CME

Geriatrics Primary Care Geriatrics

Series editors: Rosanne M. Leipzig, MD, PhD, and Helen K. Edelberg, MD

Functional assessment
Easy-to-use screening tools speed initial office work-up
Fredrick T. Sherman, MD, MSc

The mnemonic DEEP-IN stands for a series of easy-to-administer, office-

based screening tests designed to streamline initial assessment of the
geriatric patient. These screens can quickly identify signs of delirium,
dementia, depression, and adverse drug effects; vision and hearing screened for in older patients and in-
deficits; risk for future impairments in activities of daily living; clude delirium, dementia, depression,
incontinence, and malnutrition. The author developed the mnemonic polypharmacy side effects, vision and
from personal experience and from an evidence-based application of hearing impairments, decline in gen-
validated geriatric and geropsychiatric studies.
eral physical performance, inconti-
Sherman FT. Functional assessment: Easy-to-use screening tools speed initial office work- nence, and malnutrition.
up. Geriatrics 2001; 56(August):36-40.
The screens suggested by the mne-
monic are not diagnostic, but their re-
sults can indicate which patients may
benefit from further testing or inter-
vention. The “D” in the mnemonic

A
n older patient’s functional Systematic screening for common represents delirium, dementia, de-
status is one of the most im- geriatric syndromes can help reduce pression, and drugs. The subsequent
portant considerations in the the chances that a problem will go un- categories are “EE” for eyes and ears,
initial geriatric assessment. Declining treated and worsen with time.1,2 “P” for physical performance, “I” for
functional status is a marker for un- Screening results can also spur med- incontinence, and “N” for nutrition.
derlying conditions that as yet may be ical or lifestyle interventions that pro-
undiagnosed. Performing an assess- duce improvements in functional dis- Delirium
ment that reveals emerging problems abilities. Such progress is important For cases of suspected delirium in an
is challenging, however, because cer- because even incremental gains can older adult, a good rule of thumb is to
tain syndromes and conditions man- produce significant positive effects for consider any change in mental status to
ifest differently in the geriatric patient older patients and their families. be a delirium until proven otherwise.
than in younger populations. But the This article describes how to use a This is particularly true if the episode
effort can be streamlined by using series of easy-to-administer, office- occurs when a patient is in the emer-
screening tools that are tailor-made for based screening tests designed to gency department, the hospital, a nurs-
the primary care setting. streamline initial assessment of the ing home, or an assisted living setting.
geriatric patient. These screens can The Confusion Assessment Method3 can
quickly identify conditions that un- be used to identify the presence of a
dermine physical and cognitive status, delirium in a patient who meets the first
Dr. Sherman is medical director independence, and quality of life. two criteria and either of the other two:
for senior services, Mount Sinai NYU ● acute onset and fluctuating course
Health, New York; clinical associate
DEEP-IN (Is there evidence of an acute change
professor, The Brookdale Depart-
ment of Geriatrics and Adult Develop- The mnemonic DEEP-IN (table 1) can in mental status, and does the behav-
ment, Mount Sinai School of Medi- help physicians quickly identify older ior come and go?)
cine, New York; and Medical Editor of patients who are frail or at high risk ● inattention (Does the patient have
Geriatrics. This article is based on a
talk he gave during the Geriatrics
for frailty. The disorders encompassed difficulty focusing attention? Is the pa-
Educational Conference and Exposi- by DEEP-IN go beyond the medical tient easily distracted or having diffi-
tion in April 2000 in New York, NY. conditions and syndromes typically culty following what is being said?)

36 Geriatrics August 2001 Volume 56, Number 8

 ● disorganized thinking (Is the pa-
tient’s speech rambling or irrelevant, or Table 1 DEEP-IN mnemonic for geriatric functional assessment
switching from one subject to the next?) D Delirium, dementia, depression, drugs
● an altered level of consciousness. E Eyes (vision impairment)
(A normal patient should be alert; any
E Ears (hearing impairment)
other assessment of the patient’s level
of consciousness [eg, lethargic, stupor, P Physical performance and “phalls” (falls)
or hyperalert] is abnormal.) I Incontinence
Younger persons with a delirium, N Nutrition
including those experiencing alcohol
withdrawal syndrome, typically exhibit Source: Prepared for Geriatrics by Fredrick T. Sherman, MD, MSc

a hyperkinetic, hyperalert delirium. In

older persons, however, a delirium interpret, particularly in patients with level of education, but 18 and 12 are
tends to be “quiet.” It is marked by little education. An alternative is an eas- generally good cutoff points for nor-
lethargy and difficulty responding to ily administered, inexpensive, and ef- mal and abnormal results, respectively.6
stimuli, and patients are hypokinetic fective screen that allows the physician Clock completion test. This test evalu-
and hypoalert. Causes include drugs; to quickly identify patients who war- ates nondominant parietal lobe func-
electrolyte imbalance; organ-specific rant a more thorough examination. tion, which is often diminished in early-
illnesses such as lung, liver, cardiac, and Although many tools are available, stage Alzheimer’s disease. To perform
renal disease; common infections such the five discussed below evaluate re- the test, draw a circle approximately 3
as pneumonia, gallbladder sepsis, and cent memory, verbal fluency, and con- inches in diameter on an unlined piece
urinary tract infections; and pain. structional abilities. They accomplish of white paper, then ask the patient to
Lingering effects. Most clinicians pre- the goal of distinguishing the remark- write the numbers 1 through 12 in the
sume that an episode of delirium is re- able from the unremarkable patient. circle so that the result resembles a stan-
versible, but the data suggest otherwise. If the results of any of these screens dard clock face. Patients with dementia
One important study looked at the out- raise suspicions of dementia, the tend to bunch the numbers in an un-
comes of 325 hospitalized older patients MMSE should be administered. even manner, usually placing most of
with delirium who were admitted from Three-item recall. To conduct this test, them on the right half of the circle. The
either the community or a long-term tell the patient that you are going to most accurate way to score this test is to
care facility.4 Using DSM-III criteria, the name three objects (eg, ball, flag, and section the clock face into quadrants
researchers found complete resolution tree) and that you want her to re- after the patient has completed the task
of delirium symptoms in 4% of patients member them so that she can recite (figure). On a properly performed exam,
at hospital discharge, in 20% after 3 them 1 minute later. Recall of all three each quadrant will contain three num-
months, and in 17% after 6 months. Av- items suggests a low probability of de- bers (12 to 2, 3 to 5, and so on).7
erage length of hospitalization was 19 mentia, whereas recall of only one or The patient’s entries in the first three
days for patients with a delirium, com- two is associated with a moderate in- quadrants are scored as either 0 (nor-
pared with 7 days for those without one. crease in the odds of dementia.5 mal) or 1 (abnormal). The fourth
Finally, older patients with a delirium Animal-naming test. This screen is used quadrant (9 to 11), which is the most
had a seven-fold increased risk for nurs- to gauge impairment of verbal fluency sensitive indicator of dementia, is
ing home placement. and access to semantic memory. Ask scored as either 0 (normal) or 4 (ab-
Thus, when counseling the family the patient to name as many animals normal). Add up the scores of the four
of a patient with delirium, it is im- as he or she can in a 1-minute span. quadrants (maximum, 7). Any score
portant for physicians to discuss po- The typical response of a patient with of 4 or more is a good indication of
tential outcomes frankly and realisti- Alzheimer’s disease would be: “dog, the presence of dementia. Because of
cally. Complete resolution of delirium cat, cow, [long pause] dog. . . .” Then the high number of points assigned to
is often a slow process; in some cases, the patient’s attention will drift off, and it, results from the fourth quadrant
it can take months to years. he or she will lose focus. alone are often diagnostic.
Older persons without dementia The clock completion test was val-
Dementia can usually name 18 different animals idated in a study that compared its re-
The Folstein Mini-Mental State Exam- within the 1-minute span. Anything sults with those using standard men-
ination (MMSE) is widely used to help less than 12 is abnormal and correlates tal status tests such as the Short Blessed
evaluate patients with suspected de- well with an MMSE score of less than Test.7 The investigators retrospectively
mentia, but it can be time-consuming 23. Animal-naming test scores can vary reviewed the results of clock-drawing
to administer and its results difficult to depending on the patient’s age and tests taken from 76 consecutive out-

www.geri.com August 2001 Volume 56, Number 8 Geriatrics 37

CME Geriatrics
7-minute battery. The 7-minute
neurocognitive battery (also the 7-
minute screen) consists of four sets of
questions that focus on orientation to
time and date, memory, visuospatial
skills, and verbal fluency. Three of its
four tests are similar versions of the
three-item recall, animal-naming test,
and clock-completion test mentioned
above. The 7-minute screen is clini-
cally appealing because it:
● can be administered and scored
in approximately 7 minutes and 40 sec-
onds (hence the screen name)
● can be administered in the out-
patient setting by a trained assistant
● has a sensitivity of 92% and speci-
ficity of 96% for detecting Alzheimer’s
Figure. The clock completion test begins by asking the patient to write the numbers disease.9
1 through 12 in a 3-inch circle so that the result represents a standard clock face. (For more information on the 7-
The dotted lines illustrate the sectioning of the clock face into four equal quadrants minute screen, visit http://www.7min-
based on the digit that best represents the 12 of a correctly completed clock. The utescreen.com.)
quadrant error scores flank each of the clocks. As the clock in the bottom right hand
corner shows, positioning of the digits, rather than listing the correct clock face
numbers, is the focus of the test. Depression
Depression is no more common in
Source: Reprinted with permission from Watson YI, Arfken CL, Birge SJ. Clock completion: An objective
older adults than in middle-aged or
screening test for dementia. J Am Geriatr Soc 1993; 41:1235-40.
younger populations, but it can be
more devastating. Suicide is the most
patients, age 55 to 92 (mean, 76), of test. It is important to ask the primary extreme consequence of depression in
whom 40 had dementia. The re- caregiver because patients with mild older persons. An older white man
searchers found the clock completion cognitive impairment often do not re- who verbalizes suicidal intent is at
test to be a reliable method of identi- alize the extent of their disability and highest risk for carrying out a self-de-
fying dementia, with sensitivity of 87% deny any functional impairment. The structive act. Older male patients who
and specificity of 82%. The clock test, more IADLs that are impaired in a do commit suicide tend to use violent,
however, was not found to be an ac- community-residing older person, the aggressive measures. Thus statements
curate indicator of the severity of de- greater the probability that dementia of intent should be taken seriously.
mentia.7 A patient who scores a 6 is not will develop within 1 year. The depression screen should begin
necessarily more cognitively impaired
than a patient who scores a 4.
Four IADL score: Although there are Table 2 Five-item version of the Geriatric Depression Scale
seven recognized instrumental activi- 1. Are you basically satisfied with your life?
ties of daily living (IADLs), the Four 2. Do you often get bored?
IADL Score8 relies on the following
3. Do you often feel helpless?
four measures:
● money management 4. Do you prefer to stay home rather than going out and doing new things?
● medication management 5. Do you feel pretty worthless the way you are now?
● telephone use
Positive answers for depression screening are “yes” to questions 2, 3, 4, and 5
● and traveling. and “no” to question 1. A score of 0 to 1 positive answer suggests the patient
If your patient is physically able to is not depressed; a score of 2 or higher indicates possible depression
perform these IADLs but needs assis-
Sensitivity: 97%; specificity: 85%; positive predictive value: 85%; negative
tance, suspect a developing dementia.
predictive value: 97%
Simply asking a family member or
Source: Reprinted with permission from Hoyl MT, Alessi CA, Harker JO, et al. Development and
friend if the patient needs assistance is testing of a five-item version of the Geriatric Depression Scale. J Am Geriatr Soc 1999; 47(7):873-8.
all that is required to administer the

38 Geriatrics August 2001 Volume 56, Number 8

with a single question: “Do you often are going to whisper some numbers; formance, and help identify patients
feel sad or depressed?” Sensitivity and then ask the patient to close his eyes. at high risk for loss of independent
specificity for this question are 85 and Lean toward the patient to within function. Specific screens include the
65%, respectively, so it is a relatively 12 to 18 inches from his face, exhale manual counting, rapid gait, and chair
sensitive but nonspecific question. It (to standardize the volume) and whis- rise tests, and the “get up and go” test
is, however, a start. If the patient an- per four random single numbers at 1- for assessment of risk of falling.
swers affirmatively, further screening second intervals. Ask the patient to re-
can be performed using the five-item peat all the numbers you whispered. A
Geriatric Depression Scale (table 2).10 patient who does not hear at least two Results from
If results of either test are positive, of the numbers fails the test, which has
the primary care physician should per- a sensitivity of 80 to 100% and speci- performance testing
form a thorough interview that evalu- ficity of 80 to 90%.
ates neurovegetative signs, including Audioscope. The audioscope delivers
can help predict an
sleep and appetite disturbances. Anti- four high-frequency (500, 1,000, 2,000 imminent inability to
depressant therapy and referral for psy- and 4,000 Hz) tones at approximately
chotherapy should be initiated. Failure 40 dB. A patient unable to discern the live independently
of antidepressants or the expression of 1,000- or 2,000-Hz tone in either or
suicidal thoughts warrant referral to a both ears is considered to have a hear-
psychiatrist. ing impairment. Sensitivity and speci- Manual counting. The manual count-
ficity of audioscopic testing are 94 and ing test measures manual, visual, and
Drugs 72%, respectively (roughly compara- cognitive capacity. You will need a
Any older patient who is taking more ble to the whisper test). change purse that contains a $1 bill, a
than four prescribed drugs has an in- quarter, two dimes, a nickel, and three
creased risk for falls. Certain drug Eyes (vision impairment) pennies. Direct the patient to open the
classes pose more risk of falls than oth- The first step in screening for a visual purse, take out all the money, count it,
ers, including any of the “anti” agents impairment is to ask the patient, “Be- and pronounce a total. If the patient
—antipsychotics, antidepressants, and cause of your eyesight, do you have any gives an incorrect total, repeat the ex-
antihypertensives. The long-acting difficulty driving, watching television, ercise. Note the amount of time it takes
benzodiazepines have been associated reading, or performing any other daily the patient to correctly count the
with cognitive impairment and falls in activity?” Even if the patient answers money. Stop the test after three failed
older adults. Over-the-counter agents “no,” test each eye with a small Snellen attempts or if 5 minutes elapses with-
and alternative or complementary sup- eye chart to confirm the patient’s self- out a correct response.
plements can also pose risks of inter- assessment. In general, patients who perform the
actions when used with prescription Patients who use glasses or contacts task correctly within 45 seconds will
agents, so always ask patients about should wear them during testing. Hold maintain their ADL status for approx-
their use of nonprescribed remedies. the Snellen chart 14 inches from the imately 1 year, whereas those who re-
patient’s eyes for the exam. A score of quire more than 70 seconds are at risk
Ears (hearing impairment) less than 20/40 indicates a need for fur- for loss of one or more ADLs within 1
Before performing any hearing test, ther vision testing. year.11 This screen has a sensitivity of
check the patient’s ear canal for ob- 83% and a specificity of 75%.
structions and remove excess cerumen. Physical performance Rapid gait. This test requires an un-
Sometimes cerumen removal alone can Physical performance and cognitive obstructed 10-foot path in the office,
resolve hearing impairment. Patients function are intimately linked. Poor examination room, or hallway. As you
who score poorly on hearing tests physical performance often reflects keep time (using the second hand of a
should be referred to an audiologist poor cognitive status and poor ability watch or clock), ask the patient to walk
Whisper test. An audiscope is an ef- to perform activities of daily living 10 feet, turn, and walk back as quickly
fective screening tool for hearing im- (ADL). Performance testing can pre- as possible. Patients who routinely use
pairment, but an economical and dict whether your patient is facing an canes or other assistive devices should
equally effective alternative is the imminent inability to live indepen- also use them during testing. Those
“whisper test.” To perform this screen, dently. The easily administered tests who complete the test within 10 sec-
place yourself directly in front of and summarized below can be repeated onds are likely to remain stable in ADL
a few feet across from the patient (pa- and quantified, confirm or refute dis- status for at least 1 year.
tient and physician should both be crepancies between the patient’s and Chair rise. The first step in this screen
seated or standing). Explain that you caregiver’s assessments of physical per- is the qualitative chair rise. The test be-

www.geri.com August 2001 Volume 56, Number 8 Geriatrics 39

CME Geriatrics
gins with the patient seated, hands to both of these questions indicates gerontology (4th ed). New York:
McGraw-Hill, 1999:467-81.
folded on the lap. Then ask the patient that the patient has a high probability
3. Inouye SK, van Dyck CH, Alessi CA,
to stand up. A patient passes the test of being incontinent and that this Balkin S, Siegal AP, Horwitz RI. Clarifying
by rising without using his arms for probability will be borne out on uro- confusion: The confusion assessment
assistance. A patient who fails this logic assessment. Be proactive in mak- method. A new method for detection of
screen has a 40% chance of develop- ing these inquiries because patients in delirium. Ann Intern Med 1990;
113(12):941-8.
ing an ADL impairment within 1 year general are reluctant to volunteer in-
4. Levkoff SE, Evans DA, Liptzin B, et al.
if no intervention is initiated.12 formation about incontinence.
Delirium: The occurrence and
For patients who pass the first test, persistence of symptoms among elderly
the second step is the quantitative chair Nutrition hospitalized patients. Arch Intern Med
rise test. Ask the patient to stand up The screening test for malnutrition is 1992; 152(2):334-40.
and sit down three times (without one question: “Have you lost 10 5. Siu AL. Screening for dementia and
investigating its causes. Ann Intern Med
using his arms) while you time the ac- pounds over the past 6 months with-
1991; 115(2):122-32.
tivity. The cutoff for predicting ADL out trying to do so?” Any patient who
6. Morris JC, Heyman A, Mohs RC, et al.
stability is 10 seconds. Patients who answers yes is at an increased risk of The Consortium to Establish a Registry
pass both the rapid gait and chair rise mortality. Other possible indicators of for Alzheimer’s Disease (CERAD). Part I.
tests have a 96% chance of maintain- malnutrition are: Clinical and neuropsychological
ing ADL stability over the next year.12 ● body weight <100 lbs assessment of Alzheimer’s disease.
Neurology 1989; 39(9):1159-65.
Falls (“Phalls”). The first phase of falls ● body mass index <22.
7. Watson YI, Arfken CL, Birge SJ. Clock
testing consists of asking the patient Approximately 75% of all cases of completion: An objective screening test
about any history of falls. A patient who dangerous weight loss are linked to dis- for dementia. J Am Geriatr Soc 1993;
reports a recent fall should undergo ease, whereas the remainder are asso- 41(11):1235-40.
balance and gait assessment. If the pa- ciated with difficulties in obtaining 8. Barberger-Gateau P, Dartigues JF,
tient has no history of falls, one of two food, chewing or swallowing problems, Letenneur L. Four Instrumental
Activities of Daily Living Score as a
performance tests can be administered. and poor oral hygiene. predictor of one-year incident dementia.
If results for either are positive, the pa- Age Ageing 1993; 22(6):457-63.
tient is at increased risk of falling. Conclusion 9. Solomon PR, Hirschoff A, Kelly B, et al.
In the “get up and go” test, a patient Delirium, dementia, depression, poly- A 7-minute neurocognitive screening
rises from a chair, walks 10 feet, turns pharmacy, hearing and vision deteri- battery highly sensitive to Alzheimer’s
disease. Arch Neurol 1998; 55(3):349-
around, walks back to the chair, and oration, physical performance, incon-
55.
sits down again. If completion of the tinence, and malnutrition are key
10. Hoyl MT, Alessi CA, Harker JO, et al.
task takes longer than 20 seconds, the considerations in the initial assessment Development and testing of a five-item
patient is at increased risk for falls. of the older patient. Functional asssess- version of the Geriatric Depression
Testing for static (standing) balance ment need not be time-consuming. Scale. J Am Geriatr Soc 1999;
47(7):873-8.
can also predict fall risk. After the clin- How well an older person can perform
ician demonstrates each one, the pa- activities of daily living can be gauged 11. Nikolaus T, Bach M, Oster P, et al. The
timed test of money counting: A simple
tient is asked to assume three stances, quickly and effectively using screening method of recognizing geriatric patients
keeping the eyes open for each: side- tools that can be administered in the at risk for increased health care. Aging
by-side stance (feet parallel); semi-tan- office setting. The mnemonic DEEP- Clin Exp Res 1995; 7:179-83.
dem stance (feet parallel, but one of IN is easy to remember, and the screens 12. Gill TM, Richardson ED, Tinetti ME.
them ahead of the other by one-half it represents are easy to administer. In- Evaluating the risk of dependence in
activities of daily living among
of a foot length), and tandem stance tegrating DEEP-IN into the initial pa- community-living older adults with mild
(one foot directly in front of the other). tient assessment can help identify to moderate cognitive impairment. J
Patients unable to hold these stances problems that might otherwise be Gerontol A Biol Sci Med Sci 1995;
for more than 10 seconds (without missed during a routine evaluation. G 50(5):M235-41.
marked swaying) are at increased risk
for falls. References

Incontinence
1. Moore AA, Siu AL. Screening for
common problems in ambulatory
CME Exam
Screening for incontinence involves elderly: Clinical confirmation of a
screening instrument. Am J Med 1996;
To earn CME credit for
asking two basic questions: “In the past 100:438-43. reading this article, turn to
year, have you ever lost your urine and 2. Ruben D. Principles of geriatric
gotten wet? If so, have you lost it on at page 43 to take the exam
assessment. In: Hazzard WR, et al.
least six separate days?” A “yes” answer Principles of geriatric medicine and

40 Geriatrics August 2001 Volume 56, Number 8