The psychometric properties of the third version of Addenbrooke's

J Kermanshah Univ Med Sci. 2017; 21(2) (57)

Kermanshah University of

Medical Sciences

Journal of Kermanshah University of Medical Science

Journal homepage: Htpp://journals.kums.ac.ir/ojs/index.php/jkums

The psychometric properties of the third version of Addenbrooke's

Cognitive Examination (ACE-III) in a sample of Iranian older adults

Majid Mahmoud Alilou

, Zeinab Khanjani

, Touraj Hashemi

, Soheila Parvaz

1. Dept. of Psychology, Faculty of Education & Psychology, University of Tabriz, Tabriz, East Azarbaijan, Iran

Article Info

Keywords: ACE, psychometric

properties, older adult, factor

analysis

*Corresponding Author:

29 Bahman Boulevard, Faculty of

Education & Psychology,

University of Tabriz, Tabriz, East

Azarbaijan, Iran

Tel: +98 9185627601

Email:

Parvaz_soheil[email protected]

Received: 02 May, 2017

Accepted: 05 September, 2017

J Kermanshah Univ Med Sci.

2017; 21(2): 57-61

Abstract

Introduction: Cognitive problems such as dementia are common in older adults and

their prevalence increases with age. The early identification and diagnosis of patients

with dementia can help with their treatment and improve their quality of life. The

present study was conducted to investigate the psychometric properties and validate

the Addenbrooke's Cognitive Examination (ACE) in a sample comprising older adult

Iranians.

Methods: The present cross-sectional and correlational study recruited 300 older

adults in Kahrizak Geriatric Nursing Home in Iran, including 198 men and 102

women selected using simple random sampling. The data collection tools comprised

the ACE-III, the Mini–Mental State Examination (MMSE) and the Geriatric

Depression Scale (GDS). Confirmatory factor analysis was used to investigate the

construct validity of the test, and the Pearson’s correlation coefficient to examine its

convergent and divergent validity. Cronbach’s alpha was also used to investigate the

internal consistency of the items. To examine the diagnostic validity, cut-off point,

sensitivity and specificity of the test were calculated.

Results: The results found correlations between the ACE-III and other tools

(P>0.01), thus suggesting a proper convergent and divergent validity. The test-retest

reliability coefficient with a two-week interval and the Cronbach’s alpha of the

ACE-III were respectively calculated as 0.90 and 0.95. The optimal cut-off point that

struck a balance between sensitivity and specificity was found to be 75, with a

sensitivity of 0.99 and specificity of 0.95. The results of the factor analysis indicated

a good fit of the single-factor structure of this test.

Conclusion: The ACE-III has good psychometric properties and it can be used to

screen for dementia.

Introduction

Given advancements in medical sciences and

improvements in nutrition and health, today’s world is

experiencing a new phenomenon, namely population

aging (1). The population of older adults is estimated to

rise from 694 million in 1970 to 1.2 billion in 2025 and

close to 2 billion in 2050 (2). Old age is associated with

certain physiological changes that disrupt the

performance of the body systems and cause more

vulnerability to diseases (3). Cognitive impairment is a

common disorder associated with severe and progressive

disability in old ages. Despite the growing public

awareness about old-age problems, many people still

consider cognitive impairment and progressive memory

loss a normal part of the aging process; nevertheless,

given today’s promoted neuropsychological knowledge

of aging and advanced medical technology, normal

changes in the aging period can be discriminated from

brain-damaging processes. Dementia is a common

disorder associated with cognitive processes in old age

that highlights the importance of screening (4).

Dementia is a disorder which is characterized by

numerous cognitive defects, including serious memory

loss (5). Cognitive changes in dementia are normally

associated with mood, behavior and personality

disorders (6). The prevalence of cognitive impairment in

older adult populations has rarely been investigated in

Iran. A study conducted by Sharifi et al. (7) is the first

comprehensive research on the prevalence of dementia

in Iranian older adults. These researchers found the

prevalence of dementia to be 7.9%, i.e. 7.8% in women

and 6.9% in men, in the population aged over 60. They

estimated this prevalence at 8.1%, i.e. 9.6% in women

and 6.5% in men, based on age-associated WHO

standards. They also found the dementia prevalence to

be 3.7% in 60-64-year-olds, 6.2% in 65-69-year-olds,

10.4% in 70-74-year-olds, 14.4% in 75-79-year-olds and

13% in those aged over 80. This study found the lowest

prevalence of dementia to be in East Azarbaijan and the

highest to be in North Khorasan. Rashedi et al. (8) used

the MMSE to investigate the prevalence of cognitive

impairment in older adults and to examine its

relationship with demographic variables such as age and

level of education in the day care center of Hamadan,

Iran. The results showed mild cognitive impairment in

Original Article

(58) Mahmoud Alilou & et al

45.3% of the subjects, moderate cognitive impairment in

51.9% and severe cognitive impairment in 2.8% of

them. They also found that older age and lower level of

education were associated with lower the MMSE scores

and higher intensity of cognitive impairment.

Evaluating one’s performance through a mini-

cognitive test is a common method of diagnosing

dementia and a prerequisite for more tests and

investigations, which are usually called screening tools.

Compelling evidence suggests that the early diagnosis of

patients with cognitive impairment and referring them to

counseling services can help reduce stress in the families

and caregivers of these patients. It is worth mentioning

that dementia is not only the cause of much suffering

and debilitation in these patients, but also it heavily

involves families who are mainly responsible for their

care. The stress caused by taking care of patients with

dementia is high and research suggests a high

prevalence of psychological disorders in the caregivers

(9-10). Simple-to-use and reliable tools are therefore

urgently needed for daily clinical routines. Western

countries, which faced the phenomenon of population

aging for several decades, have designed and developed

different tools over the last forty years. Some of these

tools were globally accredited, translated to different

languages and standardized. Given the recent growing

population of Iranian older adults, simple tools are

needed to be developed and normalized to identify cases

with a suspected dementia and to reliably document the

cognitive changes caused by the damaging processes

and therapeutic factors (9). The benefits of developing

this type of tools are as follows: 1) Conducting

community-based epidemiological studies with the

minimum cost and time (the statistics obtained from

these studies will help health planning and contribute to

the optimal use of resources). 2) Given the introduction

of new treatments including such as acetylcholinesterase

inhibitors, which slow down the disease progression,

simplify the control of the symptoms and improve the

cognitive and psychological status of the patient (9),

highly-valid tools, which can be used to quantitatively

record the changes caused by the medication, play a key

role in determining the beginning and end of the

treatment as well as detecting responsive and non-

responsive cases to the treatment. 3) Given the high cost

of medical therapies and the fact that these medicines

are effective only in the initial and mild stages of the

disease and lose their impact with the progression of the

damaging process, the quantitative evaluation of the

cognitive status of the patients is necessary. These tools

play a key role in clinical trials and determining the

effectiveness of new medicines which are rapidly

introduced to the market. The timely diagnosis and

referring of the patient leads to the early start of the

treatment, helps the relatives and caregivers use

counseling and support services, and reduces the

financial, social, psychological and physical costs of the

disease.

Neuropsychological investigations constitute a

fundamental part of evaluating cases with a suspected

dementia. Several screening and diagnostic tests have

been developed for dementia, most of which are beyond

the cognitive evaluation range and require special

equipment or trained people. MMSE (12) is extensively

used for evaluating the psychological status. Although

MMSE can detect dementia with a relatively high

sensitivity and specificity (15-16), it is criticized for

inadequately examining cognitive functions, including

fronto-executive functions, visual-spatial functions and

semantic memory. All these functions can contribute to

fundamental defects in specific declining circumstances

such as frontotemporal dementia and Lewy body

dementia (15-16). Other weaknesses of the MMSE

include its variable accuracy for diagnosing patients

with dementia with different ages, levels of education

and ethnicity (17), and its low sensitivity for detecting

mild cases of cognitive impairment (18-19).

Given these limitations, the ACE was developed by

Mathuranath et al. (20) in the memory clinic of

Edinburgh Hospital in Cambridge, the UK to diagnose

dementia and differentiate Alzheimer’s disease from

frontotemporal dementia (21). These developers felt a

need for screening this test for three reasons as follows.

1- A large proportion of patients already diagnosed with

Alzheimer’s disease presented other declining

conditions such as Lewy body dementia and

frontotemporal dementia (22). 2- Accessibility of factors

and techniques modifying the disease highlights the

importance of the early diagnosis of dementia (20). 3-

Growing concerns about memory loss in declining years

in ordinary populations (23). ACE was revised in 2006

and ACE-R (24) was developed to increase sensitivity

and specificity of the test for detecting cognitive

impairment associated with dementia. The other purpose

of this test was to increase sensitivity to mild cognitive

impairment. Owing to some weaknesses, a few items

were replaced in ACE-R to develop ACE-III. This tool

is used today as a valid test for diagnosing dementia in

diagnostic and treatment centers. This test can enter the

diagnostic and treatment system of Iran through

preliminary investigations, and can be used as a reliable

instrument for diagnosing and screening for dementia in

Iranian older adults. Given that these patients present to

different centers, this test can be used in clinics, nursing

homes, psychiatric hospitals, health homes, outpatient

clinics and rehabilitation centers. Although Iran has a

young population, the fear is that the prevalence of

Alzheimer’s disease and dementia significantly

increases after today’s young population reach old age.

It therefore appears reasonable to upgrade our

knowledge about the screening and diagnosis of this

disease and achieve a proper tool for this purpose.

Materials and Methods

The present study is correlational in terms of

objective and descriptive in terms of data collection

method. The data collected were analyzed in SPSS-20

and LISREL-8 using the Pearson correlation coefficient

and confirmatory factor analysis. Simple random

sampling was used to select 300 older adults from

Kahrizak Geriatric Nursing Home, including 198 men

and 102 women with an age of 65-90 years. The mean

age of the study men was 75±1.92 and that of women

78±1.34.

Tools

ACE-III: The main version of ACE-III was

developed by Mathuranath et al. (20) in the memory

J Kermanshah Univ Med Sci. 2017; 21(2) (59)

clinic of Edinburgh Hospital to diagnose dementia and

differentiate Alzheimer’s disease from frontotemporal

dementia (21). Two revisions of ACE include ACE-R

and ACE-III. ACE-III comprises five domains, namely

attention/orientations, memory, verbal fluency, language

and visual-spatial skills. This test takes an average of 15

minutes to be completed, the maximum score is 100 and

higher scores denote better cognitive performance.

ACE-III has an internal consistency coefficient of 0.88,

a sensitivity of 100% and a specificity as high as 0.96

with a cut-off point of 88 (23). Research confirms the

effectiveness of this test for the early diagnosis of

dementia and differentiating frontotemporal dementia

from Lewy body dementia (25). No studies have

validated ACE-III in Iran.

MMSE: This questionnaire was introduced to

clinical experts by Folsein et al. (12) in 1975 as a

practical method of scaling the psychological status of

patients. Five to ten minutes is required to respond to its

11 items. The domains of this test include orientation to

time and place, attention/concentration, language, as

well as immediate and delayed recall. The split-half

method was used and the reliability coefficient was

reported to be 0.72, specificity 0.84 and sensitivity 0.90

(10). Foroghan et al. (26) normalized this test in Iran.

The results suggested that the internal consistency of the

test is 0.87 based on calculating the Cronbach’s alpha

and the reliability coefficient is 0.71 using the split-half

method. With a cut-off point of 21, sensitivity was

obtained as 0.90 and specificity as 0.84. This test was

used in the present study to investigate the convergent

validity of ACE-III.

GDS: Yesavage et al. (27) developed this test in 1983 as

a tool for screening depression in older adults. The long-

form GDS includes 30 items and the short-form 15

items, which are rated by Yes and No. The internal

consistency (Cronbach’s alpha) of this tool was reported

as 0.94, and its reliability coefficient was calculated as

0.94 using the split-half method. The test-retest also

suggested a reliability coefficient of 0.85 after a week.

The concurrent validity coefficients between this test

and the Beck Depression Inventory, the Zung Self-

Rating Depression Scale and the Hamilton Rating Scale

for Depression was respectively reported to be 0.73,

0.84 and 0.83 (28). A study was conducted on 300

randomly-selected older adults in Iran to examine the

reliability and factor structure of GDS (29). The internal

consistency coefficient was calculated as 0.4 and the

Cronbach’s alpha as 0.9. The reliability coefficient was

calculated as 0.89 using the split-half method and 0.58

using the test-retest after two weeks (29). This test was

used in the present study to investigate the divergent

validity. Numerous studies have examined the

relationship between depression and cognitive decline.

The findings suggest relationships between cognitive

impairment and depression, which abound in senior care

centers (30). Investing on family ties and social support

appears to help older adults resist depression and

cognitive impairment (30).

Findings

Table 1 shows the mean values of the study variables.

The test-retest coefficient of ACE-III was also obtained

by calculating the correlation between performing the

test twice with a two-week interval.

Table 1. The mean values associated with ACE-III and its dimensions and the test-retest coefficient (n=30)

Cognitive dimension

Mean

Standard Deviation

Pearson Correlation Coefficient

Attention

14.65

4.32

0.91**

Memory

22.41

3.87

0.89**

Verbal Fluency

12.20

2.09

0.93**

Language

23.50

4.32

0.92**

Visual-Spatial Skills

14.29

2.55

0.90**

Overall ACE-III Score

89.59

9.09

0.90**

** P<0.01

The findings associated with convergent validity, i.e.

correlation with MMSE, and divergent validity, i.e.

correlation with GDS, confirm the validity of ACE-III

(Table 2). Table 3 indicates that there are correlations

among all dimensions of the ACE-III (P<0.01).

Confirmatory factor analysis was used to assess the

internal constructs of the items of ACE-III subscales,

and the findings suggested that the single-factor pattern

has a good fitness in the study sample, since some of the

measured indices associated with the fit of the model

were within the acceptable range; the root mean square

Table 2. Findings associated with convergent validity

(correlation with MMSE) and divergent validity (correlation

with GDS)

Cognitive Dimension

MMSE

GDS

Attention

0.66**

-0.56**

Memory

0.75**

-0.54**

Verbal Fluency

0.65**

0.41**

Language

0.53**

0.39**

Visual-Spatial Skills

0.56**

-0.43**

Overall Score

0.69**

-0.53**

** P<0.01

Table 3. Matrix of correlation among the dimensions of ACE-III

Cognitive Dimension

1-Attention

2-Memory

0.85**

3-Verbal Fluency

0.40**

0.37**

4-Language

0.77**

0.64**

0.19**

5-Visual-Spatial Skills

0.75**

0.67**

0.43**

0.77**

** P<0.01

(60) Mahmoud Alilou & et al

error of approximation (RMSEA) was acceptable, i.e.

equal to 0.65, and the values of the fit indices, including

NFI, AGFI, GFI, CFI and NINFI, were all acceptable,

i.e. equal to 0.90. In addition, the ratio of X

/df was

found to be less than 5 and thus acceptable (P<0.01).

Moreover, investigating the internal consistency of the

items and the correlation coefficient of the score of

every item with the overall factor score suggested that

there are no needs for eliminating any of the items.

To assess ACE-III in terms of discriminating patients

with dementia from normal subjects, 30 patients with

Alzheimer's disease were matched with 30 healthy

subjects in terms of age and level of education and

compared and the ROC was obtained (Figure 1). Given

the ROC, with a cut-off point of 75, sensitivity was

found to be 0.99, specificity 0.95 and the area under the

ROC was 0.94.

Figure 1. The ROC for discriminating patients with

dementia from normal cases

Discussion

ACE-III was developed by Mathuranath et al. for the

early diagnosis and screening of people with a suspected

dementia. This tool assesses five dimensions, namely

memory, verbal fluency, language and visual-spatial

skills. The present study was conducted to investigate

the psychometric properties of this test. The results

confirmed the discriminant, convergent and divergent

validity of this test. The internal consistency coefficient

and the scale reliability were also confirmed. The

confirmatory factor analysis of the ACE-III showed that

the single-factor pattern has a good fitness. A review of

the literature suggests the lack of studies on the factor

structure of ACE-III. The results associated with the

discriminant validity of the test and its dimensions

showed that the questionnaire dimensions can well

discriminate patients with dementia from healthy

subjects. These findings are consistent with those found

in the literature (25 and 31). The diagnostic ability of

ACE-III and the closeness between the results it

produced and clinical diagnosis highlight its value as far

as the main purpose of the scale is concerned. The

relationship between depression and dementia has

frequently been emphasized in older adults (33). The

negative correlation between depression and ACE-III

can indicate an aspect of its validity given the divergent

validity shown. The correlations of ACE-III and its

dimensions with the MMSE were found to be positive.

Lower scores obtained from ACE-III indicate higher

risks of dementia. ACE-III must therefore show a

positive correlation with MMSE. The present study

findings thus confirmed the reliability and validity of

ACE-III in Iranian older adults. Given the limitations of

the present study, it can be used to examine dementia in

the Iranian older adult population. ACE-III has been

found to be significantly correlated with standard

neurological tests used to evaluate attention, language,

verbal memory and visual-spatial skills (21). The ACE-

III has also been shown to be correlated with ACE-R.

Hsieh et al. (25) validated ACE-III in patients with

frontotemporal dementia and patients with Alzheimer's

disease and compared it with other standard tests. The

results still suggested a high sensitivity and specificity

for this test. With a cut-off point of 88, sensitivity was

100 and specificity 96. The relationship between every

cognitive dimension of ACE-III and the associated tests,

e.g. the relationship between memory and digit span in

the Wechsler intelligence scale, suggested high

correlations and confirmed the test validity. The study

conducted by Mathuranath et al. (20) suggested a high

sensitivity of the test in diagnosing dementia and

discriminating Alzheimer's disease from frontotemporal

dementia. In fact, one of the strengths of ACE-III

compared to other tests is that it measures cognitive

domains that are damaged in different types of dementia.

Conclusion

ACE-III possesses favorable psychometric properties

which are adapted to the Iranian culture. It can be used

in clinical practices to differentiate patients with

dementia from normal cases. This test can also be used

as a screening method for identifying patients with a

suspected dementia in different outpatient clinics and

also in different research projects.

Acknowledgments

The authors would like to express their sincere

gratitude to all who helped conduct the present research.

It is worth noting that we received no funding for this

study.

References

1. Larson EB. Prospects for delaying the rising tide of worldwide, late-life dementias. Int Psychogeriatr. 2010; 22(8): 1196-202.

2. World Health Organization. Dementia: a public health priority. United Kingdom, 2012.

http://www.who.int/mental_health/publications/dementia_report_2012/en/

3. Denton FT, Spencer BG. Chronic health conditions: changing prevalence in an aging population and some implications for the

delivery of health care services. Can J Aging. 2010; 29(1): 11-21.

4. Chop WC, Robnett HR. Gerontology for the health care from fissional. 3th ed. Philadelphia: FA Davis Company. 1999; 92-11.

5. Kaplan HI, Sadock BJ. Synopsis of psychiatry: Behavioral sciences clinical psychiatry. 11th ed. New york: Williams & Wilkins

Company. 2015;2437-5.

6. Walash D. Dementia care training manual for staff in nursing and residential setting. 2th ed. London: Jesica Kingsley Publishers

2006; 78-9.

J Kermanshah Univ Med Sci. 2017; 21(2) (61)

7. Sharifi F, Fakhrzadeh H, Varmaghani M, Arzaghi SM, Alizadeh Khoei M, Farzadfar F,et al. Prevalence of dementia associated

factors among older adults in Iran: National Elderly Health Survey(NEHS). Arch Iran Med 2016;19(12): 838-44.

8. Rashedi V, Rezaei M, Gharbi M. Prevalence of cognitive impairment in community-dwelling older adults. Basic Clin Neurosci.

2014; 5(1): 28-30.

9. Foroghan M. [Dementia screening: validated cognitive tests in Iranian elderly (Persian)]. 1st ed. Teharan: Arjomand Publications

2003; 24-6.

10. Noroozian M. The elderly population in Iran: An ever growing concern in the health system. Iran J Psychiatry Behav Sci.

2012;6:1-6.

11. National Institute for Clinical Excellence: Guidance on the use of donepezil, rivastigmine and galantamine for the treatment of

Alzheimer’s disease. NICE Technology Appraisal Guidance 2001, No.19 (http://www.nice.org.uk/pdf/

ALZHEIMER_full_guidance.pdf).

12. Folstein MF, Folstein SE, McHugh PR. Mini-Mental State Examination. A Practical method for grading the cognitive state of

patients for the clinician. J Psychiatr Res. 1975; 12(3): 189-98.

13. Rosen WG, Mohs RC, Davis KL. A new rating scale for Alzheimer's disease. AM J Psychiatry. 1984; 141(11): 1356-64.

14. Zaudig M, Mittelhammer J, Hiller W, Pauls A, Thora C, Morinigo A, Mombour W. SIDAM–A structured interview for the

diagnosis of dementia of the Alzheimer type, multi-infarct dementia and dementias of other aetiology according to ICD-10 and

DSM-III-R. Psychol Med. 1991; 21(01): 225-36.

15. Bak TH, Mioshi E. A cognitive bedside assessment beyond the MMSE: the Addenbrooke’s Cognitive Examination. Pract

Neurol. 2007;7(4):245-9.

16. Tomburgh TN, McIntyre NJ. The mini-mental examination: A comprehensive review. J Am Geriatr Soc. 1992; 40(9): 922-35.

17. Cullen B, O’Neill B, Evans JJ, Coen RF, Lawlor BA. A review of screening tests for cognitive impairment. J Neurol Neurosurg

Psychiatry. 2007;78:790-9.

18. Boustani M, Peterson B, Hanson L, Harris R, Lohr KN. Screening for dementia in primary care: a summary of the evidence for

the U.S. Preventive Services Task Force. Ann Intern Med. 2003; 138(11): 927-37.

19. Kukull W, Larson E, Teri L, Bowen J, McCormick W, Pfanschmidt M. The mini-mental state examination score and the clinical

diagnosis of dementia. J Clin Epidemiol. 1994; 47(9): 1061-7.

20. Mathuranath PS, Nestor PJ, Berrios GE, Rakowicz W, Hodges JR. A brief cognitive test battery to differentiate Alzheimer’s

disease and Frontotemporal dementia. Neurology. 2000; 55(11): 1613-20.

21. Stokholm J, VogelA, Johannsen P, Waldemar G. Validation of the Danish Addenbrooke’s Cognitive Examination as a screening

test in a memory clinic. Dement Geriatr Cogn Disord. 2009; 27(4): 361-5.

22. McKeith I, Fairbairn A, BothwellR, Moore P, Ferrier I, Thompson P, et al. An evaluation of the predictive validity and

inter‐ rater reliability of clinical diagnostic criteria for senile dementia of Lewy body type. Neurology. 1994; 44(5): 872-7.

23. Verhey FR, Jolles J, Ponds RW, Rozendaal N, Plugge LA, De vet RC. A Comparison between a monodisciplinary and a

multidisciplinary approach. J Neuropsychiatry Clin Neurosci. 1993;5(1):78-85.

24. Mioshi E, Dawson K, Mitchell J, Arnold R, Hodges JR. The Addenbrooke’s Cognitive Examination Revised (ACE-R): a brief

cognitive test battery for dementia screening. Int J Geriatr Psychiatry. 2006;21(11):1078-85.

25. Hsieh S, Schubert S, Hoon C, Mioshi E, Hodges JR. Validation of the Addenbrooke’s Cognitive Examination III in

Frontotemporal dementia and Alzheimer’s disease. Dement Geriatr Cogn Disord. 2013; 36(3-4): 242-50.

26. Foroghan M, Jafari Z, Shirinbayan PF, Barahani ZP, Rahgozar M. [Standardization mini- mental state examination(MMSE) in

elderly in Tehran (Persian)]. J NEW Cog Sci. 2009;10(2):29-37.

27. Yesavage JA, Brink TL, Rose TL, Lum O, Huang V, Adey M, et al. Development and validation of a geriatric depression

screening scale: a preliminary report. J Psyichiatr Res. 1983;17(1):37-49.

28. Spreen O, Strauss E. A compendium of Neuropsychological test. 2th ed. Oxford.1998;113-14.

29. Malakouti k, Fathollah C, Mirabzadeh A, Salavati M, Khany SH. [Standardization of geriatric depression scale(GDS), form of 15

questions in Iranian (Persian)]. Int J Geriatr Psychiatry. 2006; 21(6): 588-93.

30. Rashedi V, Khedmati Morasae E. Depression and cognition state of older adults resorting to day care centers: are they related?.

Europ Geriatr Med. 2014; 83-158.

31. Pouretemad H, Khatibi A, Ganjavi A, Shams J, Zarei M.Validation of Addenbrooke’s Cognitive Examination(ACE) in a

Persian-speaking population. Dementia Geriatr Cog Disord. 2009; 28: 343-7

32. Gillery DW, Wilson RS, Bienias JL, Bennett DA, Evans DA. Predictors of depressive symptoms in persons with AD. J

Gerontology. 2004; 59(4): 75-83.