Association Between the Use of Hypnotics and the Risk of Alzheimer’s Disease

Abstract

Background:

Hypnotics, including benzodiazepines, are extensively and inappropriately prescribed for older people to treat anxiety and sleep disorders, despite the adverse health outcomes associated with their use.

Objective:

This study aimed to examine the association of the use of long- and short-acting hypnotics with the risk of Alzheimer’s disease.

Methods:

Data from 234,634 participants, derived from the Korean National Health Insurance Service National Sample Cohort from 2002 to 2013, were examined. Individuals over the age of 50 years were included in the study. The dependent variable was the risk of Alzheimer’s disease. Hypnotics were categorized by the period of the prescription of benzodiazepines, i.e., either till the participants were diagnosed with Alzheimer’s disease or the end of the study period (December 31, 2013). Cox regression model was built to analyze the association between variables.

Results:

Individuals who used long-acting hypnotics were found to have a higher risk of Alzheimer’s disease than non-users. Moreover, among individuals with sleep disorders, those who used hypnotics had a higher risk of Alzheimer’s disease than those who did not.

Conclusion:

This study identified an association between the use of hypnotics and the risk of Alzheimer’s disease among South Korean middle-aged and older people.

Keywords

Alzheimer’s disease benzodiazepines dementia hypnotics

INTRODUCTION

Widely prescribed hypnotics, such as benzodiazepines, and related drugs are commonly used among older people to treat anxiety and insomnia, despite the known risks [1]. In South Korea, benzodiazepines are the most frequently prescribed hypnotics, and the number of prescriptions is on the rise [2]. Benzodiazepine use requires attention, especially since long-term use by those aged≥65 years increases the risk of side effects such as cognitive disorders and falls [3]. In South Korea, the rate of prescription of long-acting benzodiazepines in patients aged≥65 years is 146.3 per 1,000 people, which is higher than the average rate of prescription in developed countries (52.0 per 1,000 people) [3, 4]. Owing to the high prescription rate not only in Korea but worldwide, the use of hypnotics including benzodiazepines is an emerging issue due to the negative health outcomes [5, 6]. Previous studies have shown that chronic use of benzodiazepines may lead to a decrease in cognitive reserve, downregulation of gamma-aminobutyric acid receptors, and degradation in the capability of utilizing alternative neural networks because of lower levels of brain activation [7 –9]. Hence, interest in the risks, safety, and adverse effects of the use of hypnotics has increased [10].

Dementia is a rapidly growing public health problem affecting around 50 million people globally in aging societies [11]. The World Health Organization’s guidelines for the reduction of the risk of cognitive decline and dementia, reported that there are nearly 10 million new cases every year and this would triple by 2050 [12]. Likewise, the number of dementia patients have increased rapidly in Korea. Nearly 6.9%of older people above 65 years of age have dementia and this is expected to increase to about 2.71 million by 2050 [13, 14]. Moreover, dementia has a huge economic impact; the total estimated worldwide cost of dementia is 1 trillion USD [11]. Similarly, in Korea, the medical costs of dementia have been increasing by 20%yearly, and the social costs have been around 12 trillion won (1%of the gross domestic product) [14]. People with dementia often live for many years after the onset of their symptoms, requiring long-term care and support; thus, dementia is a matter of concern not only for the patients but also for their caregivers and families [15]. Therefore, in dementia, it is important to provide support for both the economy of the country as well as the wellbeing of the caregivers.

Some studies have shown an association between the use of hypnotics and risk of dementia [10 , 16–18]. A previous study showed that benzodiazepines could precipitate the onset of dementia or mental decline by altering or impairing higher mental activities [19]. Additionally, benzodiazepines lower brain activity and limit an individual’s capacity to neutrally compensate and have cognitive reserve, thus precipitating dementia [20]. Likewise, one study showed that brain hypoactivity or cognitive impairment caused by the sedative effects of hypnotics could precipitate the development of dementia [10]. Hence, it is important to evaluate the relationship between the use of long- and short-acting hypnotics and dementia. Moreover, if long-acting benzodiazepines are used inappropriately and excessively in older people, the drug’s half-life may be extended, leading to excessive sedation, which can cause more problems [3]. The administration of drugs in older people requires caution; however, prescriptions for older people are often extensive and inappropriate [21].

Therefore, it is necessary to investigate the association between the use of long- and short-acting hypnotics and the risk of Alzheimer’s disease, to prevent the development of dementia. As the pathophysiology of dementia differs by type, we focused on Alzheimer’s disease, because the effect of hypnotics on the brain is more likely in Alzheimer’s disease than in other dementia types [10]. We hypothesized that the use of hypnotics would increase the risk of Alzheimer’s disease. Consequently, we present an epidemiological study including a convenience sample of claims data that aimed to examine the associations between the use of hypnotics and Alzheimer’s disease among the South Korean population.

MATERIALS AND METHODS

Data and study participants

Data for this study were obtained from the Korean National Health Insurance Service National Sample Cohort (NHIS-NSC) between 2002 and 2013. The NHIS-NSC data, obtained by random sampling, included medical claims of 1,025,340 individuals (these represent 2%of the South Korean population). As Alzheimer’s disease is age-dependent, we only included individuals who were > 50 years old. Among the beneficiaries, we excluded those who were already diagnosed with Alzheimer’s disease (International Statistical Classification of Disease and Related Health Problems, 10th Revision [ICD-10] code: F00) in 2002 or 2003. Those with vascular dementia (ICD-10: F01) were also excluded because their data were missing from the dataset. Finally, a total of 234,634 participants were included in the study (Fig. 1). The data were analyzed for the development of Alzheimer’s disease within the study period or by the end of it (December 31, 2013).

Fig. 1

Flowchart of the participant selection.

All data are available in the database of the Korean National Health Insurance Sharing Service (https://nhiss.nhis.or.kr) and can be accessed upon reasonable request. This study adhered to the tenets of the Declaration of Helsinki. Ethical approval was obtained from the NHIS-NSC committee. The study was based on routinely collected administrative and claims data of the NHIS-NSC, which are secondary data that do not contain any identifying information; therefore, additional approval was not required.

Variables

The risk of Alzheimer’s disease (ICD-10: F00) is the dependent variable in this study. As those already diagnosed with Alzheimer’s disease in 2002 or 2003 were excluded from the study, individuals who developed Alzheimer’s disease within the study period were considered to be newly diagnosed (Fig. 2).

Fig. 2

Timeline of the study.

The primary independent variable was the use of benzodiazepines. We selected 25 commonly used benzodiazepines, which were divided into two groups based on the duration of their action (that is, the half-life). Long-acting benzodiazepines (clobazam, clonazepam, chlordiazepoxide, diazepam, flunitrazepam, flurazepam, and ethyl loflazepate) were grouped as drugs with a long half-life (t^1/2≥20 h), while short-acting benzodiazepines (alprazolam, bromazepam, clotiazepam, etizolam, lorazepam, tofisopam, and triazolam) were grouped as drugs with a short half-life (t^1/2 < 20 h) [22, 23]. The drugs were classified depending on their use, based on the World Health Organization Anatomical Therapeutic Chemical classification system [22]. Benzodiazepine use was categorized by the period of the prescription, that is either till the participants were diagnosed with Alzheimer’s disease or the end of the study period (December 31, 2013). Individuals, prior to getting their first prescription of benzodiazepines or those who were never prescribed benzodiazepines, were categorized as non-users; those who were prescribed long-acting hypnotics were categorized as the long-acting hypnotics group; and those who were prescribed short-acting hypnotics were categorized as the short-acting group. Additionally, the analyses included variables such as age, sex, social security, region, disability, income, Charlson Comorbidity Index (CCI), sleep disorder (ICD-10 code: G47, F51), type 2 diabetes (ICD-10 code: E11), hypertension (ICD-10 code: I10–I15), and depression (ICD-10 code: F32). CCI was used to identify the comorbidities among the patients [24].

Statistical analysis

The chi-square test was conducted to investigate the general characteristics of the study population. The Cox proportional hazards model was used to examine the factors associated with the risk of Alzheimer’s disease [25, 26]. The Cox proportional hazards model has achieved widespread use in the analysis of time-to-event (e.g., time to death or diseases) data with censoring and covariates, which may change the values over time [27]. Considering T as the time of interest, and Z as a vector of covariates of interests. The Cox proportional hazards model specifies that λ (t|Z)= λ₀ (t) e^β′z, where β is a set of unknown regression parameters and λ₀ (t) is an unspecified baseline hazard function [28]. The Cox proportional hazard model enables the difference between study participant groups, while controlling for other factors. In this study, the hazard is the probability of the event (Alzheimer’s disease) occurring, given that the patients are at risk at that moment [29]. Survival time was measured as a disease-free survival time [30]. Differences were considered statistically significant with a p-value < 0.05. All data analyses were performed using SAS 9.4 software (SAS Institute Inc., Cary, NC, USA).

RESULTS

Baseline characteristics of the study cohort

Table 1 presents the general characteristics of the study population. Among the 234,634 participants, 15,038 (6.4%) had Alzheimer’s disease. The participants who used long-acting and short-acting hypnotics were 1,303 (0.6%) and 18,758 (8.0%), respectively. The relationship between the use of hypnotics and risk of Alzheimer’s disease was significant. Additionally, differences in other covariates between groups were significant.

Table 1

General characteristics of the study population

Variables		Total		Alzheimer’s disease				p
				Yes		No
Total		234,634	(100.0)	15,038	(6.4)	219,596	(93.6)
Hypnotics								0.0002
	Long-acting	1,303	(0.6)	112	(8.6)	1,191	(91.4)
	Short-acting	18,758	(8.0)	1,282	(6.8)	17,476	(93.2)
	Non-user	214,573	(91.5)	13,644	(6.4)	200,929	(93.6)
Sex								< 0.0001
	Male	106,323	(45.3)	4,704	(4.4)	101,619	(95.6)
	Female	128,311	(54.7)	10,334	(8.1)	117,977	(91.9)
Age								< 0.0001
	50–59	104,501	(44.5)	1,380	(1.3)	103,121	(98.7)
	60–65	42,612	(18.2)	2,016	(4.7)	40,596	(95.3)
	65–70	33,378	(14.2)	2,978	(8.9)	30,400	(91.1)
	70–75	23,670	(10.1)	3,305	(14.0)	20,365	(86.0)
	75–80	15,792	(6.7)	2,863	(18.1)	12,929	(81.9)
	≥80	14,681	(6.3)	2,496	(17.0)	12,185	(83.0)
Social security								< 0.0001
	Health insurance (corporate)	102,398	(43.6)	5,014	(4.9)	97,384	(95.1)
	Health insurance (regional)	132,236	(56.4)	10,024	(7.6)	122,212	(92.4)
Region								< 0.0001
	Metropolitan	89,156	(38.0)	4,851	(5.4)	84,305	(94.6)
	City	56,952	(24.3)	3,615	(6.3)	53,337	(93.7)
	Rural	88,526	(37.7)	6,572	(7.4)	81,954	(92.6)
Disability								< 0.0001
	Yes	18,807	(8.0)	1,482	(7.9)	17,325	(92.1)
	No	215,827	(92.0)	13,556	(6.3)	202,271	(93.7)
Income								< 0.0001
	Low	47,207	(20.1)	3,643	(7.7)	43,564	(92.3)
	Middle	94,621	(40.3)	5,053	(5.3)	89,568	(94.7)
	High	92,806	(39.6)	6,342	(6.8)	86,464	(93.2)
Charlson Comorbidity Index (CCI)								< 0.0001
	0–2	82,830	(35.3)	2,369	(2.9)	80,461	(97.1)
	3–4	79,066	(33.7)	3,985	(5.0)	75,081	(95.0)
	≥5	72,738	(31.0)	8,684	(11.9)	64,054	(88.1)
Sleep disorder								< 0.0001
	Yes	40,061	(17.1)	2,015	(5.0)	38,046	(95.0)
	No	194,573	(82.9)	13,023	(6.7)	181,550	(93.3)
Diabetes								< 0.0001
	Yes	8,466	(3.6)	684	(8.1)	7,782	(91.9)
	No	226,168	(96.4)	14,354	(6.3)	211,814	(93.7)
Hypertension								< 0.0001
	Yes	26,525	(11.3)	2,245	(8.5)	24,280	(91.5)
	No	208,109	(88.7)	12,793	(6.1)	195,316	(93.9)
Depression								< 0.0001
	Yes	791	(0.3)	110	(13.9)	681	(86.1)
	No	233,843	(99.7)	14,928	(6.4)	218,915	(93.6)

Association between the use of hypnotics and the risk of Alzheimer’s disease

Table 2 shows the association between the use of hypnotics and the risk of Alzheimer’s disease after controlling for all covariates. Model 1 presents the association between the use of hypnotics and the risk of Alzheimer’s disease. Model 2 presents the association after adjusting for additional factors such as age, sex, social security, region, disability, and income; while model 3 shows the relationship after adjusting for age, sex, social security, region, disability, income, CCI score, sleep disorder, diabetes, hypertension, and depression. In model 3, individuals who used long-acting hypnotics had a higher risk of Alzheimer’s disease than non-users (hazard ratio [HR] = 1.28, 95%confidence interval [CI] = 1.09–1.51). Participants using short-acting hypnotics showed a higher risk of Alzheimer’s disease than non-users but a lower risk than individuals who used long-acting hypnotics (HR = 1.08, 95%CI = 1.02–1.15). Compared to individuals aged 60–65 years, individuals with the age of 65 or older demonstrated a significant increase in the risk of Alzheimer’s disease (age 65–70 years; HR = 1.85, 95%CI = 1.75–1.95, age 70–75 years; HR = 2.99, 95%CI = 2.83–3.15, age 75–80 years; HR = 4.26, 95%CI = 4.03–4.50, age ≥80 years; HR = 5.00, 95%CI = 4.73–5.29). Moreover, compared to those with 0–2 score of CCI, the risk of Alzheimer’s disease increased according to the CCI score (CCI score: 3–4: HR = 2.04, 95%CI = 1.94–2.14; CCI score≥5: HR = 4.49, 95%CI = 4.29–4.70). Individuals with hypertension had a higher risk of Alzheimer’s disease than those without (HR = 1.06, 95%CI = 1.01–1.10). Further, individuals with depression had a higher risk of Alzheimer’s disease than those without (HR = 2.15, 95%CI = 1.80–2.57).

Table 2

Results of association between use of hypnotics and the risk of Alzheimer’s disease

Variables		Alzheimer’s disease
	Model 1^a		Model 2^b		Model 3^c
	HR	95%CI	HR	95%CI	HR	95%CI
Hypnotics
Long-acting	1.28	(1.09–1.50)	1.03	(0.88–1.21)	1.28	(1.09–1.51)
Short-acting	1.03	(0.98–1.09)	0.84	(0.80–0.89)	1.08	(1.02–1.15)
No	1.00		1.00		1.00
Sex
Male			1.00		1.00
Female			1.51	(1.46–1.57)	1.58	(1.53–1.64)
Age
50–59			0.27	(0.26–0.29)	0.32	(0.30–0.34)
60–65			1.00		1.00
65–70			1.87	(1.77–1.97)	1.85	(1.75–1.95)
70–75			2.94	(2.78–3.10)	2.99	(2.83–3.15)
75–80			3.85	(3.65–4.07)	4.26	(4.03–4.50)
≥80			3.70	(3.50–3.91)	5.00	(4.73–5.29)
Social security
Health insurance (corporate)			1.00		1.00
Health insurance (regional)			1.03	(0.99–1.07)	1.01	(0.97–1.05)
Region
Metropolitan			1.00		1.00
City			1.18	(1.14–1.23)	1.19	(1.14–1.23)
Rural			1.14	(1.11–1.19)	1.09	(1.05–1.13)
Disability
Yes			1.13	(1.07–1.19)	1.23	(1.17–1.29)
No			1.00		1.00
Income
Low			1.00		1.00
Middle			0.94	(0.90–0.98)	0.87	(0.83–0.91)
High			0.99	(0.94–1.03)	0.90	(0.87–0.94)
Charlson Comorbidity Index (CCI)
0–2					1.00
3–4					2.04	(1.94–2.14)
≥5					4.49	(4.29–4.70)
Sleep disorder
Yes					1.00
No					2.32	(2.23–2.43)
Diabetes
Yes					0.96	(0.89–1.03)
No					1.00
Hypertension
Yes					1.06	(1.01–1.10)
No					1.00
Depression
Yes					2.15	(1.80–2.57)
No					1.00

^aAdjusted only hypnotics. ^bAdjusted for hypnotics, age, sex, social security, region, disability, and income. ^cAdjusted for hypnotics, age, sex, social security, disability, income, CCI, sleep disorder, diabetes, hypertension, and depression.

Table 3 shows the results of the subgroup analyses by sex, region, disability, income, and sleep disorder. Among women, those using long-acting hypnotics had a higher risk of Alzheimer’s disease than non-users (HR = 1.26, 95%CI = 1.04–1.53). Among individuals living in rural areas, those using long-acting hypnotics had a higher risk of Alzheimer’s disease than non-users (HR = 1.46, 95%CI = 1.14–1.86). Among individuals with low-income levels, the risk of Alzheimer’s disease was higher in those who used long-acting hypnotics than non-users (HR = 1.41, 95%CI = 1.02–1.95). Among individuals with sleep disorders, those who used hypnotics had a higher risk of Alzheimer’s disease than non-users (long-acting: HR = 1.38, 95%CI = 1.13–1.69; short-acting: HR = 1.09, 95%CI = 1.00–1.18).

Table 3

Subgroup analysis of the association between risk of Alzheimer’s disease and covariates, according to use of hypnotics

Variables		Hypnotics
	Long acting		Short acting		No
	Adjusted HR	95%CI lower - upper	Adjusted HR	95%CI lower - upper	Adjusted HR
Sex
Male	1.33	(0.99–1.80)	1.06	(0.94–1.19)	1.00
Female	1.26	(1.04–1.53)	1.09	(1.02–1.17)	1.00
Region
Metropolitan	1.19	(0.88–1.61)	1.10	(1.00–1.22)	1.00
City	1.15	(0.84–1.58)	1.05	(0.93–1.18)	1.00
Rural	1.46	(1.14–1.86)	1.08	(0.99–1.18)	1.00
Income
Low	1.41	(1.02–1.95)	1.05	(0.93–1.19)	1.00
Middle	1.28	(0.95–1.71)	1.11	(1.00–1.23)	1.00
High	1.22	(0.95–1.56)	1.08	(0.99–1.17)	1.00
Disability
Yes	1.37	(0.89–2.12)	1.12	(0.94–1.32)	1.00
No	1.27	(1.06–1.51)	1.08	(1.01–1.14)	1.00
Sleep disorder
Yes	1.38	(1.13–1.69)	1.09	(1.00–1.18)	1.00
No	1.15	(0.87–1.51)	1.08	(0.99–1.17)	1.00

DISCUSSION

For aging societies, it has become increasingly important to prevent or manage dementia by alleviating the symptoms [11, 13]. The purpose of this study was to determine the association between the use of hypnotics, classified into long- and short-acting drugs, and the risk of Alzheimer’s disease. The findings provide suggestive evidence that could be used to develop policies to protect the older population from Alzheimer’s disease by avoiding inappropriate and extensive use of hypnotic drugs [21]. Our findings suggest that the use of long-acting hypnotics increased the risk of Alzheimer’s disease. Additionally, the use of short-acting hypnotics also increased the risk of Alzheimer’s disease but to a lesser degree. Hence, it is crucial to direct attention towards the possible development of symptoms of Alzheimer’s disease, in individuals who use hypnotics.

However, it is unclear how hypnotics could increase the risk of Alzheimer’s disease. One study showed that benzodiazepines could lower the brain activity, which could affect cognitive functions [20]. Moreover, there are studies that found an association between the use of hypnotics and cognitive decline [31, 32]. A previous study has shown that patients with Alzheimer’s disease who were taking hypnotics were more likely to have a faster rate of deterioration than those who were not taking these drugs [32]. In corroboration with the previous findings, our study found an association between the use of hypnotics and the risk of Alzheimer’s disease.

The association between the use of long- and short-acting hypnotics and an increased risk of dementia has been reported previously [33 –36]. Partly, in agreement with our findings, a study found that compared to non-users, individuals who used either long- or short-acting hypnotics had a higher risk of dementia [33]. Similarly, it was reported that ever-use of long-acting benzodiazepines is associated with an increased risk of Alzheimer’s disease, based on Canadian data [34]. Moreover, long-acting benzodiazepines were considered potentially as inappropriate medications, especially for older people [35, 36]. According to the Beer’s criteria, the use of benzodiazepines is specifically related to an increased risk of falls and fractures due to prolonged sedation or increased sensitivity [37, 38]. As long-acting benzodiazepines accumulate in the blood of older people more easily, the risk of dementia increases [39]. Therefore, more cautious use is warranted in older adults than in younger adults.

The risk of insomnia is greatest in the elderly, which leads to poor sleep quality, difficulty in falling asleep, and frequent awakening during the night [40]. Furthermore, some studies have shown that insomnia is associated with early manifestation of Alzheimer’s disease [41, 42]. People with insomnia are treated with hypnotics, which could lead to dementia [32]. In South Korea, medication using hypnotics was the most frequently used treatment for those who had sleep problems like insomnia [43]. Moreover, previous studies have shown that long-acting hypnotics increase the risk of depression and anxiety disorders [43, 44]. Our study also found that individuals with sleep disorders who used long-acting hypnotics had a higher risk of Alzheimer’s disease than non-users.

It should be noted that the current study has several limitations. First, factors that could affect the risk of Alzheimer’s disease, such as smoking status and frequency of alcohol consumption, were not included due to missing data, as this study used claims data that were collected to make payments to both patients and providers based on medical utilization. However, by identifying the presence of diabetes and hypertension, the health status of individuals could be evaluated. Second, because we used drug prescription data, purchase of hypnotics without prescription could not be controlled for. Lastly, because we focused only on benzodiazepines in the study, we could not analyze the effects of other hypnotics. Therefore, further studies including non-benzodiazepines are warranted.

Despite the limitations, this study has several strengths, including that our research used NHIS-NSC data to assess the association between the use of hypnotics and risk of Alzheimer’s disease. The data represent almost all the affected individuals in South Korea, and provide real evidence for the development of policies, to prevent the unnecessary use of hypnotics. Moreover, this study included data collected over a period of nearly 10 years and considering that dementia takes a long time to develop, this study could demonstrate the association more accurately. In addition, we divided the hypnotics into long- and short-acting to obtain more specific information.

In conclusion, the current study identified an association between the use of hypnotics and the risk of Alzheimer’s disease among the middle-aged and older South Korean population. Our findings revealed that individuals who use benzodiazepines, especially long-acting benzodiazepines, have a higher risk of Alzheimer’s disease. This implies that caution should be exercised when prescribing long-acting hypnotics, especially for older people. Future studies to understand the risk of Alzheimer’s disease associated with the use of hypnotics are needed.

Footnotes

ACKNOWLEDGMENTS

We would like to thank our colleagues from the Department of Public Health, Graduate School of Yonsei University, who provided valuable advice regarding this manuscript.

This research was supported by a grant of the Korea Health Technology R&D Project through the Korea Health Industry Development Institute (KHIDI), funded by the Ministry of Health & Welfare, Republic of Korea (grant number: HI20C1130).

Authors’ disclosures available online ().

References

Hwang

, Han

, Choi

, Park

, Kim

(2017) Trends in the prescription of benzodiazepines for the elderly in Korea. BMC Psychiatry 17, 303.

Ministry of Health and Welfare (2016) Final report on the treatment status and health effects on insomnia in Korea. Korea Centers for Disease Control and Prevention.

Ministry of Health and Welfare (2019) Report of Ministry of Health andWelfare: The overall improvement in the quality of health care in Korea. Korea Centers for Disease Control and Prevention.

Organization for Economic Cooperation and Development (2019) OECD Health Statistics 2019. Organization for Economic Cooperation and Development.

, Oh

, Lee

, Woo

, Lee

, Hwang

, Park

, Kwak

, Kwon

(2014) In-depth investigation for prescribing trends of benzodiazepines in South Korea. Int J Clin Pharmacol Ther 52, 460–470.

Kaufmann

, Spira

, Alexander

, Rutkow

, Mojtabai

(2016) Trends in prescribing of sedative-hypnotic medications in the USA: 1993–2010. Pharmacoepidemiol Drug Saf 25, 637–645.

Limon

, ReyesRuiz

, Miledi

(2012) Loss of functional GABA a receptors in the Alzheimer diseased brain. Proc Natl Acad Sci U S A 109, 10071–10076.

Stranks

, Crowe

(2014) The acute cognitive effects of zopiclone, zolpidem, zaleplon, and eszopiclone: A systematic review and meta-analysis. J Clin Exp Neuropsychol 36, 691–700.

Baek

, Lee

, Kim

, Chung

, Pratt

, KalischEllett

, Shin

(2020) Uncertain association between benzodiazepine use and the risk of dementia: A cohort study. J Am Med Direct Assoc 21, 201–211.

10.

Lee

, Jung

, Choi

, Shin

, Lee

(2018) Use of sedative-hypnotics and the risk of alzheimer’s dementia: A retrospective cohort study. PloS One 13, e0204413.

11.

World Health Organizaion (2015) World alzheimer report 2015: The global impact of dementia.

12.

World Health Organizaion (2019) Risk reduction of cognitive decline and dementia: WHO guidelines.

13.

Ministry of Health and Welfare (2017) Development of a model for early predict of dementia. Korea Centers for Disease Control and Prevention.

14.

Ministry of Health and Welfare (2015) 3rd comprehensive plan for dementia(’16 ’20). Korea Centers for Disease Control and Prevention.

15.

World Health Organizaion (2019) Dementia in theWestern Pacific.

16.

BilliotideGage

, Bégaud

, Bazin

, Verdoux

, Dartigues

, Pérès

, Kurth

, Pariente

(2012) Benzodiazepine use and risk of dementia: Prospective population based study. BMJ 345, e6231.

17.

Gallacher

, Elwood

, Pickering

, Bayer

, Fish

, BenShlomo

(2012) Benzodiazepine use and risk of dementia: Evidence from the Caerphilly Prospective Study (CaPS). J Epidemiol Community Health 66, 869.

18.

, Wang

, Chang

, Lin

(2009) The association between dementia and long-term use of benzodiazepine in the elderly: Nested case–control study using claims data. Am J Geriatr Psychiatry 17, 614–620.

19.

Islam

, Iqbal

, Walther

, Atique

, Dubey

, Nguyen

, Poly

, Masud

JHB

, Li

, Shabbir

(2016) Benzodiazepine use and risk of dementia in the dlderly population: A systematic review and meta-analysis. Neuroepidemiology 47, 181–191.

20.

Pariente

, DeGage

, Moore

, Bégaud

(2016) The benzodiazepine–dementia disorders link: Current state of knowledge. CNS Drugs 30, 1–7.

21.

Olsson

, Bergman

, Carlsten

, Oké

, Bernsten

, Schmidt

, Fastbom

(2010) Quality of drug prescribing in dlderly people in nursing homes and special care units for dementia. Clin Drug Invest 30, 289–300.

22.

Park

, Kim

, Baek

, Sung

, Park

, Lee

, Shin

(2018) Signal detection of benzodiazepine use and risk of dementia: Sequence symmetry analysis using South Korean national healthcare database. Int J Clin Pharm 40, 1568–1576.

23.

Kang

, Park

, Lee

, Kim

, Namkoong

, Kim

(2017) Association between the use of benzodiazepines and the occurrence of acute angle-closure glaucoma in the elderly: A population-based study. Eur Psychiatry 41, S655–S656.

24.

Sundararajan

, Henderson

, Perry

, Muggivan

, Quan

, Ghali

(2004) New ICD-10 version of the charlson comorbidity index predicted in-hospital mortality. J Clin Epidemiol 57, 1288–1294.

25.

Andersen

, Gill

(1982) Cox’s regression model for counting processes: A large sample study. Ann Stat 10, 1100–1120.

26.

Wolkewitz

, Allignol

, Harbarth

, de Angelis

, Schumacher

, Beyersmann

(2012) Time-dependent study entries and exposures in cohort studies can easily be sources of different and avoidable types of bias. J Clin Epidemiol 65, 1171–1180.

27.

Fisher

, Lin

(1999) Time-dependent covariates in the cox proportional-hazards regression model. Ann Rev Public Health 20, 145–157.

28.

Cox

(1972) Regression models and life tables. J R Stat Soc Ser B 34, 187–220.

29.

Bewick

, Cheek

, Ball

(2004) Statistics review 12: Survival analysis. Crit Care 8, 389–394.

30.

Clark

, Bradburn

, Love

, Altman

(2003) Survival analysis part I: Basic concepts and first analyses. Br J Cancer 89, 232–238.

31.

Barker

, Greenwood

, Jackson

, Crowe

(2004) Persistence of cognitive effects after withdrawal from long-term benzodiazepine use: A meta-analysis. Arch Clin Neuropsychol 19, 437–454.

32.

Chen

, Lee

, Sun

, Oyang

, Fuh

(2012) Risk of dementia in patients with insomnia and long-term use of hypnotics: A population-based retrospective cohort study. PLoS One 7, e49113.

33.

Tseng

, Huang

, Peng

, Chen

, Hsiao

(2020) Benzodiazepines, z-hypnotics, and risk of dementia: Special considerations of half-lives and concomitant use. Neurotherapeutics 17, 156–164.

34.

DeGage

, Moride

, Ducruet

, Kurth

, Verdoux

, Tournier

, Pariente

, Bégaud

(2014) Benzodiazepine use and risk of alzheimer’s disease: Case-control study. BMJ 349, g5205.

35.

Laroche

, Charmes

, Merle

(2007) Potentially inappropriate medications in the elderly: A French consensus panel list. Eur J Clin Pharmacol 63, 725–731.

36.

O’Mahony

, Gallagher

, Ryan

, Byrne

, Hamilton

, Barry

, O’Connor

, Kennedy

(2010) STOPP & START criteria: A new approach to detecting potentially inappropriate prescribing in old age. Eur Geriatr Med 1, 45–51.

37.

Beers

, Ouslander

, Rollingher

, Reuben

, Brooks

, Beck

(1991) Explicit criteria for determining inappropriate medication use in nursing home residents. UCLA Division of Geriatric Medicine. Arch Intern Med 151, 1825–1832.

38.

Ackroyd-Stolarz

, Mackinnon

, Sketris

, Sabo

(2009) Potentially inappropriate prescribing of benzodiazepines for older adults and risk of falls during a hospital stay: A descriptive study. Canad J Hosp Pharm 62, 276–283.

39.

Bogunovic

, Greenfield

(2004) Practical geriatrics: Use of benzodiazepines among elderly patients. Psychiatr Serv 55, 233–235.

40.

Srinivasan

, Brzezinski

, PandiPerumal

, Spence

, Cardinali

, Brown

(2011) Melatonin agonists in primary insomnia and depression-associated insomnia: Are they superior to sedative-hypnotics? Prog Neuropsychopharmacol Biol Psychiatry 35, 913–923.

41.

Osorio

, Pirraglia

, AgüeraOrtiz

, During

, Sacks

, Ayappa

, Walsleben

, Mooney

, Hussain

, Glodzik

, Frangione

, MartínezMartín

, DeLeon

(2011) Greater risk of Alzheimer’s disease in older adults with insomnia. J Am Geriatr Soc 59, 559–562.

42.

, Chen

, Jinn

(2010) Alzheimer’s disease: Aging, insomnia and epigenetics. Taiwan J Obstet Gynecol 49, 468–472.

43.

Ministry of Health and Welfare (2017) A survey on the treatment and health impact on insomnia in South Korea. Ministry of Health and Welfare.

44.

Chung

, Li

, Kuo

, Sithole

, Liu

, Chung

Risk of psychiatric disorders in patients with chronic insomnia and sedative-hypnotic prescription: A nationwide population-based follow-up study. J Clin Sleep Med 11, 543–551.