Nationwide Database Analysis of Risk Factors Associated with Decreased Activities of Daily Living in Patients with Alzheimer’s Disease

Abstract

Background:

Preserving activities of daily living (ADL) is the key issue for Alzheimer’s disease (AD) patients and their caregivers.

Objective:

To clarify the ADL level of AD patients at diagnosis and the risk factors associated with decreased ADL during long-term care (≤3 years).

Methods:

Medical records of AD patients in a Japanese health insurance claims database were analyzed retrospectively to determine ADL using the Barthel Index (BI) and identify the risk factors associated with decreased ADL.

Results:

A total of 16,799 AD patients (mean age at diagnosis: 83.6 years, 61.5% female) were analyzed. Female patients were older (84.6 versus 81.9 years; p < 0.001) and had lower BI (46.8 versus 57.6; p < 0.001) and body mass index (BMI) (21.0 versus 21.7 kg/m²; p < 0.001) than male patients at diagnosis. Disability (BI≤60) increased at age≥80 years and was significantly higher in females. Complete disability was most frequent for bathing and grooming. Risk factors for decreased ADL were determined separately by sex through comparing the ADL-preserved and ADL-decreased groups using propensity score matching by age and BI and multivariable logistic regression analysis. In males, decreased ADL was significantly associated with BMI < 21.5 kg/m², stroke, and hip fracture, and inversely associated with hyperlipidemia. In females, decreased ADL was significantly associated with BMI < 21.5 kg/m² and vertebral and hip fractures, and inversely associated with lower back pain.

Conclusion:

AD patients with low BMI, stroke, and fractures had increased risks of decreased ADL; such patients should be identified early and managed appropriately, including rehabilitation to preserve ADL.

Keywords

Activities of daily living Alzheimer’s disease baths body mass index bone fractures stroke

INTRODUCTION

With the increase of the aging population worldwide, Alzheimer’s disease (AD) and other dementias have been increasing rapidly and have become a public health concern because they are the major causes of disability and dependency in older people [1, 2]. From 1990 to 2019, the incidence and prevalence of dementia including AD increased by 147.95% and 160.84%, respectively [3]. The global number of people with dementia is forecast to increase from 57.4 million cases in 2019 to 152.8 million cases in 2050 [4]. In Japan, the number of people with dementia was 4.1 million in 2019 and is forecast to reach 5.2 million in 2050 [4] with the rapid increase of the population aged≥65 years [5].

AD progresses through three broad phases, consisting of preclinical AD, mild cognitive impairment (MCI) due to AD, and dementia due to AD [1, 6]. Individuals diagnosed with AD have different symptoms, including memory impairment, apathy, disorientation, behavior changes, and even difficulty walking and speaking [1, 2]. Worsening disease in AD patients decreases the scores for activities of daily living (ADL) and quality of life (QOL) in patients and increases caregiver burden and caregiving costs [7 –9]. In the year after diagnosis, health service use was found to be dramatically increased in dementia patients [10].

The shortage of nursing care and caregiving services for long-term care are considered serious issues in super-aged countries, including Japan, because a more than 10-year gap exists between life expectancy and healthy life expectancy in these countries [5]. The progression of AD has a major impact on caregivers, in addition to patients. Caregivers for AD patients often assist with one or more ADL, provide emotional support, and manage health conditions [1, 11]. Decreased ADL in AD patients was found to be associated with increased caregiver burden, whereas cognitive decline had less effect on it [12 –14]. Therefore, ADL preservation is the key issue for both AD patients and their caregivers.

It is important to recognize the levels of ADL and the risk factors for decreased ADL in AD patients and to make efforts to preserve and improve ADL during long-term care. In the present study, a nationwide database analysis was performed to clarify the level of ADL in AD patients at diagnosis and the risk factors associated with decreased ADL during long-term care for up to 3 years.

MATERIALS AND METHODS

Study design and data source

This was a retrospective case-control study conducted to determine the ADL level of AD patients at diagnosis and the risk factors associated with decreased ADL. This epidemiological study was conducted in accordance with the Guidelines for Good Pharmacoepidemiology Practices [15] and the ethical standards of the institutional review board of Eisai Co. Ltd. and in accordance with the Helsinki Declaration of 1975.

The Japanese health insurance claims database provided by the Japan Medical Data Center (JMDC) was used. The collected data include medical claims (for hospitalization and outpatient treatment) and diagnosis procedure combination (DPC) assessment forms, which are documents that the Japanese Ministry of Health, Labour and Welfare requests medical institutions to provide [16, 17]. The forms consist of a comprehensive evaluation of diagnostic categories and are a method of calculating medical expenses for defining daily hospitalization expenses for each diagnostic category. To protect personal information, data from medical institutions are collected as anonymized information on the basis of a personal identifier for each medical institution with which individuals cannot be identified. The database is widely available, on a fee-paying basis, for use in surveys, research, and commercial purposes and can be accessed after completing the contract for use of specific data. The database contains survey data for more than 9 million cases from more than 200 acute care hospitals in Japan, and publication of studies using this database has been increasing in recent years [16, 17]. The data extracted included administrative claims and clinical information for all inpatients discharged from participating hospitals. The database contains an anonymized patient identifier, sex, age, hospitalization summaries, disease history, medical claims data, and other data related to patients’ conditions.

Disease definitions and analysis population

According to the International Statistical Classification of Diseases: 10th Revision (ICD-10), AD patients were identified with a definitive diagnosis code of F00 and G30 in the data collected from April 1, 2014 to May 31, 2022.

To understand disease comorbidities, the records of relevant comorbid diseases were collected for 6 months before and after diagnosis using ICD-10 codes for cancer (code: C), diabetes mellitus (E10-14), hyperlipidemia (E78), depression (F32-33), insomnia (F51, G47), stroke (G45, I60–64, I69), hypertension (I10–15 codes), pneumonia (J12–18), chronic obstructive pulmonary disease (COPD: J40–47), lower back pain (M40–48, M50–54), osteoporosis (M80–82), vertebral fractures (S30-39), and hip fractures (S70-79).

In the analysis of BMI, the cut-off for low BMI was defined as 21.5 kg/m², because the target BMI recommended for the population aged≥65 years ranged from 21.5 to 24.9 kg/m² in the 2020 version of the Dietary Reference Intakes for Japanese officially announced by the Ministry of Health, Labour and Welfare, Japan [18].

Evaluation of ADL decrease and analysis of risk factors associated with decreased ADL

Records for ADL at diagnosis and during the observation period of up to 3 years were collected from the in-hospital database recorded at discharge. The ADL level was evaluated by the Barthel Index (BI), which was essential for evaluating disability of self-care activities in the BI items, including feeding, transfers, grooming, toilet use, bathing, ambulation, stair climbing, dressing, bowel care, and bladder care. BI was evaluated at diagnosis and during the observation period, and patients without a record of the ADL score at baseline or during the observation period were excluded. The BI score ranged from 0 to 100. Patients with a BI≤60 were defined as patients with ADL disability [19, 20]. In the analysis of disability changes in BI items, complete disability was defined as a zero score for each BI item.

In the separate analysis of decreased ADL by sex, patients with BI≥65 at baseline were divided into the ADL-preserved group and the ADL-decreased group, based on the lowest BI during the observation period of up to 3 years. The ADL-preserved group included patients with the lowest BI≥65, whereas the ADL-decreased group included patients with the lowest BI≤60 during the observation period of up to 3 years. The ADL-decreased group was compared with the ADL-preserved group to identify the risk factors associated with decreased ADL after propensity score matching by age and BI at baseline. Matching was performed on a 1 : 1 basis, separately for male and female patients, between the ADL-decreased group and the ADL-preserved group without replacement using a caliper width of 0.15 standard deviations of the logit of each propensity score. Adequate variable balance after matching was confirmed with a standardized mean difference (SMD) of less than 0.1 [21].

Statistical analysis

Statistical analysis was performed using R 4.2.1. Student’s t-test for continuous variables and Fisher’s exact test for categorical variables were used to assess differences between two groups. A logistic regression model was used to perform multivariable analysis with the lowest BI status during the observation period as the outcome and low BMI (<21.5 kg/m²) and the presence of 13 major comorbidities as covariates. Age and baseline BI were considered as confounders for matching, as mentioned above. The database used for this analysis is constrained by the fact that ADL data were only obtained at the time of admission; therefore, logistic regression analysis was used with the lowest BI status as the endpoint, rather than an analysis method with an assumption of time-series constancy, such as a Cox proportional-hazards model-based analysis.

RESULTS

A total of 16,799 AD patients (mean age at diagnosis: 83.6 years, 61.5% female) were analyzed (Table 1). Female patients were older (84.6 versus 81.9 years; p < 0.001) and had lower BI (46.8 versus 57.6; p < 0.001) and body mass index (BMI) (21.0 versus 21.7 kg/m²; p < 0.001) than male patients at diagnosis. Average BI was lower in female patients (46.8) than in male (57.6) patients, and the percentage of patients with BI≤60 was significantly higher in female (64.0%) than in male (52.0%) patients. BI was evaluated only once during the observation period in 63.9% of patients, and the rate was higher in female (65.2%) than male (61.8%) patients.

Table 1

Characteristics and comorbidities of AD patients

	All	Male	Female	p
Number (%)	16,799 (100.0)	6,475 (38.5)	10,324 (61.5)
Age (y): Ave (SD)	83.6 (6.6)	81.9 (6.6)	84.6 (6.3)	<0.001
BMI (kg/m²): Ave (SD)	21.2 (3.2)	21.7 (3.0)	21.0 (3.3)	<0.001
BI (score): Ave (SD)	51.0 (38.0)	57.6 (38.2)	46.8 (37.3)	<0.001
65 to 100: N (%)	6,830 (40.7)	3,111 (48.0)	3,719 (36.0)	<0.001
0 to 60: N (%)	9,969 (59.3)	3,364 (52.0)	6,605 (64.0)
BI assessments during the observation period: Number (%)
1	10,729 (63.9)	3,999 (61.8)	6,730 (65.2)	<0.001
≥2	6,070 (36.1)	2,476 (38.2)	3,594 (34.8)
Ave (SD)	1.7 (1.4)	1.8 (1.5)	1.7 (1.4)
Min - Max	1 - 36	1 - 36	1 - 27
Comorbidity: N (%)
Hypertension	11,026 (65.6)	4,218 (65.1)	6,808 (65.9)	0.293
Insomnia	5,970 (35.5)	2,294 (35.4)	3,676 (35.6)	0.816
Stroke	5,659 (33.7)	2,527 (39.0)	3,132 (30.3)	<0.001
Diabetes mellitus	5,503 (32.8)	2,506 (38.7)	2,997 (29.0)	<0.001
Hyperlipidemia	5,331 (31.7)	1,993 (30.8)	3,338 (32.3)	0.036
Lower back pain	4,897 (29.2)	1,912 (29.5)	2,985 (28.9)	0.392
Osteoporosis	3,772 (22.5)	632 (9.8)	3,140 (30.4)	<0.001
Cancer	3,224 (19.2)	1,806 (27.9)	1,418 (13.7)	<0.001
Pneumonia	2,929 (17.4)	1,344 (20.8)	1,585 (15.4)	<0.001
COPD	2,838 (16.9)	1,363 (21.1)	1,475 (14.3)	<0.001
Hip fracture	2,156 (12.8)	365 (5.6)	1,791 (17.3)	<0.001
Depression	1,759 (10.5)	590 (9.1)	1,169 (11.3)	<0.001
Vertebral fracture	1,682 (10.0)	500 (7.7)	1,182 (11.4)	<0.001

Age, BMI, and BI were extracted from the patient records at the time of diagnosis and are shown as average (Ave) and standard deviation (SD) values. Disease comorbidity was extracted for 6 months before and after diagnosis and is shown as patient number (N) and the rate (%). AD: Alzheimer’s disease, BI: Barthel index, BMI: body mass index, COPD: chronic obstructive pulmonary disease.

The rates of stroke, diabetes mellitus, cancer, pneumonia, and COPD were higher in male than in female patients, whereas the rates of hyperlipidemia, osteoporosis, depression, vertebral fracture, and hip fracture were higher in female than in male patients (Table 1). It is important to note that the rates of comorbid osteoporosis, hip fracture, and vertebral fracture were 3.1-fold, 3.1-fold, and 1.5-fold higher, respectively, in female than in male patients.

Figure 1 shows the age distribution of male and female patients at AD diagnosis, which peaked at 83 years in male and 86 years in female patients. The number of AD patients was higher in female than in male patients at age≥78 years. The percentage of patients with disability (BI≤60) increased at age≥80 years and was significantly higher in female than in male patients at age≥75 years (Fig. 1).

Fig. 1

Age distribution of AD patients and the rate of AD patients with ADL disability. A) The numbers of male (black column) and female (gray column) AD patients are shown by age at diagnosis. B) The rate of male (black circle) and female (white circle) patients with ADL disability (BI≤60) by age at diagnosis. AD, Alzheimer’s disease; ADL, activities of daily living, *p <0.05, **p <0.01.

The increase of complete disability in each BI item with the decrease of the total BI score was evaluated in AD patients (Table 2). At BI scores of 45-60, BI items showing complete disability in more than 50% of patients were bathing and grooming in male and female patients. At BI scores of 25-40, additional BI items using the same criteria were stair climbing in male and female patients and ambulation in female patients. At BI scores of 0-20, all BI items showed complete disability in more than 50% of patients.

Table 2

Increase of complete disability of ADL in each BI item with the decrease of the total BI score

BI	100 to 85	80 to 65	60 to 45	40 to 25	20 to 0
Male, Total: N (%)	2,530 (100.0)	581 (100.0)	1,019 (100.0)	646 (100.0)	1,699 (100.0)
BI 01: feeding	2 (0.1)	1 (0.2)	1 (0.1)	19 (2.9)	1,034 (60.9)
BI 02: transfers	0 (0.0)	0 (0.0)	3 (0.3)	20 (3.1)	1,143 (67.3)
BI 03: grooming	33 (1.3)	201 (34.6)	858 (84.2)	605 (93.7)	1,689 (99.4)
BI 04: toilet use	3 (0.1)	6 (1.0)	29 (2.8)	227 (35.1)	1,653 (97.3)
BI 05: bathing	323 (12.8)	514 (88.5)	1,000 (98.1)	642 (99.4)	1,697 (99.9)
BI 06: ambulation	0 (0.0)	3 (0.5)	45 (4.4)	297 (46.0)	1,620 (95.4)
BI 07: stair climbing	33 (1.3)	115 (19.8)	358 (35.1)	531 (82.2)	1,692 (99.6)
BI 08: dressing	0 (0.0)	8 (1.4)	52 (5.1)	251 (38.9)	1,646 (96.9)
BI 09: bowel care	0 (0.0)	7 (1.2)	37 (3.6)	263 (40.7)	1,672 (98.4)
BI 10: bladder care	2 (0.1)	15 (2.6)	48 (4.7)	279 (43.2)	1,669 (98.2)
Female, Total: N (%)	2,682 (100.0)	1,037 (100.0)	1,815 (100.0)	1,112 (100.0)	3,678 (100.0)
BI 01: feeding	0 (0.0)	1 (0.1)	4 (0.2)	25 (2.2)	2,272 (61.8)
BI 02: transfers	0 (0.0)	0 (0.0)	6 (0.3)	35 (3.1)	2,621 (71.3)
BI 03: grooming	48 (1.8)	379 (36.5)	1,526 (84.1)	1,034 (93.0)	3,668 (99.7)
BI 04: toilet use	1 (0.0)	4 (0.4)	49 (2.7)	365 (32.8)	3,581 (97.4)
BI 05: bathing	528 (19.7)	944 (91.0)	1,789 (98.6)	1,106 (99.5)	3,675 (99.9)
BI 06: ambulation	0 (0.0)	13 (1.3)	151 (8.3)	591 (53.1)	3,553 (96.6)
BI 07: stair climbing	68 (2.5)	275 (26.5)	753 (41.5)	966 (86.9)	3,670 (99.8)
BI 08: dressing	2 (0.1)	5 (0.5)	91 (5.0)	468 (42.1)	3,541 (96.3)
BI 09: bowel care	2 (0.1)	9 (0.9)	32 (1.8)	330 (29.7)	3,589 (97.6)
BI 10: bladder care	2 (0.1)	9 (0.9)	42 (2.3)	333 (29.9)	3,586 (97.5)

Patient numbers (%) showing complete disability of ADL were summarized for each BI item with respect to the total BI score. Underlined patient number (%) indicates complete disability of ADL in more than 50% of patients. ADL, activities of daily living; BI, Barthel index.

Because the age distribution and BI at diagnosis and comorbidities differed between male and female patients, risk factors for decreased ADL were determined separately by sex. In the separate analysis of decreased ADL for male and female patients, a total of 3,111 male patients and 3,719 female patients with BI≥65 at baseline were divided into two groups based on the lowest BI during the observation period of up to 3 years: the ADL-preserved group and the ADL-decreased group (Table 3). Because patients in the ADL-decreased group were older and showed lower BI than the ADL-preserved group, propensity score matching by age and BI was conducted separately for male and female patients. SMD values of less than 0.1 in age and BI clearly indicated adequate variable balance after matching in male and female patients. The average age of the ADL-preserved group and the ADL-decreased group was 81.3 years and 81.0 years in male patients and 82.6 years and 83.0 years in female patients, respectively.

Table 3

Propensity score matching by age and BI between the ADL-preserved group and the ADL-decreased group

		Before matching by age and BI			After matching by age and BI
		ADL-preserved	ADL-decreased	SMD	ADL-preserved	ADL-decreased	SMD
Male	N	1,679	1,432	1,208	1,208
	Age (y): Ave (SD)	79.7 (6.4)	81.7 (6.2)	0.332	81.3 (6.0)	81.0 (6.2)	0.047
	BI: Ave (SD)	95.5 (8.8)	90.2 (12.3)	0.430	93.9 (9.9)	92.9 (11.0)	0.079
Female	N	1,792	1,927	1,498	1,498
	Age (y): Ave (SD)	81.3 (6.1)	83.7 (5.9)	0.416	82.6 (5.4)	83.0 (6.0)	0.067
	BI: Ave (SD)	92.9 (10.6)	87.1 (13.0)	0.445	91.7 (11.2)	91.0 (11.4)	0.059

Patients with BI≥65 at baseline were divided into the ADL-preserved group and the ADL-decreased group, based on the lowest BI during the observation period of up to 3 years. Age and BI are shown as average (Ave) and standard deviation (SD) values before and after propensity score matching. SMD values less than 0.1 for age and BI clearly indicate adequate variable balance after matching in male and female patients, respectively. ADL, activities of daily living; BI, Barthel index; SMD, standardized mean difference.

The risk factors associated with decreased ADL were identified by comparing the ADL-decreased group and the ADL-preserved group by multivariable logistic regression analysis, separately for male and female patients (Table 4). In male patients, decreased ADL was significantly associated with BMI < 21.5 kg/m², stroke, and hip fracture, and inversely associated with hyperlipidemia. In female patients, decreased ADL was significantly associated with BMI < 21.5 kg/m², vertebral fracture, and hip fracture, and inversely associated with lower back pain (Fig. 2).

Table 4

Multivariable analysis of risk factors associated with decreased ADL during long-term care in AD patients

(A) Male	ADL-preserved	ADL-decreased	Univariable	Multivariable
			p	Odds ratio	95% CI	p.adj
N (%)	1,208 (100.0)	1,208 (100.0)
BMI (<21.5 kg/m²)	462 (38.2)	554 (45.9)	<0.001	1.33	1.12 – 1.57	<0.001
Hypertension	854 (70.7)	809 (67.0)	0.053	0.90	0.75 – 1.08	0.265
Diabetes mellitus	537 (44.5)	511 (42.3)	0.305	1.00	0.85 – 1.19	0.962
Hyperlipidemia	474 (39.2)	393 (32.5)	0.001	0.77	0.65 – 0.92	0.004
Insomnia	451 (37.3)	412 (34.1)	0.107	0.89	0.74 – 1.06	0.181
Cancer	430 (35.6)	418 (34.6)	0.639	0.96	0.81 – 1.14	0.630
Stroke	396 (32.8)	445 (36.8)	0.040	1.28	1.07 – 1.52	0.005
Lower back pain	407 (33.7)	381 (31.5)	0.278	0.90	0.75 – 1.08	0.259
COPD	267 (22.1)	258 (21.4)	0.693	0.95	0.77 – 1.16	0.612
Pneumonia	184 (15.2)	196 (16.2)	0.539	1.04	0.83 – 1.31	0.728
Osteoporosis	113 (9.4)	114 (9.4)	1.000	0.99	0.74 – 1.32	0.952
Depression	98 (8.1)	96 (7.9)	0.940	1.01	0.75 – 1.37	0.941
Vertebral fracture	56 (4.6)	75 (6.2)	0.106	1.40	0.97 – 2.03	0.077
Hip fracture	19 (1.6)	36 (3.0)	0.029	1.95	1.12 – 3.51	0.020
(B) Female	ADL-preserved	ADL-decreased	Univariable	Multivariable
			p	Odds ratio	95% CI	p.adj
N (%)	1,498 (100.0)	1,498 (100.0)
BMI (<21.5 kg/m²)	734 (49.0)	817 (54.5)	0.003	1.21	1.04 – 1.40	0.010
Hypertension	1,036 (69.2)	1,046 (69.8)	0.721	1.09	0.93 – 1.29	0.284
Hyperlipidemia	658 (43.9)	612 (40.9)	0.096	0.88	0.76 – 1.03	0.108
Insomnia	587 (39.2)	570 (38.1)	0.548	0.98	0.84 – 1.14	0.775
Lower back pain	609 (40.7)	543 (36.2)	0.015	0.82	0.70 – 0.96	0.011
Diabetes mellitus	522 (34.8)	517 (34.5)	0.878	1.06	0.91 – 1.24	0.467
Osteoporosis	506 (33.8)	504 (33.6)	0.969	0.95	0.80 – 1.12	0.534
Stroke	349 (23.3)	383 (25.6)	0.161	1.16	0.98 – 1.37	0.087
Cancer	302 (20.2)	256 (17.1)	0.035	0.85	0.71 – 1.03	0.095
COPD	235 (15.7)	217 (14.5)	0.386	0.95	0.77 – 1.17	0.649
Depression	204 (13.6)	183 (12.2)	0.276	0.90	0.72 – 1.13	0.371
Vertebral fracture	170 (11.3)	210 (14.0)	0.032	1.35	1.07 – 1.70	0.011
Pneumonia	154 (10.3)	164 (10.9)	0.593	1.09	0.86 – 1.38	0.496
Hip fracture	117 (7.8)	171 (11.4)	0.001	1.51	1.18 – 1.95	0.001

The incidences of low BMI and the comorbidities are summarized as patient numbers (N) and rates (%). A logistic regression model was used to perform multivariable analysis, and the results are shown as odds ratios. AD, Alzheimer’s disease; ADL, activities of daily living; BMI, body mass index; CI, confidence interval; COPD, chronic obstructive pulmonary disease.

Fig. 2

Forest plots of multivariable analysis identifying risk factors associated with decreased ADL. A logistic regression model was used to perform multivariable analysis in male (A) and female (B) patients, and odds ratios and 95% confidence intervals (CIs) are shown. AD, Alzheimer’s disease; ADL, activities of daily living; BMI, body mass index; COPD, chronic obstructive pulmonary disease, *p <0.05, **p <0.01.

The increase of BI items showing complete loss of ADL during the observation period of up to 3 years was evaluated in patients in the ADL-preserved group and the ADL-decreased group (Table 5). The major BI items showing complete loss of ADL in more than 90% of patients were bathing and grooming in male and female patients in the ADL-decreased group. No relevant changes were seen in patients in the ADL-preserved group during the observation period.

Table 5

Increase of complete disability of ADL in AD patients during long-term care

	ADL-preserved		ADL-decreased
	At diagnosis	During long-term care	At diagnosis	During long-term care
Male, Total: N (%)	1,208 (100.0)	1,208 (100.0)	1,208 (100.0)	1,208 (100.0)
BI 01: feeding	1 (0.1)	1 (0.1)	2 (0.2)	394 (32.6)
BI 02: transfers	0 (0.0)	1 (0.1)	0 (0.0)	446 (36.9)
BI 03: grooming	73 (6.0)	99 (8.2)	85 (7.0)	1,118 (92.5)
BI 04: toilet use	1 (0.1)	2 (0.2)	7 (0.6)	672 (55.6)
BI 05: bathing	306 (25.3)	362 (30.0)	325 (26.9)	1,198 (99.2)
BI 06: ambulation	1 (0.1)	3 (0.2)	2 (0.2)	701 (58.0)
BI 07: stair climbing	53 (4.4)	71 (5.9)	62 (5.1)	928 (76.8)
BI 08: dressing	4 (0.3)	6 (0.5)	3 (0.2)	717 (59.4)
BI 09: bowel care	1 (0.1)	2 (0.2)	4 (0.3)	686 (56.8)
BI 10: bladder care	8 (0.7)	5 (0.4)	7 (0.6)	704 (58.3)
Female, Total: N (%)	1,498 (100.0)	1,498 (100.0)	1,498 (100.0)	1,498 (100.0)
BI 01: feeding	1 (0.1)	0 (0.0)	0 (0.0)	404 (27.0)
BI 02: transfers	0 (0.0)	0 (0.0)	0 (0.0)	498 (33.2)
BI 03: grooming	139 (9.3)	175 (11.7)	143 (9.5)	1,390 (92.8)
BI 04: toilet use	2 (0.1)	3 (0.2)	0 (0.0)	756 (50.5)
BI 05: bathing	503 (33.6)	594 (39.7)	564 (37.7)	1,490 (99.5)
BI 06: ambulation	4 (0.3)	5 (0.3)	3 (0.2)	851 (56.8)
BI 07: stair climbing	122 (8.1)	131 (8.7)	111 (7.4)	1,162 (77.6)
BI 08: dressing	1 (0.1)	5 (0.3)	4 (0.3)	807 (53.9)
BI 09: bowel care	8 (0.5)	2 (0.1)	1 (0.1)	754 (50.3)
BI 10: bladder care	8 (0.5)	4 (0.3)	0 (0.0)	759 (50.7)

Patient numbers (%) showing complete disability of ADL are summarized in the ADL-preserved group and the ADL-decreased group. Underlined patient numbers (%) indicate complete disability of ADL in more than 90% of patients. AD, Alzheimer’s disease; ADL, activities of daily living; BI, Barthel index.

DISCUSSION

The results of this study provide nationwide, real-world evidence in Japan for the age distribution of AD patients and decreased ADL with age. The diagnosis of AD peaked at age 83 years in male and 86 years in female patients. The total number of AD patients was higher in female than in male patients. In addition, the percentage of AD patients with disability increased at age≥80 years in male and female patients and was higher in female than in male patients. The appropriate timing of medical checkups to evaluate cognitive function and the ADL level and optimal management may differ between male and female AD patients.

The present study demonstrated that low BMI at diagnosis was the independent risk factor for decreased ADL during the observation period for up to 3 years in both male and female patients. Decreased ADL in low BMI individuals has been reported in other studies. The cohort study with 12,376 Japanese individuals with dementia aged > 65 years showed that BMI < 23 kg/m² was a risk factor for disability due to dementia [22]. The nationwide database study in Japan found that underweight BMI (<18.5 kg/m²) was significantly associated with low BI of ADL and a higher overall complication rate in 13,348 elderly patients with femoral fractures [23]. The significant association of low BMI with decreased ADL was also reported in studies of nursing home residents [24, 25]. Importantly, poor nutritional status was significantly correlated with low BMI, low muscle power, and low BI of ADL in elderly individuals [26, 27], and a randomized, controlled trial showed significant improvement of ADL by nutritional supplements [28]. In Japan, the optimal BMI range of 21.5 to 24.9 kg/m² was recommended for the population aged≥65 years in the 2020 version of Dietary Reference Intakes for Japanese [18]. Appropriate education and management of nutritional status and BMI may preserve ADL in AD patients.

In the present study, decreased ADL was associated with stroke in male but not in female patients. Stroke has been reported to increase the risks of decreases in both cognitive function and ADL. Fifty-seven percent of stroke patients without pre-existing cognitive impairment developed cognitive decline one year after stroke [29]. The coincidence of stroke with cognitive impairment increased the severity of ADL decline, compared with either alone [30, 31]. A systematic review of time course studies showed that patients with impaired motor function of the leg were at higher risk of decreased ADL after their first stroke [32]. The small impact of stroke on decreased ADL in female patients may be due to their lower rate of comorbid stroke than male patients.

The present study also showed that decreased ADL in AD patients was associated with hip fractures in both male and female patients. AD was significantly associated with a 2.5-fold higher risk of hip fracture compared with non-AD controls [33]. In addition, an increased incidence of dementia was reported in patients who developed hip fractures, but not vertebral or non-hip non-vertebral fractures [34]. Because older individuals with dementia have a greater decrease in motor function and impaired neurological reflexes during falls, they may land on the hip and have a higher risk of hip fracture [34]. In contrast, decreased ADL in AD patients was associated with vertebral fractures in female patients, but not in male patients in the present study. Though frailty progression was increased in osteoporosis patients with hip fractures in male and female patients, it was increased with vertebral fractures in female, but not in male patients [35]. Physical frailty and dysfunction were associated with decreased ADL in AD patients and were more noticeable in female than in male patients [36]. Because female AD patients showed higher rates of comorbid osteoporosis, vertebral fractures, and hip fractures than male patients in the present study, female AD patients with fractures and their caregivers may need to pay more attention to decreased ADL and obtain appropriatemanagement.

There was an inverse correlation of decreased ADL with hyperlipidemia in male patients and with lower back pain in female patients. Medication for hyperlipidemia was initiated at average ages in the 60 s and earlier [37, 38]. The prevalence of lower back pain peaked at the age of 45–54 years [39], and patients with lower back pain living with disability peaked at 45-49 years in male and female patients [40]. Treatment, education, and management for these comorbidities at younger ages may reduce the ADL decrease in aged AD patients. Further studies are warranted to determine the relationships of comorbidities and decreased ADL in AD patients.

The present study showed that complete loss of ADL was seen in bathing and grooming the most frequently in patients with disability at diagnosis. Bathing and grooming were also the major BI items showing complete loss in patients in the ADL-decreased group during the observation period of up to 3 years. Decreased ADL in bathing was associated with increased caregiver burden and the risk of acute hospitalization of AD patients [12, 41]. Dementia patients spend time undressing and dressing for bathing and instead refuse bathing due to high levels of anxiety [42], and the aggressive behaviors of dementia patients were most frequently seen during bathing [43]. In addition, decreased ADL in bathing was reported as the optimal indicator for discriminating between participants with MCI and those with dementia and between participants with moderate dementia and those with severe dementia [44]. The decreased ability to perform ADL functions such as bathing may confer a greater risk of total ADL decrease in aged individuals and impose an increased burden on caregivers. Careful observation of aged AD patients is necessary to identify their functional changes for the early detection and management of decreased ADL.

Rehabilitation with exercise can adequately protect against decreased ADL by improving muscle strength, muscle mass, and balance. The recent meta-analysis showed significant improvement of ADL by physical activity in AD patients [45]. In Japan, commuting services of outpatient rehabilitation (OR) are available for older adults living at home, and OR users showed less ADL decrease and less deterioration in care levels, compared with users of outpatient day care services [46]. However, cognitive impairment impacted negatively on rehabilitation outcomes. Patients with severe cognitive impairment received shorter therapy hours and gained less on the BI compared with patients without cognitive impairment [47]. Baseline ADL and cognitive function and the risk factors for decreased ADL are the essential components to determine the optimal management of ADL in AD patients. In the present study, BI was evaluated only once during the observation period in 65.2% of female patients and in 61.8% of male patients. Routine assessments of baseline ADL and their changes in AD patients are required in all medical facilities. The increased allocation of healthcare resources to caregivers may lead to preserving ADL and extending healthy life expectancy of AD patients.

Limitations

There is difficulty in data linkage among medical care facilities in the JMDC database. If a patient receives treatment in another hospital, information from the other hospital cannot be identified. It is also important to recognize that the JMDC database consists of information about patients treated at large acute care hospitals in Japan, and, therefore, the population may not be representative of the general population of AD patients. In addition, large effect estimates may result from special conditions in a database analysis, if the confidence interval of the odds ratio is too wide in a small subpopulation [48]. Despite these limitations, the findings of the present study provide physicians insight based on real-world evidence regarding the appropriate management of AD patients.

Conclusion

In conclusion, male AD patients with low BMI, stroke, and hip fracture and female AD patients with low BMI and vertebral and hip fractures had increased risks of decreased ADL. It is important to make a reliable ADL evaluation in early stages and to optimize rehabilitation to preserve ADL in AD patients with higher risks for decreased ADL.

Footnotes

ACKNOWLEDGMENTS

The authors have no acknowledgments to report.

FUNDING

This work was funded by Eisai Co. Ltd. The funder was involved in the collection, analysis, and interpretation of data, the writing of the report, and the decision to submit the paper for publication.

CONFLICT OF INTEREST

All authors report personal fees from Eisai Co. Ltd. during the conduct of the study and personal fees from Eisai Co. Ltd. outside the submitted work.

DATA AVAILABILITY

All data disclosed in this manuscript were generated from the Japanese health insurance claims database provided by JMDC. All information referred to in the manuscript text is disclosed in the figures and tables provided.

References

The Alzheimer’s Association (2022) Alzheimer’s disease facts and figures. Alzheimers Dement 18, 700–789.

Livingston

, Huntley

, Sommerlad

, Ames

, Ballard

, Banerjee

, Brayne

, Burns

, Cohen-Mansfield

, Cooper

, Costafreda

, Dias

, Fox

, Gitlin

, Howard

, Kales

, Kivimäki

, Larson

, Ogunniyi

, Orgeta

, Ritchie

, Rockwood

, Sampson

, Samus

, Schneider

, Selbæk

, Teri

, Mukadam

(2020) Dementia prevention, intervention, and care: 2020 report of the Lancet Commission. Lancet 396, 413–446.

, Feng

, Sun

, Hou

, Han

, Liu

(2022) Global, regional, and national burden of Alzheimer’s disease and other dementias, 1990-2019. Front Aging Neurosci 14, 937486.

GBD 2019 Dementia Forecasting Collaborators (2022) Estimation of the global prevalence of dementia in 2019 and forecasted prevalence in 2050: An analysis for the Global Burden of Disease Study 2019. Lancet Public Health 7, e105–e125.

Okamoto

, Komamura

(2022) Towards universal health coverage in the context of population ageing: A narrative review on the implications from the long-term care system in Japan. Arch Public Health 80,, 210.

Monfared

AAT

, Byrnes

, White

, Zhang

(2022) 4. Alzheimer’s disease: Epidemiology and clinical progression. Neurol Ther 11, 553–569.

Nakanishi

, Igarashi

, Ueda

, Brnabic

AJM

, Treuer

, Sato

, Kahle-Wrobleski

, Meguro

, Yamada

, Mimura

, Arai

(2020) Costs and resource use associated with community-dwelling patients with Alzheimer’s disease in Japan: Baseline results from the prospective observational GERAS-J study. J Alzheimers Dis 74, 127–138.

Ashizawa

, Igarashi

, Sakata

, Azuma

, Fujimoto

, Kobayashi

, Takase

, Ikeda

(2021) Impact of the severity of Alzheimer’s disease on the quality of life, activities of daily living, and caregiving costs for institutionalized patients on anti-Alzheimer medications in Japan. J Alzheimers Dis 81, 367–374.

Fukuda

, Ono

, Maeda

, Murata

(2021) Medical care and long-term care expenditures attributable to Alzheimer’s disease onset: Results from the LIFE study. J Alzheimers Dis 84, 807–817.

10.

Kosteniuk

, Osman

, Quail

, Islam

, O’Connell

, Kirk

, Stewart

, Morgan

(2022) Health service use before and after dementia diagnosis: A retrospective matched case-control study. BMJ Open 12, e067363.

11.

Bachmann

, Hruska

(2022) Alzheimer caregiving problems according to ADLs: Evidence from Facebook Support Groups. Int J Environ Res Public Health 19, 6423.

12.

Kawaharada

, Sugimoto

, Matsuda

, Tsuboi

, Sakurai

, Ono

(2019) Impact of loss of independence in basic activities of daily living on caregiver burden in patients with Alzheimer’s disease: A retrospective cohort study. Geriatr Gerontol Int 19, 1243–1247.

13.

Reed

, Belger

, Andrews

, Tockhorn-Heidenreich

, Jones

, Wimo

, Dodel

, Haro

(2020) Factors associated with long-term impact on informal caregivers during Alzheimer’s disease dementia progression: 36-month results from GERAS. Int Psychogeriatr 32, 267–277.

14.

Tsai

C-F

, Hwang

W-S

, Lee

J-J

, Wang

W-F

, Huang

L-C

, Huang

L-K

, Lee

W-J

, Sung

P-S

, Liu

Y-C

, Hsu

C-C

, Fuh

J-L

(2021) Predictors of caregiver burden in aged caregivers of demented older patients. BMC Geriatr 21, 59.

15.

Public Policy Committee, International Society of Pharmacoepidemiology (2016) Guidelines for good pharmacoepidemiology practice (GPP). Pharmacoepidemiol Drug Saf 25, 2–10.

16.

Nagai

, Tanaka

, Kodaira

, Kimura

, Takahashi

, Nakayama

(2020) Data resource profile: JMDC claims databases sourced from Medical Institutions. J Gen Fam Med 21, 211–218.

17.

Fujinaga

, Fukuoka

(2022) A review of research studies using data from the administrative claims databases in Japan. Drugs Real World Outcomes 9, 543–550.

18.

Ministry of Health, Labour and Welfare, Japan. Overview of the Dietary Reference Intakes for Japanese, 2020. https://www.mhlw.go.jp/content/10900000/000862500.pdf.

19.

Suzuki

, Jin

, Ito

, Yoshie

, Ishizaki

, Iijima

, Tamiya

(2022) Factors affecting care-level deterioration among older adults with mild and moderate disabilities in Japan: Evidence from the Nationally Standardized Survey for Care-Needs Certification. Int J Environ Res Public Health 19, 3065.

20.

Sun

, Iwagami

, Sakata

, Ito

, Inokuchi

, Uda

, Hamada

, Ishimaru

, Komiyama

, Kuroda

, Yoshie

, Ishizaki

, Iijima

, Tamiya

(2022) Development and validation of a risk score to predict the frequent emergency house calls among older people who receive regular home visits. BMC Prim Care 23, 132.

21.

Austin

, Yu

AYX

, Vyas

, Kapral

(2021) Applying propensity score methods in clinical research in neurology. Neurology 97, 856–863.

22.

Zhang

, Tomata

, Sugiyama

, Kaiho

, Honkura

, Watanabe

, Tanji

, Sugawara

, Tsuji

(2016) Body mass index and the risk of incident functional disability in elderly Japanese: The OHSAKI Cohort 2006 Study. Medicine (Baltimore) 95,, e4452.

23.

Nishioka

, Wakabayashi

, Maeda

, Shamoto

, Taketani

, Kayashita

, Momosaki

(2020) Body mass index and recovery of activities of daily living in older patients with femoral fracture: An analysis of a national inpatient database in Japan. Arch Gerontol Geriatr 87, 104009.

24.

Bürge

, von Gunten

, Berchtold

(2013) Factors favoring a degradation or an improvement in activities of daily living (ADL) performance among nursing home (NH) residents: A survival analysis. Arch Gerontol Geriatr 56, 250–257.

25.

Amankwaa

, Nelson

, Rook

, Hales

(2022) Association between body mass index, multi-morbidity and activities of daily living among New Zealand nursing home older adults: A retrospective analysis of nationwide InterRAI data. BMC Geriatr 22,, 62.

26.

Kamo

, Nishida

(2014) Direct and indirect effects of nutritional status, physical function and cognitive function on activities of daily living in Japanese older adults requiring long-term care. Geriatr Gerontol Int 14, 799–805.

27.

Mugica-Errazquin

, Zarrazquin

, Seco-Calvo

, Gil-Goikouria

, Rodriguez-Larrad

, Virgala

, Arizaga

, Matilla-Alejos

, Irazusta

, Kortajarena

(2021) The nutritional status of long-term institutionalized older adults is associated with functional status, physical performance and activity, and frailty. Nutrients 13, 3716.

28.

Yoshimura

, Uchida

, Jeong

, Yamaga

(2016) Effects of nutritional supplements on muscle mass and activities of daily living in elderly rehabilitation patients with decreased muscle mass: A randomized controlled trial. J Nutr Health Aging 20, 185–191.

29.

Ihle-Hansen

, Thommessen

, Wyller

, Engedal

, Øksengård

, Stenset

, Løken

, Aaberg

, Fure

(2011) Incidence and subtypes of MCI and dementia 1 year after first-ever stroke in patients without pre-existing cognitive impairment. Dement Geriatr Cogn Disord 32, 401–407.

30.

Stamm

, Burke

, Lin

, Price

, Skolarus

(2019) Disability in community-dwelling older adults: Exploring the role of stroke and dementia. J Prim Care Community Health 10, 2150132719852507.

31.

Lee

P-H

, Yeh

T-T

, Yen

H-Y

, Hsu

W-L

, Chiu

VJ-Y

, Lee

(2021) Impacts of stroke and cognitive impairment on activities of daily living in the Taiwan longitudinal study on aging. Sci Rep 11, 12199.

32.

Wondergem

, Pisters

, Wouters

, Olthof

, de Bie

, Visser-Meily

JMA

, Veenhof

(2017) The course of activities in daily living: Who is at risk for decline after first ever stroke? Cerebrovasc Dis 43, 1–8.

33.

Liang

, Wang

(2017) Alzheimer’s disease is an important risk factor of fractures: A meta-analysis of cohort studies. Mol Neurobiol 54, 3230–3235.

34.

Hsu

, Bleicher

, Waite

, Naganathan

, Blyth

, Handelsman

, Le Couteur

, Seibel

, Cumming

(2018) Community-dwelling older men with dementia are at high risk of hip fracture, but not any other fracture: The Concord Health and Aging in Men Project. Geriatr Gerontol Int 18, 1479–1484.

35.

Gajic-Veljanoski

, Papaioannou

, Kennedy

, Ioannidis

, Berger

, Wong

AKO

, Rockwood

, Kirkland

, Raina

, Thabane

, Adachi

, CaMos Research Group (2018) Osteoporotic fractures and obesity affect frailty progression: A longitudinal analysis of the Canadian multicentre osteoporosis study. BMC Geriatr 18, 4.

36.

Ohta

, Nomura

, Hatanaka

, Osakada

, Matsumoto

, Sasaki

, Tsunoda

, Takemoto

, Tadokoro

, Hishikawa

, Wakutani

, Yamashita

, Sato

, Omote

, Abe

(2019) Female dominant association of sarcopenia and physical frailty in mild cognitive impairment and Alzheimer’s disease. J Clin Neurosci 70, 96–101.

37.

Toth

, Worthy

, Gandra

, Sattar

, Bray

, Cheng

L-I

, Bridges

, Worth

, Dent

, Forbes

, Deshpande

, Ross

, Kleijnen

, Stroes

ESG

(2017) Systematic review and network meta-analysis on the efficacy of evolocumab and other therapies for the management of lipid levels in hyperlipidemia. J Am Heart Assoc 6, e005367.

38.

Wake

, Onishi

, Guelfucci

, Oh

, Hiroi

, Shimasaki

, Teramoto

(2018) Treatment patterns in hyperlipidaemia patients based on administrative claim databases in Japan. Atherosclerosis 272, 145–152.

39.

Chen

, Chen

, Wu

, Lin

, Tao

, Cao

, Shao

, Xiao

(2021) Global, regional and national burden of low back pain 1990-2019: A systematic analysis of the Global Burden of Disease study 2019. J Orthop Translat 32, 49–58.

40.

, March

, Zheng

, Huang

, Wang

, Zhao

, Blyth

, Smith

, Buchbinder

, Hoy

(2020) Global low back pain prevalence and years lived with disability from 1990 to 2017: Estimates from the Global Burden of Disease Study 2017. Ann Transl Med 8, 299.

41.

Andrieu

, Reynish

, Nourhashemi

, Shakespeare

, Moulias

, Ousset

, Sagnier

, Richard

, Albarede

J-L

, Vellas

(2002) Predictive factors of acute hospitalization in 134 patients with Alzheimer’s disease: A one year prospective study. Int J Geriatr Psychiatry 17, 422–426.

42.

Kobayashi

, Yamamoto

(2004) Impact of the stage of dementia on the time required for bathing-related care: A pilot study in a Japanese nursing home. Int J Nurs Stud 41, 767–774.

43.

Schreiner

(2001) Aggressive behaviors among demented nursing home residents in Japan. Int J Geriatr Psychiatry 16, 209–215.

44.

Lee

M-T

, Jang

, Chang

W-Y

(2019) How do impairments in cognitive functions affect activities of daily living functions in older adults? PLoS One 14, e0218112.

45.

Zhou

, Chen

, Liu

, Zhang

, Zhao

, Li

(2022) Physical activity improves cognition and activities of daily living in adults with Alzheimer’s disease: A systematic review and meta-analysis of randomized controlled trials. Int J Environ Res Public Health 19, 1216.

46.

Maruta

, Tabira

, Makizako

, Sagari

, Miyata

, Yoshimitsu

, Han

, Yoshiura

, Kawagoe

(2019) Impact of outpatient rehabilitation service in preventing the deterioration of the care-needs level among Japanese older adults availing long-term care insurance: A propensity score matched retrospective study. Int J Environ Res Public Health 16, 1292.

47.

Dutzi

, Schwenk

, Kirchner

, Jooss

, Bauer

, Hauer

(2021) Influence of cognitive impairment on rehabilitation received and its mediating effect on functional recovery. J Alzheimers Dis 84, 745–756.

48.

Tzeng

(2021) To handle the inflation of odds ratios in a retrospective study with a profile penalized log-likelihood approach. J Clin Lab Anal 35, e23849.