Decision-Making Capacity for Informed Consent in the Older Population

13 Pages·2005·0.56 MB·English

Decision-Making Capacity for
Informed Consent in the Older
Population
Katherine Christensen, MD, Ansar Haroun, MD, Lawrence J. Schneiderman,
MD, and Dilip V. Jeste, MD
We discuss key concepts and review 12 published research studies relevant to
informed consent and decision-making capacity in the older population. The
literature suggests that aging is associated with impaired decision-making capac-
ity; the following additional factors amplify the detrimental effect of aging: lower
vocabulary level, lower educational level, chronic medical illness (as in nursing
home residents), and acute medical illness. Aging may be associated particularly
with impaired comprehension of consent forms. We discuss guidelines for clini-
cians and researchers for improving the process of obtaining a truly informed
consent.
Ensuring that the process of informed several issues, such as how well the in-
consent occurs adequately for all the in- formation is presented, whether the recip-
dividuals is an important but difficult ient is hampered by mental or sensory
matter for clinicians and researchers impairments, the emotional tone of the
alike. How informed someone becomes situation, and the rapport with the exam-
through "informed" consent depends on iner. Although the process can be modi-
fied to suit individual needs, there may
still be some individuals for whom the
Dr. Christensen is a Fellow in Geriatric Psychiatry, San
Diego VA Medical Center and University of California, information is unclear or unintelligible,
San Diego, CA. Dr. Haroun is Supervising Forensic
Psychiatrist, Superior Court of California, San Dicgo and who lack the competence to give an
County Courthouse; Assistant Clinical Professor of Psy- appropriate informed consent in certain
chiatry and Pediatrics, University of California, San
Diego School of Medicine; and adjunct professor, Uni- situations.
versity of San Diego School of Law, San Diego, CA. Dr.
In geriatric psychiatry, there is partic-
Schneiderman is Professor of Community and Family
Medicine, Division of Health Care Sciences, and Pro- ular concern about the process of in-
fessor of Medicine, University of California, San Diego,
formed consent and about decision-mak-
CA. Dr. Jeste is Professor of Psychiatry and Neuro-
sciences, and Director, Geriatric Psychiatry Clinical Re- ing capacity-which is the clinical
search Center, University of California, San Diego and
equivalent of competence. The older pop-
San Diego VA Medical Center, San Diego, CA. Address
correspondence to: Dilip V. Jeste, MD, Psychiatry Ser- ulation as a whole is a heterogenous
vice (I IbA), VA Medical Center, 3350 La Jolla Village
Drive, San Diego, CA 92 16 1. group, but includes a high proportion of
Bull Am Acad Psychiatry Law, Vol. 23, No. 3, 1995 353
Christensen et a/.
subjects at risk for impairments in deci- deserve a brief comment. Informed con-
sion-making capacity (e.g., those with sent cannot occur without an adequate
cognitive impairments1-5 and/or psychi- presentation of the information. This
atric the groups most com- brings into play such concerns as the
monly seen by geriatric psychiatrists). readability levels of consent forms and
The older population is also at a greater standardization of verbal information-
risk for physical illness, and consequently giving processes that may occur without
more often faces important decisions consent forms (e.g., for routine medical
about medical treatment, long-term care, treatment decisions).14 That subjects
and life-sustaining measure^.^ The chal- should participate voluntarily may seem
lenge is to identify subjects at risk for self-evident, but coercion can be sub-
''
impaired decision making and to ensure tle.', Peer group pressure, the desire to
an adequate process of informed consent please authority figures or caregivers, and
for them-while not infringing upon their fear that care or treatment might be with-
rights to autonomous decision making. held can be powerful and not necessarily
First, we will briefly review relevant obvious influences."'
key concepts. Precise definitions in this Competence and Decision-Making
area are particularly important, in part Capacity Legally, any adult is compe-
due to the presence of both legal and tent to make decisions for himself or her-
medical definitions. We will next review self unless the person is declared incom-
current research literature related to deci- petent by a court of law. Because the term
sion-making capacity and the process of "competence" is a legal term, the use of
informed consent in older patients. We the term decision-making cczpacit?.l has
will then offer guidelines for the clinician come to represent what the clinician as-
and researcher to improve the process of sesses. An adequate assessment of deci-
achieving informed consent. Finally, we sion-making capacity is an estimate of
will make suggestions for future research. what a court would term compe-
rence.'3, '5, 16
Numerous articles have been written as
Definitions guides for the clinician to assess decision-
Hidden within the process of obtaining making capacity, based on the courts'
"
informed consent for research or for med- definition of competence.8,
13, l7
ical treatment are the related but different One of the representative explications of
concepts of informed consent, compe- the legal standards on which a determina-
tence, and decision-making capacity. tion of competency is based (and there-
Informed Consent Informed consent fore also a guideline for the assessment
includes three aspects: full information, of decision-making capacity) is the fol-
voluntary participation, and competence lowing: choice, comprehension, conse-
to make the decision at hand.63 The quences, rationality, and reasonable out-
7' l3
last of these is the most represented in the come of choice.2, 7'9, 15, 16, 18, 19
research in this area, but the first two To satisfy the criterion of choice, a
354 Bull Am Acad Psychiatry Law, Vol. 23, No. 3, 1995
Informed Consent in Older Population
person must indicate a choice, and the cation of understanding of the situation and
choice indicated must remain consistent, its potential risks.'. Others argue for
I" 16, I"
at least long enough for the choice to be a more autonomous stance, wishing to err
enacted.", To meet the criterion of on the side of protecting the individual's
I"
comprehe~lsiolza, person must be deemed right to make decisions, even ones that
able to comprehend the matter at hand in might be viewed by others as .'bad" deci-
sions.~, Potential influences in all these
sufficient detail so that the main items are 10. 16
understood.*,"' A person who can dem- discussions are matters of ethics, public
onstrate an understanding of "conse- sentiment, and public policy, as well as
quences'' is able to identify his or her own protection of legal rights.
role in the situation and to appreciate the Not all the decisions are of the same
personal implications. l5 Rntiolzality is level of complexity or difficulty. There-
present when a person can give evidence fore, a person may have the capacity to
of having considered the benefits and make one decision and not anoth-
risks of the choices."^ In its purest er. 13. 16, 17 The criteria noted above
I"
sense, this criterion should not include the should be closely tied to an individual
examiner's assessment of whether the de- decision. Furthermore, the basis for de-
cision is wis~.~," An additional crite- termining whether someone has intact de-
rion, considered by some examiners, but cision-making capacity is not the same as
debated by others, is reasonable outcome the basis for determining whether some-
of choice.*, 7, his criterion expressly one is cognitively impaired, or whether
considers the examiner's sense (as re- someone has impaired judgement, or
flecting what any reasonable person even impaired memory. Yet at some point
might believe) of the wisdom of choice on a continuum of cognitive impairment,
based upon its potential outcome. individuals are likely to cross a threshold
There is often no single correct way to beyond which they have difficulty with
look at these criteria. If they are all met, decision-making capacity. Just where this
then decision-making capacity is intact. If threshold may fall is, however, often dif-
only some of them are met, however, the ficult to determine.
issue becomes more cloudy. One point of
view is that the criteria for competence
listed above can be regarded in a hierarchial
fashion, and can be differentiated by the Materials and Methods
level of "protection" provided to the indi- For this article we searched the En-
vidual. Some authors advocate that a differ- glish-language medical and psychiatric
ent level of decision-making capacity literature for studies that would be rele-
should be present for different levels of vant to decision-making capacity and in-
decision. Decisions with more risk, or those formed consent in older populations. Our
that do not necessarily benefit the individ- search identified 12 studies addressing
ual, such as some research protocols, may the concept of decision-making capacity
need to be accompanied by a greater indi- for informed consent that would be rele-
Bull Am Acad Psychiatry Law, Vol. 23, No. 3, 1995 355
Table 1
Literature Review: Methodology
Mean Age (range
Investi- Patient Characteristics or standard Specific Decision Aspect@) of Decision- Statistical
gators (N = sample size) Comparison Group deviation) Considered Making Capacity Other Scales or Tests Tests
Taub, 1. Community-dwelling 2. Community- I.7 0 (55-83) None Comprehension of prose WAlS vocabulary and two ANOVA
2
197g2' older women (N = 27) dwelling younger 2. 27.3 (19-36) materials memory tests and
women (N = 27)
Soskis Hospitalized men with None Not given Knowledge of their Factual information about None
and schizophrenia (N = 25) own antipsychotic medications
Jaffe, medication
19 7gz4
Stanley 1. Psychiatric inpatients 2. Medical Not given Consent for Reasonableness of decision Brief Psychiatric Rating t tests
et a/., with various psychiatric inpatients hypothetical (wiilingness to enroll in low Scaie and
19816 illnesses (N = 27) (N = 38) research of risklhigh benefit studies) correla-
differing tions
risks/benefits
Cassiieth Cancer patients receiv- None Median 59 Consent for actual Recall of essential None ANOVA
2
eta/., ing chemotherapy, (20-82) treatment of information of treatment and
1980'4 radiation therapy, or [85% 1 451 differing types consent one day after
surgery (N = 200) consent
Taub, I. Community-dwelling 2. Community- 1. 71.3 (57-83) Actual study Memory of main points of WAlS vocabulary ANOVA
2
198OZ5 older women (N = 56) dwelling younger 2. 28.5 (22-35) consent study, assessed two to and
women (N = 34) three weeks after initial
consent
Taub et Community volunteers Community (57-87) Actual study 1. Comprehension of main 1. WAIS-R vocabulary ANOVA
2
a (N = 42) tested for volunteers consent points of study and
198lZ0 comprehension and (N = 45) not 2. Memory of same points
given immediate tested for 2-3 weeks later
feedback on consent comprehension or
form given feedback
Taub, 1. Community-dwelling 2. Community- 1. 70 (55-83) None Comprehension of prose WAlS vocabulary and two ANOVA
2
1979" ooler women (N = 27) dwelling younger 2. 27.3 (19-36) materials memory tests and
0
women (N = 27)
Taub Community volunteers Community 71.3 (59-88) Actual study 1. Comprehension of main 1. WAIS-R vocabulary ANOVA %
and (N = 50) given one volunteers consent points of study
B19a8k3e"r, caonmd pfereehdebnasciko nin titriiaalli y (uNp t=o 530 ) given 2. 2M-3e mwoerye kosf siaatmere points Vt3D )
3
comprehension
trials initially
Mean Age (range
Investi- Patient Characteristics or standard Specific Decision Aspect(s) of Decision- Statistical
gators (N = sample size) Comparison Group deviation) Considered Making Capacity Other Scales or Tests Tests
Stanley 1. Older medical 2. Younger medical I. 69.2 (5.3) Consent for 1. Comprehension of 1. Quick Test for verbal IQ t tests,
2,
eta/., inpatients and inpatients and 2. 33.7 (6.6) hypothetical consent information 2. Attention: five-item and
1984' outpatients (N = 39) outpatients research of (nature of procedure, questionnaire partial
(N = 41) differing risks, benefits) correla-
riskslbenefits 2. Quality of reasoning tions
(weighing risks, benefits)
3. Reasonableness of
decision (choice of low
risk, high benefit studies)
Taub Medical patients None Consent for Comprehension of range of None ANOVA
et a/., scheduled for cardiac treatment (cardiac facts about treatment
19 8623 catheterization catheterization)
(N = 108)
Taub, I. Community-dwelling 2. Community- 1. 70 (55-83) None Comprehension of prose WAIS vocabulary and two ANOVA
2
1979" older women (N = 27) dwelling younger 2. 27.3 (19-36) materials memory tests and
women (N = 27)
Stanley 1. Dementia patients 2. Major depression I. 70.0 (6.2) Consent for 1. Comprehensive of 1. Guild memory test ANOVA
eta/., (N = 38) (N = 45) 2. 67.8 (6.2) treatment (various consent information 2. WAIS-R vocabulary
19 887 3. Non-psychiatric 3. 68.7 (6.4) medication trials) (purpose, method, risks, 3. Attention-questionnaire
controls (N = 20) benefits) 4. Global Deterioration
2. Quality of reasoning Scale
(weighing risks, benefits) 5. Brief Psychiatric Rating
Scale
6. Hamilton Depression
Scale
Fitten 1. Veterans Affairs 2. Community 1. (>60) Consent for 1. Comprehension of key 1. Mini-Mental State Exam t tests
2
et a/., nursing-home volunteers 2. similarly aged hypothetical points 2. Short Portable mental and
1990' residents (N = 51) (N = 15) treatments 2. Understanding risks, status Questionnaire
benefits 3. Activities of Daily Living
3. Quality of reasoning Scale
4. Instrumental Activities of
Daily Living Scale
Fitten 1. Acutely (but not 2. Community I.6 8.4 (6.5) Consent for 1. Comprehension of Mini-Mental State Exam t tests
2
and critically) ill volunteers 2. 70.9 (9.2) hypothetical consent information and
Waite, hospitalized medical (N 25) treatments (medical condition,
-
19 903 and surgical patients treatment, purpose, risks,
(N = 25) benefits)
2. Quality of reasoning
WAIS, Wechsler Adult Intelligence Scale; WAIS-R, WAIS-Revised.
Christensen et a/.
vant to the older population, either for impairment in decision-making capacity.
research or for clinical purposes. They may therefore reflect the success of
the informed consent process by the usual
Results
standard procedures, independent of im-
The above-mentioned 12 studies are pairments in decision-making capacity.
summarized in Table 1 (methodology) Other measures of decision-making ca-
and Table 2 (results). More specific re- pacity used in different investigations
sults arc discussed below. were rationality or quality of reason-
Measures of Decision-Making Capac- ing'p3,7 and apparent reasonableness of
ity Used Eight of the 12 studies gave choice, as discussed earlier under "Defi-
results regarding an assessment of con?- itions" 2, 7'
pr-ehe~zsion,e ither as their sole assess- Reported Predictors of Decisioiz-Mak-
ment or as a part of their overall assess- irzg Capacity. Age Five out of the 12
ment.
I-" 7, 20-23 Not all of the studies studies included populations that were
evaluated cornprehension in the same both "older" and "younger," although the
manner. Perhaps the greatest difference age cut-offs used varied across the stud-
ies.',
was in terms of whether or not the study 14, 21, 23, 25 Age by itself and/or age
considered differences in memory (i.e., in plus another factor were found to be as-
the ability to recall essential information sociated with some aspect of poorer de-
about the consent days or weeks later). cision-making capacity in four of these
Nonetheless, all eight investigations five st~dies.2~''. 25 In three studies,
237
found comprehension to be impaired to older age by itself was associated with
'"
some degree, at least in the experimental poorer decision-making capacity.'-
",
group. The report by Cassileth et a1.I4 was the
Two reports evaluated recall of key only one that found no difference due to
items about informed consent for treat- age alone that was independent of educa-
ment. Soskis and ~affe~f"ou nd that tional level. This study included a wide
knowledge of antipsychotic medication range of ages, from 20 to 82 years old,
among hospitalized chronic schizo- and a preponderance of older patients, but
phrenic men varied, ranging from know- used age 45 as a cut-off, with 85 percent
ing potential side effects (84% with cor- of the patients being over age 45. Five
rect information), to knowing how the other studies were performed entirely in
"',
medication worked (40%), to knowing populations of older subjects,l>', 73 "
their correct dosage (36%). Similarly, and two additional studies did not report
Cassileth et a/. reported that knowledge age, but their sample populations ap-
l4
of treatmcnt among 200 patients with peared to include at least some older sub-
cancer varied, ranging from 60 percent ject~.""~I n two of these studies some
who could describe what the treatmcnt aspect of decision-making capacity was
would involve, to only 27 percent who found to be impaired in both the experi-
could name even one available altcrna- mental group and the control older
tive. These studies did not directly assess groups, suggesting that aging was some-
358 Bull Am Acad Psychiatry Law, Vol. 23, No. 3, 1995
Informed Consent in Older Population
how involved in the items assessing decision-making capac-
Vocabulary Levels Vocabulary levels ity, Stanley et ~11."~va' luated reasonable-
correlated with comprehension of consent ness of decision. The authors organized
forms and/or prose materials in the inves- hypothetical studies on the basis of risk
tigations by Taub et a1.2"p22,2s All of versus benefit (high riskllow benefit to
these studies found a significant positive low risklhigh benefit) and compared an
correlation between performance on com- experimental group to a control group.
prehension of consent forms and vocabu- No significant differences were observed
lary levels. Two of the studies compared in the assessment of riskiness of decision
different age groups stratified by vocab- in psychiatric inpatients versus medical
ulary In both studies, older
inpatients (ages were not given)." In a
subjects with poorer vocabulary levels
similar study, which looked at depressed
had significantly poorer comprehension
versus demented versus normal control
than younger subjects with comparable
older subjects, Stanley et al.' found that
vocabulary levels, suggesting an indepen-
only the dementia group was impaired in
dent contribution of aging to impaired
its reasonableness of decisions, but that
comprehension.
there was no significant difference be-
Educational Level When both items
tween elderly depressed patients and eld-
were assessed, vocabulary level corre-
erly control subjects in this respect.
lated with educational ~evel.~'~'"n the
Recrdubility Levels Two studies com-
Stanley et nl. (2) study, older and younger
pared results when using consent forms of
groups were not significantly different in
substantially different readability levels:
verbal IQ or in education, yet older pa-
~aub"c ompared forms written to a 12th
tients performed worse on measures of
grade versus a 6th to 7th grade compre-
comprehension. In contrast, Cassileth et
hension level; ~ a u b ~al's o compared
a1.I4 found that low education, but not
forms written to a college versus a 7th
older age per se, was associated with poor
grade level. Contrary to their hypotheses,
recall of consent information. Taub et
neither study found consistent differences
n1.23 concluded that both older age and
attributable to the readability level of the
lower education contributed to poor com-
forms.
prehension of consent forms.
Other Factors Cassileth et n1.l' re-
Chronic m d Acute Mediccrl Illness
Using their devised construct of decision- ported that the care with which the
making capacity (see Table 1 for details), patients thought that they had read the
the two studies by Fitten et al.13' found consent forms before signing was signif-
significantly higher impairment in deci- icantly associated with subsequent recall
sion-making capacity in their two test of information provided. Taub and col-
groups: nursing-home residents (chroni- leagues2o, 7.2'23 found that immediate test-
cally n~edicallyi ll) and acutely medically ing for comprehension of the consent
ill hospitalized patients. form with corrective feedback improved
Psychiatric I1l~ies.s As one of their comprehension significantly.
Bull Am Acad Psychiatry Law, Vol. 23, No. 3, 1995 359
Christensen et al.
Table 2
Literature Review: Results
Investigator Results
-- . - - - -- -
Taub, 1979'' Older subjects made significantly more errors than middle-, but not high-
vocabulary groups. Comprehension scores were significantly better than
memory tasks for both age groups and at all vocabulary levels. Vocabulary
scores varied directly with educational level.
Soskis and Patients' knowledge about different aspects of their psychiatric medication
Jaffe, 19 7924 ranged from knowing potential side effects (84%) to knowing how the drugs
worked (40%) and the actual dosage (36%).
Stanley et a/., Psychiatric patients and controls were not significantly different in choices to
19816 enroll in hypothetical research studies that differed in risks and benefits.
Cassileth et a/., Only 60% could describe what treatment would involve, 59% could list a single
1980'~ major risk or complication, and 27% could name an alternative treatment. Poor
recall was associated with low education, being bedridden (nonambulatory),
and inadequate care with which the patients thought that they had read the
consent form. Age, race, gender, or treatment modality had no significant
effect.
Taub. 19 8025 Most patients answered one or more questions incorrectly when tested two to
three weeks later. Number of incorrect responses was higher in those with
lower levels of vocabulary and education.
Taub et a/., Recall was poorer in patients with low vocabulary level, and in those with worse
I981" O initial comprehension. Corrected feedback improved recall significantly at all
vocabulary levels.
Taub and Comprehension and memory performance increased directly with vocabulary
Baker, level. Comprehension performance (but not memory) was enhanced by
1983" multiple trials at all vocabulary levels.
Stanley eta/., Compared with younger patients, the older patients did (1) significantly worse on
19 842 comprehension of consent form items, and (2) nonsignificantly worse on
quality of reasoning, but (3) did not differ on reasonableness of decision in
most cases.
Taub et a/., Older age and lower education were significantly associated with impaired
19 8623 performance on measures of comprehension on consent forms. The effects of
readability of consent form were inconsistent. Corrective feedback improved
performance on repeat trials.
Stanley et a/., 1. Dementia patients, but not those with major depression, had significantly
19 887 poorer comprehension of consent forms than controls.
2. The 3 groups did not differ on quality of reasoning, although the quality was
only moderate in all groups.
Compared with controls, only one-third of nursing home residents had intact
decision-making capacity, with one-third being severely and another third
moderately impaired. Impairment in decision-making capacity correlated with
cognitive, but not functional (activities of daily living), impairment.
Fitten and Hospitalized patients scored significantly lower than controls on assessment of
Waite, 19903 comprehension and quality of reasoning about hypothetical research studies.
Per study's definition, 7 of 25 subjects were impaired compared with 1 of 25
control subjects.
Bull Am Acad Psychiatry Law, Vol. 23, No. 3, 1995
Informed Consent in Older Population
Table 2 (Continued)
Literature Review: Results
Comments
The authors concluded that inadequate acquisition (assessed here by comprehension) may
represent one factor underlying age-related differences in immediate retention of information. No
true decision-making capacity was assessed, but cognitive concepts involved have clear bearing
on decision-making capacity.
Information on age was not given. It was not clear how subjects' knowledge reflected decision-
making capacity. Not all subjects had true choices about taking medications.
Information on age was not given. Outcome of decision was the only area examined relevant to
decision-making capacity.
24.9% said they did not read the consent form at all. Hence their performance might not reflect
decision-making capacity.
Immediate correction of incorrect answers led to improved recall of items at a later date. Quality of
reasoning was not assessed. Correct responses may not truly reflect comprehension or ability to
appreciate implications of information.
Correct responses may not truly reflect comprehension or ability to appreciate implications of
information. Quality of reasoning was not assessed.
Correct responses may not truly reflect comprehension or ability to appreciate implications of
information. Quality of reasoning was not assessed.
Groups were not different in IQ or education level. Despite significantly poorer comprehension in
older subjects, most decisions reached were reasonable.
Correct responses may not reflect ability to appreciate implications of information.
Consent forms used to assess subjects were not standardized, because subjects were undergoing
different medical treatments.
Control scores were adjusted until the reference group achieved nearly a perfect score as the
control subjects were assumed a priori to have largely intact decision-making capacity. Nearly
one-third of nursing home residents had impaired decision-making capacity for some tasks, but
intact capacity for others.
Mini-Mental State Examination (MMSE) scores were significantly lower in patients than in controls.
Bull Am Acad Psychiatry Law, Vol. 23, No. 3, 1995 361
Christensen et a/.
Discussion suggested. Aging appears to be associated
with impaircd capacity for decision mak-
Metlzodological Limitations The
ing.2.
14, 21, 2'3, 25 The following factors
studies cited above evaluated different
amplify the detrimental effects of aging
sample populations with different meth-
on decision making: (1) lower vocabulary
ods. They used varying definitions of de-
leve1,20-22, (2) lower educational lev-
25
cision-making capacity and/or assessed
el,I43 (3) chronic medical illness (as in
different components of decision-making 2'3
nursing home residents),' (4) acute med-
capacity. The investigators either em-
ical illness,%nd (5) cognitive impair-
ployed only a few tests of cognitive abil-
ment.7 Aging may be particularly detri-
ity, or did not define or assess cognitive
mental for the comprehension of consent
deficits at all. Some of the reports had
forms.~-x On the other hand, the
7, 2-23
small sample sizes. Also, some investiga-
following factors were not shown to have
tors studied actual treatment or research
great effects on decision-making capac-
protocols, while others studied hypo-
ity: (1) depression,7 (2) other psychiatric
thetical treatment or research protocols.
illness," and (3) readability level of in-
This distinction between treatment and ''
formed consent form~.~" Such "nega-
research, and between actual and hypo-
tive" results must, however, be inter-
thetical situations, may have some impor-
preted cautiously in view of the study
tance. Thus, conditions from which one
limitations. It is thus likely that major
may actually derive bcncfit or harm may
differences in severity of depression or in
have more personal meaning and might
readability of consent forms would influ-
lead to better attention and recall for such
ence the process of informed consent. Fi-
information as opposed to hypothetical
nally, informed consent may not be all
situations. On the other hand, there is a
that informed, even in subjects with pre-
potential for denial and repression as psy-
sumed intact decision-making capacity.
chological confounders in cases of actual
Other factors in addition to decision-
treatment situations, particularly for the making capacity (such as careful reading
seriously ill person. Finally, a majority of of the consent form) may also affect the
the studies listed in Tables 1 and 2 were process of informed consent. l4
performed by a small number of research- Overall, the studies that were reviewed
''
ers. Taub er d2'-'" accounted for five examined a very limited number of di-
of the studies, Stanley et aL2' 6' for three, mensions of cognitive function. When
and Fitten et al. "or two. Hence, there factors in addition to age were consid-
were no multiple diverse samples in ered, they seemed to have an impact in
which these results were replicated using further separating the older from younger
similar methods. groups, or the experimental groups of
Summary of Reszslts The above- older patients from older controls. This
mentioned methodological limitations re- may suggest that as one ages there is less
strict our intcrpretation of the data. None- "decision-making reserve." Most proba-
theless, some conclusions may be bly, multiple cognitive skills are utilized
362 Bull Am Acad Psychiatry Law, Vol. 23, No. 3, 1995