| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
| ||||||||||||||||||||||||
Correspondence: Address correspondence to Kara Bottiggi Dassel, PhD, 222 W. Thomas Road, Suite #310, Phoenix, AZ 85013. E-mail: kara.bottiggi{at}chw.edu
 |
Abstract |
|---|
 TOP Abstract MethodsResultsDiscussionReferences |
|---|
Key Words: Activities of daily living • Executive functions • Patient performance
Executive dysfunction has also been linked to impairment in instrumental activities of daily living (IADL), such as driving, cooking, and managing finances (Bell-McGinty, Podell, Franzen, Baird, & Williams, 2002; Cahn-Weiner, Boyle, & Malloy, 2002; Grigsby, Kaye, Baxter, Shetterly, & Hamman, 1998). The successful completion of IADLs is heavily dependent upon a combination of cognitive skills (e.g., EF) that allow an individual to adapt to continuous changes within his or her surroundings (Bryan & Luszcz, 2000; Kray, Eber, & Lindenberger, 2004). EF in older adults not only impacts the achievement of IADLs but may also influence the ability for such deficits to be observed by older caregivers.
Recent studies have revealed that executive deficits are associated with impaired personal judgment regarding a variety of situations within one's environment. Rabin and colleagues (2007) found that executive deficits within the older adult population result in impaired practical judgment involving safety as well as medical, social, ethical, and financial issues. Furthermore, poor judgment in social situations has been found to correlate with frontal and temporal cortical atrophy, which are age-related physiological changes, as well as with executive dysfunction (Eslinger et al., 2007). Additionally, several studies have confirmed significant cognitive impairments in older adult caregivers compared with noncaregivers in areas of attention, monitoring, (Mackenzie, Smith, Hasher, Leach, & Behl, 2007), global cognition, immediate and delayed memory (Lee, Kawachi, & Grodstein, 2004), and vocabulary (Vitaliano et al., 2005). This body of literature provides evidence of cognitive impairments in older adults and caregivers in particular in areas of attention, executive functioning, and judgment, all of which may be exacerbated by the caregiving role. Therefore, it is likely that these cognitive impairments may impact caregivers' ability to observe and accurately report on functional impairments exhibited by their spouses.
The capacity to detect change in another individual who functions within an overlapping environment has significant clinical implications. Family members of older adult patients are commonly relied upon (e.g., by physicians) to provide information regarding the patient's health, physical functioning, affective status, or cognitive abilities. This information is used in the diagnosis and treatment of various health conditions. Unfortunately, it is assumed that the proxy is capable of providing accurate and complete information. There is a large body of evidence revealing inconsistent and generally poor to moderate levels of agreement (
= 0.26–0.81) between informant reports, self-reports, and patient performance on a variety of functional tasks (Magaziner, 1997; Reuben, Siu, & Kimpau, 1992; Rubenstein, Schairer, Wieland, & Kane, 1984; Yasuda et al., 2004). Several informant characteristics such as age (Von Dras et al., 1997), affective status (Sink, Covinsky, Barnes, Newcomer, & Yaffe, 2006), gender (Von Dras et al.), time spent with the patient (Epstein, Hall, Tognetti, Son, & Conant, 1989; Loewenstein et al., 2001), and relationship to the patient (Karagiozis, Gray, Sacco, Shapiro, & Kawas, 1998; Magaziner, Bassett, Hebel, & Gruber-Baldini, 1996; Von Dras et al.) have been identified as influencing the relationship between activity of daily living (ADL) reports. However, even though these studies identified these biases, they did not explore these biases as predictors of ADL report discrepancies.
Spouses are the most common proxy for older adult patients (Spector & Fleishman, 1998), who because of their age, may be experiencing subtle, nonclinical, cognitive deficits in EF that are not immediately recognized by family, friends, or clinicians. These age-related changes in EF could, in turn, have an impact on the accuracy of their reports of patients' health and functioning. The ability to detect subtle impairments in ADLs, as the indicator of patient status, is of particular clinical interest in persons being evaluated for memory disorders such as mild cognitive impairment (MCI; Petersen et al., 1999) or early Alzheimer's disease (AD). MCI is thought by many to represent preclinical AD and is characterized by subtle but measurable cognitive impairment(s) in one or more cognitive domains that are not severe enough to meet the criteria for a diagnosis of dementia (Petersen et al., 1999). MCI is a growing concern among clinicians, because persons with MCI are at increased risk for progression to dementia compared with the general population (6% to 25% vs 1% to 2% per year; Petersen et al., 1999, 2001). Because degenerative diseases are progressive in nature, distinguishing when a person transitions from MCI to dementia can be a complex task for health professionals.
A key diagnostic criterion that distinguishes MCI from AD or other dementias involves the degree to which cognitive decline has an impact on the individual's ability to function within his or her environment, including complex ADLs such as cooking, shopping, and managing medications. By definition, persons with MCI generally maintain the ability to carry out day-to-day tasks necessary for independent living (Petersen et al., 1999) whereas persons with AD have daily functioning that is significantly impaired as a result of more severe cognitive deficits (McKhann et al., 1984). Therefore, it is crucial that the reports provided to physicians regarding the patient's daily functioning be as precise as possible in order for the physicians to reach an accurate diagnosis, identify and adapt the appropriate treatment strategies, plan for future care, and implement the proper safety precautions (e.g., driving cessation and supervision of medications).
To summarize, informants of persons undergoing evaluation for memory disorders are likely to exhibit these common age-related changes in EF. These changes in EF, in areas such as problem solving, attention, concept formation, and judgment, may influence a person's ability to observe and properly identify subtle impairments in intricate daily living tasks, which may determine whether a referral to a physician is made. Further, executive dysfunction may also impact on the precision of informant reports, which are often relied on in the medical community. Thus, our goal in this study was to examine the degree to which mild, nonclinical differences in EF influence the accuracy and reliability of functional reports by spousal caregivers of persons with MCI or mild to moderate AD. We hypothesized that caregivers' executive skills would explain a significant portion of variance in discrepancies between caregiver report of and patient performance on ADLs.
|
Methods |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
Inclusion criteria for cognitively impaired spouses were as follows: (a) they met the research criteria for MCI according to the guidelines of Petersen and colleagues (1999) or for mild to moderate AD according to the guidelines from the National Institute of Neurological and Communicative Disorders and Stroke Alzheimer's Disease and Related Disorders Association (McKhann et al., 1984); (b) they were at least 60 years of age; and (c) they had a Mini-Mental State Examination (MMSE; Folstein, Folstein, & McHugh, 1975) score of 16 or greater, which is indicative of mild to moderate levels of impairment (Schmitt & Wichems, 2006). We excluded those persons with a history of an additional neurological or psychiatric disorder (e.g., Parkinson's disease, bipolar disorder), current untreated depression, major physical handicap, or severe sensory impairment (e.g., glaucoma or deafness).
Caregiver inclusion criteria were as follows: (a) they were at least 60 years of age; (b) they had an Adult Lifestyle Functional Interview–MMSE (Fischbach, 1990) score of 19 or greater, which is equivalent to an MMSE score of 27, suggesting normal mental status; and (c) they had an in-person MMSE score of 27 or greater, which allowed us to rule out any underlying cognitive impairments. We excluded those caregivers who were depressed according to the Geriatric Depression Scale (Yesavage et al., 1982), who used medication(s) that could potentially interfere with cognitive acuity, or who had a history of a neurological or psychiatric disorder.
We screened a total of 47 potential participant dyads for eligibility and found 7 of them to be ineligible. In all cases, it was a caregiver characteristic that excluded the dyad from entering the study. We excluded two caregivers because they had moderate to high levels of depressive symptomatology as expressed on the Geriatric Depression Scale (scores of 12 and 16). Two caregivers reported a history of stroke and, therefore, were excluded. The remaining three caregivers had active medical conditions (possible Parkinson's disease, thyroid and sleep disorders, or orthopedic injury) that excluded them from participation.
Procedures
Data collection consisted of an initial telephone screen for study eligibility in which the Geriatric Depression Scale, the Adult Lifestyle Functional Interview–MMSE, and health questionnaire were administered to the caregiver. The health questionnaire inquired about the caregiver's and care recipient's age, education level, current or previous diagnosis of a psychiatric or neurological disorder, current medication use, and presence of any sensory impairment (e.g., vision or hearing). The caregiver portion contained an additional question in which the caregiver was asked to rate his or her overall health status, and the care-recipient portion included a question that inquired about the approximate date at which the care recipient was diagnosed with a memory disorder. If the couple was found to be eligible for study inclusion, then the interviewer scheduled two home visits to obtain written consent for the study along with caregiver cognitive, affective, and burden information in addition to patient ADL data.
We counterbalanced the order of administration of cognitive measures and the Screen for Caregiver Burden (Vitaliano, Russo, Young, Becker, & Maiuro, 1991) during Visits 1 and 2. We did this to control for potential effects of the order of administration of the cognitive tests in addition to any emotional stress that may have been elicited by the screen. Analyses, based on independent t tests, revealed an absence of order effects on EF measures (raw scores) in either the MCI or AD caregiver groups. During the first home visit, the interviewer had the caregiver complete the Caregiver portion of the Test of Everyday Functional Abilities (TEFA; Cullum et al., 2001), known as the CrTEFA, which asked the caregiver to report on the care recipient's average performance on a variety of ADLs over the previous 4 weeks. The caregiver also completed another ADL report, the Alzheimer's Disease Cooperative Study ADL Scale (ADCS-ADL; Galasko et al., 1997), which we included for validation purposes. During the second home visit the interviewer administered the Patient portion of the TEFA, known as PtTEFA, to the cognitively impaired spouse as a way to rate his or her ADL performance. Each visit took approximately 2 hours to complete and all measures were obtained by the first author (K. Dassel). The average number of days between the telephone screen and the first home visit was 13.8 (SD = 12.4), and that between the first and second home visits was 3.9 (SD = 3.6).
Evaluation Instruments
Study evaluation instruments are listed and briefly described in Table 1 and are divided into the following domains: measures of global cognitive status, measures of executive functioning, measures of physical performance, and measures of caregiver affect and burden.
|
The ADL difference score between the CrTEFA and the PtTEFA served as the outcome variable in tests of the primary study hypothesis. We subtracted the transformed CrTEFA score from the PtTEFA score in order to derive the difference score. In the presence of no functional impairment, perfect agreement between caregiver and patient TEFA scores would be indicated by a score of 1. In the presence of functional impairments, acceptable agreement ranges between –1 and +1, based on the 1-point difference between the scales.
We completed group differences on cognitive, functional, and basic demographic information by using t tests. We used Pearson correlations to examine the relationship between the three primary ADL measures: ADCS-ADL, CrTEFA, and PtTEFA. We analyzed the study hypothesis, that is, that caregivers' EF would account for a significant portion of variance in the ADL difference scores, by using a linear regression model. Specifically, we conducted a backwards, stepwise linear regression in which the caregiver EF composite scores and education levels were entered as predictors of the difference score between PtTEFA performance and CrTEFA ratings. We used this same regression model to analyze associations by group (MCI and AD). Using a conservative approach, we set the significance level for entry into the model at p =.10 and that for removal at p <.05.
|
Results |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
|
|
|
|
|
|
The mean ADL difference score for the total group was –6.13 (SD = 6.31) with discrepancy values ranging from –18.0 to 13.0. Negative ADL difference scores would indicate that, on average, caregivers tended to underestimate the level of functional disability displayed by their spouses. The mean ADL difference score for the AD dyads was –7.00 (SD = 5.90) with a range of –18.0 to 2.0. The mean ADL difference score for the MCI group was –5.25 (SD = 6.7) with a range of –17.0 to 13.0. An independent samples t test revealed no significant differences in the ADL difference score between the two dyad groups: t(2, 38) = –0.87; p =.39.
On the MMSE, we found patients to differ significantly, t(2, 38) = –5.23, p <.001, with the AD group performing more poorly (difference, M = 4.15 points). We found the time since diagnosis to differ between patient groups, t(2, 38) = 3.61, p <.01, with a mean difference of 1.7 years (the AD group had longer time since diagnosis). Participants were recruited into the study within 1 month of physician referral; therefore, it is unlikely that any MCI patients converted to mild AD within that limited time period.
Results of the primary study analyses using a backwards, multiple linear regression model that included caregivers' EF composite scores and education as predictors of the ADL difference score were significant (R2 = 0.11). We found the caregivers' composite EF scores (β = 0.33, p =.04) and education levels to be significant (β = –0.11; p <.05) individual predictors of the ADL difference score. Specifically, results showed that, as caregivers' EF composite scores increased, there was a trend toward greater caregiver ADL report accuracy as reflected by smaller ADL difference score. These results suggest that those caregivers with higher EF scores were more accurate in their ratings of patients' daily functioning than were caregivers with poorer executive skills, thereby supporting our primary study hypothesis.
We examined this pattern in which caregiver EF predicted discrepancies between ADL performance and caregiver report by diagnostic group, using the same backwards, multiple linear regressions. In the AD spousal dyads, we found the regression model, predicting the difference score between CrTEFA and PtTEFA, to be nonsignificant (R2 = 0.05; p =.67), and we removed both education (β = –0.17; p =.56) and composite EF scores (β = 0.25; p =.39) from the model. The same analysis conducted using the MCI spousal dyads was significant (R2 = 0.24; p =.03). Specifically, caregiver composite EF scores remained in the model (β = 0.49; p =.03), whereas we again found education to be a nonsignificant predictor (β = –0.72; p =.74).
|
Discussion |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
Caregivers of both MCI and mild AD spouses who had poorer executive skills tended to underestimate functional disability within their respective spouses, which is consistent with previous studies that have found that spousal caregivers are less accurate compared with other informant types (e.g., children) on their reports of ADL performance and tend to underestimate disability (Karagiozis et al., 1998; Von Dras et al., 1997; Wadley, Harrell, & Marson, 2003). Although the current study cannot address the accuracy of spousal versus other informant types, it does provide a potential explanation for the pattern of underestimation—that is, the influence of caregiver executive skills on ADL reports.
The general pattern of caregiver underestimation of patient disability was consistent between the MCI and mild AD groups (the total group average difference score was –6.13; SD = 6.31); however, the impact of caregiver education and executive skills on these inaccuracies differed by group, according to our secondary analyses. Executive functions were found to significantly predict the ADL difference score in the MCI group, but not in the AD group. This pattern may be explained by the severity of functional impairments displayed by persons with each disorder. Impairments in IADL, such as financial management, telephone use, driving, cooking, medication management, and reading, have been confirmed in persons with MCI (Farias et al., 2006; Griffith et al., 2003; Okonkwo, Wadley, Griffith, Ball, & Marson, 2006; Peres et al., 2006; Perneczky et al., 2006) as well as those with dementia (Glosser et al., 2002; Kounti, Tsolaki, & Kiosseoglou, 2006; Mahurin, DeBettignies, & Pirozzolo, 1991; Marson et al., 2000; Mendez, Mendez, Martin, Smyth, & Whitehouse, 1990; Willis, 1996). However, because of differences in disease severity, it is likely that IADL impairments in persons with MCI are more novel, subtle, and inconsistent compared with those exhibited by persons with mild AD, thereby requiring astute cognitive skills (i.e., EF) to be able to detect such changes. It is possible that an extended disease course (i.e., greater severity) would have provided caregivers of AD patients with a greater opportunity to observe ADL performance as well as to educate themselves regarding the disease and the expected cognitive and functional outcomes. Nevertheless, it is possible that the discrepant findings between the groups may be a result of the small study sample size, which limited the statistical power of our post hoc analyses. These discrepancies were not likely explained by variations in the ADL difference scores or in caregiver EF performance, because our analyses did not reveal any significant group difference between the MCI and AD dyads on these measures.
The discovery that executive skills impact the ability to observe and accurately report on IADL impairments extends the current ecological validity (i.e., real-world impact) of EF within the older adult population. Previous studies have confirmed detrimental real-world consequences of executive dysfunction on IADL tasks such as driving, using the telephone, shopping, and cooking in community-dwelling nondemented older adults (Bell-McGinty et al., 2002; Cahn-Weiner et al., 2002; Dodge, Du, Saxton, & Ganguli, 2006; Grigsby et al., 1998). Results of the current study suggest that not only do deficits in executive skills influence one's ability to perform IADLs successfully, but they may impact judgment of such events occurring within one's environment (i.e., ADL performance of a spouse). These findings are in conjunction with those of Rabin and colleagues (2007) and Eslinger and colleagues (2007), who found that executive deficits were associated with impaired judgment in a variety of social, financial, ethical, and medical issues; it extends their findings by showing that these decrements also impair one's judgment of another person's functional performance.
The identification of caregiver executive skills as a source of potential bias in reports of ADL performance in persons with memory disorders contributes to both research and clinical endeavors. First, the results of the current study support recent literature confirming the presence of functional impairments in persons with MCI. Earlier diagnostic criteria (Petersen et al., 1999) for MCI stated that ADL performance should be relatively preserved in persons with this disorder. There is growing evidence to suggest that IADL impairments are more pronounced than previously described in persons with MCI and extend to areas of managing finances, driving, using the telephone, shopping, and preparing meals (Farias et al., 2006; Griffith et al., 2003; Okonkwo et al., 2006; Peres et al., 2006; Perneczky et al., 2006). In agreement with these previous studies, the persons with MCI in the current study were impaired on IADLs including telephone use and financial and medication management. Furthermore, reports of functional impairments in persons with MCI by their spouses may be underestimating the severity of disability. A more recent view of MCI diagnostic criteria describes IADL impairments exhibited by persons with MCI as "mild" (Winblad et al., 2004). Results of the current study suggest that these impairments may be more advanced than previously noted as a result of the underestimation of disability by spousal caregivers. Many of the investigations of functional impairments in persons with MCI relied on caregiver ADL reports opposed to the direct assessment of patient functioning (Farias et al.; Griffith et al.; Perneczky et al.). The executive skills of spousal caregivers may have influenced the detection of impairments, resulting in a general underestimation of functional disability within the MCI population.
Second, the ability of caregivers to detect functional deficits in persons with memory disorders has significant implications for the concept and validity of early detection of preclinical dementia (i.e., MCI). Early detection of subtle functional impairments displayed by persons with preclinical dementia (as shown by the results of the current study) is dependent on the cognitive skills of spousal informants. The lack of identification of these changes may delay referral to a neurologist, resulting in delayed diagnosis and treatment. Early identification of functional impairments also provides greater diagnostic stability in MCI cases as well as identification of those older adults at risk for future cognitive decline. Peres and colleagues (2006) found that those MCI persons who exhibited functional in addition to cognitive impairments had a fourfold greater risk of progression to dementia compared with those who displayed no functional deficits. Furthermore, those persons who were cognitively unimpaired at baseline, but were impaired on IADL, had a higher risk of dementia over a 2-year period than those with MCI and no functional impairment at baseline. Regarding the stability of MCI diagnoses, MCI persons with IADL deficits had a lower chance of returning to previous higher levels of functioning than did those without functional impairments.
Early detection of functional impairments and the subsequent diagnosis of a memory disorder provides the opportunity for early therapeutic intervention, potentially slowing the progression of cognitive and functional decline from a few months to several years (Borkowska, Ziolkowska-Kochan, & Rybakowski, 2005; Farlow, Anand, Messina, Hartman, & Veach, 2000; Galasko, Kershaw, Schneider, Zhu, & Tariot, 2004). There is currently no treatment approved by the Food and Drug Administration for MCI. However, a recent examination of the efficacy of donepezil, an acetylcholinesterase inhibitor, in persons with MCI revealed that early intervention was able to slow the progression to AD (Petersen et al., 2005). As pharmaceutical research advances in the field of neurodegenerative disorders, early detection and therapeutic treatment will become fundamental in delaying irreversible cognitive and functional losses that accompany neurodegenerative conditions such as MCI and AD.
Lastly, the current study findings are pertinent to clinical drug trials for antidementia drugs. Pharmaceutical treatments for dementia are evaluated for clinical efficacy through multicenter clinical trials and are highly dependent on informant reports of patients' cognitive and physical functioning. Variations in executive skills of caregivers may result in an underestimation of baseline disability or the functional improvements provided by the treatment, subsequently impacting ratings of drug efficacy. In order to prevent this potential negative impact on drug efficacy caused by bias interjected by caregivers in addition to increasing the sensitivity of detection of functional changes in study participants, pharmaceutical companies may benefit from replacing or supplementing caregiver reports with direct ADL assessment tools.
There are several options that may increase the precision of functional assessment of persons undergoing evaluation for memory performance. First, although there is no gold standard of functional assessment, clinicians may want to consider including an objective performance-based functional assessment tool (i.e., TEFA) as a part of their clinical evaluation in order to supplement or replace informant ADL reports. Second, educational programs that provide information to caregivers regarding ADL impairments associated with various stages of the disease course may increase awareness and subsequently reduce the dependency on cognitive skills to detect such changes. Third, the cognitive training of older adults, caregivers in particular, may ameliorate the impact of executive differences or age-related changes in EF on ADL reports. For example, results of a cognitive training clinical trial showed that 10 sessions of cognitive training were able to improve reasoning abilities (i.e., EF) and IADL performance in a group of older adults over a 5-year period (Willis et al., 2006). Cognitive training may not only improve executive skills and personal IADL performance, but may also enhance one's ability to monitor and detect change in IADLs in one's spouse, resulting in earlier referrals, diagnoses, and implementations of therapeutic treatments.
A limitation of the current study was the small sample size. Although the analyses were powered for such a number, the results should still be considered as preliminary in nature. Our sample was also highly homogenous, consisting of Caucasian, highly educated, healthy older adults from the Lexington area, and it may not be representative of the older adult population as a whole. Lastly, variations in symptom duration or a diagnosis of MCI prior to AD may have biased caregiver ratings by increasing their awareness of functional deficits. Future studies should investigate additional factors that may influence caregiver reports of spousal ADL performance, such as caregiver denial of the disease or symptoms, differences between habitual and clinical ADL performance, and the issue of role reversal (i.e., tasks previously performed by the patient may have been taken over by the caregiver), thereby forcing the caregiver to conjecture on spousal ADL performance on nonobservable tasks.
Summary
The current study found that caregivers' executive skills predict discrepancies between caregiver reports of and patient performance on ADL in persons with MCI or mild to moderate AD. The identification of an additional source of bias in caregiver reports has significant implications for the early diagnosis and treatment of memory disorders. The implementation of direct-assessment ADL tools in research studies, within clinical practice, and in clinical drug trials may counteract the impact of informant executive skills on the accuracy of reports of patient daily functioning.
|
Footnotes |
|---|
1 Department of the Graduate Center for Gerontology, University of Kentucky, Lexington.
2 Department of Neurology and Sanders-Brown Center on Aging, University of Kentucky, Lexington.
Decision Editor: William J. McAuley, PhD
Received for publication September 17, 2007. Accepted for publication January 31, 2008.
|
References |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
| ||||||||||||||||||||||||
| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
