COGNITIVE

ASSESSMENT TOOLKIT
A guide to detecting cognitive impairment during the Medicare
Annual Wellness Visit
TABLE OF CONTENTS

Overview 3

Medicare Annual Wellness Visit Algorithm for

Assessment of Cognition 4

General Practitioner Assessment of Cognition (GPCOG) 5

Mini-Cog™ © 8

Short Form of the Informant Questionnaire on

Cognitive Decline in the Elderly (Short IQCODE) 10

Eight-Item Informant Interview to Differentiate Aging

and Dementia (AD8) 12

Alzheimer’s Association Recommendations for Operationalizing

the Detection of Cognitive Impairment During the
Medical Annual Wellness Visit in a Primary Care Setting 15

800.272.3900 | alz.org®
Overview 3

The Alzheimer’s Association®, the leading voluntary health organization in Alzheimer’s care, support and research, is
dedicated to driving early detection and diagnosis of dementia. To help, the Association has created an easy-to-implement
process to assess cognition during the Medicare Annual Wellness Visit. Developed by a group of clinical dementia experts,
the recommended process outlined on Page 4 allows you to efficiently identify patients with probable cognitive impairment
while giving you the flexibility to choose a cognitive assessment tool that works best for you and your patients.

This Cognitive Assessment Toolkit contains:

• The Medicare Annual Wellness Visit Algorithm for Assessment of Cognition, incorporating patient history,
clinician observations, and concerns expressed by the patient, family or caregiver

• Two validated patient assessment tools: the General Practitioner Assessment of Cognition (GPCOG) and
the Mini-Cog™©. Both tools:

› Can be administered in five minutes or less

1. 3
› Are equal or superior to the Mini-Mental State Exam (MMSE) for detecting dementia

› Are easily administered by medical staff who are not physicians

› Are relatively free from educational, language and/or cultural bias

• Three validated informant assessment of patient tools: the Short Form of the Informant Questionnaire
on Cognitive Decline in the Elderly (Short IQCODE), the Eight-Item Informant Interview to Differentiate
Aging and Dementia (AD8) and the GPCOG

• The “Alzheimer’s Association Recommendations for Operationalizing the Detection of Cognitive Impairment
During the Medical Annual Wellness Visit in a Primary Care Setting,” as published in Alzheimer’s & Dementia®:
The Journal of the Alzheimer’s Association

For more information on the detection, diagnosis and treatment of Alzheimer’s disease, as well as direct access to patient
and caregiver resources, please visit our Health Systems and Clinicians Center at alz.org/hcps.

800.272.3900 | alz.org®
Alzheimer’s Association® 4

Medicare Annual Wellness Visit Algorithm for Assessment of Cognition

A. Review HRA, clinician observation, self-reported concerns, responses to queries

YES Signs/symptoms present NO

NO Informant available to confirm

B.* Conduct brief structured assessment

• Patient assessment: Mini-Cog or GPCOG YES
• Informant assessment of patient: Short IQCODE, AD8 or GPCOG

1. 4 Follow-up during subsequent AWV

Brief assesment(s) triggers concerns: Patient: Mini-Cog ≤3

or GPCOG <5 (5-8 score is indeterminate without informant)
or Informant: Short IQCODE ≥ 3.38 or AD8 ≥ 2 or GPCOG NO
informant score ≤3 with patient score <8

YES

C. Refer OR conduct full dementia evaluation

* No one tool is recognized as the best brief assessment to determine if a full dementia
evaluation is needed. Some providers repeat patient assessment with an alternate tool
(e.g., SLUMS, or MoCA) to confirm initial findings before referral or initiation of full
dementia evaluation.

AD8 = Eight-Item Informant Interview to Differentiate Aging and Dementia; AWV = Annual Wellness
Visit; GPCOG = General Practitioner Assessment of Cognition; HRA = Health Risk Assessment;
MoCA = Montreal Cognitive Assessment; SLUMS = St. Louis University Mental Status Exam;
Short IQCODE = Short Informant Questionnaire on Cognitive Decline in the Elderly

Cordell CB, Borson S, Boustani M, Chodosh J, Reuben D, Verghese J, et al. Alzheimer’s Association recommendations for operationalizing
the detection of cognitive impairment during the Medicare Annual Wellness Visit in a primary care setting. Alzheimers Dement. 2013;9(2):141-150.
Available at https://alz-journals.onlinelibrary.wiley.com/journal/15525279

800.272.3900 | alz.org®
 5

Patient name: _________________________

Testing date: _________________________

STEP 1 – PATIENT EXAMINATION

GENERAL
Unless specified,PRACTITIONER ASSESSMENT
each question should only be asked once.
OF COGNITION
Name and address for(GPCOG)
subsequent recall test
IAam going to GPCOG
web-based give youand
a name and address.
downloadable After I have said
paper-and-pencil it, I want
versions of theyou to repeat it.
Remember this name and address because I am going to ask you to tell it to me again in a
GPCOG (in many languages) are available at gpcog.com.au. Both ask the
few minutes: John Brown, 42 West Street, Kensington. (Allow a maximum of 4 attempts.)
same questions, the only difference being the web-based GPCOG
automatically scores the test.
Time orientation Correct Incorrect
1. What is the
Preparation date? (exact only)
& Training
Clock
Beforedrawing (use blank
you administer page)for the first time, please review the following:
GPCOG
2. Please mark in all the numbers to indicate
1. the hours
Make sureofyou
a clock.
have(correct
read thespacing required)
instructions (on the first page of the test)
3. Please mark in hands to show 10 minutes past
2. eleven
Watch the training
o’clock. 1. 5
video (approx. 5 minutes)
(11.10)

https://www.youtube.com/watch?v=If7nv2_B89M
Information
4. Can you tell me something that happened in the news recently?
(Recently = in the last week. If a general answer is given, e.g.
“war”, “lot of rain”, ask for details. Only specific answer scores.)

Recall
5. What was the name and address I asked you to remember?
John
Brown
42
West (St)
Kensington

Add the number of items answered correctly: Total score: out of 9

/
9
9 No significant cognitive impairment
Further testing is not necessary
5–8 More information required
Proceed with informant interview in step 2 on next page
0–4 Cognitive impairment is indicated
Conduct standard investigations

© University of New South Wales as represented by the Dementia Collaborative Research Centre – Assessment and Better Care;
Brodaty et al, JAGS 2002; 50:530-534
 6

Patient name: _________________________

Testing date: _________________________

STEP 1 – PATIENT EXAMINATION

Unless specified, each question should only be asked once.

Name and address for subsequent recall test

I am going to give you a name and address. After I have said it, I want you to repeat it.
Remember this name and address because I am going to ask you to tell it to me again in a
few minutes: John Brown, 42 West Street, Kensington. (Allow a maximum of 4 attempts.)

Time orientation Correct Incorrect

1. What is the date? (exact only)

Clock drawing (use blank page)

2. Please mark in all the numbers to indicate
the hours of a clock. (correct spacing required)
3. Please mark in hands to show 10 minutes past
eleven o’clock. (11.10) 1. 6

Information
4. Can you tell me something that happened in the news recently?
(Recently = in the last week. If a general answer is given, e.g.
“war”, “lot of rain”, ask for details. Only specific answer scores.)

Recall
5. What was the name and address I asked you to remember?
John
Brown
42
West (St)
Kensington

Add the number of items answered correctly: Total score: out of 9

/
9
9 No significant cognitive impairment
Further testing is not necessary
5–8 More information required
Proceed with informant interview in step 2 on next page
0–4 Cognitive impairment is indicated
Conduct standard investigations

© University of New South Wales as represented by the Dementia Collaborative Research Centre – Assessment and Better Care;
Brodaty et al, JAGS 2002; 50:530-534
 7
Patient name: _________________________
Testing date: _________________________

STEP 2: INFORMANT INTERVIEW

Informant name: _______________________

Relationship to patient, i.e. informant is the patient’s: _________________________

Ask the informant:

Don’t
Compared to 5–10 years ago, YES NO
know
N/A

1. Does the patient have more trouble remembering things

that have happened recently than s/he used to?

2. Does s/he have more trouble recalling conversations

a few days later?

3. When speaking, does s/he have more difficulty in

finding the right word or tend to use the wrong words
1. 7
more often?

4. Is s/he less able to manage money and financial

affairs (e.g. paying bills and budgeting)?

5. Is s/he less able to manage his or her medication

independently?

6. Does s/he need more assistance with transport

(either private or public)?
(If the patient has difficulties only due to physical
problems, e.g. bad leg, tick ‘no’.)

Add the number of items answered Total score: out of 6

with ‘NO’, ‘Don’t know’ or ‘N/A’:

4–6 No significant cognitive impairment

Further testing is not necessary
0–3 Cognitive impairment is indicated
Conduct standard investigations

When referring to a specialist, mention the individual scores for the two GPCOG test steps:
STEP 1 Patient examination: __ / 9
STEP 2 Informant interview: __ / 6 or N/A

© University of New South Wales as represented by the Dementia Collaborative Research Centre – Assessment and Better Care;
Brodaty et al, JAGS 2002; 50:530-534
 8

Mini-Cog™ Instructions for Administration & Scoring

ID: ______________ Date: ________________________

Step 1: Three Word Registration

Look directly at person and say, “Please listen carefully. I am going to say three words that I want you to repeat back
to me now and try to remember. The words are [select a list of words from the versions below]. Please say them for
me now.” If the person is unable to repeat the words after three attempts, move on to Step 2 (clock drawing).

The following and other word lists have been used in one or more clinical studies.1-3 For repeated administrations,
use of an alternative word list is recommended.

Version 1 Version 2 Version 3 Version 4 Version 5 Version 6

Banana Leader Village River Captain Daughter
Sunrise Season Kitchen Nation Garden Heaven
Chair Table Baby Finger Picture Mountain

Step 2: Clock Drawing

Say: “Next, I want you to draw a clock for me. First, put in all of the numbers where they go.” When that is completed,
say: “Now, set the hands to 10 past 11.”

Use preprinted circle (see next page) for this exercise. Repeat instructions as needed as this is not a memory test.
Move to Step 3 if the clock is not complete within three minutes.

Step 3: Three Word Recall

Ask the person to recall the three words you stated in Step 1. Say: “What were the three words I asked you to
remember?” Record the word list version number and the person’s answers below.

Word List Version: _____ Person’s Answers: ___________________ ___________________ ___________________

Scoring

Word Recall: ______ (0-3 points) 1 point for each word spontaneously recalled without cueing.

Normal clock = 2 points. A normal clock has all numbers placed in the correct
sequence and approximately correct position (e.g., 12, 3, 6 and 9 are in anchor
Clock Draw: ______ (0 or 2 points) positions) with no missing or duplicate numbers. Hands are pointing to the 11
and 2 (11:10). Hand length is not scored.
Inability or refusal to draw a clock (abnormal) = 0 points.

Total score = Word Recall score + Clock Draw score.

A cut point of <3 on the Mini-Cog™ has been validated for dementia screening,
Total Score: ______ (0-5 points)
but many individuals with clinically meaningful cognitive impairment will score
higher. When greater sensitivity is desired, a cut point of <4 is recommended as
it may indicate a need for further evaluation of cognitive status.

Mini-Cog™ © S. Borson. All rights reserved. Reprinted with permission of the author solely for clinical and educational purposes.
May not be modified or used for commercial, marketing, or research purposes without permission of the author ([email protected]).
v. 01.19.16
 9

Clock Drawing ID: ______________ Date: ________________________

References
1. Borson S, Scanlan JM, Chen PJ et al. The Mini-Cog as a screen for dementia: Validation in a population-based
sample. J Am Geriatr Soc 2003;51:1451–1454.
2. Borson S, Scanlan JM, Watanabe J et al. Improving identification of cognitive impairment in primary care. Int J
Geriatr Psychiatry 2006;21: 349–355.
3. Lessig M, Scanlan J et al. Time that tells: Critical clock-drawing errors for dementia screening. Int
Psychogeriatr. 2008 June; 20(3): 459–470.
4. Tsoi K, Chan J et al. Cognitive tests to detect dementia: A systematic review and meta-analysis. JAMA Intern
Med. 2015; E1-E9.
5. McCarten J, Anderson P et al. Screening for cognitive impairment in an elderly veteran population:
Acceptability and results using different versions of the Mini-Cog. J Am Geriatr Soc 2011; 59: 309-213.
6. McCarten J, Anderson P et al. Finding dementia in primary care: The results of a clinical demonstration
project. J Am Geriatr Soc 2012; 60: 210-217.
7. Scanlan J & Borson S. The Mini-Cog: Receiver operating characteristics with the expert and naive raters. Int J
Geriatr Psychiatry 2001; 16: 216-222.

Mini-Cog™ © S. Borson. All rights reserved. Reprinted with permission of the author solely for clinical and educational purposes.
May not be modified or used for commercial, marketing, or research purposes without permission of the author ([email protected]).
v. 01.19.16
 10

Short Form of the Informant Questionnaire on Cognitive Decline

in the Elderly (Short IQCODE)1

by A. F. Jorm

Centre for Mental Health Research

The Australian National University
Canberra, Australia

There is no copyright on the Short IQCODE. However, the author appreciates

being kept informed of research projects which make use of it.

Note: As used in published studies, the IQCODE was preceded by questions to the
informant on the subject’s sociodemographic characteristics and physical health.
 11

Now we want you to remember what your friend or relative was like 10 years ago and to compare it with what he/she is
like now. 10 years ago was in 20__.* Below are situations where this person has to use his/her memory or intelligence and
we want you to indicate whether this has improved, stayed the same or got worse in that situation over the past 10 years.
Note the importance of comparing his/her present performance with 10 years ago. So if 10 years ago this person always
forgot where he/she had left things, and he/she still does, then this would be considered “Hasn’t changed much”. Please
indicate the changes you have observed by circling the appropriate answer.

Compared with 10 years ago how is this person at:

1 2 3 4 5
1. Remembering things about family and friends Much A bit Not much A bit Much
e.g. occupations, birthdays, addresses improved improved change worse worse

2. Remembering things that have happened Much A bit Not much A bit Much
recently improved improved change worse worse
Much A bit Not much A bit Much
3. Recalling conversations a few days later improved improved change worse worse
4. Remembering his/her address and telephone Much A bit Not much A bit Much
number improved improved change worse worse
Much A bit Not much A bit Much
5. Remembering what day and month it is improved improved change worse worse
Much A bit Not much A bit Much
6. Remembering where things are usually kept improved improved change worse worse
7. Remembering where to find things which have Much A bit Not much A bit Much
been put in a different place from usual improved improved change worse worse

8. Knowing how to work familiar machines Much A bit Not much A bit Much
around the house improved improved change worse worse

9. Learning to use a new gadget or machine Much A bit Not much A bit Much
around the house improved improved change worse worse
Much A bit Not much A bit Much
10. Learning new things in general improved improved change worse worse
Much A bit Not much A bit Much
11. Following a story in a book or on TV improved improved change worse worse
Much A bit Not much A bit Much
12. Making decisions on everyday matters improved improved change worse worse
Much A bit Not much A bit Much
13. Handling money for shopping improved improved change worse worse
14. Handling financial matters e.g. the pension, Much A bit Not much A bit Much
dealing with the bank improved improved change worse worse

15. Handling other everyday arithmetic problems

Much A bit Not much A bit Much
e.g. knowing how much food to buy, knowing improved improved change worse worse
how long between visits from family or friends

16. Using his/her intelligence to understand what’s Much A bit Not much A bit Much
going on and to reason things through improved improved change worse worse

*The original tool was published in 1994.

The Alzheimer’s Association updated the year 19__ as published in the original tool to 20__ .
Tool Reference: Jorm AF. A short form of the Informant Questionnaire on Cognitive Decline in the
Elderly (IQCODE): development and cross-validation. Psychol Med 1994; 24: 145–153.
 12

AD8® Dementia Screening Interview  Patient ID#:  

CS ID#:  
Date:  

Remember, “Yes, a change” indicates that  YES,  NO,  N/A, 

there has been a change in the last several  A change  No change  Don’t know 
years caused by cognitive (thinking and 
memory) problems. 

1. Problems with judgment (e.g., problems
making decisions, bad financial 
decisions, problems with thinking) 

2. Less interest in hobbies/activities

3. Repeats the same things over and over
(questions, stories, or statements) 

4. Trouble learning how to use a tool,
appliance, or gadget (e.g., VCR, 
computer, microwave, remote control) 

5. Forgets correct month or year

6. Trouble handling complicated financial
affairs (e.g., balancing checkbook, income 
taxes, paying bills) 

7. Trouble remembering appointments

8. Daily problems with thinking and/or
memory 

TOTAL AD8 SCORE 

Adapted from Galvin JE et al, The AD8, a brief informant interview to detect dementia, Neurology 2005:65:559‐564.
Copyright ©2005 by Washington University in St. Louis, MO. All Rights Reserved. 
 Copyright ©2005 Washington University, 13
St. Louis, Missouri. All Rights Reserved.

The AD8® Administration and Scoring Guidelines 

A spontaneous self‐correction is allowed for all responses without counting as an error. 
The questions are given to the respondent on a clipboard for self–administration or can be 
read aloud to the respondent either in person or over the phone. It is preferable to 
administer the AD8 to an informant, if available. If an informant is not available, the AD8 
may be administered to the patient. 
When administered to an informant, specifically ask the respondent to rate change 
in the patient. 
When administered to the patient, specifically ask the patient to rate changes in his/her 
ability for each of the items, without attributing causality. 
If read aloud to the respondent, it is important for the clinician to carefully read the phrase 
as worded and give emphasis to note changes due to cognitive problems (not physical 
problems). There should be a one second delay between individual items. 
No timeframe for change is required. 
The final score is a sum of the number items marked “Yes, A change”. 
Interpretation of the AD8  (Adapted from Galvin JE et al, The AD8, a brief informant interview to detect
dementia, Neurology 2005:65:559‐564) 

A screening test in itself is insufficient to diagnose a dementing disorder. The AD8 is, 
however, quite sensitive to detecting early cognitive changes associated many common 
dementing illness including Alzheimer disease, vascular dementia, Lewy body dementia and 
frontotemporal dementia. 
Scores in the impaired range (see below) indicate a need for further assessment. Scores in 
the “normal” range suggest that a dementing disorder is unlikely, but a very early disease 
process cannot be ruled out. More advanced assessment may be warranted in cases where 
other objective evidence of impairment exists. 
Based on clinical research findings from 995 individuals included in the development 
and validation samples, the following cut points are provided: 
0 – 1:  Normal cognition 
2 or greater: Cognitive impairment is likely to be present 

Administered to either the informant (preferable) or the 
patient, the AD8 has the following properties: 
Sensitivity > 84% 
Specificity > 80% 
Positive Predictive Value > 85% 
Negative Predictive Value > 70% 
Area under the Curve: 0.908; 95%CI: 0.888‐0.925 
 Copyright ©2005 Washington University, 14
St. Louis, Missouri. All Rights Reserved.

Permission Statement 

Washington University grants permission to use and reproduce the Eight‐item Informant Interview to 
Differentiate Aging and Dementia exactly as it appears in the PDF available here without modification 
or editing of any kind solely for end user use in investigating dementia in clinical care or in non-profit
research (the “Purpose”). For the avoidance of doubt, the Purpose does not include the (i) use or
distribution of the Eight-item Informant Interview to Differentiate Aging and Dementia in a clinical trial
sponsored in whole or in part by a commercial entity; (ii) sale, distribution or transfer of  the Eight‐item 
Informant Interview to Differentiate Aging and Dementia or copies thereof for any consideration or 
commercial value; (iii) the creation of any derivative works, including translations; (iv) use of the Eight‐
item Informant Interview to Differentiate Aging and Dementia as a marketing tool for the sale of any drug;
and/or (v) use of the Eight-item Informant Interview to Differentiate Aging and Dementia with any
electronic medical health records sytem or electronic registry.
All copies of the AD8® shall include the following notice: “Copyright © 2005 Washington University, St.
Louis, Missouri. All Rights Reserved.” Please contact [email protected] for a commercial license, for
permission to make modifications, or for any other intended purpose. 
 15

Alzheimer’s & Dementia 9 (2013) 141–150

Alzheimer’s Association recommendations for operationalizing

the detection of cognitive impairment during the Medicare Annual
Wellness Visit in a primary care setting
Cyndy B. Cordella,*, Soo Borsonb,c, Malaz Boustanid,e,f, Joshua Chodoshg,h, David Reubenh,
Joe Verghesei, William Thiesa, Leslie B. Friedj,k; for the Medicare Detection
of Cognitive Impairment Workgroup
a
Alzheimer’s Association, Chicago, IL, USA
b
Department of Psychiatry and Behavioral Sciences, University of Washington School of Medicine, Seattle, WA, USA
c
Memory Disorders Clinic and Dementia Health Services, University of Washington School of Medicine, Seattle, WA, USA
d
Indiana University Center for Aging Research, Indianapolis, IN, USA
e
Regenstrief Institute, Inc., Indianapolis, IN, USA
f
Department of Medicine, Indiana University School of Medicine, Indianapolis, IN, USA
g
Veterans Administration Greater Los Angeles Healthcare System, Los Angeles, CA, USA
h
Division of Geriatrics, David Geffen School of Medicine, University of California Los Angeles, Los Angeles, CA, USA
i
Department of Neurology, Albert Einstein College of Medicine, Bronx, NY, USA
j
American Bar Association, Washington, DC, USA
k
Alzheimer’s Association Medicare Advocacy Project, Washington, DC, USA

Abstract The Patient Protection and Affordable Care Act added a new Medicare benefit, the Annual
Wellness Visit (AWV), effective January 1, 2011. The AWV requires an assessment to detect cog-
nitive impairment. The Centers for Medicare and Medicaid Services (CMS) elected not to recom-
mend a specific assessment tool because there is no single, universally accepted screen that
satisfies all needs in the detection of cognitive impairment. To provide primary care physicians
with guidance on cognitive assessment during the AWV, and when referral or further testing is
needed, the Alzheimer’s Association convened a group of experts to develop recommendations.
The resulting Alzheimer’s Association Medicare Annual Wellness Visit Algorithm for Assessment
of Cognition includes review of patient Health Risk Assessment (HRA) information, patient ob-
servation, unstructured queries during the AWV, and use of structured cognitive assessment tools
for both patients and informants. Widespread implementation of this algorithm could be the first
step in reducing the prevalence of missed or delayed dementia diagnosis, thus allowing for better
healthcare management and more favorable outcomes for affected patients and their families and
caregivers.
 2013 The Alzheimer’s Association. All rights reserved.
Keywords: Annual Wellness Visit; AWV; Cognitive impairment; Assessment; Screen; Dementia; Alzheimer’s disease;
Medicare; Algorithm; Patient Protection and Affordable Care Act

1. Introduction (AWV), effective January 1, 2011. The AWV includes

routine measurements such as height, weight, and blood
The Patient Protection and Affordable Care Act of 2010
pressure; a review of medical and family history; an assess-
added a new Medicare benefit, the Annual Wellness Visit
ment to detect cognitive impairment; and establishment of
a list of current medical providers, medications, and sched-
*Corresponding author. Tel.: 312-335-5867. Fax: 866-699-1246. ule for future preventive services. In addition, during the first
E-mail address: [email protected] AWVonly, beneficiaries are to be screened for depression (if

1552-5260/$ - see front matter  2013 The Alzheimer’s Association. All rights reserved.
http://dx.doi.org/10.1016/j.jalz.2012.09.011
 16

C.B. Cordell et al. / Alzheimer’s & Dementia 9 (2013) 141–150

not completed under a separate Medicare benefit) and for  Detection of cognitive impairment is a stepwise, itera-
functional difficulties using nationally recognized appropri- tive process.
ate screening questions or standardized questionnaires. Al-  Informal observation alone by a physician is not suffi-
though the U.S. Preventive Services Task Force (USPSTF) cient (i.e., observation without a specific cognitive
in 2003 concluded that there was insufficient published evi- evaluation).
dence of better clinical outcomes as a result of routine  Detection of cognitive impairment can be enhanced by
screening for cognitive impairment in older adults, the specifically asking about changes in memory, lan-
Task Force recognized that the use of cognitive assessment guage, and the ability to complete routine tasks.
tools can increase the detection of cognitive impairment  Although no single tool is recognized as the “gold stan-
[1]. As per the Centers for Medicare and Medicaid Services dard” for detection of cognitive impairment, an initial
(CMS) regulation, the AWV requires detection of cognitive structured assessment should provide either a baseline
impairment by “. assessment of an individual’s cognitive for cognitive surveillance or a trigger for further eval-
function by direct observation, with due consideration of in- uation.
formation obtained by way of patient report, concerns raised  Clinical staff can offer valuable observations of cogni-
by family members, friends, caretakers, or others” [2]. Dur- tive and functional changes in patients who are seen
ing the public comment period, several organizations, in- over time.
cluding the Alzheimer’s Association, noted that the use of  Counseling before and after cognitive assessment is an
a standardized tool for assessment of cognitive function essential component of any cognitive evaluation.
should be part of the AWV.  Informants (family member, caregiver, etc.) can pro-
These comments are supported by a number of studies vide valuable information about the presence of
showing that cognitive impairment is unrecognized in a change in cognition.
27%–81% of affected patients in primary care [3–7]. The
use of a brief, structured cognitive assessment tool 2.2. Principles specific to the AWV
correctly classifies patients with dementia or mild
cognitive impairment (MCI) more often than spontaneous  The AWV requires the completion of a Health Risk As-
detection by the patients’ own primary care physicians sessment (HRA) by the patient either before or during
(83% vs 59%, respectively) [8]. the visit. The HRA should be reviewed for any reported
In response to concerns submitted during public comment, signs and symptoms indicative of possible dementia.
CMS elected not to recommend a specific tool for the final  The AWV will likely occur in a primary care setting.
AWV benefit because “There is no nationally recognized Tools for initial cognitive assessments should be brief
screening tool for the detection of cognitive impairments at (,5 min), appropriately validated, easily administered
the present time.” [9]. However, CMS recognizes that with- by non-physician clinical staff, and available free of
out clarification, the full intended benefits of the AWV cogni- charge for use in a clinical setting.
tive assessment may not be realized [10]. CMS is working  If further evaluation is indicated based on the results of
with other governmental agencies (e.g., National Institutes the AWV, a more detailed evaluation of cognition
on Aging) on recommendations for use of specific tools. should be scheduled for a follow-up visit in primary
Understanding that, under the present regulation, each care or through referral to a specialist.
healthcare provider who conducts an AWV would have to
determine how best to “detect cognitive impairment,” the
Alzheimer’s Association convened the Medicare Detection 3. Review of available brief tools for use during the AWV
of Cognitive Impairment Workgroup to develop recommen-
3.1. Workgroup review process
dations for operationalizing the cognitive assessment com-
ponent in primary care settings. This workgroup was Although there is no single cognition assessment tool that
comprised of geographically dispersed USA experts with is considered to be the gold standard, there is a plethora of
published works in the field of detecting cognitive impair- tools in the literature. A MEDLINE (PubMed) search con-
ment during primary care visits. The focus on primary care ducted in October 2011, using the key words “screening or
was deliberate, as most Medicare beneficiaries will receive detection of dementia or cognitive impairment,” yielded
their AWV in this setting. over 500 publications. To narrow the search to tools more
applicable to the AWV, the workgroup sought to determine
whether the literature offered a consensus regarding brief
2. Guiding principles for recommendations cognitive assessment during time-limited primary care visits.
2.1. Consensus on general principles The workgroup focused on systematic evidence review
(SER) studies published since 2000 resulting in four studies
Based on their expertise, the workgroup agreed on the fol- by Lorentz et al, Brodaty et al, Holsinger et al, and Milne et al
lowing general principles to guide the development of rec- [11–14]. Although each SER had a similar objective—to
ommendations for cognitive assessment: determine which tools were best for administration during
 17

C.B. Cordell et al. / Alzheimer’s & Dementia 9 (2013) 141–150

primary care visits—different comparison criteria to select

Kansagara and Freeman, 2010 [16]
the tools were applied (Table 1). Two other studies were

 Relevance of study to the VA

 Tools identified by the VA as
also considered relevant to the development of the work-
alternatives to the MMSE
group recommendations: Ismail et al [15] conducted a litera-
ture review designed to identify widely used and most
promising newer brief cognitive tools being used in primary

 Admin time
care and geriatrics, and an SER by Kansagara and Freeman

 Specificity
 Sensitivity
[16] of six brief cognitive assessment tools that could serve
setting

 Cost
as possible alternatives to the Mini-Mental State Examina-
tion (MMSE) for use by the U.S. Department of Veterans

Abbreviations: MMSE, Mini-Mental State Examination; NPV, negative predictive value; PC, primary care; UK, United Kingdom; VA, US Department of Veteran Affairs.
Affairs (VA). Neither study was designed to determine which
 Tools recommended or
 Tools most frequently

brief tool is the “best,” but both provided evidence related to

studies and strength/

 Newer tools that ad-

weaknesses of tools
Ismail et al, 2010 [15]

 Summary of other
newly used in PC

dress weaknesses

primary care use and performance characteristics of brief

assessments of cognition (Table 1).
used in PC

3.2. Workgroup review results

Of the five publications that focused specifically on
identifying brief cognitive assessments most suitable or
 Admin time suitable for

 Geriatric PC screens for

most used in primary care settings [11–15], all selected

Psychometric proper-
Milne et al, 2008 [14]

the Memory Impairment Screen (MIS), and four of these

cognitive change

publications [11,12,14,15] also selected the General

Applicability
Practicality

Practitioner Assessment of Cognition (GPCOG) and the

Feasibility
PC in UK

Mini-Cog (Table 2).

ties

The following attributes of the GPCOG, Mini-Cog, and

the MIS contributed to their selection as most suited for rou-





tine use in primary care:

Holsinger et al, 2007 [13]

 Criterion to diagnose
dementia acceptable

 Requires 5 minutes or less to administer.

 Studied in patients

Likelihood ratios

Utility in special
Domains tested

 Is validated in a primary care or community setting.

Study quality
Admin time

 Is easily administered by medical staff members who

60 years

situations

are not physicians.

Review articles of brief cognitive assessment tools—select inclusion and comparison criteria

 Has good to excellent psychometric properties.

 Is relatively free from educational, language, and/or






culture bias.
 Validated in community or PC

 Administration characteristics

 Can be used by clinicians in a clinical setting without

 Admin 5 min and simple

payment for copyrights.

 Psychometric properties
Brodaty et al, 2006 [12]

 Misclassification rate

 Applicability to PC

Charging a fee for clinical use of brief cognitive assess-

 NPV  MMSE

ment tool has become an issue because of increased enforce-

 Study validity

ment of the MMSE copyright. First published in 1975 [17],

 MMSE

the MMSE copyright is now held by Psychological Assess-

ment Resources, Inc., which charges a fee for each use (for
exact fees see www.parinc.com). The comparative SER
within the VA [16] evaluated alternatives to the proprietary
 Ease of use by nonspecialists
 Face validity, sensitivity, and
evaluated in 1 community
 Performance characteristics

MMSE, including the GPCOG and the Mini-Cog, along

 Sociodemographic biases
 Comparison with MMSE

with four other brief tools (Table 2). The Mini-Cog and
Lorentz et al, 2002 [11]

MIS are copyrighted, but the owners, Soo Borson, MD,

or clinical setting
 Admin 10 min

and Albert Einstein College of Medicine, respectively, allow

 Acceptability

free use by clinicians as clinical tools with distribution re-

specificity

strictions for other entities (e.g., commercial companies).

The GPCOG has similar use rules.

3.3. Patient structured cognitive assessment tools

Comparison

recommended for AWV

criteria

criteria
Inclusion
Table 1

In alignment with the workgroup’s guiding principles

and supported by data in the six selected SERs/reviews,
 18

C.B. Cordell et al. / Alzheimer’s & Dementia 9 (2013) 141–150

Table 2
Brief cognitive assessment tools evaluated in multiple review articles
Lorentz et al, Brodaty et al, Holsinger et al, Milne et al, Ismail et al, Kansagara and
Assessment Tool 2002 [11] 2006 [12] 2007 [13] 2008 [14] 2010 [15] Freeman, 2010* [16]
7-Minute Screener X X X X
AMT X X X X
CAMCOG X Suitedy
CDT X X Suitedz X X
GPCOG Most suited Most suited X Most suited Most suited X
Mini-Cog Most suited Most suited X Most suited Most suited X
MIS Most suited Most suited Suitedz Most suited Most suited
MMSE X X Suitedx X X
MoCA Suitedy X X
RUDAS X X
SAS-SI X X X
SBT (BOMC, 6-CIT) X X X X X
SPMSQ X X
STMS X X X X
T&C X X
Abbreviations: 6-CIT, 6-Item Cognitive Impairment Test; AMT, Abbreviated Mental Test; BOMC, 6-item Blessed Orientation-Memory-Concentration Test;
CAMCOG, Cambridge Cognitive Examination; CDT, Clock Drawing Test; GPCOG, General Practitioner Assessment of Cognition; MIS, Memory Impairment
Screen; MMSE, Mini-Mental State Examination; MoCA, Montreal Cognitive Assessment; RUDAS, Rowland Universal Dementia Assessment; SAS-SI, Short
and Sweet Screening Instrument; SBT, Short Blessed Test; SLUMS, St Louis Mental Status; SPMSQ, Short Portable Mental Status Questionnaire; STMS, Short
Test of Mental Status; T&C, Time and Change Test.
X 5 assessment reviewed, but not identified as most suited for general use in primary care.
Suited 5 tool appropriate for the following clinical issue: y available time is not limited; z available time is limited; and x cognitive impairment is at least
moderate. Most suited 5 tool identified as most suited for routine use in primary care.
*Kansagara and Freeman evaluated six tools, including the SLUMS, which was not evaluated in any other review.

the GPCOG, Mini-Cog, and MIS are brief structured 4. Recommended algorithm for detection of cognitive
tools that are suitable for assessment of cognitive func- impairment during the AWV
tion during the AWV. Each tool has unique benefits.
The GPCOG has patient and informant components 4.1. Incorporating assessment of cognition during the
that can be used alone or together to increase specificity AWV
and sensitivity [18]. The Mini-Cog has been validated in The Alzheimer’s Association Medicare Annual Wellness
population-based studies and in community-dwelling Visit Algorithm for Assessment of Cognition for consistency
older adults heterogeneous with respect to language, cul- (Figure 1) illustrates a stepwise process. The process is in-
ture, and education [19–22]. The MIS is a verbally tended to detect patients with a high likelihood of having de-
administered word-recall task that tests encoding as mentia. The AWV algorithm includes both structured
well as retrieval [23], and is an option for patients who assessments discussed previously and other less structured
have motor impairments that prevent use of paper and patient- and informant-based evaluations. By assessing and
pencil. documenting cognitive status on an annual basis during the
AWV, clinicians can more easily determine gradual cogni-
3.4. Structured cognitive assessment tools for use with tive decline over time in an individual patient—a key crite-
informants rion for diagnosing dementia due to Alzheimer’s disease and
other progressive conditions affecting cognition.
Cognitive assessment combined with informant- For patients with a previous diagnosis of MCI or demen-
reported data improves the accuracy of assessment tia, this should be documented and included in their AWV
[24–27]. If an informant is present during the AWV, list of health risk factors. Annual unstructured and structured
use of a structured informant tool is recommended. cognitive assessments could be used to monitor significant
Similar to cognitive assessment tools for use with changes in cognition and potentially lead to a new diagnosis
patients, there is no single “gold standard” informant of dementia for those with MCI or new care recommenda-
tool; however, relatively few brief informant tools tions for those with dementia.
have been validated in community and/or primary care
settings. Brief tools appropriately validated include the
4.2. Detection of cognitive impairment during the AWV—
Short IQCODE [25], the AD8 [28], which can be ad-
initial HRA review, conversations, and observations
ministered in-person or by telephone, and the aforemen-
tioned GPCOG [18], which has both patient and The first step in detection of cognitive impairment during
informant components. the AWV (Fig. 1, Step A), involves a conversation between
 19

C.B. Cordell et al. / Alzheimer’s & Dementia 9 (2013) 141–150

Medicare Annual Wellness Visit (HCPCS codes G0438 or G0439)

A Review HRA (especially reports of functional deficits), clinician observations, and self-
reported concerns; and query patient and, if available, informant

Yes Signs/symptoms of cognitive No

impairment present

No
Informant
available to
confirm
Yes

* Conduct brief structured assessment

B Patient Assessment: GPCOG or Mini-Cog or MIS
Informant assessment of patient: AD8 or GPCOG or Short IQCODE

Follow-up during
subsequent AWV
Brief assessment(s) triggers concerns:
-8 score is indeterminate without
informant) or Mini-
Informant: AD8 No
patient score <8 3.38

Yes

C Refer for full dementia evaluation or

Conduct full dementia evaluation
If informant is available during AWV can follow
- up same day as AWV and bill for E/M service with
CPT codes 99201-99215. If not, schedule new visit for evaluation and request presence of
family/companion to facilitate assessment.

* No one tool is recognized as the best brief assessment to determine if a full dementia evaluation is
needed. Alternate tools (eg, MMSE, SLUMS, or MoCA) can be used at the discretion of the clinician.
Some providers use multiple brief tools prior to referral or initiation of a full dementia evaluation.

AWV = Annual Wellness Visit; GPCOG = General Practitioner Assessment of Cognition; HRA = Health Risk Assessment;
MIS = Memory Impairment Screen; MMSE = Mini Mental Status Exam; MoCA = Montreal Cognitive Assessment; SLUMS =
St. Louis University Mental Status Exam; Short IQCODE = short Informant Questionnaire on Cognitive Decline in the Elderly

Fig. 1. Alzheimer’s Association Medicare Annual Wellness Visit Algorithm for Assessment of Cognition.

a clinician and the patient and, if present, any family member lation through the Behavioral Risk Factor Surveillance System
or other person who can provide collateral information. This or presented as HRA example questions:
introduces the purpose and content of the AWV, which in-
1. During the past 12 months, have you experienced con-
cludes: a review of the HRA; observations by clinicians
fusion or memory loss that is happening more often or
(medical and associated staff); acknowledgment of any self-
is getting worse [30]?
reported or informant-reported concerns; and conversational
2. During the past 7 days, did you need help with others
queries about cognition directed toward the patient and others
to perform everyday activities such as eating, getting
present. If any concerns are noted, or if an informant is not
dressed, grooming, bathing, walking, or using the toi-
present to provide confirmatory information, further evalua-
let [29]?
tion of cognition with a structured tool should be performed.
3. During the past 7 days, did you need help from others
Patient completion of an HRA is a required element of the
to take care of things such as laundry and housekeep-
AWVand can be accomplished with the help of a family mem-
ing, banking, shopping, using the telephone, food
ber or other knowledgeable informants, including a profes-
preparation, transportation, or taking your own medi-
sional caregiver. Published CMS guidance offers healthcare
cations [29]?
professionals flexibility as to the specific format, questions,
and delivery methods that can be used for an AWV HRA A noted deficit in activities of daily living (ADLs) (e.g.,
[29]. The following questions may be suitable for the AWV eating and dressing) or instrumental activities of daily living
HRA and have been tested and evaluated in the general popu- (IADLs) (e.g., shopping and cooking) that cannot be
 20

C.B. Cordell et al. / Alzheimer’s & Dementia 9 (2013) 141–150

attributed to physical limitations should prompt concern, as the patient can be normalized with a statement such as,
there is a strong correlation between decline in function and “This is something I do for all of my older patients as part
decline in cognitive status across the full spectrum of demen- of their annual visit.” When the initial assessment prompts
tia [31]. In addition to clinically observed concerns, any pa- further evaluation, explanation of results should be deferred
tient- or informant-reported concerns should trigger further until a more comprehensive evaluation has been completed.
evaluation [13]. Positive responses to conversational “There are many reasons for not getting every answer cor-
queries, such as “Have you noticed any change in your mem- rect. More evaluation will help us determine that,” is an ex-
ory or ability to complete routine tasks, such as paying bills ample statement that may encourage patients to pursue
or preparing a meal?” should be followed up with a struc- further testing.
tured assessment of cognition.
Upon realizing the time constraints of a typical primary
5. Full dementia evaluation
care visit, if no cognitive concerns surface during the initial
evaluation and this information is corroborated by an infor- Patients with assessments that indicate cognitive im-
mant, the clinician may elect not to perform a structured cog- pairment during the AWV should be further evaluated to
nitive assessment and assume that the patient is not currently determine appropriate diagnosis (e.g., MCI, Alzheimer’s
demented. This approach is supported by studies in popula- disease) or to identify other causes. As reflected in the algo-
tions with low rates of dementia that suggest the absence of rithm (Figure 1, Step C), initiation of a full dementia evalu-
memory difficulties reported by informants and patients re- ation is outside the scope of the AWV, but can occur in
duces the likelihood that dementia is present [32,33]. a separate visit either on the same day, during a newly sched-
uled visit, or through referral to a specialist. Specialists who
4.3. Structured cognitive assessment tools for use with have expertise in diagnosing dementia include geriatricians,
patients and informants during the AWV geriatric psychiatrists, neurologists, and neuropsychologists.
The two-visit approach has been cited as a time-effective
The second step in detection of cognitive impairment dur- process to evaluate suspected dementia in primary care
ing the AWV (Figure 1, Step B) requires cognitive assess- [34] and is consistent with the two-step approach widely
ment using a structured tool. Based on synthesis of data used in epidemiologic research on dementia. Regardless of
from the six review articles previously discussed, patient the timing and setting, clinicians are encouraged to counsel
tools suitable for the initial structured assessment are the patients to include an informant in the diagnostic process.
GPCOG, Mini-Cog, and MIS. Components of a full dementia evaluation can vary de-
Recognizing that there is no single optimal tool to detect pending on the presentation and include tests to rule in or
cognitive impairment for all patient populations and set- out the various causes of cognitive impairment and establish
tings, clinicians may select other brief tools to use in their its severity. Diagnostic evaluations include a complete med-
clinical practice, such as those listed in Table 3. The 15 brief ical history; assessment of multiple cognitive domains, in-
tools listed were evaluated in multiple review articles cluding episodic memory, executive function, attention,
(passed through at least two review search criteria for tools language, and visuospatial skills; neurologic exam (gait, mo-
possibly suited for primary care) or are used in the VA. Tools tor function, reflexes); ADL and IADL functioning; assess-
listed in Table 3 are subject to the inclusion/exclusion crite- ment for depression; and review for medications that may
ria of each review and do not represent the entire listing of adversely affect cognition. Standard laboratory tests include
the .100 brief cognitive assessment tools that may be suit- thyroid-stimulating hormone (TSH), complete blood count
able for primary care practices. (CBC), serum B12, folate, complete metabolic panel, and,
If an informant is present, defined as someone who can if the patient is at risk, testing for sexually transmitted dis-
attest to a patient’s change in memory, language, or function eases (human immunodeficiency virus, syphilis). Structural
over time, it is suitable to use the AD8, the informant com- brain imaging, including magnetic resonance imaging
ponent of the GPCOG, or the Short IQCODE, during the (MRI) or computed tomography (CT), is a supplemental
AWV. aid in the differential diagnosis of dementia, especially if
neurologic physical exam findings are noted. An MRI or
CT can be especially informative in the following cases: de-
4.4. Primary care workflow considerations
mentia that is of recent onset and is rapidly progressing;
According to the algorithm, any patient who does not younger onset dementia (,65 years of age); history of
have an informant present should be assessed with a struc- head trauma; or neurologic symptoms suggesting focal
tured tool. For such patients (and for practices that imple- disease.
ment structured assessments during all AWVs), completion
of this structured assessment can be administered by trained 6. Discussion
medical staff as the first step for cognitive impairment detec-
tion. This could improve office efficiency. To increase ac- Unfortunately, up to 81% of patients who meet the crite-
ceptance of a structured assessment, the reason provided to ria for dementia have never received a documented diagnosis
 21

C.B. Cordell et al. / Alzheimer’s & Dementia 9 (2013) 141–150

Table 3
Key advantages and limitations of brief cognitive assessment tools evaluated in multiple reviews and/or for use in the VA
Assessment* Time (wmin) Advantages Limitations
7-Minute Screener [48] 7–12 � Little or no education bias � Difficult to administer
� Validated in primary care � Complex logarithmic scoring
AMT [49] 5–7 � Easy to administer � Education/language/culture bias
� Verbal memory test (no writing/drawing) � Limited use in US (mostly used in Europe)
� Does not test executive function or visuospatial
skills
CAMCOG [50] 20 � Tests many separate domains (7) � Difficult to administer
� Long administration time
CDT [51] �1 � Very brief administration time � Lacks standards for administration and scoring
� Minimal education bias
GPCOGy [18]
Patient 2–5 � Developed for and validated in primary care � Patient component scoring has an indeterminate
Informant 1–3 � Informant component useful when initial range that requires an informant score to assess as
complaint is informant-based pass or fail
� Little or no education bias � Informant component alone has low specificity
� Multiple languages accessible at www.gpcog. � Lacks data on any language/culture biases
com.au
Mini-Cogy [8, 19] 2–4 � Developed for and validated in primary care � Use of different word lists may affect failure rates
and multiple languages/cultures � Some study results based on longer tests with the
� Little or no education/language/race bias Mini-Cog elements reviewed independently
� Short administration time
MIS [23,52] 4 � Verbal memory test (no writing/drawing) � Does not test executive function or visuospatial
� Little or no education bias skills
MMSE [17] 7–10 � Most widely used and studied worldwide � Education/age/language/culture bias
� Often used as reference for comparative eval- � Ceiling effect (highly educated impaired subjects
uations of other assessments pass)
� Required for some drug insurance reimburse- � Proprietary—unless used from memory, test needs
ments to be purchased at www.parinc.com
� Best performance for at least moderate cognitive
impairment
MoCAy [53] 10–15 � Designed to test for mild cognitive impairment � Lacks studies in general practice settings
� Multiple languages accessible at www. � Education bias (�12 years)
mocatest.org � Limited use and evidence due to published data
� Tests many separate domains (7) relatively new (2005)
� Admin time �10 min
RUDAS [54] 10 � Designed for multicultural populations � Validated in Australian community
� Little or no education/language bias � Limited use and evidence due to published data
relatively new (2004)
SAS-SI [55] 10 � Detected dementia better than neuropsycho- � Does not test memory
logic testing in a community population � Lacks data on any education/language/culture
biases
SBT (BOMCy and 4–6 � Verbal test (no writing/drawing) � Education/language/cultural/race bias
6-CIT) [56,57] � Scoring can be cumbersome
� Does not test executive function
SLUMSy [58] 7 � No education bias � Limited use and evidence due to published data
� Tests many separate domains (7) relatively new (2006)
� Available at: http://aging.slu.edu/pdfsurveys/ � Studied in VA geriatric clinic (predominantly white
mentalstatus.pdf males)
SPMSQ [59] 3–4 � Verbal test (no writing/drawing) � Scoring can be cumbersome
� Does not test short-term memory
STMSy [60] 5 � Validated in primary care � Education/language/race bias
� Tests many separate domains (7) � Studied in relatively educated subjects, may not be
applicable to general population
T&C [61] �1 � Very brief administration time � Strong language/cultural bias
� Little or no education bias
Abbreviations: 6-CIT, 6-Item Cognitive Impairment Test; AMT, Abbreviated Mental Test; BOMC, 6-item Blessed Orientation-Memory-Concentration Test;
CAMCOG, Cambridge Cognitive Examination; CDT, Clock Drawing Test; GPCOG, General Practitioner Assessment of Cognition; MIS, Memory Impairment
Screen; MMSE, Mini-Mental State Examination; MoCA, Montreal Cognitive Assessment; RUDAS, Rowland Universal Dementia Assessment; SAS-SI, Short
and Sweet Screening Instrument; SBT, Short Blessed Test; SLUMS, St Louis University Mental Status; SPMSQ, Short Portable Mental Status Questionnaire;
STMS, Short Test of Mental Status; T&C, Time and Change Test.
*References provide descriptions of assessments.
y
Brief tools used in the VA healthcare system reviewed by Kansagara and Freeman.
 22

C.B. Cordell et al. / Alzheimer’s & Dementia 9 (2013) 141–150

[35]. Delayed or missed diagnosis deprives affected individ- for Alzheimer’s disease. Detection of cognitive impairment
uals of available treatments, care plans, and services that can during the AWV is further supported by previously pub-
improve their symptoms and help maintain independence. lished quality indicators that state all vulnerable elders (de-
Studies show that interventions tailored to patients with de- fined as persons 65 years who are at risk for death or
mentia can improve quality of care, reduce unfavorable functional decline) should be evaluated annually for cogni-
dementia-related behaviors, increase access to community tive and functional status [46].
services for both the patient and their caregivers, and result There are limitations to these recommendations. They
in less caregiver stress and depression [36–42]. Early are based on assessment of recommendations from review
diagnosis of dementia also provides families and patients articles and on expert opinion, not on a new, comprehensive
an opportunity to plan for the future while the affected review of original research to define the optimal approach
individual is still able to participate in the decision-making to detection of cognitive impairment or review of emerging
processes. technologies that could assist in testing (e.g., use of online
Early detection and medical record documentation may or electronic tablet applications). Further complicating
improve medical care. The medical record could inform SERs of brief cognitive assessment tools is that sensitivity
all clinicians, including those who may be managing comor- and specificity will vary depending on the dementia preva-
bidities on a sporadic basis, that treatment and care should be lence of the study population, the tool(s) used, and the cut
adjusted to accommodate cognitive impairment. According score selected for each tool. Brodaty et al [12] recognized
to a 2004 Medicare beneficiary survey, among patients that published research concerning cognitive impairment
with dementia, 26% had coronary heart disease, 23% had di- screening tools is uneven in quantity and quality. The liter-
abetes, and 13% had cancer [43]. ature also is lacking in comparative validity of brief cogni-
It is important to note that the unstructured and structured tive assessment tools in low-education or illiterate
cognitive assessments being recommended for the AWV are populations.
only the first steps in diagnosing dementia, and cognitive as- The Alzheimer’s Association Medicare Annual Wellness
sessment is best as an iterative process. For example, clini- Visit Algorithm for Assessment of Cognition is based
cians concerned with HRA information about decline in on current validated tools and commonly used rule-out
function may proceed directly to a structured assessment assessments. The use of biomarkers (e.g., CSF tau and
or continue to query the patient for additional information; beta amyloid proteins, amyloid tracer positron emission
a self-reported memory concern coupled with a failed struc- tomography scans) was not considered as these measures
tured cognitive assessment should always result in a full de- are not currently approved or widely available for clinical
mentia evaluation. use.
Not all who are referred for further assessment will ul- In 2011, greater than two million Medicare beneficiaries
timately receive a dementia diagnosis. In a USA primary received their AWV preventive service [47]. There are no
care population aged 65 years (N 5 3340), 13% failed data available as to what methods were used to detect cogni-
a brief screen for cognitive impairment and approxi- tive impairment or how many beneficiaries were assessed
mately half (n 5 227) agreed to be further evaluated as having cognitive impairment. For future AWVs, the
for dementia [7]. Among the 107 patients ultimately di- Alzheimer’s Association Medicare Annual Wellness Visit
agnosed with dementia, 81% were newly diagnosed Algorithm for Assessment of Cognition provides guidance
based on the absence of any medical record of dementia, to primary care practices on a process to operationalize
thus facilitating appropriate medical and psychosocial in- this required AWV element. With widespread implementa-
terventions [7]. tion of the algorithm, the AWV could be the first step in re-
Despite the many advantages of early dementia diagno- ducing the prevalence of missed or delayed dementia
sis, several barriers to diagnosis still exist. These include diagnoses, thus allowing for better healthcare management
physician concerns of the time burden resulting from testing and more favorable outcomes for affected patients and their
and counseling [35] and stigma concerns among physicians, families and caregivers.
patients, and caregivers [35,44,45]. Despite these barriers,
successful widespread implementation of a brief cognitive
7. Author Disclosures
assessment has been reported. McCarten et al [22] evaluated
the Mini-Cog for routine cognitive assessment of veterans Soo Borson is the developer of the Mini-Cog and is the
presenting for primary care. Of the 8342 veterans ap- owner of its copyrights.
proached, .96% agreed to be assessed and those that failed Over the past 5 years, Malaz Boustani has received re-
the brief assessment exhibited no serious reactions upon dis- search support for investigator- initiated projects from Forest
closure of test results. Pharmaceutical and Novartis; honoraria from Norvartis and
The AWV provides an unprecedented opportunity to Pfizer, Inc.; and research support for investigator-initiated
overcome current barriers and initiate discussions about cog- projects from the NIH and AHRQ. Dr Boustani was a mem-
nitive function among the growing population most at risk ber of the US Preventive Services Task Force that published
 23

C.B. Cordell et al. / Alzheimer’s & Dementia 9 (2013) 141–150

the systematic evidence review, Dementia Screening, for the [8] Borson S, Scanlan JM, Watanabe J, Tu S-P, Lessig M. Improving iden-
AHRQ in 2003. tification of cognitive impairment in primary care. Int J Geriatr Psychi-
atry 2006;21:349–55.
[9] Anonymous. Medicare coverage of Annual Wellness Visit providing
a personalized prevention plan. Fed Regist 2010;75:73401.
RESEARCH IN CONTEXT [10] US Department of Health and Human Services. Advisory Council on
Alzheimer’s research, care, and services: opportunities and gaps. 2011.
Available at: http://aspe.hhs.gov/daltcp/napa/092711/Mtg1-Slides3.
pdf. Accessed October 14, 2011.
1. Systematic review: Our research included comparing [11] Lorentz WJ, Scanlan JM, Borson S. Brief screening tests for dementia.
Can J Psychiatry 2002;47:723–33.
five systematic evidence reviews (SER) of brief de-
[12] Brodaty H, Low L-F, Gibson L, Burns K. What is the best dementia
mentia screening tools published since 2000 and screening instrument for general practitioners to use? Am J Geriatr
a 2010 literature review of newer brief assessments Psychiatry 2006;14:391–400.
of cognition. Our research focused on determining [13] Holsinger T, Deveau J, Boustani M, Williams JW Jr. Does this patient
if there was a consensus among the published SERs have dementia? JAMA 2007;297:2391–404.
[14] Milne A, Culverwell A, Guss R, Tuppen J, Whelton R. Screening for
as to which tool is most suited for primary care and
dementia in primary care: a review of the use, efficacy and quality of
if there were any common results across the publica- measures. Int Psychogeriatr 2008;20:911–26.
tions. [15] Ismail Z, Rajji TK, Shulman KI. Brief cognitive screening instru-
ments: an update. Int J Geriatr Psychiatry 2010;25:111–20.
2. Interpretation: Our research concluded there is a con- [16] Kansagara D, Freeman M. A systematic evidence review of the signs and
sensus in the literature concerning suitable tools for symptoms of dementia and brief cognitive tests. Available at: http://
screening for dementia in primary care. We also reaf- www.ncbi.nlm.nih.gov/pubmed/21155200. Accessed June 7, 2011.
firmed that many validated tools are available, and [17] Folstein MF, Folstein SE, McHugh PR. “Mini-mental state.” A practi-
cal method for grading the cognitive state of patients for the clinician.
that screening for dementia should not be solely J Psychiatr Res 1975;12:189–98.
based on a tool, but should be a stepwise process to [18] Brodaty H, Pond D, Kemp NM, Luscombe G, Harding L, Berman K,
include other assessments. et al. The GPCOG: a new screening test for dementia designed for gen-
eral practice. J Am Geriatr Soc 2002;50:530–4.
3. Future directions: Further validation of existing and [19] Borson S, Scanlan J, Brush M, Vitaliano P, Dokmak A. The Mini-Cog:
emerging screening tools (e.g., iPad applications, a cognitive “vital signs” measure for dementia screening in multi-
gait monitoring) may result in newer tools being rec- lingual elderly. Int J Geriatr Psychiatry 2000;15:1021–7.
ognized more suitable and practical for primary care [20] Borson S, Scanlan JM, Watanabe J, Tu S-P, Lessig M. Simplifying de-
tection of cognitive impairment: comparison of the Mini-Cog and
settings. Mini-Mental State Examination in a multiethnic sample. J Am Geriatr
Soc 2005;53:871–4.
[21] Borson S, Scanlan JM, Chen P, Ganguli M. The Mini-Cog as a screen
for dementia: validation in a population-based sample. J Am Geriatr
Soc 2003;51:1451–4.
References [22] McCarten JR, Anderson P, Kuskowski MA, McPherson SE, Borson S.
Screening for cognitive impairment in an elderly veteran population:
[1] Boustani M, Peterson B, Harris R, Lux L, Krasnov C, Sutton S, et al. acceptability and results using different versions of the Mini-Cog.
Screening for dementia. Rockville, MD: Agency for Healthcare Re- J Am Geriatr Soc 2011;59:309–13.
search and Quality. Available at: http://www.ncbi.nlm.nih.gov/books/ [23] Buschke H, Kuslansky G, Katz M, Stewart WF, Sliwinski MJ,
NBK42773/; 2003. Accessed September 3, 2011. Eckholdt HM, et al. Screening for dementia with the memory impair-
[2] Anonymous. Patient Protection and Affordable Care Act of 2010, 42 ment screen. Neurology 1999;52:231–8.
CFR. x410.15(a). 2010. Available at: http://ecfr.gpoaccess.gov/cgi/t/ [24] Mackinnon A, Mulligan R. Combining cognitive testing and informant
text/text-idx?c5ecfr&sid56b50669da0f96db4eea346533db23747& report to increase accuracy in screening for dementia. Am J Psychiatry
rgn5div8&view5text&node542:2.0.1.2.10.2.35.4&idno542. 1998;155:1529–35.
[3] Chodosh J, Petitti DB, Elliott M, Hays RD, Crooks VC, [25] Jorm AF. A short form of the Informant Questionnaire on Cognitive
Reuben DB, et al. Physician recognition of cognitive impairment: Decline in the Elderly (IQCODE): development and cross-validation.
evaluating the need for improvement. J Am Geriatr Soc 2004; Psychol Med 1994;24:145–53.
52:1051–9. [26] Ayalon L. The IQCODE versus a single-item informant measure to
[4] Camicioli R, Willert P, Lear J, Grossmann S, Kaye J, Butterfield P. discriminate between cognitively intact individuals and individuals
Dementia in rural primary care practices in Lake County, Oregon. with dementia or cognitive impairment. J Geriatr Psychiatry Neurol
J Geriatr Psychiatry Neurol 2000;13:87–92. 2011;24:168–73.
[5] Callahan CM, Hendrie HC, Tierney WM. Documentation and evalua- [27] Galvin JE, Roe CM, Morris JC. Evaluation of cognitive impairment in
tion of cognitive impairment in elderly primary care patients. Ann In- older adults: combining brief informant and performance measures.
tern Med 1995;122:422–9. Arch Neurol 2007;64:718–24.
[6] Valcour VG, Masaki KH, Curb JD, Blanchette PL. The detection of [28] Galvin JE, Roe CM, Xiong C, Morris JC. Validity and reliability
dementia in the primary care setting. Arch Intern Med 2000; of the AD8 informant interview in dementia. Neurology 2006;
160:2964–8. 67:1942–8.
[7] Boustani M, Callahan CM, Unverzagt FW, Austrom MG, [29] Goetzel R, Staley P, Ogden L, Strange P, Fox J, Spangler J, et al. A
Perkins AJ, Fultz BA, et al. Implementing a screening and diagnosis framework for patient-centered health risk assessments—providing
program for dementia in primary care. J Gen Intern Med 2005; health promotion and disease prevention services to Medicare benefi-
20:572–7. ciaries. Atlanta, GA: US Department of Health and Human Services,
 24

C.B. Cordell et al. / Alzheimer’s & Dementia 9 (2013) 141–150

Centers for Disease Control and Prevention. Available at: http://www. [45] Boustani MA, Justiss MD, Frame A, Austrom MG, Perkins AJ,
cdc.gov/policy/opth/hra/; 2011. Cai X, et al. Caregiver and noncaregiver attitudes toward dementia
[30] Anonymous. Behavioral Risk Factor Surveillance System Survey screening. J Am Geriatr Soc 2011;59:681–6.
Questionnaire. Bethesda, MD: Centers for Disease Control and [46] Feil DG, MacLean C, Sultzer D. Quality indicators for the care of de-
Prevention. Available at: http://www.cdc.gov/brfss/questionnaires/ mentia in vulnerable elders. J Am Geriatr Soc 2007;55(Suppl
pdf-ques/2011brfss.pdf; 2011. Accessed January 10, 2012. 2):S293–301.
[31] Njegovan V, Hing MM, Mitchell SL, Molnar FJ. The hierarchy of [47] Anonymous. Preventive New Media. Centers for Medicare &
functional loss associated with cognitive decline in older persons. Medicaid Services. Available at: http://www.cms.gov/NewMedia/
J Gerontol A Biol Sci Med Sci 2001;56:M638–43. 02_preventive.asp; 2011. Accessed January 10, 2012.
[32] Carr DB, Gray S, Baty J, Morris JC. The value of informant versus in- [48] Solomon PR, Pendlebury WW. Recognition of Alzheimer’s disease:
dividual’s complaints of memory impairment in early dementia. Neu- the 7 Minute Screen. Fam Med 1998;30:265–71.
rology 2000;55:1724–6. [49] Hodkinson HM. Evaluation of a mental test score for assessment of
[33] Tobiansky R, Blizard R, Livingston G, Mann A. The Gospel Oak Study mental impairment in the elderly. Age Ageing 1972;1:233–8.
stage IV: the clinical relevance of subjective memory impairment in [50] Roth M. CAMDEX: the Cambridge examination for mental disorders
older people. Psychol Med 1995;25:779–86. of the elderly. Cambridge: Cambridge University Press; 1988.
[34] Simmons BB, Hartmann B, Dejoseph D. Evaluation of suspected de- [51] Shulman KI. Clock-drawing: is it the ideal cognitive screening test? Int
mentia. Am Fam Physician 2011;84:895–902. J Geriatr Psychiatry 2000;15:548–61.
[35] Bradford A, Kunik ME, Schulz P, Williams SP, Singh H. Missed and [52] Kuslansky G, Buschke H, Katz M, Sliwinski M, Lipton RB. Screening
delayed diagnosis of dementia in primary care: prevalence and contrib- for Alzheimer’s disease: the memory impairment screen versus the
uting factors. Alzheimer Dis Assoc Disord 2009;23:306–14. conventional three-word memory test. J Am Geriatr Soc 2002;
[36] Bass DM, Clark PA, Looman WJ, McCarthy CA, Eckert S. The Cleve- 50:1086–91.
land Alzheimer’s managed care demonstration: outcomes after 12 [53] Nasreddine ZS, Phillips NA, B�edirian V, Charbonneau S,
months of implementation. Gerontologist 2003;43:73–85. Whitehead V, Collin I, et al. The Montreal Cognitive Assessment,
[37] Callahan CM, Boustani MA, Unverzagt FW, Austrom MG, MoCA: a brief screening tool for mild cognitive impairment. J Am
Damush TM, Perkins AJ, et al. Effectiveness of collaborative care Geriatr Soc 2005;53:695–9.
for older adults with Alzheimer disease in primary care: a randomized [54] Storey JE, Rowland JTJ, Basic D, Conforti DA, Dickson HG. The
controlled trial. JAMA 2006;295:2148–57. Rowland Universal Dementia Assessment Scale (RUDAS): a multicul-
[38] Fortinsky RH, Unson CG, Garcia RI. Helping family caregivers by tural cognitive assessment scale. Int Psychogeriatr 2004;16:13–31.
linking primary care physicians with community-based dementia [55] Belle SH, Mendelsohn AB, Seaberg EC, Ratcliff G. A brief cognitive
care services: the Alzheimer’s Service Coordination Program. Demen- screening battery for dementia in the community. Neuroepidemiology
tia 2002;1:227–40. 2000;19:43–50.
[39] Reuben DB, Roth CP, Frank JC, Hirsch SH, Katz D, McCreath H, et al. [56] Katzman R, Brown T, Fuld P, Peck A, Schechter R, Schimmel H. Val-
Assessing care of vulnerable elders—Alzheimer’s disease: a pilot idation of a short Orientation-Memory-Concentration Test of cogni-
study of a practice redesign intervention to improve the quality of de- tive impairment. Am J Psychiatry 1983;140:734–9.
mentia care. J Am Geriatr Soc 2010;58:324–9. [57] Brooke P, Bullock R. Validation of a 6 item cognitive impairment test
[40] Vickrey BG, Mittman BS, Connor KI, Pearson ML, Della Penna RD, with a view to primary care usage. Int J Geriatr Psychiatry 1999;
Ganiats TG, et al. The effect of a disease management intervention on 14:936–40.
quality and outcomes of dementia care: a randomized, controlled trial. [58] Tariq SH, Tumosa N, Chibnall JT, Perry MH 3rd, Morley JE. Compar-
Ann Intern Med 2006;145:713–26. ison of the Saint Louis University mental status examination and the
[41] Olazar�an J, Reisberg B, Clare L, Cruz I, Pe~na-Casanova J, Del Ser T, mini-mental state examination for detecting dementia and mild neuro-
et al. Nonpharmacological therapies in Alzheimer’s disease: a system- cognitive disorder—a pilot study. Am J Geriatr Psychiatry 2006;
atic review of efficacy. Dement Geriatr Cogn Disord 2010;30:161–78. 14:900–10.
[42] Auclair U, Epstein C, Mittelman M. Couples counseling in Alz- [59] Pfeiffer E. A short portable mental status questionnaire for the assess-
heimer’s disease: additional clinical cindings from a novel intervention ment of organic brain deficit in elderly patients. J Am Geriatr Soc
study. Clin Gerontol 2009;32:130–46. 1975;23:433–41.
[43] Thies W, Bleiler L. 2011 Alzheimer’s disease facts and figures. Alz- [60] Kokmen E, Naessens JM, Offord KP. A short test of mental status: de-
heimers Dement 2011;7:208–44. scription and preliminary results. Mayo Clin Proc 1987;62:281–8.
[44] Justiss MD, Boustani M, Fox C, Katona C, Perkins AJ, Healey PJ, et al. [61] Inouye SK, Robison JT, Froehlich TE, Richardson ED. The time and
Patients’ attitudes of dementia screening across the Atlantic. Int J Ger- change test: a simple screening test for dementia. J Gerontol A Biol
iatr Psychiatry 2009;24:632–7. Sci Med Sci 1998;53:M281–6.