Received: 19 August 2023

DOI: 10.1002/gps.6029

RESEARCH ARTICLE
- Accepted: 12 November 2023

Association of anticholinergic drug exposure with the risk of

dementia among older adults in Japan: The LIFE Study

Yuki Okita1 | Tetsuhisa Kitamura1 | Sho Komukai2 | Ling Zha1 |

Masayo Komatsu1 | Nobuhiro Narii1 | Fumiko Murata3 | Maeda Megumi3 |
Yasufumi Gon4 | Yasuyoshi Kimura4 | Kosuke Kiyohara5 | Tomotaka Sobue1 |
Haruhisa Fukuda3

1
Division of Environmental Medicine and
Population Sciences, Department of Social Abstract
Medicine, Graduate School of Medicine, Osaka
Objectives: Several studies have investigated that anticholinergic drugs cause
University, Suita, Japan
2 cognitive impairment. However, the risk of dementia associated with anticholiner-
Biomedical Statistics, Department of
Integrated Medicine, Graduate School of gics has not been extensively investigated in the super‐aging society of Japan. We
Medicine, Osaka University, Suita, Japan
conducted this study to assess the association between anticholinergic drugs and
3
Department of Health Care Administration
and Management, Kyushu University Graduate
the risk of dementia in older adults in Japan.
School of Medical Sciences, Fukuoka, Japan Methods: This nested case‐control study used data from the Longevity Improve-
4
Department of Neurology, Graduate School of ment & Fair Evidence Study, which includes claim data in Japan from 2014 to 2020.
Medicine, Osaka University, Suita, Japan
5
We included 66,478 cases of diagnosed dementia and 328,919 matched controls
Department of Food Science, Faculty of Home
Economics, Otsuma Women's University, aged ≥65 years, matched by age, sex, municipality, and cohort entry year. Primary
Chiyoda‐ku, Japan exposure was the total cumulative anticholinergic drugs prescribed from cohort

Correspondence
entry date to event date or matched index date, which was the total standardized
Tetsuhisa Kitamura, Division of Environmental daily doses for each patient, calculated by adding the total dose of different types
Medicine and Population Sciences,
of anticholinergic drugs in each prescription, divided by the World Health
Department of Social Medicine, Graduate
School of Medicine, Osaka University, Suita Organization‐defined daily dose values. Odds ratios for dementia associated with
565‐0871, Japan.
cumulative exposure to anticholinergic drugs were calculated using conditional lo-
Email: [email protected]
gistic regression adjusted for confounding variables.
Funding information Results: The mean (standard deviation) age at index date was 84.3 (6.9), and the
Japan Agency for Medical Research and
Development, Grant/Award Number: percentage of women was 62.1%. From cohort entry date to event date or matched
JP21nf0101635; Grants‐in‐Aid for Scientific index date, 18.8% of the case patients and 13.7% of the controls were prescribed at
Research, Fundamental Research A from the
Japan Society for the Promotion of Science, least one anticholinergic drug. In the multivariable‐adjusted model, individuals with
Grant/Award Numbers: JP19K21590, anticholinergic drugs prescribed had significantly higher odds of being diagnosed
P20H00563; JST FOREST Program, Grant/
Award Number: JPMJFR205J with dementia (adjusted odds ratio, 1.50 [95% confidence interval, 1.47–1.54]).
Among specific types of anticholinergic drugs, a significant increase in risk was
observed with the use of antidepressants, antiparkinsonian drugs, antipsychotics,
and bladder antimuscarinics in a fully multivariable‐adjusted model.
Conclusions: Several types of anticholinergic drugs used by older adults in Japan are
associated with an increased risk of dementia. These findings suggest that the un-
derlying risks should be considered alongside the benefits of prescribing anticho-

-
linergic drugs to this population.

Int J Geriatr Psychiatry. 2023;e6029. wileyonlinelibrary.com/journal/gps © 2023 John Wiley & Sons Ltd. 1 of 12
https://doi.org/10.1002/gps.6029
 10991166, 2023, 12, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/gps.6029 by Kinneret Academic College, Wiley Online Library on [18/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
2 of 12
- OKITA ET AL.

KEYWORDS
anticholinergic drugs, dementia, elderly population, nested case‐control study

Key points
� The risk of dementia associated with anticholinergics has not been extensively investigated
in the super‐aging society of Japan.
� This nested case‐control study found a statistically significant association between anti-
cholinergic drug use and the risk of dementia, especially for antipsychotics, antiparkinsonian
agents, antidepressants, and bladder antimuscarinics, after adjusting for confounding
variables.
� The prescription of specific types of anticholinergic drugs to older adults should be made
with caution due to the associated risk of dementia.

1 | INTRODUCTION Longevity Improvement & Fair Evidence (LIFE) Study to investigate

the association of anticholinergic drugs with the risk of dementia in
Dementia is a major cause of disability and dependency among older older adults in Japan.
adults worldwide. Approximately 50 million people are living with
dementia,1 globally. As of 2012, approximately 4.6 million people
have dementia in Japan,2 and the number of people with dementia in 2 | MATERIALS AND METHODS
2025 and 2060 is estimated to be approximately 6.5–7 million and
8.5–11.5 million, respectively.3 2.1 | Study data
Several factors that increase the risk of dementia, including age,
genetics, and lifestyle have been identified.4–6 Recently, there has This was a nested case‐control study using data from the LIFE Study;
been growing concern regarding the role of medications in the the detailed design and participant background are presented
development of dementia. Anticholinergic drugs comprise a diverse elsewhere.16
class of medications that exert pharmacological effects by inhibiting The LIFE Study is a longitudinal cohort study that links admin-
the action of the neurotransmitter acetylcholine in both the central istrative claims data, health checkups, and residence‐related infor-
and peripheral nervous systems, and are commonly prescribed to mation collected from the National Health Insurance (NHI), the
treat a variety of conditions such as allergies, overactive bladder, and Latter Stage Older Persons Health Care (LSHC) system, and the
depression.7 In the older population, these drugs may induce short‐ Long‐Term Care (LTC) insurance systems through participating mu-
term negative effects such as cognitive impairment and memory nicipalities. The NHI and LSHC systems cover medical care services.
deficits, owing to their effects on the cholinergic system. Several NHI enrollees comprise mainly retirees aged ≤74 years, self‐
studies have investigated the association between anticholinergic employed people, non‐regularly employed people, and primary in-
drug use and cognitive impairment in older adults.7–10 A meta‐ dustry workers, and LSHC enrollees comprise all residents aged
analysis and multiple case‐control studies reported that the use of ≥75 years. The LTC insurance system covers LTC services, and the
anticholinergic drugs was associated with an increased risk of de- enrollees comprise residents aged ≥65 years with certified needs and
mentia,11–13 especially when taking higher cumulative doses and residents aged 40–64 years with debilitating disease. The number of
specific types of anticholinergic drugs.10 municipalities participating in the LIFE Study varies monthly.
As the global population ages, the number of people living with The study was approved by the Kyushu University Institutional
dementia increases, and the associated societal and economic burden Review Board for Clinical Research (approval no. 2021‐70) and the
increases.4 The number of patients with dementia in Japan is pre- Osaka University Institutional Review Board (approval no. 21107).
dicted to increase to 38.4 per 1000 population in 2037, compared to
the Organisation for Economic Co‐operation and Development
(OECD) average of 17.3.14,15 As Japan's older population represents 2.2 | Study participants
a potential future scenario for other countries, it is important to
investigate the association of anticholinergic drugs with dementia We used data from 17 municipalities in Japan that provided medical
risk in clinical practice in the older population. However, this asso- and LTC services. The study period was from April 2014 to December
ciation has not been extensively investigated in Japan. We conducted 2020, and 1,693,933 individuals were included. The number of study
a nested case‐control study using large‐scale claims data from the participants and duration of claims data, including the initial and last
 10991166, 2023, 12, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/gps.6029 by Kinneret Academic College, Wiley Online Library on [18/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
OKITA ET AL.
- 3 of 12

claims data for each municipality, are shown in Supplemental Table 1. period for each case. A random selection of five controls was con-
We set a time period of 6 months from the date of the initial claims ducted based on the date of case identification by age at the cohort
data (lookback period) to define baseline covariates, medical history, entry date, sex, municipality (17 municipalities), and year of cohort
and previous medication for each individual. Therefore, the cohort entry (2014–2020). The final analysis included 66,478 cases and
entry date for each individual was recorded 7 months after the first 328,919 controls (Figure 1).
claim was recorded. The cohort was followed up until the last claim
was recorded (due to migration outside the study area, enrollment in
another insurance plan, or death) or until the end date for each 2.3 | Exposures
municipality (Supplemental Figure 1).
A flowchart of the study participants and the selection of cases The identification of drugs with anticholinergic properties lacks
and controls is shown in Figure 1. The exclusion criteria were as consensus. We adopted the method developed by Gray et al.,19 which
follows: (1) individuals with less than six months of health insurance involves selecting drugs identified as having strong anticholin-
coverage were excluded (n = 219,933) because their baseline cova- ergic properties and including additional drugs20,21 referring to the
riates and medical history could not be defined, (2) individuals approach by Coupland et al.10 Our study included 42 anticholinergic
younger than 65 years at the cohort entry date (n = 576,887), (3) drugs prescribed to the study participants that were approved
those with a recorded diagnosis of dementia or Parkinson's disease for prescription in Japan during the study period (Supplemental
during the lookback period (n = 84,733), and (4) individuals who were Table 2).
prescribed the anticholinergic drugs specified in Table 1 as previous The primary exposure variable was the total cumulative expo-
medications during the lookback period were excluded (n = 73,641), sure to anticholinergic drugs, which combined different types of
leaving 735,739 individuals for matching. Individuals with Parkinson's anticholinergic medications, based on a previous study.19 The total
disease were excluded to reduce indication bias because cognitive dose of each prescription was calculated by multiplying the number
impairment is common in patients with Parkinson's disease17 and of prescribed units (e.g., tablets, injections, and solutions) by the dose
they are frequently treated with anticholinergic drugs. Medical his- included in each unit. The total dose was then divided by WHO ‐
tory was defined using the International Classification of Diseases defined daily dose values22 to standardize the prescribed doses.
and Related Health Problems, 10th Revision (ICD‐10) codes for de- The standardized values were summed for all anticholinergic pre-
mentia (ICD‐10 codes: F00 (G30), F01‐03, and G31.8) and Parkin- scriptions during the follow‐up period to obtain the total standard-
son's disease (ICD‐10 code: G20). ized daily dose (TSDD) for each patient. For profenamine and
The case group included patients diagnosed with dementia (ICD‐ piperidolate, the defined daily dose values were not available in the
10 codes F00 (G30), F01‐03, and G31.8) during follow‐up. The database. These values were sourced by referring to the dosage and
diagnosis was based on the WHO definition; G31.8 was added to administration of each drug in the package inserts.
include Lewy body dementia. Five controls for each case were We divided the anticholinergic exposure variables into five cat-
18
sampled from the participants for analysis by risk set sampling. The egories: (non‐use, 1–10, 11–30, 31–90, and >90 TSDDs). The asso-
date of dementia diagnosis for case was the index date, and matched ciations for each of the 10 different types of anticholinergic drugs
controls were assigned from all other subjects that has not yet been were also examined using three categories (non‐use, 1–30, and >30
diagnosed with dementia on the index date with the same follow‐up TSDDs).

F I G U R E 1 Flowchart of the selection

of the study participants and cases and
their controls. For each case, we performed
a random selection of five event‐free
controls based on the identification date.
The matching criteria were age at the
cohort entry date, sex, municipality (17
municipalities), and year of cohort entry
(2014, 2015, 2016, 2017, 2018, 2019, or
2020). AD, Alzheimer's disease; LIFE,
Longevity Improvement & Fair Evidence.
 10991166, 2023, 12, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/gps.6029 by Kinneret Academic College, Wiley Online Library on [18/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
4 of 12
- OKITA ET AL.

TABLE 1 Characteristics of cases with dementia and their controls.

Case Control
(n = 66,478) (n = 328,919)

n % n % p‐value

Age at index date, mean (SD), y 84.3 (6.9) 84.2 (6.8)

Age at cohort entry date, y 0.091

65–74 9214 13.9 46,033 14.0

75–84 31,633 47.6 157,547 47.9

˃85 25,631 38.6 125,339 38.1

Sex 0.841

Male 25,169 37.9 124,395 37.8

Female 41,309 62.1 204,524 62.2

Municipality 0.844

A 20,875 31.4 104,301 31.7

B 8640 13.0 43,067 13.1

C 8690 13.1 43,221 13.1

D 8302 12.5 41,107 12.5

E 3456 5.2 17,069 5.2

F 1692 2.6 8280 2.5

G 2758 4.2 13,483 4.1

H 2042 3.1 10,012 3.0

I 1475 2.2 7156 2.2

J 1537 2.3 7402 2.3

K 1467 2.2 7108 2.2

L 1435 2.2 6913 2.1

M 1187 1.8 5723 1.7

N 743 1.1 3508 1.1

O 1135 1.7 5525 1.7

P 589 0.9 2850 0.9

Q 455 0.7 2194 0.7

Cohort entry year <0.001

2014 13,968 21.0 69,706 21.2

2015 27,572 41.5 137,182 41.7

2016 2502 3.8 11,446 3.5

2017 21,015 31.6 104,329 31.7

2018 1035 1.6 4633 1.4

2019 350 0.5 1510 0.5

2020 36 0.1 113 0.0

Comorbidity (yes)

Chronic heart failure 15,331 23.1 69,493 21.1 <0.001

Cardiac arrhythmias 13,074 19.7 63,807 19.4 0.112

Valvular disease 5585 8.4 27,683 8.4 0.898

Pulmonary circulation disorders 367 0.6 1464 0.4 <0.001

 10991166, 2023, 12, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/gps.6029 by Kinneret Academic College, Wiley Online Library on [18/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
OKITA ET AL.
- 5 of 12

T A B L E 1 (Continued)

Case Control
(n = 66,478) (n = 328,919)

n % n % p‐value

Peripheral vascular disease 9480 14.3 47,715 14.5 0.100

Hypertension 43,202 65.0 221,060 67.2 <0.001

Chronic pulmonary disease 13,773 20.7 69,616 21.2 0.010

Diabetes 6080 9.2 25,751 7.8 <0.001

Renal failure 4595 6.9 18,314 5.6 <0.001

Liver disease 12,956 19.5 65,492 19.9 0.013

Cancer 8021 12.1 41,286 12.6 0.001

Psychosis 1164 1.8 2221 0.7 <0.001

Depression 4399 6.6 14,803 4.5 <0.001

Number of months with outpatient visit in lookback period <0.001

0 686 1.0 2470 0.8

1 318 0.5 833 0.3

2 554 0.8 1553 0.5

3 1036 1.6 3461 1.1

4 2428 3.7 9582 2.9

5 4484 6.8 18,973 5.8

6 56,972 85.7 292,047 88.8

Certified LTC need level <0.001

No support 45,739 68.8 264,782 80.5

Support need level 1 or 2 6433 9.7 29,444 9.0

LTC need level 1 4328 6.5 9783 3.0

LTC need level 2 3698 5.6 9610 2.9

LTC need level 3 2623 4.0 6503 2.0

LTC need level 4 2335 3.5 5552 1.7

LTC need level 5 1322 2.0 3245 1.0

Abbreviation: LTC, Long‐Term Care.

2.4 | Covariates during the lookback period was calculated to measure the individual
behavior of healthcare provider visits.
We set a lookback period to define the baseline covariates for each
individual. We used ICD‐10 codes to define comorbidities during the
lookback period based on algorithms using administrative data.23,24 2.5 | Statistical analysis
According to previous studies, we included patients with chronic
heart failure, cardiac arrhythmias, valvular disease, pulmonary cir- We used conditional logistic regression to assess the association
culation disorders, peripheral vascular disease, hypertension, chronic between anticholinergic drugs and dementia by calculating odds ra-
pulmonary disease, diabetes, renal failure, liver disease, cancer, psy- tios (ORs) with 95% confidence intervals (CIs). We compared in-
chosis, and depression.5,10,12,25 LTC need levels (support need levels dividuals who were prescribed anticholinergic drugs from the cohort
1–2 and care need levels 1–5) were certified by the municipality entry date to the event date or matched index date with those who
according to the degree of disability and instrumental activities of were not prescribed anticholinergic drugs. In the unadjusted model,
26
daily living. The LTC need level of the participant was defined as matching factors (age at cohort entry date, sex, municipality, and year
the maximum LTC need level during the lookback period. In addition, of cohort entry) were the only factors. In the multivariable‐adjusted
the number of months through which outpatient visits were observed model, we adjusted for chronic heart failure, cardiac arrhythmias,
 10991166, 2023, 12, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/gps.6029 by Kinneret Academic College, Wiley Online Library on [18/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
6 of 12
- OKITA ET AL.

valvular disease, pulmonary circulation disorders, peripheral vascular higher in cases than in controls, whereas the prevalence of hyper-
disease, hypertension, chronic pulmonary disease, diabetes, renal tension was higher in controls. The demographic characteristics of
failure, liver disease, cancer, psychosis, and depression (yes or no) as the patients with AD are shown in Supplemental Table 3.
comorbidities, the number of months with outpatient visits over a 6‐ From the cohort entry date to the event date or matched index
month period (0, 1, 2, 3, 4, 5, and 6), and LTC need levels (no support, date, 18.8% of the cases and 13.7% of the controls were pre-
support 1–2, care 1, care 2, care 3, care 4, and care 5). In the fully scribed at least one anticholinergic drug (Supplemental Table 4). The
adjusted model used for the analysis of each of the 10 different types most frequently prescribed anticholinergic drugs were antihistamines
of anticholinergic drugs, the use of other types of anticholinergic (6.9% of cases and 5.4% of controls), bladder antimuscarinics (4.7% of
drugs was used as a confounding factor. cases and 3.4% of controls), and antipsychotics (3.5% of cases and
We conducted the following two sensitivity analyses. 0.7% of controls). Supplemental Table 4 presents the TSDDs for each
type of anticholinergic drug in patients with one or more pre-
(1) We assessed this association in patients diagnosed with Alz- scriptions. Bladder antimuscarinics had the highest total cumulative
heimer's disease (AD) (ICD‐10 code: F00 (G30)). use (median (IQR) 63.0 (18.7–203.0) for cases and 57.0 (16.3–199.0)
(2) To eliminate protopathic bias, we excluded individuals with a for controls), followed by antidepressants and antiarrhythmic drugs.
recorded diagnosis of psychosis or depression during the look- The risk of dementia associated with the use of any type of
back period. anticholinergic drug is shown in Table 2. In the multivariable‐adjusted
model, the OR was significant (adjusted OR [aOR], 1.50 [95% CI,
All hypothesis testing in this study were two‐sided, and the sig- 1.47–1.54]). Subgroup analysis by sex or age group also showed re-
nificance level was set at p < 0.05. Statistical analysis was performed sults similar to those of the total population. The risk was higher in
using STATA software (version 16.0; Stat Corporation, College Sta- men than in women (aOR, 1.62 [95% CI, 1.56–1.68] for men and aOR,
tion, TX, USA). 1.43 [95% CI, 1.39–1.47] for women), and the highest risk was
observed in the category of 65–74 years (aOR, 1.78 [95% CI, 1.67–
1.89] for 65–74 years of age, and aOR, 1.45 [95% CI, 1.39–1.51] for
3 | RESULTS ≥85 years of age). Significant trends were observed in all analyses of
the total population and subgroups, and the interaction between sex
A total of 735,739 individuals from 17 municipalities included in the and age group was statistically significant (p < 0.001 for both). Similar
LIFE Study data were eligible for the analysis (Figure 1). Among results were obtained in the AD group (Supplemental Table 5).
these, 66,478 (9.0%) individuals were matched with 328,919 controls We also investigated the risks associated with the total cumu-
without dementia during the follow‐up period (median 1.8 years, IQR lative use of any type of anticholinergic drug (Table 3). The aOR
0.9–3.0). The demographic characteristics are shown in Table 1. increased from 1.46 (95% CI, 1.42–1.50) for 1–10 TSDDs to 1.72
Cases had a mean (SD) age of 84.3 (6.9) years at diagnosis, and 62.1% (95% CI, 1.63–1.82) for >90 TSDDs (P for trend <0.001). A similar
were women. The prevalence of almost all comorbidities was slightly statistical significance was observed in the subgroup analysis by sex

TABLE 2 Association between anticholinergic drug (any) use and risk of dementia.

Unadjusted model Adjusted model

Case (n = 66,478) Control (n = 328,919) P For
n (%) n (%) OR 95% CI OR 95% CI interaction

All 12,516 (18.8) 45,042 (13.7) 1.52 (1.48–1.55) 1.50 (1.47–1.54)

Sex <0.001

Male 5100 (20.3) 17,308 (13.9) 1.64 (1.58–1.70) 1.62 (1.56–1.68)

Female 7416 (17.8) 27,734 (13.6) 1.44 (1.40–1.48) 1.43 (1.39–1.47)

Age groups <0.001

65–74 2085 (22.6) 6439 (14.0) 1.89 (1.78–2.00) 1.78 (1.67–1.89)

75–84 6514 (20.6) 24,043 (15.3) 1.48 (1.44–1.53) 1.45 (1.41–1.50)

≥85 3917 (15.3) 14,560 (11.6) 1.42 (1.36–1.48) 1.45 (1.39–1.51)

Note: The ‘adjusted’ model was adjusted for chronic heart failure, cardiac arrhythmias, valvular disease, pulmonary circulation disorders, peripheral
vascular disease, hypertension, chronic pulmonary disease, diabetes, renal failure, liver disease, cancer, psychosis, and depression (yes or no) for
comorbidities, the number of months with outpatient visits over a 6‐month period (0, 1, 2, 3, 4, 5, and 6), and LTC needs (no support, support 1–2, care 1,
care 2, care 3, care 4, and care 5).
Abbreviations: CI, confidence interval, OR, odds ratio.
 10991166, 2023, 12, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/gps.6029 by Kinneret Academic College, Wiley Online Library on [18/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
OKITA ET AL.
- 7 of 12

TABLE 3 Overall, sex‐based, and age‐based associations between cumulative use of anticholinergic drugs (any) and risk of dementia.

Case Control
Unadjusted model Adjusted model
(n = 66,478) (n = 328,919) P for P for
n (%) n (%) OR 95% CI OR 95% CI trend interaction

All participants

Non‐use 53,962 (81.2) 283,877 (86.3) Ref Ref <0.001

1–10 8150 (12.3) 30,307 (9.2) 1.47 (1.43–1.51) 1.46 (1.42–1.50)

11–30 1452 (2.2) 5343 (1.6) 1.48 (1.39–1.57) 1.47 (1.38–1.56)

31–90 1165 (1.8) 3912 (1.2) 1.62 (1.52–1.73) 1.57 (1.47–1.68)

>90 1749 (2.6) 5480 (1.7) 1.76 (1.66–1.86) 1.72 (1.63–1.82)

Sex <0.001

Male

Non‐use 20,069 (79.7) 107,087 (86.1) Ref Ref <0.001

1–10 3417 (13.6) 12,049 (9.7) 1.58 (1.51–1.64) 1.56 (1.49–1.63)

11–30 550 (2.2) 1826 (1.5) 1.66 (1.51–1.83) 1.66 (1.50–1.83)

31–90 443 (1.8) 1474 (1.2) 1.67 (1.50–1.86) 1.64 (1.47–1.84)

>90 690 (2.7) 1959 (1.6) 1.98 (1.81–2.17) 1.97 (1.80–2.16)

Female

Non‐use 33,893 (82.1) 176,790 (86.4) Ref Ref <0.001

1–10 4733 (11.5) 18,258 (8.9) 1.40 (1.35–1.45) 1.40 (1.35–1.45)

11–30 902 (2.2) 3517 (1.7) 1.38 (1.28–1.49) 1.37 (1.27–1.48)

31–90 722 (1.8) 2438 (1.2) 1.59 (1.46–1.73) 1.54 (1.41–1.68)

>90 1059 (2.6) 3521 (1.7) 1.64 (1.52–1.76) 1.59 (1.48–1.71)

Age groups <0.001

65–74

Non‐use 7129 (77.4) 39,594 (86.0) Ref Ref <0.001

1–10 1348 (14.6) 4657 (10.1) 1.69 (1.58–1.81) 1.62 (1.51–1.74)

11–30 261 (2.8) 686 (1.5) 2.21 (1.91–2.56) 2.11 (1.81–2.45)

31–90 198 (2.2) 519 (1.1) 2.20 (1.87–2.60) 1.98 (1.66–2.36)

>90 278 (3.0) 577 (1.25) 2.83 (2.44–3.27) 2.45 (2.10–2.86)

75–84

Non‐use 25,119 (79.4) 133,504 (84.7) Ref Ref <0.001

1–10 4245 (13.4) 16,161 (10.3) 1.44 (1.38–1.49) 1.42 (1.37–1.48)

11–30 723 (2.3) 2907 (1.9) 1.36 (1.25–1.48) 1.31 (1.21–1.43)

31–90 613 (1.9) 2081 (1.3) 1.61 (1.47–1.77) 1.55 (1.41–1.70)

>90 933 (3.0) 2894 (1.8) 1.78 (1.65–1.92) 1.70 (1.58–1.84)

≥85

Non‐use 21,714 (84.7) 110,779 (88.4) Ref Ref <0.001

1–10 2557 (10.0) 9489 (7.6) 1.42 (1.36–1.49) 1.45 (1.38–1.52)

11–30 468 (1.8) 1750 (1.4) 1.40 (1.26–1.55) 1.45 (1.30–1.61)

31–90 354 (1.4) 1312 (1.1) 1.42 (1.26–1.60) 1.43 (1.26–1.61)

>90 538 (2.1) 2009 (1.6) 1.43 (1.30–1.58) 1.47 (1.33–1.62)

Note: The ‘adjusted’ model was adjusted for chronic heart failure, cardiac arrhythmias, valvular disease, pulmonary circulation disorders, peripheral
vascular disease, hypertension, chronic pulmonary disease, diabetes, renal failure, liver disease, cancer, psychosis, and depression (yes or no) for
comorbidities, the number of months with outpatient visits over a 6‐month period (0, 1, 2, 3, 4, 5, and 6), and LTC needs (no support, support 1–2, care 1,
care 2, care 3, care 4, and care 5).
Abbreviations: CI, confidence interval; OR, odds ratio.
 10991166, 2023, 12, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/gps.6029 by Kinneret Academic College, Wiley Online Library on [18/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
8 of 12
- OKITA ET AL.

and age. The risk was higher in men than in women and lower in the subgroup analysis stratified by age, and a significant interaction was
older age group. Analyses of patients diagnosed with AD were similar observed among the age groups. Consistent with the findings of
(Supplemental Table 6). When the categorization of total cumulative previous studies, the younger group in elderly population had the
use was changed to three categories (non‐use, 1–30, and >30 highest risk.10,34,35
TSDDs), the results did not change (Supplemental Table 7). In the analysis of cumulative use of any type of anticholinergic
Among the specific types of anticholinergic drugs, a significant drug, we found that the dementia risk associated with the use of
increase in risk was observed with the use of antidepressants, anti- anticholinergic drugs increased by approximately 70% for the cate-
parkinsonian agents, antipsychotics, and bladder antimuscarinics in gory of the highest TSDD (>90 TSDDs), and a dose‐dependent
the fully multivariable‐adjusted model (Table 4). Fully adjusted odds relationship existed. This finding is consistent with those of previ-
ratios (faORs) in the highest exposure category (>30 TSDDs) ous studies, in which approximately 50% increase in risk was
compared with those of non‐use were 1.84 (95% CI, 1.59–2.13) for observed.10,19 The potential mechanisms by which anticholinergics
antidepressants, 4.73 (95% CI, 3.19–7.01) for antiparkinsonian affect dementia risk have been reported; anticholinergic drugs may
agents, 2.98 (95% CI, 2.43–3.65) for antipsychotics, and 1.47 (95% CI, reduce brain glucose metabolism, increase brain atrophy,36,37 and
1.40–1.55) for bladder antimuscarinics. Statistically significant trends reduce cholinergic transmission in the brain.38 Based on these
in the TSDD category were observed for all four types. In addition, mechanisms, higher doses of anticholinergic drugs may have a
the trends for antihistamines, antivertigo/antiemetics, and antiepi- greater effect on the nervous system and potentially increase the risk
leptics were statistically significant in the TSDD category. There was of dementia. In the subgroup analysis of sex and age, higher risk was
a significantly increased risk associated with 1–30 TSDD, whereas observed in the higher TSDD group.
the highest exposure category of these types did not show a signif- A large sample size from the LIFE Study enabled the analysis of
icant increase in risk. The results were similar when the analysis was anticholinergic drug‐associated dementia risk. This analysis is
conducted on the outcomes of diagnosed AD (Supplemental Table 8). important since bladder antimuscarinics occupy a large proportion of
In the sensitivity analysis excluding patients diagnosed with psy- the total cumulative use of any anticholinergic drug used for this
chosis or depression during the lookback period, the results were not analysis due to long‐term prescription for chronic disease of the
significantly different (Supplemental Table 9). overactive bladder. Statistically significant associations with demen-
tia were observed for antipsychotics, antiparkinsonian agents, anti-
depressants, and bladder antimuscarinics, which is consistent with
4 | DISCUSSION previous reports.10,39–42 The issue of reverse causality must be
considered for antipsychotics, antiparkinsonian agents, and antide-
This nested case‐control study of a large cohort, based on claims pressants. The degenerative process in dementia probably starts
data, found an increased risk of dementia associated with the use of before cognitive impairment or dementia can be diagnosed,43 and
anticholinergic drugs, in older adults aged ≥65 years in Japan. Sub- some medications may be prescribed because it is difficult to
group analyses of sex and age reported that the risk of dementia distinguish between prodromal symptoms and symptoms related to
associated with anticholinergic drug use was higher in men and those diseases of the nervous system, such as psychosis and depression. In
aged 65–74 years. Among anticholinergic drugs, a stronger associa- addition to the exclusion of participants with Parkinson's disease,
tion was observed with antipsychotics, antiparkinsonian agents, an- participants diagnosed with psychosis and depression during the
tidepressants, and bladder antimuscarinics, whereas no significant lookback period were excluded from the sensitivity analysis, and
increase in the risk was observed for skeletal muscle relaxants, there were no significant differences in findings from this analysis.
gastrointestinal antispasmodics, and antiarrhythmics. The criteria Another sensitivity analysis was conducted with the outcome of an
and guidelines issued in other countries suggest that anticholinergic AD diagnosis, which is the most common form of dementia, and
drugs should be used with caution in older adults due to the poten- similar results were obtained.
tial risk of cognitive impairment.27,28 The guidelines in Japan Our study had several limitations. First, some patients may not
also recommend considering dose reduction or withdrawal of anti- have taken their medications or doses as prescribed, leading to
cholinergic prescriptions in older adults.29 Further investigations of exposure misclassification. This misclassification might potentially
the effects of anticholinergic drugs using Japanese data can reinforce reduce the aORs/faORs and may explain the lack of association for
the evidence in the guidelines. some drug types. Second, the follow‐up period was shorter than
Through the subgroup analysis of sex, our findings suggest that previous studies10 due to the limited duration of claim data collec-
the dementia risk associated with anticholinergics is increased in tion. A longer follow‐up period may increase the number of cases and
both sexes, and the risk may be more pronounced in men than cumulative use of anticholinergic drugs for each participant, which
women based on the significant interaction observed between sexes, could enable analysis of the long‐term and dose‐dependent effects of
which is consistent with the results of previous studies.30,31 However, specific types of anticholinergics. Third, the duration of lookback
sex differences in the causes of cognitive impairment have been period was 6 months in this study, since it is difficult to set longer
inconsistent in previous studies32 and needs further investigation.33 lookback period considering the median of the follow‐up period was
In addition, all age groups showed a significant increase in risk in the 1.8 years. This short period of lookback period could cause the
 10991166, 2023, 12, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/gps.6029 by Kinneret Academic College, Wiley Online Library on [18/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
OKITA ET AL.
- 9 of 12

TABLE 4 Association between cumulative use of type‐wise anticholinergic drugs and risk of dementia.

Case Control
Unadjusted model Adjusted model Fully adjusted model
(n = 66,478) (n = 328,919) P For
n (%) n (%) OR 95% CI OR 95% CI OR 95% CI trend

Antihistamines

Non‐use 61,907 (93.1) 311,175 (94.6) Ref Ref Ref <0.001

1–30 4398 (6.6) 16,974 (5.2) 1.32 (1.28–1.37) 1.31 (1.26–1.35) 1.21 (1.17–1.25)

> 30 173 (0.3) 770 (0.2) 1.14 (0.97–1.35) 1.05 (0.89–1.24) 1.02 (0.86–1.21)

Antidepressants

Non‐use 65,712 (98.9) 326,907 (99.4) Ref Ref Ref <0.001

1–30 491 (0.7) 1365 (0.4) 1.80 (1.63–2.00) 1.74 (1.57–1.94) 1.58 (1.42–1.76)

> 30 275 (0.4) 647 (0.2) 2.14 (1.86–2.47) 1.97 (1.71–2.28) 1.84 (1.59–2.13)

Antivertigo/antiemetic

Non‐use 65,711 (98.9) 326,429 (99.2) Ref Ref Ref <0.001

1–30 747 (1.1) 2430 (0.7) 1.55 (1.42–1.68) 1.54 (1.42–1.68) 1.40 (1.29–1.53)

> 30 20 (0.0) 60 (0.0) 1.68 (1.01–2.78) 1.32 (0.79–2.21) 1.15 (0.68–1.94)

Antiparkinsonian agents

Non‐use 66,135 (99.5) 328,584 (99.9) Ref Ref Ref <0.001

1–30 276 (0.4) 290 (0.1) 4.74 (4.02–5.60) 4.18 (3.52–4.95) 3.67 (3.09–4.35)

> 30 67 (0.1) 45 (0.0) 7.52 (5.15–10.98) 5.29 (3.58–7.83) 4.73 (3.19–7.01)

Antipsychotics

Non‐use 64,182 (96.6) 326,453 (99.3) Ref Ref Ref <0.001

1–30 2108 (3.2) 2245 (0.7) 4.84 (4.56–5.14) 4.36 (4.09–4.64) 4.10 (3.85–4.37)

> 30 188 (0.3) 221 (0.1) 4.39 (3.61–5.34) 3.17 (2.59–3.87) 2.98 (2.43–3.65)

Bladder antimuscarinics

Non‐use 63,238 (95.3) 317,736 (96.6) Ref Ref Ref <0.001

1–30 1126 (1.7) 4361 (1.3) 1.31 (1.23–1.40) 1.34 (1.25–1.43) 1.29 (1.21–1.38)

> 30 2024 (3.0) 6822 (2.1) 1.51 (1.44–1.59) 1.52 (1.44–1.60) 1.47 (1.40–1.55)

Skeletal muscle relaxants

Non‐use 64,821 (97.5) 320,684 (97.5) Ref Ref Ref 0.651

1–30 1561 (2.4) 7798 (2.4) 1.00 (0.94–1.06) 1.05 (0.99–1.11) 1.01 (0.95–1.07)

> 30 96 (0.1) 437 (0.1) 1.10 (0.88–1.37) 1.10 (0.88–1.38) 1.05 (0.84–1.32)

Gastrointestinal antispasmodics

Non‐use 64,629 (97.2) 320,817 (97.5) Ref Ref Ref 0.532

1–30 1831 (2.8) 8009 (2.4) 1.14 (1.09–1.20) 1.15 (1.09–1.22) 1.02 (0.97–1.08)

> 30 18 (0.0) 93 (0.0) 0.97 (0.59–1.61) 0.88 (0.53–1.47) 0.82 (0.49–1.36)

Antiarrhythmic

Non‐use 66,378 (99.9) 328,407 (99.8) Ref Ref Ref 0.939

1–30 68 (0.1) 365 (0.1) 0.93 (0.71–1.20) 0.95 (0.73–1.24) 0.91 (0.70–1.18)

> 30 32 (0.1) 147 (0.0) 1.09 (0.74–1.60) 1.15 (0.78–1.69) 1.10 (0.74–1.62)

(Continues)
 10991166, 2023, 12, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/gps.6029 by Kinneret Academic College, Wiley Online Library on [18/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
10 of 12
- OKITA ET AL.

T A B L E 4 (Continued)

Case Control
Unadjusted model Adjusted model Fully adjusted model
(n = 66,478) (n = 328,919) P For
n (%) n (%) OR 95% CI OR 95% CI OR 95% CI trend

Antiepileptics

Non‐use 66,185 (99.6) 328,035 (99.7) Ref Ref Ref <0.001

1–30 242 (0.4) 714 (0.2) 1.69 (1.46–1.95) 1.60 (1.38–1.86) 1.52 (1.31–1.77)

> 30 41 (0.1) 170 (0.1) 1.50 (1.10–2.06) 1.30 (0.94–1.79) 1.20 (0.87–1.66)

Note: The ‘adjusted’ model was adjusted for chronic heart failure, cardiac arrhythmias, valvular disease, pulmonary circulation disorders, peripheral
vascular disease, hypertension, chronic pulmonary disease, diabetes, renal failure, liver disease, cancer, psychosis, and depression (yes or no) for
comorbidities, the number of months with outpatient visits over a 6‐month period (0, 1, 2, 3, 4, 5, and 6), and LTC needs (no support, support 1–2, care 1,
care 2, care 3, care 4, and care 5). The ‘fully adjusted’ model was further adjusted for the use of other types of anticholinergic drugs (yes or no).
Abbreviations: CI, confidence interval; OR, odds ratio.

misclassification of the presence of medical history and potentially A CK NO W L E D GE M E NT S

prevented the exclusion of the relevant individuals. Fourth, although This study was funded by grants from the Japan Agency for Medi-
we consider that the claims from insurance medical care facilities in cal Research and Development (grant number: JP21nf0101635),
Japan are reviewed by payment agencies and the diagnosis in claims Grants‐in‐Aid for Scientific Research, Fundamental Research A
data is reliable,44 the definition of diseases relies only on ICD‐10 from the Japan Society for the Promotion of Science (grant numbers:
code, which may have resulted in the misclassification of medical JP19K21590 and JP20H00563), and the JST FOREST Program (grant
history and outcome. Fifth, the changes in covariate status between number: JPMJFR205J). We gratefully acknowledge all staff associ-
cohort entry date and index date were not considered. The covariate ated with the LIFE Study. We thank our colleagues from the Osaka
status at cohort entry was used in this study to minimize bias because University Center of Medical Data Science and the Advanced Clinical
the status at cohort entry is consistent between cases and controls. Epidemiology Investigator's Research Project for providing insights
However, the changes of covariates over time might affect the study and expertise for carrying out our study.
results. Lastly, as this was an observational study, we could not adjust
for unmeasured confounders. For instance, this study could not CO N F L I C T O F I N T E R E S T S T A T E M E N T
adjust for lifestyle factors (e.g., tobacco use, drinking, diet, exercise The authors have no conflicts of interest to declare.
habits, and body mass index) and socioeconomic status, which are
potential risk factors for dementia. In addition, there might be re- DA T A A V A I L A B I L I T Y S T A T E M E N T
sidual confounding by indication, as patients without diagnoses of Research data are not shared.
depression or psychosis still received antidepressants or antipsy-
chotics for symptoms that were not included as covariates in the OR CI D
study. Tetsuhisa Kitamura https://orcid.org/0000-0003-0107-0580
Fumiko Murata https://orcid.org/0000-0002-8751-3401

5 | CONCLUSION
R EF ER E N CE S
This study provided evidence of the potential risk of dementia 1. Global Action Plan on the Public Health Response to Dementia 2017 ‐
associated with the use of anticholinergic drugs in older adults in 2025, World Health Organization. 2017.
2. Asada T. Prevalence of Dementia and Response to Life Dysfunction
Japan. These findings suggest that the underlying risks should be
Due to Dementia in Urban Areas. Health Labour Sciences Research
considered alongside the benefits of anticholinergic drugs such as Grants, Dementia Countermeasures Comprehensive Research
antipsychotics, antiparkinsonian agents, antidepressants, and bladder Project 2011‐2012 Research Report; 2013.
antimuscarinics while prescribing them to older adults. 3. Ninomiya T. Study on Future Projection of the Elderly Population
with Dementia in Japan, Health and Labor Sciences Research Grants,
Special Health and Labor Sciences Research Project 2014 Research
A U TH O R C O N TR IB UT I O N Report; 2015.
HF designed the LIFE‐based study and acquired data. FM and MM 4. Livingston G, Sommerlad A, Orgeta V, et al. Dementia prevention,
managed the LIFE data. YO, TK, SK, LZ, MK, NN, YG, YK, KK and TS intervention, and care. Lancet (London, Engl. 2017;390(10113):
designed the study. YO performed statistical analyses and drafted 2673‐2734. https://doi.org/10.1016/s0140‐6736(17)31363‐6
5. Livingston G, Huntley J, Sommerlad A, et al. Dementia prevention,
the manuscript. All authors have read and revised the manuscript
intervention, and care: 2020 report of the Lancet Commission.
critically for important intellectual content and have agreed to the Lancet (London, Engl. 2020;396(10248):413‐446. https://doi.org/10.
submission of the manuscript for publication. 1016/s0140‐6736(20)30367‐6
 10991166, 2023, 12, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/gps.6029 by Kinneret Academic College, Wiley Online Library on [18/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
OKITA ET AL.
- 11 of 12

6. Baumgart M, Snyder HM, Carrillo MC, Fazio S, Kim H, Johns H. Med Care. 2005;43(11):1130‐1139. https://doi.org/10.1097/01.mlr.
Summary of the evidence on modifiable risk factors for cognitive 0000182534.19832.83
decline and dementia: a population‐based perspective. Alzheimer's & 25. Kjaergaard AD, Johannesen BR, Sørensen HT, Henderson VW,
dementia J Alzheimer's Assoc. 2015;11(6):718‐726. https://doi.org/10. Christiansen CF. Kidney disease and risk of dementia: a Danish
1016/j.jalz.2015.05.016 nationwide cohort study. BMJ open. 2021;11(10):e052652. https://
7. Gerretsen P, Pollock BG. Drugs with anticholinergic properties: doi.org/10.1136/bmjopen‐2021‐052652
a current perspective on use and safety. Expet Opin Drug Saf. 26. Tsutsui T, Muramatsu N. Care‐needs certification in the long‐
2011;10(5):751‐765. https://doi.org/10.1517/14740338.2011.57 term care insurance system of Japan. J Am Geriatr Soc. 2005;
9899 53(3):522‐527. https://doi.org/10.1111/j.1532‐5415.2005.53
8. Fox C, Richardson K, Maidment ID, et al. Anticholinergic medication 175.x
use and cognitive impairment in the older population: the medical 27. National Institute for Health and Care Excellence. Guidelines. Na-
research council cognitive function and ageing study. J Am Geriatr tional Institute for Health and Care Excellence (NICE); 2018.
Soc. 2011;59(8):1477‐1483. https://doi.org/10.1111/j.1532‐5415. 28. American Geriatrics Society 2023 updated AGS Beers Criteria® for
2011.03491.x potentially inappropriate medication use in older adults. J Am Geriatr
9. Fox C, Smith T, Maidment I, et al. Effect of medications with anti‐ Soc. 2023.
cholinergic properties on cognitive function, delirium, physical 29. Guidelines for Medical Treatment and its Safety in the Elderly 2015,
function and mortality: a systematic review. Age ageing. 2014;43(5): The Japan Geriatrics Society.
604‐615. https://doi.org/10.1093/ageing/afu096 30. Carrière I, Fourrier‐Reglat A, Dartigues JF, et al. Drugs with anti-
10. Coupland CAC, Hill T, Dening T, Morriss R, Moore M, Hippisley‐Cox cholinergic properties, cognitive decline, and dementia in an elderly
J. Anticholinergic drug exposure and the risk of dementia: a nested general population: the 3‐city study. Archives Intern Med. 2009;
case‐control study. JAMA Intern Med. 2019;179(8):1084‐1093. 169(14):1317‐1324. https://doi.org/10.1001/archinternmed.2009.
https://doi.org/10.1001/jamainternmed.2019.0677 229
11. Zheng YB, Shi L, Zhu XM, et al. Anticholinergic drugs and the risk of 31. Dmochowski RR, Thai S, Iglay K, et al. Increased risk of incident
dementia: a systematic review and meta‐analysis. Neurosci Biobehav dementia following use of anticholinergic agents: a systematic
Rev. 2021;127:296‐306. https://doi.org/10.1016/j.neubiorev.2021. literature review and meta‐analysis. Neurourol Urodyn. 2021;40(1):
04.031 28‐37. https://doi.org/10.1002/nau.24536
12. Richardson K, Fox C, Maidment I, et al. Anticholinergic drugs and 32. Burke SL, Hu T, Fava NM, et al. Sex differences in the development
risk of dementia: case‐control study. BMJ Clin Res ed). 2018;361: of mild cognitive impairment and probable Alzheimer's disease as
k1315. https://doi.org/10.1136/bmj.k1315 predicted by hippocampal volume or white matter hyperintensities. J
13. Pieper NT, Grossi CM, Chan WY, et al. Anticholinergic drugs and Women Aging. 2019;31(2):140‐164. https://doi.org/10.1080/
incident dementia, mild cognitive impairment and cognitive decline: 08952841.2018.1419476
a meta‐analysis. Age ageing. 2020;49(6):939‐947. https://doi.org/10. 33. Ferretti MT, Iulita MF, Cavedo E, et al. Sex differences in Alz-
1093/ageing/afaa090 heimer disease ‐ the gateway to precision medicine. Nat Rev
14. OECD. Care Needed: Improving the Lives of People with Dementia, Neurol. 2018;14(8):457‐469. https://doi.org/10.1038/s41582‐018‐
OECD Health Policy Studies. OECD Publishing; 2018. 0032‐9
15. OECD. Trends Shaping Education 2019. OECD Publishing; 2019. 34. Hsu WH, Wen YW, Chen LK, Hsiao FY. Comparative associations
16. Fukuda H, Ishiguro C, Ono R, Kiyohara K. The longevity improve- between measures of anti‐cholinergic burden and adverse clinical
ment & Fair evidence (LIFE) study: overview of the study design and outcomes. Ann Fam Med. 2017;15(6):561‐569. https://doi.org/10.
baseline participant profile. J Epidemiol. 2022;33(8):428‐437. https:// 1370/afm.2131
doi.org/10.2188/jea.je20210513 35. Joung KI, Kim S, Cho YH, Cho SI. Association of anticholinergic use
17. Aarsland D, Batzu L, Halliday GM, et al. Parkinson disease‐ with incidence of Alzheimer's disease: population‐based cohort st-
associated cognitive impairment. Nat Rev Dis Prim. 2021;7(1):47. udy. Sci Rep. 2019;9(1):6802. https://doi.org/10.1038/s41598‐019‐
https://doi.org/10.1038/s41572‐021‐00280‐3 43066‐0
18. Lubin JH, Gail MH. Biased selection of controls for case‐control 36. Geula C. Abnormalities of neural circuitry in Alzheimer's disease:
analyses of cohort studies. Biometrics. 1984;40(1):63‐75. https:// hippocampus and cortical cholinergic innervation. Neurology. 1998;
doi.org/10.2307/2530744 51(1 (Suppl 1)):S18‐S29. Discussion S65‐17. https://doi.org/10.
19. Gray SL, Anderson ML, Dublin S, et al. Cumulative use of strong 1212/wnl.51.1_suppl_1.s18
anticholinergics and incident dementia: a prospective cohort study. 37. Risacher SL, McDonald BC, Tallman EF, et al. Association between
JAMA Intern Med. 2015;175(3):401‐407. https://doi.org/10.1001/ anticholinergic medication use and cognition, brain metabolism,
jamainternmed.2014.7663 and brain atrophy in cognitively normal older adults. JAMA Neurol.
20. By the American Geriatrics Society 2015 Beers Criteria Update 2016;73(6):721‐732. https://doi.org/10.1001/jamaneurol.2016.
Expert Panel. American geriatrics society 2015 updated beers 0580
criteria for potentially inappropriate medication use in older adults. J 38. Roher AE, Kuo YM, Potter PE, et al. Cortical cholinergic dener-
Am Geriatr Soc. 2015;63(11):2227‐2246. https://doi.org/10.1111/jgs. vation elicits vascular A beta deposition. Ann N Y Acad Sci. 2000;
13702 903(1):366‐373. https://doi.org/10.1111/j.1749‐6632.2000.tb06
21. Durán CE, Azermai M, Vander Stichele RH. Systematic review of 388.x
anticholinergic risk scales in older adults. Eur J Clin Pharmacol. 2013; 39. Ehrt U, Broich K, Larsen JP, Ballard C, Aarsland D. Use of drugs with
69(7):1485‐1496. https://doi.org/10.1007/s00228‐013‐1499‐3 anticholinergic effect and impact on cognition in Parkinson's disease:
22. ATC/DDD Index, WHO Collaborating Centre for Drug Statistics a cohort study. J Neurol Neurosurg Psychiatr. 2010;81(2):160‐165.
Methodology, Accessed April, 2023. https://doi.org/10.1136/jnnp.2009.186239
23. Elixhauser A, Steiner C, Harris DR, Coffey RM. Comorbidity mea- 40. Heath L, Gray SL, Boudreau DM, et al. Cumulative antidepressant
sures for use with administrative data. Med Care. 1998;36(1):8‐27. use and risk of dementia in a prospective cohort study. J Am Geriatr
https://doi.org/10.1097/00005650‐199801000‐00004 Soc. 2018;66(10):1948‐1955. https://doi.org/10.1111/jgs.15508
24. Quan H, Sundararajan V, Halfon P, et al. Coding algorithms for 41. Averbeck MA, Altaweel W, Manu‐Marin A, Madersbacher H. Man-
defining comorbidities in ICD‐9‐CM and ICD‐10 administrative data. agement of LUTS in patients with dementia and associated
 10991166, 2023, 12, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/gps.6029 by Kinneret Academic College, Wiley Online Library on [18/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
12 of 12
- OKITA ET AL.

disorders. Neurourol Urodyn. 2017;36(2):245‐252. https://doi.org/10. S U P P O R T I N G I N F O RM A T I O N

1002/nau.22928 Additional supporting information can be found online in the Sup-
42. Reinold J, Palese F, Romanese F, Logroscino G, Riedel O, Pisa FE.
porting Information section at the end of this article.
Anticholinergic burden before and after hospitalization in older adults
with dementia: increase due to antipsychotic medications. Int J Geriatr
Psychiatr. 2019;34(6):868‐880. https://doi.org/10.1002/gps.5084
43. Jack CR, Jr., Knopman DS, Jagust WJ, et al. Tracking pathophysio-
logical processes in Alzheimer's disease: an updated hypothetical How to cite this article: Okita Y, Kitamura T, Komukai S, et al.
model of dynamic biomarkers. Lancet Neurology. 2013;12(2):207‐ Association of anticholinergic drug exposure with the risk of
216. https://doi.org/10.1016/s1474‐4422(12)70291‐0 dementia among older adults in Japan: the LIFE Study. Int J
44. Yamana H, Konishi T, Yasunaga H. Validation studies of Japa-
Geriatr Psychiatry. 2023;e6029. https://doi.org/10.1002/gps.
nese administrative health care data: a scoping review. Pharmacoe-
6029
pidemiol Drug Saf. 2023;32(7):705‐717. https://doi.org/10.1002/pds.
5636