Associations between Anticholinergic Use and Dementia Risk in a Prevalent-User Cohort: Analysis of Japanese Medication Histories as Real-World Data

Abstract

Medication history from Japanese pharmacies is potentially rich in useful information regarding drug-related events in diverse populations; however, this information has rarely been utilized as real-world data. This study aimed to evaluate the feasibility of using medication history as real-world data to assess associations between anticholinergic medications for overactive bladder (OAB) and increased risk of dementia. A retrospective prevalent-user cohort study was performed to analyze medication history using a logistic regression model. The primary exposure was anticholinergic drug use for OAB (vs. beta-3 agonists only). The risk associated with each specific anticholinergic drug was also analyzed. Adjustments included age, sex, comorbidities (diabetes, hyperlipidemia, and hypertension), and psychotropic drug use. OAB anticholinergics and age were significant predictors. The adjusted odds ratio (OR) and 95% confidence interval (CI) for OAB anticholinergics were 1.53 (95% CI: 1.10–2.14, p = 0.0110), and those for age were 1.07 (95% CI: 1.04–1.09, p < 0.0001). Among individual OAB anticholinergics, the ORs for oxybutynin (4.53; 95% CI: 2.18–9.42, p < 0.0001) and tolterodine (5.64; 95% CI: 2.42–13.17, p < 0.0001) were significant, whereas ORs for other OAB anticholinergics were not. These findings suggest the potential to identify agents with lower dementia risk. In conclusion, analysis in this prevalent-user cohort study using medication history data showed a significant association between OAB anticholinergic use and increased dementia risk, with variation by drug. These results support the feasibility of utilizing medication history as real-world data for pharmacovigilance.

INTRODUCTION

With the aging population worldwide, minimizing the side effects of medications on geriatric syndromes, such as dementia and frailty, is becoming an urgent issue.^1,2) For the safety evaluation of geriatric drug treatment, it is desirable to utilize real-world data obtained from daily clinical practice. The International Council for Harmonisation (ICH) provides the ICH M14 (R1) guideline overarching principles for using real-world data and highlights the importance of leveraging large datasets from sources such as electronic health records, registries, and secondary data.

Recently, in addition to adverse event reporting databases such as FAERS, real-world data sources, such as clinical information databases derived from electronic health records, insurance claims data, and disease registry information, have been utilized in the post-marketing safety evaluations of pharmaceuticals. However, adverse event reporting databases, clinical information databases, and disease registry information do not comprehensively cover the elderly patient population. Furthermore, it is difficult to capture subjective adverse reactions from standardized data in insurance claims or electronic health records. Therefore, a database that covers the entire patient population and accumulates both subjective and objective patient information is necessary for a complementary drug safety evaluation infrastructure.^3,4)

Japanese medication history is a continuous record of each patient visiting a pharmacy in Japan and includes drug preparation, drug administration instructions, basic patient information, general health status, and illnesses.⁵⁾ Under Japan’s universal health insurance system, more than 60000 dispensing pharmacies process approximately 800 million prescriptions annually, generating a vast medication history database that covers nearly all elderly patients.⁶⁾ Japanese medication history has several unique characteristics not found in other forms of medical information: (i) it covers the majority of patients with a wide range of illnesses, age groups, and care settings including home and older adult care facilities; (ii) it includes subjective information about patients obtained through dialogue, allowing evaluation of their physical condition and any side effects; (iii) it records drug administration status and lifestyle habits from a pharmacological perspective; and (iv) it allows continuous information collection from individual patients. Therefore, we focused on using medication history as real-world data and previously demonstrated a significant increase in the risk of dementia onset associated with anticholinergic use for overactive bladder (OAB) by univariate analysis of Japanese medication history data.⁷⁾ These results were in agreement with those in previous reports.^6,8,9)

Large-scale clinical trials in Europe and the United States have shown that an increased risk of dementia is associated with the use of anticholinergic drugs.^6,8,9) The types of OAB medications used in Japan differ from those used in the United States and Europe. First-line medications for OAB include 5 anticholinergic drugs and 2 beta-3 adrenergic receptor agonists.^10,11) To prevent dementia associated with anticholinergic use, identifying the risks associated with each anticholinergic drug is important.

This study aimed to evaluate the feasibility of using medication history as real-world data to assess associations between anticholinergic medications for OAB and increased risk of dementia. Specifically, we conducted a multivariable analysis, adjusting for factors associated with dementia onset, including age, underlying diseases, and psychotropic drug use, to assess both the overall risk of OAB anticholinergics and the risk associated with individual anticholinergic agents.

MATERIALS AND METHODS

Ethics Approval

In this study, we analyzed the provided medication history data without collecting any new information or conducting patient interviews; thus, there was no direct patient contact or intervention. The medication history data did not contain any personally identifiable information such as names, addresses, or insurance numbers; therefore, the requirement for obtaining individual patient consent was waived. This study was conducted in compliance with the Japanese Personal Information Protection Law and the Ethical Guidelines for Medical and Health Research Involving Human Subjects. This study underwent ethical review and was approved by the Teikyo University Ethical Review Board for Medical and Health Research Involving Human Subjects (Approval Number: 21-138-4).

Study Design and Participants

In this study, the risk of dementia associated with OAB anticholinergic drugs and each OAB anticholinergic drug was evaluated by analyzing medication history. The medication history dataset was provided by SUNDRUG Co., Ltd. (Tokyo, Japan), a community pharmacy chain in Japan. Medication history is a Japanese-specific medical record system comprising logs of prescription content, patient compliance instructions, medication adherence, and physical changes. The dataset contains the following data fields of “Patient No.,” “Age Group,” “Sex,” “Dispensing Date,” “Prescription Date,” “Medical Institution,” “Department,” “Physician Name,” “Adherence,” “Physical Condition,” “Occurrence of Side Effects,” “Visits to Other Departments,” “Concurrent Medications,” “Medical History,” “Diet,” “Remaining Medications,” “Intent to Use Generic Drugs,” “Allergy History,” “Constitution,” “Pregnancy and Lactation Status,” “History of Side Effects,” “Patient-Specified Contraindicated Drugs,” “Habits,” “Medication Record Book,” “Consultation Items,” “Pharmacist No.,” “Subjective,” “Objective,” “Assessment,” “Plan,” “Prescription Details,” “Test Values,” and “Inquiry for Clarification.” The “Age” field, which was recorded using age groups rather than individual patient ages, was treated as a continuous variable.

In this study, 4547 patients undergoing OAB treatment were analyzed. This study employed a prevalent-user design, meaning that patients already receiving OAB medications at the time of their first pharmacy visit were included, rather than restricting the cohort to new initiators of therapy. Patient data were collected from 20 branches of SUNDRUG Co., Ltd. Visits were conducted from April 1, 2013, to March 31, 2024. The inclusion criteria were OAB medication use and age >55 years. The exclusion criteria were having visited the pharmacy fewer than twice; having a medical history of Huntington’s disease, Parkinson’s disease, Creutzfeldt–Jakob disease, or human immunodeficiency virus infection; and having dementia prior to the first prescription of OAB medication.

Exposures

(1) OAB anticholinergics: Patients who were prescribed any OAB anticholinergics (oxybutynin, propiverine, tolterodine, fesoterodine, imidafenacin, or solifenacin) at least once were classified into the exposed group. Participants prescribed only a beta-3 agonist (mirabegron or vibegron) were allocated to the control group.

(2) Each anticholinergic drug: Patients prescribed an anticholinergic drug for OAB were classified into the exposed group. Among these patients, those who were prescribed two or more OAB anticholinergics were defined as undergoing switching therapy. Participants prescribed only a beta-3 agonist were allocated to the control group.

The prescribed OAB anticholinergics were identified from the “Prescription Details” data field and included the insurance-approved anticholinergic drugs for OAB: oxybutynin, propiverine, tolterodine, fesoterodine, solifenacin, and imidafenacin.¹²⁾ The prescribed beta-3 agonists for OAB were identified from the “Prescription Details” data field and included the insurance-approved beta-3 agonist drugs for OAB: mirabegron (Betanis^®) and vibegron (Beova^®).¹²⁾

Outcome

The initiation of a prescription for any of the insurance-approved anti-dementia drugs (donepezil, galantamine, rivastigmine, or memantine) was defined as the onset of dementia. Prescription of anti-dementia drugs was identified when the “Prescription Details” or “Concurrent Medications” fields included any of these medications for dementia treatment.¹²⁾ As described above, patients who had anti-dementia drug prescriptions before their first OAB medication prescription were excluded from the analysis.

Variables for Adjustment

Variables for adjustment included age, sex, and prescription status of medications for diabetes, hyperlipidemia, hypertension, and psychotropic drugs, since these conditions could be reliably identified from pharmacy dispensing records through consistent prescription patterns of disease-specific medications (antidiabetic drugs, lipid-lowering agents, antihypertensive medications, and psychotropic drugs, respectively). These cardiovascular and metabolic conditions are well-established risk factors for dementia and have high prevalence in older adults taking OAB medications, making them important potential confounders to adjust for in the analysis.^{1,6,9,13–15)} Age and sex data were obtained from the dataset. Participants were considered to have diabetes, hyperlipidemia, or hypertension if they were prescribed drugs for these conditions. A prescription for diabetes was defined when the “Prescription Details” or “Concurrent Medications” field included any insurance-approved medications for diabetes (Supplementary Appendix List 1). A prescription for hyperlipidemia was defined when the “Prescription Details” or “Concurrent Medications” field included any insurance-approved medications for hyperlipidemia (Supplementary Appendix List 2). A prescription for hypertension was defined when the “Prescription Details” or “Concurrent Medications” field included any insurance-approved medications for hypertension (Supplementary Appendix List 3). Prescription of psychotropic drugs was defined when the field of “Prescription Details” or “Concurrent Medications” field included any insurance-approved psychotropic drug (Supplementary Appendix List 4).

Statistical Analysis

Continuous variables are presented as means ± standard deviations (S.D.). Categorical variables are presented as numbers with percentages. We performed a multivariable logistic regression analysis on patients treated for OAB. The outcome was dementia onset, defined as the initiation of anti-dementia drug prescriptions. The primary exposure was the use of anticholinergic drugs for OAB. Two types of exposure were considered: (1) ever prescribed any anticholinergic drug (vs. prescribed a beta-3 agonist only), and (2) each specific anticholinergic drug prescribed (vs. prescribed a beta-3 agonist only). Variables for adjustment included prescriptions for comorbidities such as diabetes, hyperlipidemia, and hypertension, and the use of psychotropic drugs.

In addition, in the sensitivity analysis, we excluded patients who had been prescribed drugs that showed a significant increase in dementia risk in the drug-specific analysis. Using this restricted dataset, we re-evaluated the association between OAB anticholinergics as a group and dementia risk compared with beta-3 agonists. The risk of dementia onset was evaluated using the odds ratio (OR) and the associated 95% confidence interval (CI).

Furthermore, we conducted time-to-event analysis using restricted mean survival time (RMST) and the log-rank test. To visualize the dementia-free survival over time, Kaplan–Meier curves were generated. RMST represents the average event-free time within a prespecified follow-up period. The follow-up period started at the initiation of OAB medications and ended at the initiation of anti-dementia medication or the last recorded visit, whichever occurred first. The time horizon τ was defined as the shorter of the maximum follow-up times observed in the two groups to ensure comparability. Between-group differences and ratios were calculated with their 95% CIs. RMST was chosen as the primary summary measure because it does not require the proportional hazards assumption, which may not hold over the entire follow-up period in this study, and provides a clinically interpretable measure of average event-free time.^16–18)

Statistical significance was set at a p-value <0.05. Data handling, logistic regression, and Kaplan–Meier curves were generated using Python version 3.11.9 (https://www.python.org), and analysis of RMST was conducted using R version 4. 2. 2 (https://www.r-project.org/).

RESULTS

Between April 1, 2013, and March 31, 2024, of the 174849 patients who visited 20 branches of SUNDRUG Co., Ltd., 4547 patients were prescribed an OAB drug. We performed logistic regression analysis to evaluate the risk of dementia associated with the prescription of OAB anticholinergics and each anticholinergic drug.

Evaluating the Dementia-Inducing Risk of Being Prescribed OAB Anticholinergics

Among the 4547 patients, 2397 were ever prescribed an anticholinergic agent at least once (exposed group), of whom 101 (4.21%) were subsequently prescribed an anti-dementia drug. Additionally, 2150 patients were prescribed a beta-3 agonist only (control group), of whom 60 (2.79%) were subsequently prescribed an anti-dementia drug. In the exposed group of patients receiving OAB anticholinergics (ever prescribed an anticholinergic for OAB at least once), 1110 patients (46.31%) were male; mean age was 72.67 (S.D.: 8.68) years; 250 (10.43%) had diabetes; 504 (21.03%) had hyperlipidemia; 923 (38.51%) had hypertension; and 589 (24.57%) were taking a psychotropic drug (Table 1).

Table 1. Characteristics of the Participants

	Total (N = 4547)	Beta-3 (n = 2150)	Anticholinergic (n = 2397)
Sex, male	2375 (52.23)	1265 (58.84)	1110 (46.31)
Age (years)	72.96 (8.73)	73.27 (8.77)	72.67 (8.68)
Basal disease
Diabetic	426 (9.37)	176 (8.19)	250 (10.43)
Hyperlipidemia	887 (19.51)	383 (15.52)	504 (21.03)
Hypertension	1573 (34.59)	650 (30.23)	923 (38.51)
Psychotropic drug use	995 (21.88)	406 (18.88)	589 (24.57)
Onset dementia	161 (3.54)	60 (2.79)	101 (4.21)

According to the logistic regression model, OAB anticholinergic use and age were significant predictors. The adjusted OR of being prescribed OAB anticholinergics compared with being prescribed a beta-3 agonist only was 1.53 (95% CI: 1.10–2.14; p = 0.0110). The adjusted OR for age was 1.07 (95% CI: 1.04–1.09; p < 0.0001) (Fig. 1).

Fig. 1. Forest Plot of Odds Ratios and 95% Confidence Intervals for Being Prescribed Anticholinergic Drugs and Other Basic Factors

Evaluating the Dementia-Inducing Risk of Being Prescribed Each Anticholinergic Drug

Of the 4547 patients, 2397 were prescribed anticholinergic drugs, of whom 101 (4.21%) were subsequently prescribed anti-dementia drugs. The number of patients prescribed each anticholinergic drug, along with the number subsequently prescribed anti-dementia drugs, was as follows: solifenacin (712 prescribed, 23 anti-dementia), fesoterodine (485 prescribed, 18 anti-dementia), imidafenacin (414 prescribed, 18 anti-dementia), propiverine (176 prescribed, 9 anti-dementia), oxybutynin (86 prescribed, 10 anti-dementia), tolterodine (58 prescribed, 8 anti-dementia), and switching therapy (466 prescribed, 16 anti-dementia). The control group consisted of 2150 individuals who were prescribed only a beta-3 agonist (Supplementary Table S1).

According to the logistic regression model, prescriptions of oxybutynin and tolterodine, along with age, were significant factors. The adjusted ORs of each anticholinergic drug compared with being prescribed beta-3 agonists were as follows: solifenacin (1.20; 95% CI: 0.73–1.96; p = 0.4750), fesoterodine (1.40; 95% CI: 0.81–2.41; p = 0.2230), imidafenacin (1.52; 95% CI: 0.88–2.62; p = 0.1290), propiverine (1.64; 95% CI: 0.79–3.41; p = 0.1820), oxybutynin (4.53; 95% CI: 2.18–9.42; p < 0.0001), tolterodine (5.64; 95% CI: 2.42–13.17; p < 0.0001), and switching therapy (1.25; 95% CI: 0.71–2.21; p = 0.4570) (Fig. 2).

Fig. 2. Forest Plot of Odds Ratios and 95% Confidence Intervals for Being Prescribed Each Anticholinergic Drug and Other Basic Factors

Sensitivity Analysis

To examine whether the overall association was driven by specific agents, we conducted an additional analysis excluding patients who had been prescribed oxybutynin and tolterodine, which were significantly associated with dementia in the drug-specific analysis. Among the 4403 patients, 2253 were ever prescribed an anticholinergic, and 2150 were prescribed a beta-3 agonist only. In the exposed group, 1055 patients (46.83%) were male; mean age was 72.71 years (S.D.: 8.64); 238 patients (10.56%) had diabetes; 474 (21.04%) had hyperlipidemia; 857 (38.04%) had hypertension; and 553 (24.55%) were taking a psychotropic drug (Supplementary Table S2). Logistic regression analysis revealed that sex, age, and psychotropic drug use were significant covariates. However, anticholinergic exposure was not significantly associated with the dementia risk in this restricted dataset (Fig. 3).

Fig. 3. Forest Plot of Odds Ratios and 95% Confidence Intervals for Being Prescribed Anticholinergic Drugs except Oxybutynin and Tolterodine and Other Basic Factors

Time-to-Event Analysis

Kaplan–Meier survival curves for time to initiation of anti-dementia medication are shown in Supplementary Fig. S1. The log-rank test showed no significant difference in time-to-event between the groups (χ² = 1.26, df = 1, p = 0.26). The RMST was evaluated up to a time horizon of τ = 3986 d, defined as the shorter of the maximum follow-up times in the two groups. The estimated RMST was 3698.1 d (95% CI, 3638.1–3758.0) for OAB anticholinergic drugs and 3791.5 d (95% CI, 3732.4–3850.6) for OAB beta-3 agonists. The RMST difference between groups, calculated as (OAB anticholinergic drugs − OAB beta-3 agonist), was −93.5 d (95% CI, −177.7 to −9.3; p = 0.030), indicating a shorter average event-free time for patients receiving OAB anticholinergic drugs. The RMST ratio, calculated as (OAB anticholinergic drugs/OAB beta-3 agonist), was 0.975 (95% CI, 0.954–0.998; p = 0.030), consistent with a modest but statistically significant reduction in average event-free survival time associated with OAB anticholinergic drugs.

DISCUSSION

We performed a retrospective prevalent-user cohort study with multivariable analysis of medication history data from patients undergoing OAB treatment and found an association between OAB anticholinergic use and an increased risk of developing dementia, with variation depending on the individual drug.

Among 174849 patients who visited 20 branches of a community pharmacy chain in Japan, 4547 patients taking OAB medications were included in this study. Multivariable analysis, which included age, sex, underlying medical conditions, and psychotropic drug use, showed that age and OAB anticholinergic use were associated with an increased risk of dementia (Fig. 1). We detected a significant increase in the adjusted OR (1.53; 95% CI: 1.10–2.14) for OAB anticholinergics compared with beta-3 agonists. Large-scale clinical trials have reported relative risks of dementia associated with anticholinergic drug use ranging from 1.48 to 1.65.^6,8,9) The results of this study are consistent with those of previous large-scale clinical trials, suggesting the feasibility of utilizing medication history as real-world data for pharmacovigilance.

Among all the anticholinergic drugs prescribed, the adjusted ORs for oxybutynin and tolterodine were significantly higher than those for beta-3 agonists (Fig. 2). However, no significant increase in OR was observed for solifenacin, fesoterodine, imidafenacin, or propiverine. Previous studies have similarly reported heterogeneity among anticholinergics; Malcher et al. found significantly increased risks with oxybutynin and solifenacin but not with trospium or fesoterodine, while other reports confirmed the increased risk of oxybutynin and the absence of a significant risk with fesoterodine.^19,20) Shiho et al. demonstrated that these differences can be explained by pharmacokinetics, showing that oxybutynin exhibits high brain distribution and significant muscarinic receptor occupancy in the rat brain, whereas 5-HMT, the active metabolite of fesoterodine, shows extremely low brain distribution and barely blocks brain muscarinic receptors.²¹⁾ Sensitivity analysis showed no significant differences in the dementia risk of OAB anticholinergics other than oxybutynin and tolterodine (Fig. 3), suggesting that the association with increased risk of dementia observed in the primary analysis was largely attributable to oxybutynin and tolterodine. These findings suggest the potential to identify agents with lower dementia risk.

Our survival analysis using RMST demonstrated that anticholinergic use was associated with significantly shorter restricted mean dementia-free survival time (difference of approximately 93 d), despite the nonsignificant log-rank test (χ² = 1.26, p = 0.262). This discrepancy is likely due to violation of the proportional hazards assumption, which is common in real-world data and prevalent-user designs. Unlike the log-rank test, RMST provides a more robust and clinically interpretable measure when treatment effects vary over time. The consistency between this RMST result and our logistic regression finding (adjusted OR = 1.53; 95% CI: 1.10–2.14) reinforces the observed association. While this difference in RMST suggests a potential reduction in dementia-free survival time, these findings should be interpreted as preliminary associations within a prevalent-user cohort rather than evidence of causality, as discussed in the context of study limitations below.

This study had some limitations. A fundamental limitation of pharmacy-based datasets is the difficulty in capturing complete medication histories when patients use multiple pharmacies—a common practice in Japan—leading to potential exposure and outcome misclassification with unpredictable bias direction that is difficult to resolve through statistical adjustment alone. A major limitation is the prevalent-user design, which has several implications for interpreting our results. By including patients already taking OAB medications, our cohort may have excluded individuals who developed early dementia and discontinued therapy before our observation period (depletion of susceptibles), potentially underestimating the true risk. Additionally, the temporal relationship between exposure and outcome cannot be definitively established, as some patients may have had early cognitive decline that led to prescription of OAB medications (protopathic bias) rather than the medications causing dementia. Prevalent users have, by definition, survived without developing the outcome during their prior exposure period (immortal time bias) and may differ systematically from new users in terms of disease severity and comorbidity burden, factors that could confound the association between anticholinergic use and dementia risk.

Second, the 2024 Lancet Commission on dementia prevention identified 14 modifiable risk factors accounting for approximately 45% of dementia cases worldwide.¹⁾ Our study could not adjust for some key factors relevant to older adults.^22–29) Third, of the 2397 patients taking OAB anticholinergic drugs, 176–712 received solifenacin, fesoterodine, imidafenacin, or propiverine alone, but the number of patients taking oxybutynin and tolterodine was small, at 86 and 58, respectively. Therefore, analysis of the risk of developing dementia using a larger number of cases is necessary for both drugs in future studies. In 2020, the number of prescriptions for oxybutynin and tolterodine in Japan had decreased to 0.5 to 0.6% of the total number of prescriptions for OAB medications.³⁰⁾ While it may be difficult to accumulate sufficient case data for these two drugs in the future, it is possible to utilize the large amount of medication history data already accumulated by pharmaceutical companies.

To prevent dementia associated with anticholinergic use, it is important to analyze the risks associated with individual drugs using a larger amount of medication history data. Furthermore, medication history contains not only medication information but also information on drug treatment outcomes and it is expected to be used for a wider range of drug safety assessments in future studies.

In conclusion, this prevalent-user cohort study found an association between anticholinergic medications for OAB and increased dementia risk in Japanese pharmacy patients, with particularly notable associations observed for oxybutynin and tolterodine. Despite limitations such as incompleteness of pharmacy-based datasets, prevalent-user bias and potential confounding, the observed associations were consistent with those reported in previous large-scale studies. These findings warrant further validation using medication history data for pharmacovigilance in real-world clinical practice settings.

Acknowledgments

SUNDRUG Co., Ltd. provided a wealth of support, including access to the medication history data analyzed in this study. We express our deepest gratitude to everyone who cooperated in this project.

DECLARATIONS

Funding

This study was funded by SUNDRUG Co., Ltd.; JSPS KAKENHI (Grant Number: JP25K02431); and ACRO Research Grants from Teikyo University (TeTe22-22).

Author Contributions

HI designed the study, analyzed the data, and drafted the manuscript. ST, EN, SM, TY, and TO contributed to the design and revision; SM analyzed the data. JY and MK acquired the data. TO supervised and secured the funding. All the authors approved the final manuscript. HI and TO are the guarantors of this study.

Conflict of Interest

TO received a research grant from SUNDRUG Co., Ltd. (Tokyo, Japan). The other authors declare no conflict of interest.

Data Availability

The medication history data used in this analysis will not be disclosed, as determined by the provider company. We agree with the company’s opinion that medication history, which contains detailed medical information classified as personal, should be managed at the highest level of security.

Supplementary Materials

This article contains supplementary materials.

REFERENCES

• 1)  Livingston G, Huntley J, Liu KY, et al. Dementia prevention, intervention, and care: 2024 report of the Lancet standing Commission. Lancet, 404, 572–628 (2024).
• 2)  Fang M, Hu J, Weiss J, Knopman DS, Albert M, Windham BG, Walker KA, Sharrett AR, Gottesman RF, Lutsey PL, Mosley T, Selvin E, Coresh J. Lifetime risk and projected burden of dementia. Nat. Med., 31, 772–776 (2025).
• 3)  Seidel D, Thyrian JR. Burden of caring for people with dementia - comparing family caregivers and professional caregivers. A descriptive study. J. Multidiscip. Healthc., 12, 655–663 (2019).
• 4)  Gulle BT, Tozduman B. Population attributable fraction of modifiable risk factors for dementia in Turkey. Dementia, 2025, 14713012251357162 (2025).
• 5)  Kinoshita S, Kishimoto T. Dementia in Japan: a societal focus, Lancet Neurol., 22, 1101–1102 (2023). <https://pubmed.ncbi.nlm.nih.gov/37977703/>.
• 6)  Gray SL, Anderson ML, Dublin S, Hanlon JT, Hubbard R, Walker R, Yu O, Crane PK, Larson EB. Cumulative use of strong anticholinergics and incident dementia: a prospective cohort study. JAMA Intern. Med., 175, 401–407 (2015).
• 7)  Tanaka S, Miyata S, Yamato J, Kumagai S, Kogawa M, Yamori T, Irie M, Nakatani E, Okura T. Evaluation of dementia risk in patients taking medication for overactive bladder using medication history in Japan. Translational and Regulatory Sciences, 6, 10–14 (2024).
• 8)  Coupland CAC, Hill T, Dening T, Morriss R, Moore M, Hippisley-Cox J. Anticholinergic drug exposure and the risk of dementia: a nested case-control study. JAMA Intern. Med., 179, 1084–1093 (2019).
• 9)  Richardson K, Fox C, Maidment I, Steel N, Loke YK, Arthur A, Myint PK, Grossi CM, Mattishent K, Bennett K, Campbell NL, Boustani M, Robinson L, Brayne C, Matthews FE, Savva GM. Anticholinergic drugs and risk of dementia: case-control study. BMJ, 361, k1315 (2018).
• 10)  Lightner DJ, Gomelsky A, Souter L, Vasavada SP. Diagnosis and treatment of overactive bladder (non-neurogenic) in adults: AUA/SUFU guideline amendment 2019. J. Urol., 202, 558–563 (2019).
• 11)  Nambiar AK, Bosch R, Cruz F, Lemack GE, Thiruchelvam N, Tubaro A, Bedretdinova DA, Ambühl D, Farag F, Lombardo R, Schneider MP, Burkhard FC. EAU guidelines on assessment and nonsurgical management of urinary incontinence. Eur. Urol., 73, 596–609 (2018).
• 12)  MANUAL OF THERAPEUTIC AGENTS 2024. Igakushoin, Tokyo (2024).
• 13)  Ohara T, Doi Y, Ninomiya T, Hirakawa Y, Hata J, Iwaki T, Kanba S. Glucose tolerance status and risk of dementia in the community: the Hisayama study. Neurology, 77, 1126–1134 (2011).
• 14)  Solomon A, Kivipelto M, Wolozin B, Zhou J, Whitmer RA. Midlife serum cholesterol and increased risk of Alzheimer’s and vascular dementia three decades later. Dement. Geriatr. Cogn. Disord., 28, 75–80 (2009).
• 15)  Whitmer RA, Sidney S, Selby J, Johnston SC, Yaffe K. Midlife cardiovascular risk factors and risk of dementia in late life. Neurology, 64, 277–281 (2005).
• 16)  Uno H, Claggett B, Tian L, Inoue E, Gallo P, Miyata T, Schrag D, Takeuchi M, Uyama Y, Zhao L, Skali H, Solomon S, Jacobus S, Hughes M, Packer M, Wei L-J. Moving beyond the hazard ratio in quantifying the between-group difference in survival analysis. J. Clin. Oncol., 32, 2380–2385 (2014).
• 17)  Royston P, Parmar MKB. Restricted mean survival time: an alternative to the hazard ratio for the design and analysis of randomized trials with a time-to-event outcome. BMC Med. Res. Methodol., 13, 152 (2013).
• 18)  Pharmaceuticals and Medical Devices Agency. “The 4th Data Science Roundtable Conference.”: <https://www.pmda.go.jp/review-services/symposia/0060.html>, accessed 28 December, 2025.
• 19)  Malcher MF, Droupy S, Berr C, Ziad A, Huguet H, Faillie J-L, Serrand C, Mura T. Dementia associated with anticholinergic drugs used for overactive bladder: a nested case-control study using the french national medical-administrative database. J. Urol., 208, 863–871 (2022).
• 20)  Wagg A, Khullar V, Marschall-Kehrel D, Michel MC, Oelke M, Darekar A, Bitoun CE, Weinstein D, Osterloh I. Flexible-dose fesoterodine in elderly adults with overactive bladder: results of the randomized, double-blind, placebo-controlled study of fesoterodine in an aging population trial. J. Am. Geriatr. Soc., 61, 185–193 (2013).
• 21)  Shiho M, Tanaka S, Nakatani E, Kurosawa T, Kubo Y, Deguchi Y, Okura T. Brain distribution and muscarinic receptor occupancy of an active metabolite of fesoterodine in rats. J. Pharmacol. Exp. Ther., 392, 100058 (2025).
• 22)  Sofi F, Valecchi D, Bacci D, Abbate R, Gensini GF, Casini A, Macchi C. Physical activity and risk of cognitive decline: a meta-analysis of prospective studies. J. Intern. Med., 269, 107–117 (2011).
• 23)  Anstey KJ, von Sanden C, Salim A, O’Kearney R. Smoking as a risk factor for dementia and cognitive decline: a meta-analysis of prospective studies. Am. J. Epidemiol., 166, 367–378 (2007).
• 24)  Kuiper JS, Zuidersma M, Oude Voshaar RC, Zuidema SU, van den Heuvel ER, Stolk RP, Smidt N. Social relationships and risk of dementia: A systematic review and meta-analysis of longitudinal cohort studies. Ageing Res. Rev., 22, 39–57 (2015).
• 25)  Qizilbash N, Gregson J, Johnson ME, Pearce N, Douglas I, Wing K, Evans SJW, Pocock SJ. BMI and risk of dementia in two million people over two decades: a retrospective cohort study. Lancet Diabetes Endocrinol., 3, 431–436 (2015).
• 26)  Xu W, Wang H, Wan Y, Tan C, Li J, Tan L, Yu J-T. Alcohol consumption and dementia risk: a dose-response meta-analysis of prospective studies. Eur. J. Epidemiol., 32, 31–42 (2017).
• 27)  Meng X, D’Arcy C. Education and dementia in the context of the cognitive reserve hypothesis: a systematic review with meta-analyses and qualitative analyses. PLOS ONE, 7, e38268 (2012).
• 28)  Diniz BS, Butters MA, Albert SM, Dew MA, Reynolds CF 3rd. Late-life depression and risk of vascular dementia and Alzheimer’s disease: systematic review and meta-analysis of community-based cohort studies. Br. J. Psychiatry, 202, 329–335 (2013).
• 29)  Peters R, Ee N, Peters J, Booth A, Mudway I, Anstey KJ. Air pollution and dementia: a systematic review. J. Alzheimers Dis., 70(s1), S145–S163 (2019).
• 30)  Suzuki M, Yamamoto M, Shiratori T, Naito A, Nakamura M. Evaluation of trends in prescription drugs for overactive bladder in Japanese patients. Int. J. Urol., 30, 331–333 (2023).