Azar AR, Moye J. Agree- Geriatr Soc. 1997;45(4):453-457. 2092.
ment between instruments for rating treatment de- 48. Martin RC, Okonkwo OC, Hill J, et al. Medical 60. Tymchuk AJ, Ouslander JG, Rahbar B, Fitten J.
cisional capacity. Am J Geriatr Psychiatry. 2007; decision-making capacity in cognitively impaired Par- Medical decision-making among elderly people in long
15(2):168-173. kinson’s disease patients without dementia. Mov term care. Gerontologist. 1988;28(suppl):59-63.
37. Holzer JC, Gansler DA, Moczynski NP, Folstein Disord. 2008;23(13):1867-1874. 61. Vellinga A, Smit JH, Van Leeuwen E, Van Tilburg
MF. Cognitive functions in the informed consent evalu- 49. Molloy DW, Silberfeld M, Darzins P, et al. Mea- W, Jonker C. Decision-making capacity of elderly pa-
ation process: a pilot study. J Am Acad Psychiatry Law. suring capacity to complete an advance directive. J Am tients assessed through the vignette method: imagi-
1997;25(4):531-540. Geriatr Soc. 1996;44(6):660-664. nation or reality? Aging Ment Health. 2005;9(1):
38. Janofsky JS, McCarthy RJ, Folstein MF. The Hop- 50. Morris CD, Niederbuhl JM, Mahr JM. Determin- 40-48.
kins Competency Assessment Test: a brief method for ing the capability of individuals with mental retarda- 62. Vollmann J, Bauer A, Danker-Hopfe H, Helmchen
evaluating patients’ capacity to give informed consent. tion to give informed consent. Am J Ment Retard. 1993; H. Competence of mentally ill patients: a compara-
Hosp Community Psychiatry. 1992;43(2):132- 98(2):263-272. tive empirical study. Psychol Med. 2003;33(8):
136. 51. Moye J, Karel MJ, Azar AR, Gurrera RJ. Capacity 1463-1471.
39. Karlawish JH, Casarett DJ, James BD, Xie SX, Kim to consent to treatment: empirical comparison of three 63. Community tools: Aid to Capacity Evaluation
SY. The ability of persons with Alzheimer disease (AD) instruments in older adults with and without dementia. (ACE). University of Toronto Joint Centre for Bioeth-
to make a decision about taking an AD treatment. Gerontologist. 2004;44(2):166-175. ics. http://www.jointcentreforbioethics.ca/tools/ace
Neurology. 2005;64(9):1514-1519. 52. Okonkwo O, Griffith HR, Belue K, et al. Medical .shtml. Accessed November 8, 2010.
40. Kitamura F, Tomoda A, Tsukada K, et al. Method decision-making capacity in patients with mild cog- 64. Grisso T, Applebaum PS. Manual for Understand-
for assessment of competency to consent in the men- nitive impairment. Neurology. 2007;69(15):1528- ing Treatment Disclosures. Worcester: University of
tally ill: rationale, development, and comparison with 1535. Massachusetts Medical School; 1992.
the medically ill. Int J Law Psychiatry. 1998;21 53. Palmer BW, Dunn LB, Appelbaum PS, et al. As- 65. Wang P, Ennis K, Copland S. CCT Manual. To-
(3):223-244. sessment of capacity to consent to research among ronto, ON: Dept of Psychology, Mount Sinai Hospi-
41. Lai JM, Gill TM, Cooney LM, Bradley EH, Hawkins older persons with schizophrenia, Alzheimer disease, tal; 1987.
KA, Karlawish JH. Everyday decision-making ability in or diabetes mellitus: comparison of a 3-item ques- 66. Kitamura T, Kitamura F. Structured Interview for
older persons with cognitive impairment. Am J Geri- tionnaire with a comprehensive standardized capac- Competency/Incompetency Assessment Testing and
atr Psychiatry. 2008;16(8):693-696. ity instrument. Arch Gen Psychiatry. 2005;62(7): Ranking Inventory (SICIATRI). Ichikawa, Japan: Dept
42. Marson DC, Cody HA, Ingram KK, Harrell LE. Neu- 726-733. of Sociocultural Environmental Research, National In-
ropsychologic predictors of competency in Alzhei- 54. Pruchno RA, Smyer MA, Rose MS, Hartman-Stein stitute of Mental Health; 1993.
mer’s disease using a rational reasons legal standard. PE, Henderson-Laribee DL. Competence of long- 67. Holleman DR Jr, Simel DL. Does the clinical ex-
Arch Neurol. 1995;52(10):955-959. term care residents to participate in decisions about amination predict airflow limitation? JAMA. 1995;
43. Marson DC, Ingram KK, Cody HA, Harrell LE. As- their medical care: a brief, objective assessment. 273(4):313-319.
sessing the competency of patients with Alzheimer’s Gerontologist. 1995;35(5):622-629. 68. Uniform Health-Care Decisions Act. National Con-
disease under different legal standards: a prototype 55. Raymont V, Bingley W, Buchanan A, et al. Preva- ference of Commissioners on Uniform State Laws, Au-
instrument. Arch Neurol. 1995;52(10):949-954. lence of mental incapacity in medical inpatients and gust 1993. http://www.law.upenn.edu/bll/archives
44. Marson DC, Chatterjee A, Ingram KK, Harrell LE. associated risk factors: cross-sectional study. Lancet. /ulc/fnact99/1990s/uhcda93.htm. Accessed March 31,
Toward a neurologic model of competency: cogni- 2004;364(9443):1421-1427. 2011.
tive predictors of capacity to consent in Alzheimer’s 56. Rutman D, Silberfeld M. A preliminary report on 69. Welie JV, Welie SP. Patient decision making com-
disease using three different legal standards. Neurology. the discrepancy between clinical and test evaluations petence: outlines of a conceptual analysis. Med Health
1996;46(3):666-672. of competence. Can J Psychiatry. 1992;37(9):634- Care Philos. 2001;4(2):127-138.
45. Marson DC, Earnst KS, Jamil F, Bartolucci A, Harrell 639. 70. Grisso T, Appelbaum PS. Assessing Competence
LE. Consistency of physicians’ legal standard and per- 57. Schmand B, Gouwenberg B, Smit JH, Jonker C. to Consent to Treatment. New York, NY: Oxford Uni-
sonal judgments of competency in patients with Alz- Assessment of mental competency in community- versity Press; 1998.
heimer’s disease. J Am Geriatr Soc. 2000;48(8): dwelling elderly. Alzheimer Dis Assoc Disord. 1999; 71. Wong JG, Clare ICH, Gunn MJ, Holland AJ. Ca-
911-918. 13(2):80-87. pacity to make health care decisions: its importance
46. Marson DC, Dreer LE, Krzywanski S, Huthwaite 58. Tomoda A, Yasumiya R, Sumiyama T, et al. Va- in clinical practice. Psychol Med. 1999;29(2):437-
JS, Devivo MJ, Novack TA. Impairment and partial re- lidity and reliability of Structured Interview for Com- 446.
covery of medical decision-making capacity in trau- petency Incompetency Assessment Testing and Rank- 72. Betancourt JR, Green AR, Carillo JE. The chal-
matic brain injury: a 6-month longitudinal study. Arch ing Inventory. J Clin Psychol. 1997;53(5):443- lenges of cross-cultural health care: diversity, ethics,
Phys Med Rehabil. 2005;86(5):889-895. 450. and the medical encounter. Bioethics Forum. 2000;
47. Marson DC, McInturff B, Hawkins L, Bartolucci 59. Triebel KL, Martin RC, Nabors LB, Marson DC. 16(3):27-32.

©2011 American Medical Association. All rights reserved. JAMA, July 27, 2011—Vol 306, No. 4 427

Downloaded From: https://jamanetwork.com/ Biblioteca Virtual del SSPA by Jose-Maria Dominguez-Roldan on 12/18/2019