Research at Banner Sun Health Research Institute  

Learning from our Elders:
Research results

 

The Contribution of ADL/IADL Scores in Calculating a Frailty Index in Community Dwelling Octogenarians, Nonagenarians and Centenarians
Doppalapudi1, M. Ahmadi1,  K. O’Connor1, L. Evans1, J. Ceimo1,  D. Coon2, and W. Nieri1
1BSHRI Center for Healthy Aging, Sun City, AZ; 2College of Nursing and Health Innovation, AZ State University, Phoenix, AZ
 
Background: Frailty is best thought of as a syndrome with multiple components:  medical, functional and psychological. Many frailty indices (FI) exist, some quite long and clinically cumbersome.  Functional assessments (FA) combined with medical parameters have been shown to be predictive of outcomes in hospitalized elderly.  We investigated the impact of FA (ADL/IADL scores) on frailty scores in a cohort of community-dwelling elders > 80 years old.
 
Methods: This cross-sectional study of community-dwelling 80+ year olds is part of a larger longitudinal study on healthy aging (O’Connor et al, 2009).  Participants lived in Arizona, were interviewed in person, and had an MMSE >17.  ADL/IADL scores were derived from initial examination data.  A frailty index (FI) was derived in year two based on 34 variables including medical conditions, FA scores and the CEDS (Ceimo et al, 2012).  FI was calculated using medical criteria with and without both functional and psychological components.  Descriptive and non-parametric statistical analysis was performed using SPSS software and Mann Whitney U.
 
Results:  127 subjects living independently (65% female, mean age 90) were included.  They ranged in age from 80 to 104 years of age. The correlation coefficients between age and basic ADLs (0.385) or IADLs (0.5) were statistically significant (P<0.0001); age, however, did not correlate with frailty (correlation coefficient 0.012).  Correlation coefficients between frailty scores and basic ADL and IADL scores were weak (0.260 and 0.284 respectively) and not statistically significant.
 
Conclusions:  There was a weak and non-significant correlation between frailty and functional status as measured by ADL/IADL scores.  The ADL/IADL scores appear to decrease in concert.  Gender had no significant impact on either IADLs or ADLs in this cohort, nor did age correlate with frailty.  Other investigators (Brandstetter et al) looked at the contribution of psychological factors in the calculation of a frailty index.  Our results suggest that a streamlined FI may yield the same results in this population as the 34-item one that is widely used.  

Depression and Frailty in Community Dwelling Octogenarians, Nonagenarians and Centenarians
K. Brandstetter1, M. Ahmadi1, K. O’Connor1, L. Evans1, J. Ceimo1 D. Coon2, and W. Nieri1
1BSHRI Center for Healthy Aging, Sun City, AZ; 2College of Nursing and Health Innovation, AZ State University, Phoenix, AZ
  
Background:  Depression is not uncommon in the elderly and has been associated with both increased mortality and impaired functional status.  Measures of depression are common components in global estimates of performance such as a frailty index (FI), which also includes functional and medical batteries.    We looked at the association of depression with frailty in a community-dwelling population of elders > 80 years old.
 
 Methods: This cross-sectional study on community dwelling 80+ year olds is part of a larger longitudinal study on healthy aging (O’Connor et al, 2009).  Participants lived in Arizona, were interviewed in person, and had an MMSE >17.  We used the Center for Epidemiologic Studies Depression Scale (CES-D) to screen for depression at the initial interview and at every subsequent annual examination.  A 34-item FI was obtained at the time of the first annual examination (Ceimo et al, 2012).  We then calculated FI using medical criteria with and without both functional and psychological components.  Descriptive and non-parametric statistical analysis was performed using SPSS software and Mann Whitney U.
 
Results:  127 subjects living independently (65% female, mean age 90) were included.  They ranged in age from 80 to 104 years of age.  The correlation coefficient between CESD scores and age was weak (0.127) and not significant (p = 0.16).  Similarly, the correlation between CESD scores and frailty was also weak (0.143) and not significant. 
 
Conclusions:  In this cohort of the oldest old, depression did not appear to be associated with the degree of frailty.  Gender had no significant impact on depression.  We observed a weak and non-significant association between age and depression.   Other investigators (Doppalapudi et al, personal communication) have looked at the contribution of functional parameters in the calculation of a frailty index.  Our results suggest that a streamlined frailty index may yield the same results in this population as the 34-item one that is widely used.

Younger Age Identities in Healthy Octogenarians, Nonagenarians, and Centenarians are Associated with Better Cognition and Work History
[Abstract/poster accepted for the Presidential Poster session at American Geriatrics Society conference in May 2011)Brian W. Leonard1*†, Walter J. Nieri1,2, Kathleen O’Connor1,2, Linda R. Evans2, David W. Coon3 
1Banner Family Medicine Geriatric Fellowship Program, Banner Sun Health Research Institute, Sun City, AZ, 2Center for Healthy Aging, Banner Sun Health Research Institute, Sun City, AZ, 3College of Nursing and Health Innovation, Arizona State University, Phoenix, AZ

Background: Expansion of the average lifespan of U.S. citizens by over 27 years in the last century has stimulated much research on the determinants of longevity. While longevity is undoubtedly associated with particular genetic attributes, psychological and behavioral factors also contribute significantly.

Previous research shows that positive self-perceptions of aging lead to increased longevity, and younger age identities in older individuals are associated with more positive dispositions about one’s cognitive abilities.

Over the past three years, we have anecdotally noted that subjects in our research program on healthy aging (O’Connor, 2009) report feeling much younger than their chronological age. In the present study, we measured age identity in the oldest old (80+ years) and determined for the first time if younger age identities are associated with better performance on the Mini-Mental Status Exam, not just better dispositions about one’s cognition. We were also interested in determining whether current age identity was related to gender, category of prior employment (white vs. blue collar) and other sociodemographic factors.

Methods: This cross-sectional study is part of a larger longitudinal investigation characterizing factors associated with healthy aging. Subjects in this larger study are 50-108 years old, communicate adequately, and obtain a MMSE>17. In the present study, we focused on 80+ year old subjects, as previous studies have looked only at the younger old (i.e., 55-74 years). Medical data as well as sociodemographic, and neurophysical data were collected directly during in-person interviews.

One test we administer is the Mini-Mental Status Exam (MMSE), a widely used instrument that assesses an individual’s orientation, attention, calculation, recall, language and motor skills, and is often used to screen for cognitive decline (Folstein, Folstein, & McHugh, 1975). The age-identity question (i.e., “Many people feel older or younger than they actually are. What age do you feel most of the time?”) was posed either by telephone (N=52) or at the in-person interview (N=73). Descriptive and nonparametric statistical analyses were performed with PASW 17;was set at 0.05.

Results: The sample (n=125; mean age=90.6 years; 64% women) comprised 61 octogenarians, 55 nonagenarians, and 14 centenarians. More than 91% of subjects rated their health as being good or better (i.e., very good, excellent). Age identities ranged from +44 to -83 years, with an average of -16.98 years. A negative age identity indicates feeling younger than one’s chronological age. On average, males reported feeling younger (-18.71 years) than females (-15.96 years), but this did not reach statistical significance. Ceiling effects of MMSE scores preclude a meaningful correlation to be calculated between MMSE score and age identity. However, cognitive performance of our subjects was at a high level (modal MMSE score=29; mean 27.5; 30 points possible), with most (56%) subjects performing in the 90th percentile or better of age norms for the MMSE (Dufouil, et al., 2000). There was no statistically reliable difference in mean age identity among married (N=48), divorced (N=8), and widowed subjects (N=69). Work history emerged as a potential variable that appears to influence age identity in the oldest old individuals. White collar workers (N=96) reported significantly (p=0.048) younger identities than blue collar workers (N=29), averaging -18.48 years vs. -12.00 years for blue collar workers.

Conclusions: Our preliminary data on age identity in the oldest old subjects show that feeling much younger than one’s chronological age is related not only to one’s perceptions about one’s cognitive abilities but also to objectively measured cognitive function, even into the 11th decade of life. The nature of older individuals’ jobs (white vs. blue collar) while they were working appears to modify older individuals’ age identity.

 

Frailty and Body Mass Index in Community Dwelling Octogenarians, Nonagenarians and Centenarians
J. Ceimo1, N. Bravo1, B. Leonard1, T. Minani1, K. O’Connor1, L. Evans1, D. Coon2, and W. Nieri1
1BSHRI Center for Healthy Aging, Sun City, AZ; 2College of Nursing and Health Innovation, AZ State University, Phoenix, AZ

(Submitted to American Geriatrics Society for May 2012 and Arizona Geriatrics Society for October 2012)

Background: Research in frailty has attempted to determine both a consensual diagnosis and its component parts.  Prior studies show frailty associated with extremes of BMI (U-shaped curve). Not everyone with BMIs < 20 or > 30 is frail.  We propose that stable BMI through mid- and late adult life (age 50 and up) may identify a subset > 80 years in whom outlying BMIs were a less significant contributor to frailty.

Methods: This cross-sectional study on community dwelling 80+ year olds is part of a larger longitudinal study on healthy aging (O’Connor et al, 2009).  Participants lived in Arizona, were interviewed in person, and had an MMSE >17.

Height/weight was taken from the initial and second annual exams (year three).  Height/weight data at age 50 was based on participants’ recall.  Body Mass Index (BMI) was divided into four categories: 1) < 20, 2) 20-25, 3) 26-30, and 4) > 30.  Frailty was determined from 34 selected deficits associated with physiologic decline and increased mortality. An individual’s total points divided by total number of deficits yielded a Frailty Index (FI) from 0.0 to 1.0; higher scores indicated greater frailty.  Descriptive and non-parametric statistical analysis was performed using SPSS software and Mann Whitney U.

Results:  202 subjects living independently (62% female, mean age 92) were included.  61 were octogenarians, 69 nonagenarians, and 10 centenarians.  Initial mean BMI ranged from 25.2 in octogenarians to 22.4 in centenarians with no significant drift between years one and three.  Correlated to BMI at age 50, those with <one SD were less likely to be frail or were less frail.  We confirmed prior observations that strength training correlated with decreased frailty across all ages.  A U shaped curve correlating frailty/BMI held true for subjects up to age 89. BMI flattened after age 90, suggesting the impact of BMI on frailty is less.

Conclusions:  Frailty correlates with advancing age; BMI does not.  Much effort is spent to obtain/maintain normal BMIs in the elderly.  Our study identifies a subset of elders with outlying BMIs in whom BMI is not a significant contributor to frailty.  Prolonged activity at a stable BMI may produce physical conditioning that is beneficial.  Avoidance of frailty in this group should stress continued activity over caloric manipulation.

 

MOCA & MMSE Comparison in Detection of Cognitive Impairment and Change in Very Old Age
Ashraf Mohammed, Sara Kabsoun, Brian Leonard, Walter Nieri, Kathleen O’Connor, Linda Evans
Banner Family Medicine Geriatric Fellowship Program, Banner Sun Health Research Institute, Center for Healthy Aging, Sun City, AZ

(Abstract/poster submitted to 2011 Arizona Geriatrics Society)

Background
The Mini Mental State Exam (MMSE) is the most widely used screening test for dementia. However, research shows that it can underestimate cognitive impairment, varies within the population by age & education, and probably has limitations in assessing progressive cognitive decline. The Montreal Cognitive Assessment (MOCA) test may be superior to the MMSE in early detection of Alzheimer’s disease, but little research has been done with it using very old subjects, the most vulnerable group for developing Alzheimer’s disease. We determined the efficacy of the MMSE and MOCA in detecting early signs of cognitive impairment and change in healthy subjects 70-105 years old. We also determined whether level of physical activity was related to cognitive performance.

Methods
Subjects were in the longevity project at the Center for Healthy Aging, BSHRI. They were 70 years old or older with no dementia at enrollment, and had completed baseline, 2nd and 3rd annual assessments in cognitive, medical, and psychosocial domains. One hundred and forty-eight subjects met these criteria and were stratified by age decade (70s, 80s, 90s, and 100s). To assess relative efficacy, MMSE & MOCA scores were plotted for each subject, and the proportion performing at, below, or above standard cutoff scores was determined, respectively, for the MMSE & MoCA.  To assess decline over 1-year, difference scores were calculated separately for the MMSE & MoCA. Level of activity was measured with the Rapid Assessment of Physical Activity (RAPA), and scores were correlated with MoCA performance.

Results
Fair number of healthy 70, 80, 90 and 100 year-old subjects performed above the cutoffs on both the MMSE & MoCA. With increasing age, however, an increasing proportion performed below the MoCA cutoff while scoring above the cutoff for the MMSE, supporting previous findings of superiority of the MoCA over the MMSE in the younger old. In the same groups, 1-year change was negligible for the MMSE, while for the MoCA, scores were on average 1.3 points lower.  Subjects who were more physically active scored higher on the MoCA, but only for 90- and 100-year olds.

Conclusion
This study suggests that the MoCA is a better instrument for detecting cognitive impairment compared to MMSE, even in the oldest old. The MoCA can be a good tool for monitoring cognitive changes over time, and thus can aid care planning in early dementia. Physical activity may be associated with better cognitive function in the very old.  Few subjects scored highly on both tests, which can be an area of future research to look for protective factors against development of Alzheimer’s Dementia.

Banner Sun Health Research Institute
10515 W. Santa Fe Drive
Sun City, AZ 85351
(623) 832-6500
 
 
 
Jump to top links