Skip to main content
NCBI home page
The Gerontologist logoLink to The Gerontologist
. 2010 Oct 25;51(2):179–189. doi: 10.1093/geront/gnq087

Regression-Based Estimates of Observed Functional Status in Centenarians

Meghan B Mitchell 1,*, L Stephen Miller 1, John L Woodard 2, Adam Davey 3, Peter Martin 4, Molly Burgess 1, Leonard W Poon 1; for the Georgia Centenarian Study†
PMCID: PMC3058129  PMID: 20974657

Abstract

Purpose of the Study: There is lack of consensus on the best method of functional assessment, and there is a paucity of studies on daily functioning in centenarians. We sought to compare associations between performance-based, self-report, and proxy report of functional status in centenarians. We expected the strongest relationships between proxy reports and observed performance of basic activities of daily living (BADLs) and instrumental activities of daily living (IADLs). We hypothesized that the discrepancy between self-report and observed daily functioning would be modified by cognitive status. We additionally sought to provide clinicians with estimates of centenarians’ observed daily functioning based on their mental status in combination with subjective measures of activities of daily living (ADLs). Design and Methods: Two hundred and forty-four centenarians from the Georgia Centenarian Study were included in this cross-sectional population-based study. Measures included the Direct Assessment of Functional Status, self-report and proxy report of functional status, and the Mini-Mental State Examination (MMSE). Results: Associations between observed and proxy reports were stronger than between observed and self-report across BADL and IADL measures. A significant MMSE by type of report interaction was found, indicating that lower MMSE performance is associated with a greater discrepancy between subjective and objective ADL measures. Implications: Results demonstrate associations between 3 methods of assessing functional status and suggest proxy reports are generally more accurate than self-report measures. Cognitive status accounted for some of the discrepancy between observed and self-reports, and we provide clinicians with tables to estimate centenarians’ performance on observed functional measures based on MMSE and subjective report of functional status.

Keywords: Daily functioning, Direct Assessment of Functional Status, Self-report, Proxy report, MMSE

Ability to execute basic activities of daily living (BADLs; e.g., dressing, bathing; Katz, Ford, Moskowitz, Jackson, & Jaffe, 1963) and instrumental activities of daily living (IADLs; e.g., grocery shopping, financial management; Lawton & Brody, 1979) and cognitive impairment are intimately linked (American Psychiatric Association, 2000; Tabert et al., 2002), and impairment of these abilities brings increased risk of outcomes, such as malnutrition, falls, automobile accidents, and prescription medication noncompliance (Edelberg, Shallenberger, & Wei, 1999; Hall, 1997). Functional impairment is a necessary criterion for dementia diagnosis (American Psychiatric Association, 2000), and older adults with unimpaired cognition or mild cognitive impairment can also experience functional decline (Tabert et al., 2002). In addition to the importance of maintaining older adults’ physical safety and health when functional impairment becomes an issue, emotional well-being is also relevant. Older adults commonly link a loss of independence to a loss of personal well-being (Salmoni, Viverais-Dressler, Porter, & Garg, 1997), and those who live in assisted living facilities tend to have smaller social networks and spend more time indoors relative to independently living older adults (Broese van Groenou & Thomese, 1996). Thus, early detection of functional impairment is important to provide older adults with services necessary for maintaining their physical and emotional well-being and their independence in the community.

There is a lack of consensus on assessment, content, and factor structure of BADLs and IADLs (Glass, 1998; Spector & Fleishman, 1998; Thomas, Rockwood, & McDowell, 1998). Studies of healthy geriatric samples and older adults with dementia typically find a hierarchical relationship between BADLs and IADLs, with IADLs affected prior to BADLs (Njegovan, Man-Son-Hing, Mitchell, & Molnar, 2001; Spector, Katz, Murphy, & Fulton, 1987; Stuck et al., 1999). Self-report and proxy report measures can provide large amounts of information in a short time, but they are susceptible to reporter biases and error (Lum, Lin, & Kane, 2005). Self-reporters generally minimize deficits and proxy reporters generally magnify deficits (Magaziner, Zimmerman, GruberBaldini, Hebel, & Fox, 1997; Miller, Mitchell, Woodard, & Poon, 2008; Rubenstein, Schairer, Wieland, & Kane, 1984) with self-report bias attributed to factors including cognitive decline and lack of insight (Mitchell & Miller, 2008; Royall et al., 2007) and proxy bias attributed to factors including caregiver burden, depression, and limitations in the proxy’s exposure to observing performance on specific functional tasks (Loewenstein et al., 2001; Zanetti, Geroldi, Frisoni, Bianchetti, & Trabucchi, 1999).

Performance-based measures of BADLs and IADLs have greater face validity than report-based measures, as older adults are evaluated during their actual performance on daily living tasks (Myers, Holliday, Harvey, & Hutchinson, 1993). Reliability of these measures is increased relative to questionnaire measures because several components underlying each functional task are assessed, while questionnaires typically only ask one question for each functional task (Sherman & Reuben, 1998). However, performance-based functional measures have also been critiqued for having limited sensitivity to small decrements in functional ability in community-based samples (Curb et al., 2006). Among clinicians, however, the main critique of performance-based measures is that they are often time consuming. Report-based measures should thus be preferred when they can be completed accurately, but it is difficult to know in advance just how accurate a given respondent might be (Myers et al., 1993; Royall et al., 2007). Few studies have investigated all three methods of assessment (i.e., self, proxy, and performance-based measures) across identical domains (Curb et al., 2006; Dorevitch et al., 1992; Loewenstein et al., 2001; Sherman & Reuben, 1998).

Most research on functional status in older adults has assessed performance across broad age ranges (Loewenstein et al., 2001; Spector et al., 1987). However, few studies have investigated multimethod measurement of functional status in centenarians. As age is a major risk factor for functional decline, centenarians are particularly vulnerable to detrimental outcomes associated with functional decline (Poon et al., 2007). We evaluated relationships among self, proxy, and performance-based functional measures within centenarians. We chose measures that contained identical BADL and IADL tasks to directly compare three methods of assessment (observed, self-report, and proxy) while minimizing error variance due to uncorrelated functional items. We chose functional measures that we anticipated were likely to be actively performed by centenarians. We chose to include detailed assessment of several specific BADLs and IADLs at the cost of excluding some functional tasks. We hypothesized that proxy report would show stronger associations with observed performance relative to self-report. We additionally hypothesized that decrements in global cognitive functioning, as measured by the Mini-Mental State Examination (MMSE; Folstein, Folstein, McHugh, & Fanjiang, 2000), would moderate the relationship between subjective reports and observed performance, with greater cognitive impairment being associated with a greater discrepancy between functional measures. Given the difficulties with time constraints that clinicians face, we sought to use these results to create a clinical tool which enables clinicians to use MMSE and self-report or proxy report of activities of daily living (ADLs) to estimate how well they would perform on an observed measure of functional status.

Design and Methods

Participants

This study used population-based recruitment strategies to obtain a representative sample of centenarians in Georgia. Participants were 244 centenarians or near-centenarians drawn from 44 counties in Northeast Georgia using two population-based recruitment strategies (Poon et al., 2007). First, a list of all personal care homes and skilled nursing facilities (“facilities”) in the 44 county regions was compiled, and each facility was contacted by interviewers to recruit participants aged 98 years and older. In the second strategy, a list of all registered voters aged 98 years and older in the 44 county regions was compiled, and eligible participants were contacted by telephone. Participants were only enrolled if they agreed to provide blood samples, and no other exclusionary criteria were used.

Consent was obtained in a joint consent process, in which both the centenarian and caregiver provided their informed consent. We required that both centenarians and their legal proxy agree to participation in order to minimize the risk of obtaining consent from centenarians with limited decision-making capacity, while also maintaining participants’ sense of autonomy and choice in the consenting process. This study was approved by the University of Georgia Institutional Review Board. Participants were assessed in their residence over four to five 2-hr sessions, which were completed within thirty days. Physical and mental health, cognitive, functional, genetic, nutrition, resources, and personality data were collected as part of the larger study. Participants were compensated up to $600 depending on number of completed testing sessions. Self-report and proxy report of functional ability were generally assessed in a separate session prior to the session in which the performance-based measure was conducted. Eighty-eight percent of participants completed the self-report first, 5% completed the performance-based measure first, and 5% completed them both on the same day. Mean days between completion of self-report and performance-based assessment was two days (SD = 6.2). The sample sizes for each analysis described below vary somewhat due to missing data for some measures used.

Average age of centenarians was 100.6 years (SD = 2.04), and average education was 10.6 years (SD = 3.78). Centenarians were predominantly female (n = 207, 85%), Caucasian (n = 192, 79%), and 37% were living independently (n = 91).

Measures

Functional status was assessed with a modified version of the Direct Assessment of Functional Status (DAFS), a laboratory assessment of BADLs and IADLs (Loewenstein et al., 1989). The DAFS is a clinician-rated scale based on performance on time orientation, communication, transportation, preparing for grocery shopping, financial skills, grocery shopping, dressing and grooming, and eating. Transportation, preparing for grocery shopping, and grocery shopping tasks of the DAFS were omitted due to increased physical demands and a low likelihood that centenarians were currently engaged in these activities (see Figure 1). Each ADL task on the DAFS was scored on a dichotomous scale based on the participant’s successful completion of the functional task as follows: Dressing/Grooming—participants were asked to demonstrate picking up a toothbrush, taking cap off toothpaste, putting toothpaste on brush, brushing teeth, turning on water, using soap, washing face, turning off water, putting on coat, buttoning coat, zipping a garment, tying a shoe, and brushing hair (0 or 1 for each item, possible range = 0–13); Eating—participants were asked to identify a fork, knife, spoon, pour water from pitcher, and drink from glass (0 or 2 for each item, possible range = 0–10); Time Orientation—participants were asked to tell the time at four settings on an analogue clock and to state the correct date, day, month, and year (0 or 2 for each item, possible range = 0–16); Communication—participants were rated on their ability to use a large-button touchtone telephone to dial the operator, dial a number from telephone book, dial a number presented orally, dial a number written down in large print, pick up the receiver, dial, hang up the telephone, and perform the correct sequence of actions across the above listed telephone operations. They were additionally asked to prepare a letter for mailing and were rated on their ability to fold the letter, put it in an envelope, seal the envelope, stamp the envelope, write the correct mailing address, and the correct return address (0 or 1 for each item, range = 0–14 points); Financial Skills—participants were asked to identify seven different types of currency (0 or 1 for each item, range = 0–7), count the correct amount of currency when provided with currency amounts by the examiner across four test trials of increasing difficulty (0 or 2 for each item, range = 0–8), make the correct change in simulated shopping encounters in which the participant is asked provide the examiner with correct change for a specific currency amount across four test trials of increasing difficulty (0 or 2 for each item, range = 0–8), write a check and write the correct signature, pay to the order of, written amount, numeric amount, and date (0 or 1 for each item, range = 0–4), and correctly balance a checkbook across four test trials (0 or 1 for each item, range = 0–4; range for total financial skills = 0–28).

Figure 1.

Figure 1.

Open in a new tab

Domains and skill requirements for the modified Direct Assessment of Functional Status and corresponding self and proxy questions. Note: BADL = basic activities of daily living; IADL = instrumental activities of daily living. Self-report and proxy questions were answered with a score of 2 (without any help), 1 (with some help), or 0 (completely unable to do so alone).

The BADL score was calculated by summing the grooming, dressing, and eating scales (possible range = 0–23 points, higher scores represent higher functional status); the IADL score was calculated by summing the time orientation, communication, and financial skills scales (possible range = 0–58 points, higher scores represent higher functional status). Reliability analysis revealed that both BADL and IADL subscales of the modified DAFS demonstrated strong internal consistency (DAFS BADL Chronbach’s α = .926; IADL = .911). The DAFS has been validated in older adults with Alzheimer’s disease (Loewenstein et al., 1989) and in functionally independent samples of older adults (McDougall et al., 2010). Although typically administered in an outpatient medical setting (Loewenstein et al., 2001), the measure has been demonstrated to be easily adapted to home-based assessments in previous studies, the samples of which included community-dwelling and institutionalized older adults (Lewis & Miller, 2007; Mitchell & Miller, 2008). The DAFS has been demonstrated to have interrater reliabilities ranging from .91 to 1.0 and test–retest reliabilities ranging from .55 to .92 across subscales (Loewenstein et al., 1989).

In creating a functional assessment that was appropriate for centenarians, we chose to include detailed assessment of a selection of daily living tasks instead of including an exhaustive assessment of all ADLs. This approach was chosen in order to increase reliability of our functional measures, as we had several test items assessing each functional task instead of including many daily living activities with only a single test item per task. We chose specific functional tasks based on our judgment that centenarians would be actively participating in the activity in their daily lives. We additionally excluded tasks from our functional assessment that were judged to cause undue physical burden on centenarians or on which we judged performance to be likely influenced by physical or sensory limitations. Although we recognize the importance of assessing BADLs that are heavily dependent on physical functioning, our focus in this article is on daily tasks with a cognitive component that centenarians were likely to be able to perform, in spite of prevalent sensory and physical limitations.

To directly compare DAFS items with self-report and proxy report measures, we developed a modified version of the Older American Resources and Services (OARS) Activities of Daily Living Scale (Duke University, Center for the Study of Aging and Human Development, 1978) to include only those functional tasks assessed on the DAFS (see Table 1 for a summary of each daily activity assessed on functional measures). Each question was rated in separate interviews by the centenarian and the proxy using a score of 2 (without any help), 1 (with some help), or 0 (completely unable to do so alone). The theoretical range for OARS BADL items (i.e., grooming, dressing, and eating items) was 0–8 and for IADL items (i.e., time orientation, communication, and financial skill items) was 0–16, with higher scores representing higher functional status. We conducted reliability analysis on modified OARS BADL and IADL subscales, and all were internally consistent (Cronbach’s α range = .871–.930).

Table 1.

Descriptive Statisticsa

Variable n Proportion or mean value
Race (Black/White) 52/192 21%/79%
Gender (male/female) 37/207 15%/85%
Age in years 244 100.6
Years education 237 10.6
MMSE total score 244 16.2
DAFS BADL 228 16.5
DAFS IADL 228 26.0
Self-report BADL 221 6.8
Self-report IADL 208 12.0
Proxy report BADL 231 6.0
Proxy report IADL 223 9.2
Open in a new tab
a

MMSE = Mini-Mental State Examination, theoretical range = 0–30. BADL = basic activities of daily living; DAFS = Direct Assessment of Functional Status; IADL = instrumental activities of daily living. DAFS BADL theoretical range = 0–23, with higher scores representing higher functional status. DAFS IADL theoretical range = 0–58, with higher scores representing higher functional status. Self-report/proxy report BADL theoretical range = 0–8, with higher scores representing higher functional status. Self-report/proxy report IADL theoretical range = 0–16, with higher scores representing higher functional status.

The MMSE (Folstein et al., 2000) is a gross measure of cognitive status that generates a score ranging from 0 to 30. The MMSE is used to assess orientation to time and place, immediate and short-delayed recall of words, attention, calculation, language, and visual construction. All measures were administered by trained research assistants.

Data Analysis

Data were analyzed using SPSS 16.0 (SPSS Inc., Chicago, IL). Correlation and regression analyses tested our hypothesis that proxy report would show stronger relationships with observed BADL and IADL performance than self-report. Because the distribution of scores on functional measures was not normally distributed, we used Spearman’s rank-order correlations to directly compare relationships between all three functional measures. We used a z test procedure recommended by DeCoster (2007) to compare dependent correlation coefficients between functional measures. This procedure compares two measures’ (i.e., self-report vs. proxy report) correlation with a third measure (observed performance), while accounting for the shared variance between the two measures (self-report and proxy report). We additionally computed 95% confidence intervals (CIs) for correlation coefficients in order to determine if coefficient CIs were nonoverlapping. Regression analysis was used to test our hypothesis that decreased cognitive functioning would account for the discrepancy between self-report and performance-based functional measures. In all regression models, we entered living arrangement (coded as 0 = community dwelling and 1 = facility) in the first step in order to ensure that a subset of functionally dependent more cognitively impaired individuals were not systematically driving the effect of our observed relationships between independent and dependent variables.

Results

Descriptive Statistics and Missing Data Analysis

Descriptive statistics for the sample are provided in Table 1. The total sample size was 244 centenarians and near-centenarians; 91% had complete self-report BADL data (n = 221), 85% had complete self-report IADL data (n = 208), 95% had complete proxy report BADL data (n = 231), and 91% had complete proxy report IADL data. The DAFS was successfully completed for 93% of the sample (n = 228) for both BADL and IADL scales. We conducted a missing data analysis using independent samples t tests for continuous variables and Fisher’s exact test for categorical variables to determine whether the probability of being a “completer” on each functional measure was associated with demographic or gross cognitive differences. There were no significant differences in age, gender, race, or MMSE score between completers and noncompleters of the DAFS. There were similarly no group differences in age, gender, or race between completers and noncompleters of self-report and proxy report measures. Centenarians with low MMSE scores were more likely to have incomplete self-report and proxy report data for BADLs and IADLs.

Hypothesis Testing

We conducted correlation analyses to test the hypothesis that the relationship between proxy report and observed DAFS performance would be stronger than that between self-report and observed DAFS performance (Table 2). To directly compare all three functional measures, Spearman rank-order correlations were compared. The correlation between proxy report and observed BADL performance was significant, rs(216) = .72, p < .001, as was the correlation between self-report and observed BADL performance, rs(209) = .51, p < .01. The proxy–observed correlation was significantly stronger than the self–observed correlation (z = −4.64, p < .001) using a z test of dependent correlations (DeCoster, 2007). Similarly, although both the correlation between observed IADL performance and proxy report, rs(210) = .86, p < .001, and the correlation between self-report and observed IADL performance, rs(199) = .53, p < .001, were both significant, the proxy–observed correlation was stronger than the self–observed correlation (z = −8.19, p < .001). As is demonstrated in Table 2, 95% CIs for self-report and observed performance for BADL and IADL measures were not overlapping, further demonstrating the discrepancy between reporters and observed performance.

Table 2.

Spearman’s Rho Correlations between DAFS, Self-report, and Proxy Reporta

Self-report 95% CI
 Proxy report 95% CI
Lower bound Upper bound Lower bound Upper bound
DAFS BADL 0.508 0.401 0.601 0.716 0.645 0.775
DAFS IADL 0.534 0.428 0.626 0.857 0.817 0.889
DAFS domains
Eating 0.461 0.351 0.558 0.462 0.355 0.557
Dressing/grooming 0.502 0.394 0.596 0.719 0.649 0.777
Time orientation 0.485 0.378 0.579 0.732 0.667 0.786
Communication 0.352 0.229 0.464 0.745 0.682 0.797
Financial skills 0.548 0.446 0.636 0.781 0.723 0.828
Open in a new tab
a

BADL = basic activities of daily living; CI = confidence interval; DAFS = Direct Assessment of Functional Status; IADL = instrumental activities of daily living. DAFS BADL theoretical range = 0–23, with higher scores representing higher functional status. DAFS IADL theoretical range = 0–58, with higher scores representing higher functional status. Self-report/proxy report BADL theoretical range = 0–8, with higher scores representing higher functional status. Self-report/proxy report IADL theoretical range = 0–16, with higher scores representing higher functional status. Bold font indicates nonoverlapping CI.

To test our hypothesis that cognitive impairment would moderate the relationship between subjective reports and observed functional status, we conducted a series of four regressions, entering living arrangement, subjective report, MMSE score, and an MMSE × Subjective Report interaction term to estimate DAFS performance. Each of the four models differed in which subjective measure of functional status was entered as an independent variable (i.e., Model 1 = self-report of BADL, Model 2 = proxy report of BADL, Model 3 = self-report of IADL, Model 4 = proxy report of IADL; Table 3). The self-report BADL model (Model 1) accounted for approximately 57% of the variance in observed BADL performance, adjusted R2 = .57, F (4, 206) = 71.29, p < .001, whereas the proxy report BADL model (Model 2) accounted for approximately 64% of the variance in observed BADLs, adjusted R2 = .64, F (4, 213) = 97.92, p < .001. The self-report IADL model (Model 3) accounted for approximately 77% of the variance in observed IADLS, adjusted R2 = .77, F (4, 196) = 171.67, p < .001, and the proxy report IADL model (Model 4) accounted for 84% of the variance in observed IADLS, adjusted R2 = .83, F (4, 207) = 267.59, p < .001. Although self-report and MMSE each accounted for unique variance in Model 1, the Self-report × MMSE interaction term in this model was not significant (Table 3). However, in Models 2, 3, and 4, the Subjective Report × MMSE interaction term was significant, indicating that the relationship between subjective report on these measures and actual performance on functional tasks is moderated by level of cognitive impairment. The largest interaction term was demonstrated in Model 3, where the IADL Self-report × MMSE interaction term accounted for the most variance out of all variables in the model, t(196) = 6.168, β = .696, p < .001, indicating that the relationship between self-report and actual performance on IADLS is the most influenced by decrements in cognitive status. To better illustrate how MMSE moderates the relationship between subjective reports and observed performance on functional measures, we used our models to calculate regression-based estimates of observed functional ability based on MMSE scores and subjective report. By entering each possible combination of scores from the modified OARS and MMSE into our regression equations, we created tables of estimated scores on the modified DAFS. Because clinicians are unlikely to be familiar with the raw score ranges on our modified version of the DAFS BADL and IADL subscales, we converted our tables of regression-derived DAFS values to cumulative percentiles, based on the cumulative prevalence of each DAFS score in our centenarian sample (Tables 4 and 5). These tables additionally provide geriatric service providers with a clinical tool for estimating how centenarians would perform on observed functional measures based on the MMSE and subjective report of functional status. As an example, if a centenarian obtained a 10 on the MMSE and provided a self-reported IADL score of 6, their regression-derived DAFS IADL score would be at a cumulative percentile of 26, indicating that when compared with a population-based sample of centenarians, 26% performed at or below their estimated level of functional ability on IADL tasks (Table 5).

Table 3.

Regression Models Predicting Observed Functional Statusa

Variable Model 1: self-report BADL
 Model 2: proxy report BADL
Coefficient β p Coefficient β p
Living arrangement −0.064 −.004 .927 −0.740 −.045 .299
Self-report or proxy report BADL 1.364 .424 <.001 1.821 .527 <.001
MMSE total score 0.620 .693 <.001 0.824 .897 <.001
MMSE × Self-report or Proxy Report BADL −0.024 −.232 .176 −0.060 −.560 <.001
Model 3: Self-report IADL
 Model 4: proxy report IADL
Coefficient β p Coefficient β p
Living arrangement −1.583 −.043 .219 −1.689 −.045 .137
Self-report or proxy report IADL −0.436 −.118 .043 0.526 .150 .046
MMSE total score 0.599 .273 .003 0.588 .278 <.001
MMSE × Self-report or Proxy Report IADL 0.089 .696 <.001 0.061 .504 <.001
Open in a new tab
a

BADL = basic activities of daily living; DAFS = Direct Assessment of Functional Status; MMSE = Mini-Mental State Exam; β = standardized regression coefficient, Living arrangement was coded as 0 = community dwelling, 1 = residing in a facility.

Table 4.

Cumulative Percentiles for Observed DAFS Performance Predicted by MMSE and Subjective Reports of BADLSa

MMSE Self-report of BADL
 Proxy report of BADL
0 1 2 3 4 5 6 7 8 0 1 2 3 4 5 6 7 8
0 14 15 17 19 20 20 21 21 23 11 14 17 17 20 20 21 23 26
1 14 17 17 19 20 20 21 22 23 11 15 17 19 20 21 21 23 26
2 15 17 19 20 20 21 21 23 24 14 15 17 20 20 21 22 24 26
3 15 17 19 20 20 21 22 24 26 14 17 19 20 20 21 22 24 28
4 17 19 20 20 21 21 23 24 28 15 17 19 20 21 21 23 24 28
5 17 19 20 20 21 22 24 26 31 17 19 20 20 21 22 23 26 28
6 17 20 20 21 21 23 24 28 31 17 19 20 20 21 22 24 26 31
7 19 20 20 21 22 24 26 31 33 19 20 20 21 21 23 24 28 31
8 20 20 21 21 23 24 28 33 35 20 20 20 21 22 23 26 28 31
9 20 20 21 22 24 26 31 33 43 20 20 21 21 22 24 26 28 33
10 20 20 21 23 24 28 31 35 49 20 20 21 22 23 24 28 31 33
11 20 21 22 23 26 31 33 43 61 20 21 21 22 24 26 28 31 33
12 20 21 22 24 28 31 35 49 61 21 21 22 23 24 26 28 33 35
13 21 21 23 26 28 33 43 49 82 21 22 23 24 26 28 31 33 35
14 21 22 24 26 31 35 49 61 100 21 22 23 24 26 28 31 33 35
15 21 23 24 28 33 43 49 82 — 22 23 24 26 28 31 33 35 43
16 22 23 26 31 35 43 61 100 — 23 24 26 28 31 31 33 35 43
17 22 24 28 33 35 49 82 — — 24 26 28 28 31 33 35 43 43
18 23 26 28 33 43 61 100 — — 26 28 28 31 33 33 35 43 49
19 23 26 31 35 49 82 — — — 28 28 31 33 33 35 43 43 49
20 24 28 33 43 49 82 — — — 28 31 33 33 35 43 43 49 49
21 26 31 33 43 61 100 — — — 31 33 33 35 35 43 49 49 61
22 26 31 35 49 82 — — — — 33 35 35 43 43 49 49 61 61
23 28 33 43 61 100 — — — — 35 35 43 43 49 49 61 61 61
24 31 33 43 61 — — — — — 43 43 43 49 49 61 61 61 82
25 31 35 49 82 — — — — — 49 49 49 49 61 61 61 82 82
26 33 43 61 100 — — — — — 49 61 61 61 61 82 82 82 82
27 33 43 61 — — — — — — 61 61 82 82 82 82 82 100 100
28 35 49 82 — — — — — — 82 82 82 82 100 100 100 — —
29 43 61 100 — — — — — — 100 100 100 100 — — — — —
30 43 61 — — — — — — — — — — — — — — — —
Open in a new tab
a

BADL = basic activities of daily living, DAFS = Direct Assessment of Functional Status. DAFS BADL theoretical range = 0–23, with higher scores representing higher functional status. Self-report/proxy report BADL theoretical range = 0–8, with higher scores representing higher functional status.

Table 5.

Cumulative Percentiles for Observed DAFS Performance Predicted by MMSE and Subjective Report of IADLSa

MMSE Self-report of IADL
 Proxy report of IADL
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
0 14 13 13 12 12 12 12 12 12 12 12 12 12 12 12 12 12 12 13 13 14 15 15 18 18 19 19 20 20 22 22 24 24 25
1 14 14 14 13 13 13 12 12 12 12 12 12 12 12 12 12 12 13 14 14 15 15 18 19 19 20 20 22 22 24 25 25 26 26
2 15 15 14 14 14 14 13 13 13 13 12 12 12 12 12 12 12 14 14 15 18 18 19 19 20 22 22 24 25 25 26 29 29 30
3 15 15 15 15 15 15 14 14 14 14 14 14 13 13 13 13 13 14 15 18 18 19 20 20 22 24 25 25 26 29 29 30 31 33
4 18 18 18 18 18 18 18 18 15 15 15 15 15 15 15 15 15 15 18 18 19 20 22 22 24 25 26 26 29 30 31 33 33 34
5 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 15 18 19 20 22 22 24 25 26 29 30 30 31 33 34 35 39
6 19 19 19 20 20 20 20 20 20 20 20 20 20 22 22 22 22 18 19 20 22 22 24 25 26 29 30 31 33 34 35 35 39 40
7 20 20 20 20 22 22 22 22 22 24 24 24 24 24 25 25 25 19 19 20 22 24 25 26 29 30 31 33 34 35 39 40 41 42
8 20 22 22 22 24 24 24 24 25 25 25 26 26 26 26 29 29 19 20 22 24 25 26 29 30 31 33 34 35 39 40 41 42 43
9 22 22 24 24 25 25 25 26 26 26 29 29 29 30 30 31 31 20 22 24 25 26 29 30 31 33 34 35 40 41 42 43 44 47
10 24 24 25 25 25 26 26 29 29 30 30 31 31 33 33 34 34 20 22 25 26 29 30 31 33 34 39 40 41 42 43 44 47 47
11 24 25 25 26 26 29 30 30 31 31 33 33 34 34 35 35 39 22 24 25 26 30 31 33 34 35 40 41 42 43 44 47 50 52
12 25 25 26 29 29 30 31 31 33 34 34 35 39 39 40 40 41 22 25 26 29 30 33 34 35 40 41 42 43 47 47 50 52 54
13 25 26 29 30 30 31 33 34 34 35 39 39 40 41 42 42 43 24 25 29 30 31 34 35 39 41 42 43 47 47 50 53 54 54
14 26 29 30 31 31 33 34 35 39 39 40 41 42 43 43 44 47 25 26 29 31 33 35 39 40 42 43 44 47 50 53 54 54 59
15 29 30 30 31 33 34 35 39 40 41 42 43 44 44 47 47 50 25 29 30 33 34 35 40 41 43 44 47 50 53 54 54 59 61
16 29 30 31 33 34 35 39 40 41 42 43 44 47 47 50 52 53 26 29 31 33 35 39 41 42 44 47 50 52 54 54 59 61 66
17 30 31 33 34 35 39 40 41 43 44 47 47 50 52 53 54 54 26 30 33 34 39 40 42 43 47 47 52 54 54 59 61 66 68
18 30 33 34 35 39 40 41 43 44 47 47 50 53 54 54 56 59 29 31 33 35 39 41 43 44 47 52 53 54 56 61 66 68 69
19 31 33 35 39 40 41 43 44 47 47 52 53 54 54 59 61 64 30 31 34 39 40 42 44 47 50 53 54 56 61 64 68 69 72
20 33 34 35 40 41 42 44 47 47 52 53 54 56 59 64 66 68 30 33 35 39 41 43 47 47 52 54 56 59 64 68 69 71 74
21 33 35 39 41 42 43 47 47 52 53 54 56 61 64 66 68 69 31 34 35 40 42 44 47 50 53 54 59 64 66 68 71 74 76
22 34 35 40 41 43 47 47 52 53 54 56 61 64 68 68 71 72 31 34 39 41 43 47 50 52 54 56 61 66 68 71 72 75 77
23 34 39 41 42 44 47 50 53 54 56 61 64 68 69 71 74 75 33 35 40 42 44 47 50 53 54 59 64 68 69 72 75 77 80
24 35 40 41 43 47 50 52 54 56 59 64 68 69 71 74 76 77 34 39 40 42 44 47 52 54 56 61 66 68 71 74 76 78 82
25 39 40 42 44 47 52 53 54 59 64 68 68 71 74 76 78 80 34 39 41 43 47 50 53 54 59 66 68 71 74 76 78 82 84
26 39 41 43 47 50 53 54 56 61 66 68 71 74 76 78 82 83 35 40 42 44 47 52 54 56 64 68 69 72 75 77 80 83 86
27 40 42 44 47 52 54 56 61 66 68 71 74 76 78 82 84 86 35 40 43 47 50 53 54 59 66 68 71 74 76 80 83 86 89
28 40 43 47 50 53 54 59 64 68 69 72 75 77 80 83 86 89 39 41 43 47 52 54 56 61 68 69 72 75 78 82 84 86 91
29 41 43 47 52 54 56 61 66 68 71 75 77 80 83 86 89 93 39 42 44 47 53 54 59 66 68 71 74 77 80 83 86 89 93
30 42 44 — — — 59 64 68 71 74 76 78 83 86 89 91 98 40 42 47 50 — 56 61 68 69 72 76 78 82 86 89 91 98
Open in a new tab
a

DAFS = Direct Assessment of Functional Status; IADL = instrumental activities of daily living. DAFS IADL theoretical range = 0–58, with higher scores representing higher functional status. Self-report/proxy report IADL theoretical range = 0–16, with higher scores representing higher functional status.

Discussion

As is demonstrated in Table 2, if a performance-based measure is used as a reference for comparison, proxy report shows stronger correlations with observed performance than with self-report across all functional tasks in centenarians. We further demonstrate in our regression models that the discrepancy between subjective reports and observed performance on functional tasks is moderated by gross cognitive status (Table 3), with increased cognitive impairment being associated with a greater discrepancy between subjective report and observed performance, particularly when self-reports of IADLs are evaluated. This finding suggests that those functional tasks with greater cognitive demands (i.e., IADLs) are also less likely to be reported as being problematic by centenarians, particularly when they are experiencing cognitive impairment. This finding is consistent with previous literature, suggesting greater cognitive demands for IADLs in comparison to BADLs (Royall et al., 2007) and implies that self-reports of IADLs are particularly susceptible to reporter biases and should be evaluated by clinicians in combination with other sources of information. Our results provide clinicians with a time-effective method for assessing functional ability that does not require testing that is outside the scope of what is typically administered in outpatient and inpatient geriatric settings.

Consistent with previous findings in relatively younger geriatric samples, our findings demonstrate that while self-report is less consistent with actual performance on functional measures, proxy report is highly consistent (Magaziner et al., 1997; Miller et al., 2008). Previous studies have demonstrated an association between cognitive decline and decrements in both physical and cognitive components of functional tasks (Tabbarah, Crimmins, & Seeman, 2002; Williamson et al., 2009). Our results suggest that cognitive decline accounts for a portion of the minimization of functional impairment on self-report, which has also been demonstrated with younger geriatric samples (Royall et al., 2007). Previous comparisons of different methods of functional status typically use dissimilar subjective and objective measures (Kempen, Steverink, Ormel, & Deeg, 1996). A strength of our study is that we compared the relationships between all three types of functional status measures using measures that contained identical BADL and IADL items, enabling us to directly compare the three methods of assessment.

In terms of clinical application of our findings, we maintain that direct observation of performance on daily tasks is the method of assessment that is least susceptible to reporter biases and will likely provide clinicians with the most accurate assessment of functional abilities. Nonetheless, clinicians are regularly asked to make judgments about older adults’ functional status with limited time for assessment. In these situations, completion of the short functional questionnaire used in our study, either by the centenarian or by a caregiver proxy reporter, in combination with administration of the MMSE could be used to make an estimate of centenarian performance on an observation-based functional measure. Brief functional questionnaires and the MMSE are among the most commonly administered assessment tools in clinical care of geriatric patients and take a combined administration time of approximately 20 min (Duke University, Center for the Study of Aging and Human Development, 1978; Folstein et al., 2000; Pezzotti, Scalmana, Mastromattei, & Di Lallo, 2008). This brief assessment has significant advantages in that it remains quite brief yet is potentially a more valid referent of actual performance ability than a cognitive or functional screening measure used alone. Although a 20-min test administration time is perhaps longer than the time it would take to perform an unstructured clinical interview with a patient, the use of standardized measures are essential for reliable detection of early changes in cognition and function.

Limitations

Limitations of our study include the omission of some important functional abilities on our measure of observed daily functioning (e.g., walking, transferring, toileting). Similarly, the DAFS includes items measuring ability to keep track of time, which is a slightly less commonly evaluated IADL measure in the gerontology literature. This study was population based and did not include classification of centenarians based on diagnosis of dementia and thus cannot address the issue of diagnostic accuracy of the functional measures used. Another aspect of this study is its use of an observed functional measure as a reference for comparing self-report and proxy report. Some have argued that observed functional measures have questionable reliability (Tager, Swanson, & Satariano, 1998) or are measuring a different construct than self-report and proxy report measures and thus the three cannot be directly compared (Myers et al., 1993). We maintain that direct observation of functional tasks is the most ecologically valid of available methods for assessing daily functioning and is the least likely to be susceptible to reporter biases.

Future Directions

There are many questions about functional ability in centenarians that remain to be answered. Although we showed in this study that MMSE performance accounted for part of the discrepancy between self-report and observed performance on BADL and IADL measures, the MMSE is a global measure of cognitive status that has been critiqued for its lack of sensitivity and specificity (Kukull et al., 1994; Lopez, Charter, Mostafavi, Nibut, & Smith, 2005). Future investigations exploring component processes of cognition and their unique relationships with different BADL and IADL tasks in centenarians are recommended. In addition, future studies could potentially use more sophisticated statistical techniques such as structural equation modeling to examine a latent functional variable that is measured by multiple methods of assessing daily functioning. Although confirmatory factor analysis is beyond the scope of the current article, we note that sample sizes of approximately 250 would be required for adequate power to test close, not close, and exact fit in a measurement model including four BADL and eight IADL items (i.e., with 53 df; Davey & Savla, 2010). Larger sample sizes would be required to compare a nested sequence of models, such as across community and facility settings or between self-report and proxy measures. These questions remain for future research.

Funding

The Georgia Centenarian Study was supported by 1P01-AG17553 (L. W. Poon, PI, 2001–2009) from the National Institute on Aging.

Acknowledgments

Author contributions: All authors worked in collaboration to develop this paper. As principal investigator of the Georgia Centenarian Study, L. W. Poon had responsibility for the study design, collection of data, and interpretation of findings. M. B. Mitchell, L. S. Miller, J. L. Woodard, and A. Davey were involved in developing the hypotheses for this paper, data analysis, and writing of this manuscript. A. Davey was additionally responsible for database development, data management, and statistical consultation. P. Martin was involved in study design, measurement selection, and review of this manuscript. M. Burgess served as a research coordinator and worked to recruit and test research participants, in addition to review of this manuscript.

Sponsor’s role: The funding source did not have a role in the data collection, analysis, or preparation of this paper.

References

  1. American Psychiatric Association. Diagnostic and statistical manual of mental disorders. 4th ed. Washington, DC: Author; 2000. Text Revision. [Google Scholar]
  2. Broese van Groenou M, Thomese F. The relative importance of independent living for the social functioning of the elderly. Journal of Gerontology and Geriatrics (Dutch) 1996;27:150–158. [PubMed] [Google Scholar]
  3. Curb JD, Ceria-Ulep CD, Rodriguez BL, Grove J, Guralnik J, Willcox BJ, et al. Performance-based measures of physical function for high-function populations. Journal of the American Geriatrics Society. 2006;54:737–742. doi: 10.1111/j.1532-5415.2006.00700.x. [DOI] [PubMed] [Google Scholar]
  4. Davey A, Savla J. Statistical power analysis with missing data: A structural equation modeling approach. New York: Routledge/Taylor & Francis Group; 2010. [Google Scholar]
  5. DeCoster J. Applied linear regression notes set 1. 2007. Retrieved June 15, 2008, from http://www.stat-help.com/notes.html. [Google Scholar]
  6. Dorevitch MI, Cossar RM, Bailey FJ, Bisset T, Lewis SJ, Wise LA, et al. The accuracy of self and informant ratings of physical functional capacity in the elderly. Journal of Clinical Epidemiology. 1992;45:791–798. doi: 10.1016/0895-4356(92)90057-t. [DOI] [PubMed] [Google Scholar]
  7. Duke University, Center for the Study of Aging and Human Development. Multidimensional functional assessment: The OARS methodology. Durham, NC: Duke University Press; 1978. [Google Scholar]
  8. Edelberg HK, Shallenberger E, Wei JY. Medication management capacity in highly functioning community-living older adults: Detection of early deficits. Journal of the American Geriatrics Society. 1999;47:592–596. doi: 10.1111/j.1532-5415.1999.tb02574.x. [DOI] [PubMed] [Google Scholar]
  9. Folstein M, Folstein M, McHugh PR, Fanjiang G. The Mini-Mental State Examination. Odessa, FL: Psychological Assessment Resources; 2000. [Google Scholar]
  10. Glass TA. Conjugating the “tenses” of function: Discordance among hypothetical, experimental, and enacted function in older adults. The Gerontologist. 1998;38:101–112. doi: 10.1093/geront/38.1.101. [DOI] [PubMed] [Google Scholar]
  11. Hall WJ. Update in geriatrics. Annals of Internal Medicine. 1997;127:557–564. doi: 10.7326/0003-4819-127-7-199710010-00007. [DOI] [PubMed] [Google Scholar]
  12. Katz S, Ford AB, Moskowitz RW, Jackson BA, Jaffe MW. Studies of illness in the aged. The index of ADL: A standardized measure of biological and psychosocial function. Journal of the American Medical Association. 1963;185:914–919. doi: 10.1001/jama.1963.03060120024016. [DOI] [PubMed] [Google Scholar]
  13. Kempen GIJM, Steverink N, Ormel J, Deeg DJH. The assessment of ADL among frail elderly in an interview survey: Self-report versus performance-based tests and determinants of discrepancies. Journal of Gerontology. 1996;51B:254–260. doi: 10.1093/geronb/51b.5.p254. [DOI] [PubMed] [Google Scholar]
  14. Kukull WA, Larson EB, Teri L, Bowen J, McCormick W, Pfanschmidt ML. The Mini-Mental State Examination score and the clinical diagnosis of dementia. Journal of Clinical Epidemiology. 1994;47:1061–1067. doi: 10.1016/0895-4356(94)90122-8. [DOI] [PubMed] [Google Scholar]
  15. Lawton MP, Brody EM. Assessment of older people: Self maintaining and instrumental activities of daily living. The Gerontologist. 1979;9:179–186. [PubMed] [Google Scholar]
  16. Lewis M, Miller LS. Executive control functioning and functional ability in older adults. The Clinical Neuropsychologist. 2007;21:274–285. doi: 10.1080/13854040500519752. [DOI] [PubMed] [Google Scholar]
  17. Loewenstein DA, Amigo E, Duara R, Guterman R, Hurwitz D, Berkowitz N, et al. A new scale for the assessment of functional status in Alzheimer’s disease and related disorders. Journal of Gerontology. 1989;44:114–121. doi: 10.1093/geronj/44.4.p114. [DOI] [PubMed] [Google Scholar]
  18. Loewenstein DA, Arguelles S, Bravo M, Freeman R, Arguelles T, Acevedo A, et al. Caregivers’ judgments of the functional abilities of the Alzheimer's disease patient: A comparison of proxy reports and objective measures. Journal of Gerontology: Psychological Sciences. 2001;56B:78–84. doi: 10.1093/geronb/56.2.p78. [DOI] [PubMed] [Google Scholar]
  19. Lopez MN, Charter RA, Mostafavi B, Nibut LP, Smith WE. Psychometric properties of the Folstein Mini-Mental State Examination. Assessment. 2005;12:137–144. doi: 10.1177/1073191105275412. [DOI] [PubMed] [Google Scholar]
  20. Lum TY, Lin WC, Kane R. Use of proxy respondents and accuracy of minimum data set assessments of activities of daily living. Journals of Gerontology:Medical Sciences. 2005;60A:654–659. doi: 10.1093/gerona/60.5.654. [DOI] [PubMed] [Google Scholar]
  21. Magaziner J, Zimmerman SI, GruberBaldini AL, Hebel JR, Fox KM. Proxy reporting in five areas of functional status—Comparison with self-reports and observations of performance. American Journal of Epidemiology. 1997;146:418–428. doi: 10.1093/oxfordjournals.aje.a009295. [DOI] [PubMed] [Google Scholar]
  22. McDougall GJ, Becker H, Vaughan PW, Acee TW, Delville CL. The revised direct assessment of functional status for independent older adults. The Gerontologist. 2010;50:363–370. doi: 10.1093/geront/gnp139. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Miller LS, Mitchell MB, Woodard JL, Poon LW. Cognitive capacity in the oldest old. 2008, August. Presented at the Georgia Centenarian Study 20th Anniversary Conference, Atlanta, GA. [Google Scholar]
  24. Mitchell MB, Miller LS. Executive functioning and observed versus self-reported measures of functional ability. The Clinical Neuropsychologist. 2008;22:471–479. doi: 10.1080/13854040701336436. [DOI] [PubMed] [Google Scholar]
  25. Myers AM, Holliday PJ, Harvey KA, Hutchinson KS. Functional performance measures: Are they superior to self-assessments? Journal of Gerontology. 1993;48:M196–M206. doi: 10.1093/geronj/48.5.m196. [DOI] [PubMed] [Google Scholar]
  26. Njegovan V, Man-Son-Hing M, Mitchell SL, Molnar FJ. The hierarchy of functional loss associated with cognitive decline in older persons. Journal of Gerontology: Medical Sciences. 2001;56A:M638–M643. doi: 10.1093/gerona/56.10.m638. [DOI] [PubMed] [Google Scholar]
  27. Pezzotti P, Scalmana S, Mastromattei A, Di Lallo D. The accuracy of the MMSE in detecting cognitive impairment when administered by general practitioners: A prospective observational study. BMC Family Practice. 2008:1–11. doi: 10.1186/1471-2296-9-29. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Poon LW, Jazwinski M, Green RC, Woodard JL, Martin P, Rodgers WL, et al. Methodological consideration in studying centenarians. In: Poon LW, Perls TT, editors. Annual review of gerontology and geriatrics. (Vol. 27, pp. 231–264). New York: Springer Publishing Company; 2007. [PMC free article] [PubMed] [Google Scholar]
  29. Royall DR, Lauterbach EC, Kaufer D, Malloy P, Coburn KL, Black KJ. The cognitive correlates of functional status: A review from the Committee on Research of the American Neuropsychiatric Association. Journal of Neuropsychiatry and Clinical Neurosciences. 2007;19:249–265. doi: 10.1176/jnp.2007.19.3.249. [DOI] [PubMed] [Google Scholar]
  30. Rubenstein LZ, Schairer C, Wieland GD, Kane R. Systematic biases in functional status assessment of elderly adults: Effects of different data sources. Journal of Gerontology. 1984;39:686–691. doi: 10.1093/geronj/39.6.686. [DOI] [PubMed] [Google Scholar]
  31. Salmoni MR, Viverais-Dressler G, Porter E, Garg R. Perceived risks to independent living: The views of older, community-dwelling adults. The Gerontologist. 1997;37:729–736. doi: 10.1093/geront/37.6.729. [DOI] [PubMed] [Google Scholar]
  32. Sherman SE, Reuben D. Measures of functional status in community-dwelling elders. Journal of General Internal Medicine. 1998;13:817–823. doi: 10.1046/j.1525-1497.1998.00245.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Spector WD, Fleishman JA. Combining activities of daily living with instrumental activities of daily living. Journal of Gerontology: Social Sciences. 1998;53B:S46–S57. doi: 10.1093/geronb/53b.1.s46. [DOI] [PubMed] [Google Scholar]
  34. Spector WD, Katz S, Murphy JB, Fulton JP. The hierarchical relationship between activities of daily living and instrumental activities of daily living. Journal of Chronic Disease. 1987;40:481–489. doi: 10.1016/0021-9681(87)90004-x. [DOI] [PubMed] [Google Scholar]
  35. Stuck AE, Walthert JM, Nikolaus T, Büla CJ, Hohmann C, Beck JC. Risk factors for functional status decline in community-living elderly people: A systematic literature review. Social Science Medicine. 1999;48:445–469. doi: 10.1016/s0277-9536(98)00370-0. [DOI] [PubMed] [Google Scholar]
  36. Tabbarah M, Crimmins EM, Seeman TE. The relationship between cognitive and physical performance: MacArthur Studies of Successful Aging. Journal of Gerontology: Medical Sciences. 2002;57A:228–235. doi: 10.1093/gerona/57.4.m228. [DOI] [PubMed] [Google Scholar]
  37. Tabert MH, Albert SM, Borukhova-Milov L, Camacho Y, Pelton G, Liu X, et al. Functional deficits in patients with mild cognitive impairment: Prediction of AD. Neurology. 2002;58:758–764. doi: 10.1212/wnl.58.5.758. [DOI] [PubMed] [Google Scholar]
  38. Tager IB, Swanson A, Satariano WA. Reliability of physical performance and self-reported functional measures in an older population. Journal of Gerontology: Medical Sciences. 1998;53A:295–300. doi: 10.1093/gerona/53a.4.m295. [DOI] [PubMed] [Google Scholar]
  39. Thomas VS, Rockwood K, McDowell I. Multidimensionality in instrumental and basic activities of daily living. Journal of Clinical Epidemiology. 1998;51:315–321. doi: 10.1016/s0895-4356(97)00292-8. [DOI] [PubMed] [Google Scholar]
  40. Williamson JD, Espeland M, Kritchevsky SB, Newman AB, King AC, Pahor M, et al. Changes in cognitive function in a randomized trial of physical activity: Results of the Lifestyle Interventions and Independence for Elders Study. Journal of Gerontology: Medical Sciences. 2009;64A:688–694. doi: 10.1093/gerona/glp014. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Zanetti O, Geroldi C, Frisoni GB, Bianchetti A, Trabucchi M. Contrasting results between caregiver’s report and direct assessment of activities of daily living in patients affected by mild and very mild dementia: The contribution of the caregiver’s personal characteristics. Journal of the American Geriatrics Society. 1999;47:196–202. doi: 10.1111/j.1532-5415.1999.tb04578.x. [DOI] [PubMed] [Google Scholar]

Articles from The Gerontologist are provided here courtesy of Oxford University Press

RESOURCES