The Gerontologist 45:222-230 (2005)
© 2005 The Gerontological Society of America
Cognitive Impairment as a Strong Predictor of Incident Disability in Specific ADL–IADL Tasks Among Community-Dwelling Elders: The Azuchi Study
Hiroko H. Dodge, PhD1,
Takashi Kadowaki, MD, MPH, PhD2,
Takehito Hayakawa, PhD3,
Masanobu Yamakawa, PhD4,
Akira Sekikawa, MD, PhD1 and
Hirotugu Ueshima, MD, FFPH2
Correspondence: Address correspondence to Dr. Hiroko Dodge, 519 Parran Hall, 130 DeSoto Street, Department of Epidemiology, University of Pittsburgh, Pittsburgh, PA 15261. E-mail: Dodge{at}edc.pitt.edu
 |
Abstract
|
|---|
Purpose: We examined differential effects of cognitive impairment on each of the activities of daily living (ADL) and instrumental activities of daily living (IADL) tasks. Design and Methods: In a 3-year follow-up of community-dwelling elderly persons in Azuchi, Japan, we assessed cognition by using the Hasegawa Dementia Scale. We examined (a) the cross-sectional association between cognitive impairment and functional disability in each ADL–IADL item; (b) cognitive impairment as a risk factor for incident disability in each ADL–IADL item, through logistic regression models; and (c) probabilities of incident loss of ADL–IADL abilities or death in 3 years, using multinomial logistic regression models. We also calculated the population attributable risk (PAR%) of cognitive impairment on incident loss of task-specific ADL–IADL abilities. Results: Cross-sectionally, the severity of cognitive impairment was associated with disability in each ADL–IADL task, with larger effects shown for ADL items. Longitudinally, minimally or mildly cognitively impaired individuals had a significantly higher risk of losing functional abilities compared with those with intact cognition. The PAR% indicated that cognitive impairment accounts for 11% to 36% of incident disability in ADL–IADL tasks, with the highest PAR% shown for the ability to feed oneself. Implications: Cognitively impaired subjects are heterogeneous; the severity of cognitive impairment has a different impact on incident loss of task-specific ADL–IADL abilities, and comorbidities could affect disabilities differently. Consideration of these heterogeneities will enrich future studies on the impact of cognitive impairment on ADL–IADL abilities.
Key Words: Hasegawa Dementia Scale • ADL • IADL • Population attributable risk • Caregiving needs
As the population ages, cognitive disorders raise a growing public health concern. Cognitive impairment is strongly associated with functional disability measured by basic and instrumental activities of daily living (Aguero-Torres et al., 1998; Chen et al., 1995; Dodge, Shen, Pandav, DeKosky, & Ganguli, 2003; Sauvaget, Yamada, Fujiwara, Sasaki, & Mimori, 2002). However, there are several questions unanswered by past studies. First, most studies defined functional disability by total numbers of ADL and IADL disabilities, although they can be factored into more than two groups, and certain individual ADLs and IADLs are more closely associated with cognitive impairment than are others (Thomas, Rockwood, & McDowell, 1998; Wilms, Kanowski, & Baltes, 2000; Wolinsky & Johnson, 1991).
Only a few studies have examined the effect of cognitive impairment on specific ADL or IADL tasks; in one such study, Blaum, Ofstedal, and Liang (2002) examined the cross-sectional association between low cognitive performance (LCP) and task-specific disability. Even after chronic disease and conditions were controlled for, LCP had a significant association not only with all IADL tasks but also with mobility tasks, including walking across a room and climbing stairs. However, LCP was not uniformly associated with limitations in ADL tasks, and it had attenuated effects with ADL compared with IADL tasks. One potential explanation for this, as the authors point out, is the exclusion of those who required proxy informants; because the inability to respond to a questionnaire is highly related to cognitive impairment, limiting the study sample only to those with self-report would have underestimated the effect of cognitive impairment on ADL tasks.
To our knowledge, only two previous studies, both in the United States, have examined the effect of cognitive impairment on incident disability in specific ADL or IADL tasks. In one study, Furner, Rudberg, and Cassel (1995) found that cognitive impairment was not a predictor of incident disability in any of the IADL tasks in the Longitudinal Study of Aging (LSOA). However, in the LSOA, cognitive impairment was based on a self-reported measure defined as having some trouble remembering things and frequently getting confused. In the other study, Greiner, Snowdon, and Schmitt (1996) found that disabilities in all ADL tasks were strongly predicted by cognitive function objectively measured by the Mini-Mental State Examination (MMSE; Folstein, Folstein, & McHugh, 1975). Moreover, participants with low normal cognition (MMSE scores between 24 and 27) at baseline had a higher risk of losing basic ADL abilities 1 year later than those with high normal cognition. That study, however, did not examine IADL tasks.
A more comprehensive understanding of the effect of cognition on functional ability requires the examination of the full range of daily activity tasks, including both ADLs and IADLs. If we could predict the specific ADL–IADL activities with which currently independent community-dwelling elderly persons with cognitive impairment will eventually need help, we could delay institutionalization or reduce caregiver burden through program development for in-home care.
Our first purpose in the current study is to examine the effect of the range of cognitive status (normal cognition, minimal, mild, and severe cognitive impairment) on specific tasks of both ADLs and IADLs, cross-sectionally and longitudinally. Previous cross-sectional studies have shown severity of cognitive impairment was associated with higher likelihood of being disabled in aggregated ADL–IADL functions (Arling & Williams, 2003; Loewenstein et al., 2001; McConnell, Pieper, Sloane, & Branch, 2002). Using cross-sectional data, first, we verify whether this is also applicable to each of the specific ADL–IADL tasks.
Longitudinally, we hypothesize that those with minimal or mild cognitive impairment might be more vulnerable than those with severe cognitive impairment to incident loss of each activity, possibly because of progression into more severe stages of cognitive impairment or susceptibility to other cataclysmic events such as stroke. Those who can perform certain tasks despite their severe cognitive impairment might continue to do so as a result of their differential pathophysiological conditions. Finally, we examine the probability of dying and the probability of losing at least one IADL ability or losing both IADL and ADL abilities in 3 years by baseline cognitive status. These probabilities clarify the magnitude of future caregiving needs associated with different levels of cognitive impairment.
|
Methods
|
|---|
To investigate functional disabilities, quality of life, family structures, and other issues relevant to the elderly population, the Department of Health Science at the Shiga University of Medical Science (in Shiga, Japan), with funding from the Japan Ministry of Education, Culture, Sports, Science and Technology, conducted a survey of all noninstitutionalized residents aged 65 and older living in Azuchi (total population approximately 12,000). Azuchi is located near the middle of the Shiga prefecture, and 82% of households engage in farming as supplemental employment. In 1990, the proportion of elderly persons aged 65 and older living in Azuchi was 12%, very similar to the 11.6% national level. Among 1,398 community dwellers, data were collected from 1,289 (92.2%) during the 1990–1992 period (baseline). Thirty-four individuals could not be contacted (2.4%), and 75 individuals refused to participate (5.4%). After undergoing intensive training for assessment reliability, a certified nurse and a medical student conducted the interview in pairs. The survey was self-reported, supplemented by information from proxies; if a person was unable to answer the questionnaire or could not fully remember the past events, a family member or caregiver was asked to respond. For 115 out of 1,289 participants (8.9%), information was obtained from proxies for at least part of the survey. The follow-up survey was conducted 3 years later (1993–1995). The gathered information included the Hasegawa Dementia Scale (HDS; Hasegawa, 1983; baseline only), demographic variables, health conditions, and quality of life. Informed consent was obtained from each participant.
Functional Ability
The functional status items included five ADL tasks (bathing, toileting, feeding, dressing, and moving around the house) and seven IADL tasks (preparing meals, using the telephone, cleaning one's room, shopping, using transportation, walking outside with or without walking aids, and using stairs). Moving around the house, walking outside, and using stairs can be categorized as mobility functions, separately from IADL tasks. However, our exploratory factor analysis showed two clear factors (ADL and IADL), and these mobility tasks were loaded strongly on the factor with other IADL tasks. Therefore, we grouped these mobility tasks into IADL in this study. We asked participants whether they were able to perform each task independently, with some help, or not at all. We categorized responses into two groups, independently able versus partly or completely unable.
Cognitive Impairment
We assessed cognitive impairment by using the HDS (range, 0–32.5; see Hasegawa, 1983). The HDS had a correlation of 0.85 with the MMSE (Folstein et al., 1975; Otsuka & Homma, 1991), and, with a cutoff score of 25, it had 83% sensitivity and 82% specificity compared with clinical assessment for dementia (Otsuka & Homma, 1991). We categorized cognitive status into four groups as follows: intact cognition, HDS 
25; minimal cognitive impairment, 21 
HDS < 25; mild cognitive impairment, 17 HDS < 21; and severe cognitive impairment, 0 HDS < 17.
Health Conditions at Baseline
Interviewers asked participants (or proxies), "Do you have any medical conditions for which you were treated during the past year?" In Japan, everyone aged 65 and older has medical access under the national insurance system. Therefore, we assume that the nonresponse to this question as a result of lack of access to medical care was minimal. Trained nurse interviewers categorized the responses into 18 medical conditions that correspond to broad categories in the International Classification of Diseases, Ninth Revision (ICD-9). We further grouped these categories into the following six diseases and controlled for them when we examined the cross-sectional association between cognitive impairment and disabilities: 1, stroke; 2, diseases of the circulatory system except stroke (heart disease, hypertension, and other diseases of the circulatory system); 3, mental disorders, and diseases of the nervous system and sense organs; 4, endocrine, nutritional, and metabolic diseases, and immunity disorders; 5, diseases of musculoskeletal system and connective tissue; and 6, others (categories include infections and parasitic disease; neoplasm; diseases of the blood and blood-forming organs; diseases of the respiratory system; diseases of the genitourinary system; diseases of skin and subcutaneous tissue; symptoms, signs, and ill-defined conditions; and injury and poisoning).
Statistical Analysis
Cross-Sectional Analysis
We assessed the adjusted odds ratio of cognitive status on functional disability by using logistic regression, first by controlling for age, gender, education (high school education or more vs. less than high school education), and response status (self-report vs. proxies' information), and second by adding covariates indicating the six aforementioned disease categories.
Longitudinal Analysis
First, using logistic regression, we examined whether cognitive status at baseline predicted 3-year incidence of functional dependence for each ADL–IADL item separately, controlling for age, gender, education, and response status. We also examined the effect of overall cognitive impairment (HDS < 25) in order to calculate the population attributable risk (PAR%) of cognitive impairment for each ADL–IADL disability incidence. The formula used for PAR% is;
where p is the proportion of the sample with the risk factor (i.e., proportion with cognitive impairment at baseline) and r is the effect of cognitive impairment on incidence of functional dependence (Kahn & Sempos, 1989). This statistic shows the proportion of incident disability that is due to cognitive impairment.
Next, limiting the analysis to those who were free of functional disabilities (IADL–ADL tasks) at baseline, we used multinomial logistic regression models to examine whether cognitive impairment at baseline predicted three possible states: (a) developing one or more disabilities only in IADL; (b) developing one or more disabilities in both IADL and ADL; and (c) death. We ran three logistic regression models simultaneously, with the outcome being the probability of each of the aforementioned three states, in comparison with retaining all ADL and IADL abilities (reference group). Using the coefficients obtained from these models, we calculated the transitional probabilities with which those free from any disability would lose only IADL ability, lose ADL in addition to IADL ability, or die within 3 years.
We could not access ADL or IADL disability status at follow-up for some participants (5.2% of the survivors). By using their demographic information (age, gender, education, and response status) and HDS scores at baseline, we imputed the missing ADL status and IADL status at follow-up and included them in the longitudinal analysis. For each longitudinal analysis, we created three imputed data sets and combined the results by using the multiple imputation procedure (Little & Rubin, 1987). We used SAS Version 8.02 (SAS Institute Inc., 1999) and M-Plus Version 3.01 (the latter for multiple imputation; see Muthen & Muthen, 1998–2004).
|
Results
|
|---|
Among 1,289 participants at baseline, 29 (2.3%) with missing values on the HDS were excluded from further analyses. These 29 individuals were significantly older (mean age 78.7 vs. 74.0, t test, df = 1287, p <.001), but they were not different in either the percentage of women (58.6% vs. 60.4%, chi-square test, df = 1, p =.84) or of those with higher education (27.6% vs. 30.2%, chi-square test, df =1, p =.77) compared with the 1,260 individuals who were included in the analyses. Our study cohort had a mean age (SD) of 74.1 (6.4); 60.4% were female and 30.2% had at least a high school education (see Table 1). The proportions of participants with intact cognition, and minimal, mild, and severe cognitive impairment were 79%, 10.2%, 5.9%, and 4.9%, respectively. Information for 97 participants (7.7%) was obtained from proxies. Participants for whom proxy respondents were needed were significantly older (mean age 78.9 vs. 73.6, t test, df = 1259, p <.001) and were more likely to be cognitively impaired (54.6% vs. 18.2%, chi-square test, df = 1, p <.001), but they did not differ in education or gender distribution from the self-respondents. Among the six disease categories, stroke was the only illness that was significantly associated with cognitive impairment (HDS < 25; chi-square test, 1 df, p =.01).
Cross-Sectional Findings
The prevalence of ADL disabilities was very low at baseline in this cohort of community-dwelling elderly persons (Table 2), at less than 3% for each ADL task. Inability to do IADL tasks ranged from 5.5% (cleaning one's room) to 15.0% (using transportation). Those with minimal cognitive impairment were not different from individuals with intact cognition in the likelihood of being disabled for some of the functions, including toileting, moving around the house, and preparing meals. However, those with mild and severe cognitive impairment had a significantly higher likelihood of being disabled compared with individuals with intact cognition in all activities. Furthermore, the effects increased as the severity of cognitive impairment increased. Overall, we observed a larger impact of cognitive impairment on ADL abilities than IADL abilities. Because of the small number of those unable to self-feed, the confidence interval of the effect on this task was very large. We observed somewhat weaker effects for using stairs, in which the odds of being disabled were only 3.9 times higher even for severely cognitively impaired participants, compared with those with intact cognition. Even after we controlled for comorbidities, the estimates and significance levels virtually did not change. As previous studies have included diabetes mellitus as a covariate, we explored but found no association between ADL–IADL and endocrine, nutritional, metabolic, and immune system diseases (including diabetes). We therefore excluded this disease category from the logistic regression models.
Disease categories significantly associated with disabilities were stroke (associated with all of the ADL tasks and with all IADL tasks except using telephone, with odds ratios ranging 15.3–47.4 for ADL tasks and 3.5–7.1 for IADL tasks), nervous system and sense organ diseases and mental disorders (associated with all of the ADL tasks and IADL tasks except walking outside, using transportation, and using stairs, with odds ratios ranging 6.7–14.3 for ADL tasks and 1.2–3.2 for IADL tasks), and diseases of the musculoskeletal system and connective tissue (associated with moving around the house, cleaning one's room, walking outside, using transportation, and using stairs, with odds ratios ranging 2.7–4.6). The significance levels of the effects of stroke, mental disorders, and diseases of the nervous system and sense organs on these disabilities were very strong, with p <.01 in all models, whereas those of the diseases of the musculoskeletal system and connective tissue fell between p =.01 and.05. In all models, the need for proxy information was significantly associated with disabilities, with odds ratio ranging from 2.8 (using transportation) to 4.6 (dressing).
Longitudinal Findings
Of the 1,260 participants examined at baseline, 127 (10.0%) died before the follow-up interview. To verify the comparability of mortality rates in this cohort with national figures, we ran a logistic regression model with age and gender as covariates and calculated the probabilities of dying within 3 years at ages 70, 75, 80, and 85, and we compared them with the probabilities derived from the Life Table of Japan in 1995. The results are presented in the first two columns of Table 4, which is shown later. The mortality rates in our study cohort are very similar to those of national figures.
Among 1,133 survivors, we could not access ADL or IADL status at follow-up for 59 participants (5.2%), including 7 who refused to participate in the follow-up interview. These individuals did not differ from those with complete data on age, gender, or education, but they were more likely to have been cognitively impaired at baseline (17.5% vs. 32.2%, chi-square test, df = 1, p <.01). We included these 59 individuals in the analysis by using multiple imputations (Little & Rubin, 1987). Over 3 years, more individuals became impaired in the IADL tasks than in the ADL tasks (Table 3). Those with minimal cognitive impairment had an increased risk of incident loss in all task-specific ADL–IADL abilities compared with those with intact cognition, except with two IADL tasks, namely cleaning one's room and using stairs. In these two tasks, those with mild cognitive impairment had the highest risk of incident disability. All three cognitive impairment groups had an excess risk of losing the ability to prepare meals. For meal preparation, the more severe the cognitive impairment, the higher the risk of incident loss of this ability. Those with severe cognitive impairment also had a high risk of incident loss in shopping ability. The PAR% of cognitive impairment (HDS < 25) for developing disability in ADL–IADL functions ranged from 11% (using stairs) to 36% (feeding).
View this table:
[in this window]
[in a new window]
|
Table 3. Azuchi Study, Japan 1990–1995: Cognitive Impairment as a Predictor of Incident Loss of Each ADL–IADL Ability Among Those Able to Do Each Task Without Help at Baseline.
|
|
Next we selected 948 individuals among 1,260 participants at baseline, who were functionally independent in all IADL and ADL abilities at baseline, to examine whether cognitive impairment at baseline would predict incidence of disability in IADL–ADL or mortality at follow-up. The mean (standard deviation) HDS score at baseline was significantly higher for the 948 individuals (28.7; SD = 3.9) than for those with at least one ADL or IADL disability (23.1; SD = 7.5; Wilcoxon rank-sums test, p <.001). In addition, none of these 948 individuals had HDS scores below 11 at baseline, although 7.4% of the original sample had scores this low. With the exception of 1 participant, all of those who lost ADL independence at follow-up also lost IADL independence. This particular individual lost only the ability to use the toilet independently (i.e., independently gets to toilet; gets on and off toilet; arranges clothes; cleans excretory organs). By excluding this person, we created four mutually exclusive hierarchical outcome orders: retaining all IADL and ADL functional abilities, losing one or more abilities only in IADL, losing one or more ADL abilities in addition to IADL ability, and death. We simultaneously estimated three logistic regression models by using multinomial logistic regression models. Among the 947 individuals, 48 died before follow-up interview (i.e., outcome state = death). Among the remaining 899 individuals, 39 (4.3%) had missing data for at least one ADL–IADL item at follow-up. We included these people in the analysis by using multiple imputations. The significant variables and the odds ratios are as follows: The odds of losing ability both in ADL and IADL versus retaining ADL and IADL abilities were 7.6 times (p =.004) higher among those with mild cognitive impairment, and 3.4 times (p =.04) higher among those with minimal cognitive impairment, compared with those without cognitive impairment. The odds of dying within 3 years versus retaining ADL and IADL abilities were 3.2 times (p =.03) higher among those with mild cognitive impairment than those without cognitive impairment. Gender was significant only for the model predicting death compared with retaining ADL and IADL abilities, with men having 2.6 (p =.003) times higher odds than women. Age was significant in all three models, with odds ratios ranging from 1.1 (p <.001) to 1.2 (p <.001).
Finally, we estimated the transitional probabilities to each of four states (i.e., retaining both ADL and IADL abilities, losing only IADL ability, losing both ADL and IADL abilities, and death) by cognitive status, using the coefficients obtained from the multinomial logistic regression models. The results are shown in Table 4. Because no cognitive status group was significant in predicting losing only IADL abilities, and severe cognitive impairment was not significant in any models, we present in the table the transitional probabilities for retaining intact ADL–IADL abilities, losing both ADL–IADL abilities, and death for three cognitive groups: intact cognition, and minimally and mildly cognitively impaired groups. The table can be interpreted as follows; for example, among 85-year-old men who had intact abilities in all ADL–IADL tasks at baseline, 41% of those without cognitive impairment maintained their intact ADL–IADL abilities, whereas 33% and 21% of those with minimal cognitive impairment and mild cognitive impairment could do so, respectively. Only 4% of those with intact cognition lost both ADL and IADL abilities, but 12% of those with minimal cognitive impairment and 18% of those with mild cognitive impairment did so 3 years later. Finally, 25% of those with intact cognition, 28% of those with minimal cognitive impairment, and 42% of those with mild cognitive impairment died within 3 years.
|
Discussion
|
|---|
We found that, cross-sectionally, the severity of cognitive impairment was significantly associated with disabilities in each ADL–IADL task, with larger effects shown for ADL tasks. Even after we controlled for comorbidities, these effects remained. Stroke had the highest and most pervasive impact, affecting all ADL tasks and all but one IADL task. As we expected, diseases of the musculoskeletal and connective tissue had an impact on activities that involved using stairs. Longitudinally, those with minimal or mild cognitive impairment had an excess risk of incident loss of abilities in all of the ADL and IADL tasks. Overall, cognitive impairment accounted for 11% to 29% of incident loss in IADL abilities, and 18% to 36% of incident loss in ADL abilities. The highest PAR% of cognitive impairment on incident loss was shown for feeding ability (PAR% = 36%). We found those with minimal or mild cognitive impairment who had intact abilities in both ADL and IADL tasks at baseline to have significantly higher risk of losing these abilities in 3 years, implying an acceleration of caregiving needs among these groups.
In our study, we found those with minimal cognitive impairment had the highest risk of incident loss in all of the ADL abilities. Those with minimal cognitive impairment might lose the abilities either because of the progression of cognitive impairment or because of the new incidence of an acute event such as a stroke. Our finding coincides with the study by McConnell and colleagues (2002), which found that the severity of cognitive impairment was not a significant predictor of decline in ADL dependence, despite the fact that it was highly associated with ADL dependence cross-sectionally.
The high PAR% of cognitive impairment on the incident loss of feeding ability found in our study also coincides with the study by Greiner and colleagues (1996), which found that the effect of cognitive impairment is greatest on the loss of the ability to feed oneself. On the basis of our cross-sectional result that showed that stroke had the highest impact on feeding ability, and the excess risk of incident loss of this ability among those with minimal cognitive impairment, we suspect the high PAR% of cognitive impairment on feeding ability is probably due to the new stroke incidence among those with minimal cognitive impairment between assessments. However, we cannot confirm this scenario because we do not have data on disease incidence in our study.
Among those free from any disabilities in ADL and IADL tasks at baseline, minimal or mild cognitively impaired participants were significantly more likely to lose both ADL and IADL abilities within 3 years, compared with those with intact cognition. The effect is more profound in older age. For example, at age 85, only 4% of men and 3% of women lose both ADL and IADL abilities within 3 years if they have intact cognition, whereas over 10% of individuals lose these abilities if they have minimal cognitive impairment, and almost 20% lose these abilities if they are mildly cognitively impaired. As the oldest-old segment of the population grows, the overall burden of functional dependence that is due to cognitive impairment is likely to accelerate.
The present study has some limitations. We did not examine the transition to functional improvements, even though past studies have shown that regaining some abilities is possible even among the elderly population (Konno, Katsumata, Arai, & Tamashiro, 2003; Liu, Liang, Muramatsu, & Sugisawa, 1995). In a cross-sectional analysis, because of the small number of those unable to do each ADL task independently, confidence intervals for the odds ratios were large. Hip fractures and sensory impairment were known to be associated with disabilities, yet we do not have variables indicating these specific conditions in our data. As with other community-based studies, disease prevalence was based on self-report, not on diagnosis. The sample was taken from a rural area of Japan and the generalizability of the results to other population groups is unclear. We could not examine the face validity of ADL–IADL items because there were no established ADL–IADL scales when the survey was conducted by which to evaluate the measures used in this study. Finally, we ran models by including those with missing values through multiple imputations, but if the missing pattern was informative dropout (Rubin, 1976), this adjustment would not be adequate.
The implications of our findings for medical care are several fold. Minimal cognitive impairment is a marker of increased risk of future incident loss in both ADL and IADL abilities, possibly because of the progression of cognitive impairment or acute illness such as stroke. Regular and careful assessment of physical conditions of those with minimal cognitive impairment is encouraged. Stroke is one of the major diseases contributing to functional disability in Japan (Hayakawa et al., 2000). We found that stroke had the highest impact on and affected most ADL–IADL abilities. Public policies such as reduction of hypertension through nutritional management, exercise, and increased awareness, which are in effect and have had an impact on reducing stroke incidence over the past few decades (Suh, & Shah, 2001; Ueshima, Zhang, & Choudhury, 2000), are positive prevention programs that should be continued.
In the United States, researchers have found that the proportion of severe cognitive impairment as well as disabilities among the population declined over the past two decades (Freedman, Martin, & Schoeni, 2002), although this finding has yet to be confirmed (Rodgers, Ofstedal, & Herzog, 2003). The trends in other counties are not well known. Our study showed that, even among those with intact ADL and IADL abilities, depending on the status of cognitive impairment, a significant portion of individuals lose both ADL and IADL abilities within a short period. This effect is more profound among the oldest-old group. As the oldest-old segment of the population grows rapidly, careful monitoring of the effect of cognitive impairment on disability is required.
Finally, as shown in this study, cognitively impaired persons are heterogeneous; the severity of cognitive impairment has a different impact on incident loss of ADL–IADL abilities, and comorbidities could have different effects on specific ADL and IADL abilities. Consideration of these heterogeneities will enrich future studies on the impact of cognitive impairment on ADL–IADL abilities.
|
Footnotes
|
|---|
This research was supported in part by Grants-in-Aid for Scientific Research (B) 07457106 and 02454211 from the Japan Ministry of Education, Culture, Sports, Science and Technology in Japan and by Grant K01AG023014-01 from the National Institute on Aging in the United States. 
We thank Ms. Mikie Tsuzihashi, Shoko Hatano at Shiga Prefectural School of Nursing and Dental Care, and Ms. Atsuko Nishida at Shiga University of Medical Science for data collection. We thank Ms. Joni Vander Bilt; Mrs. Barbara Nydeck; Drs. Mary Ganguli, Gerda Fillenbaum, Judy Saxton, and Laurey Lavery; and three anonymous reviewers for their helpful comments.
1 Department of Epidemiology, Graduate School of Public Health, University of Pittsburgh, Pennsylvania.
2 Department of Health Science, Shiga University of Medical Science, Japan.
3 School of Medicine, Shimane University, Japan.
4 Department of Health Science, Osaka Kyoiku University, Japan.
Decision Editor: Linda S. Noelker, PhD
Received for publication February 4, 2004.
Accepted for publication October 21, 2004.
|
References
|
|---|
- Aguero-Torres, H., Fratiglioni, L., Guo, Z., Viitanen, M., von Strauss, E., & Winblad, B., (1998). Dementia is the major cause of functional dependence in the elderly: 3-year follow-up data from a population-based study. American Journal of Public Health, 88, 1452-1456.[Abstract/Free Full Text]
- Arling, G., & Williams, A. R., (2003). Cognitive impairment and resource use of nursing home residents: A structural equation model. Medical Care, 41, 802-812.[Medline]
- Blaum, C. S., Ofstedal, M. B., & Liang, J., (2002). Low cognitive performance, comorbid disease, and task-specific disability: Findings from a nationally representative survey. Journal of Gerontology: Medical Sciences, 57A, M523-M531.
- Chen, P., Yu, E. S., Zhang, M., Liu, W. T., Hill, R., & Katzman, R., (1995). ADL dependence and medical conditions in Chinese older persons: A population-based survey in Shanghai, China. Journal of the American Geriatrics Society, 43, 378-383.[Medline]
- Dodge, H. H., Shen, C., Pandav, R., DeKosky, S., & Ganguli, M., (2003). Functional transitions and active life expectancy associated with Alzheimer Disease. Archives of Neurology, 60, 253-259.[Abstract/Free Full Text]
- Folstein, M. F., Folstein, S. E., & McHugh, P. R., (1975). Mini-Mental State: A practical method for grading the cognitive state of patients for the clinician. Journal of Psychiatric Research, 12, 189-198.[Medline]
- Freedman, V. A., Martin, L. G., & Schoeni, R. F., (2002). Recent trends in disability and functioning among older adults in the United States: A systematic review. Journal of the American Medical Association, 288, 3137-3146.[Abstract/Free Full Text]
- Furner, S. E., Rudberg, M. A., & Cassel, C. K., (1995). Medical conditions differentially affect the development of IADL disability: Implications for medical care and research. The Gerontologist, 35, 444-450.[Abstract]
- Greiner, P. A., Snowdon, D. A., & Schmitt, F. A., (1996). The loss of independence in activities of daily living: The role of low normal cognitive function in elderly nuns. American Journal of Public Health, 86, 62-66.[Abstract/Free Full Text]
- Hasegawa, K., (1983). The clinical assessment of dementia in the aged: A dementia screening scale for psycho geriatric patients. In M. Bergener, U. Leher, E. Lang, & R. Schmitz-Schezer (Eds.), Aging in the eighties and beyond (pp. 207–218). New York: Springer.
- Hayakawa, T., Okayama, A., Ueshima, H., Kita, Y., Choudhury, S. R., & Tamaki, J., (2000). Prevalence of impaired activities of daily living and the impact of stroke and lower limb fracture in elderly persons in Japan. CVD Prevention, 3, 187-194.
- Kahn, H. A., & Sempos, C. T., (1989). Statistical methods in epidemiology (pp. 73–75). New York: Oxford University Press.
- Konno, K., Katsumata, Y., Arai. A., & Tamashiro, H., (2004). Functional status and active life expectancy among senior citizens in a small town in Japan. Archives of Gerontology & Geriatrics, 38, 153-166.
- Little, R. J., & Rubin, D. B., (1987). Statistical analysis with missing data. New York: Wiley.
- Liu, X., Liang, J., Muramatsu, N., & Sugisawa, H., (1995). Transitions in functional status and active life expectancy among older people in Japan. Journal of Gerontology: Social Sciences, 50B, S383-S394.
- Loewenstein, D. A., Ownby, R., Schram, L., Acevedo, A., Rubert, M., & Arguelles., T., (2001). An evaluation of the NINCDS-ADRDA neuropsychological criteria for the assessment of Alzheimer's disease: A confirmatory factor analysis of single versus multi-factor models. Journal of Clinical & Experimental Neuropsychology, 23, 274-284.
- McConnell, E. S., Pieper, C. F., Sloane, R. J., & Branch, L. G., (2002). Effects of cognitive performance on change in physical function in long-stay nursing home residents. Journal of Gerontology: Medical Sciences, 57A, M778-M784.
- Muthen, L. K., & Muthen, B. O., (1998–2004). Mplus user's guide (3rd ed.). Los Angeles, CA: Muthen & Muthen.
- Otsuka, T., & Homma, A., (1991). Koureisha no tameno chiteki-kinou kensa no tebiki [Assessment manual of intellectual function for the demented elderly] (pp. 15–19). Tokyo: World Planning Co.
- Rodgers, W. L., Ofstedal, M. B., & Herzog, A. R., (2003). Trends in scores on tests of cognitive ability in the elderly U.S. population, 1993–2000. Journal of Gerontology: Social Sciences, 58B, S338-S346.
- Rubin, D. B., (1976). Inference and missing data. Biometrika, 63, 581-592.[Abstract/Free Full Text]
- SAS Institute, Inc. (1999). SAS/STAT User's Guide (Version 8) [Computer software]. Cary, NC: Author.
- Sauvaget, C., Yamada, M., Fujiwara, S, Sasaki, H., & Mimori, Y., (2002). Dementia as a predictor of functional disability: A four-year follow-up study. Gerontology, 48, 226-233.[Medline]
- Suh, G. H., & Shah, A., (2001). A review of the epidemiological transition in dementia—cross-national comparisons of the indices related to Alzheimer's disease and vascular dementia. Acta Psychiatrica Scandinavica, 104, 4-11.[Medline]
- Thomas, V. S., Rockwood, K., & McDowell, I., (1998). Multidimensionality in instrumental and basic activities of daily living. Journal of Clinical Epidemiology, 51, 315-321.[Medline]
- Ueshima, H., Zhang, X. H., & Choudhury, S. R., (2000). Epidemiology of hypertension in China and Japan. Journal of Human Hypertension, 14, 765-769.[Medline]
- Wilms, H. U., Kanowski, S., & Baltes, M. M., (2000). Limitations in activities of daily living: Towards a better understanding of subthreshold mental disorders in old age. Comprehensive Psychiatry, 41, 19-25.
- Wolinsky, F. D., & Johnson, R. J., (1991). The use of health services by older adults. Journal of Gerontology: Social Sciences, 46, S345-S357.
This article has been cited by other articles:
|
|
|
|
 
H. H. Dodge, Y. Kita, H. Takechi, T. Hayakawa, M. Ganguli, and H. Ueshima
Healthy Cognitive Aging and Leisure Activities Among the Oldest Old in Japan: Takashima Study
J. Gerontol. A Biol. Sci. Med. Sci.,
November 1, 2008;
63(11):
1193 - 1200.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
A. P. D. Jansen, H. P. J. van Hout, G. Nijpels, H. W. J. van Marwijk, C. Gundy, H. C. W. de Vet, and W. A. B. Stalman
Self-reports on the IQCODE in Older Adults: A Psychometric Evaluation
J Geriatr Psychiatry Neurol,
June 1, 2008;
21(2):
83 - 92.
[Abstract]
[PDF]
|
|
|
|
|
|
|
C. T. Cigolle, K. M. Langa, M. U. Kabeto, Z. Tian, and C. S. Blaum
Geriatric Conditions and Disability: The Health and Retirement Study
Ann Intern Med,
August 7, 2007;
147(3):
156 - 164.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
H. H. Dodge, Y. Du, J. A. Saxton, and M. Ganguli
Cognitive Domains and Trajectories of Functional Independence in Nondemented Elderly Persons
J. Gerontol. A Biol. Sci. Med. Sci.,
December 1, 2006;
61(12):
1330 - 1337.
[Abstract]
[Full Text]
[PDF]
|
|
|
TOP
Abstract
Methods
