Presence of a Synergistic Interaction Between Current Cigarette Smoking and Diabetes Mellitus on Development of Dementia in Older Adults

Abstract

Background:

Both cigarette smoking and diabetes mellitus are well-established risk factors for development of dementia. However, the interaction between smoking and diabetes is yet unknown.

Objective:

In this study, we clarify association between smoking, diabetes, and dementia risk in older adults.

Methods:

Participants in this study included community residents aged 65 years and older who had participated in a health checkup in 2006, followed for 10 years (n = 9,403) and had long-term care insurance information data. Furthermore, the risk estimates of smoking status and diabetes diagnosis on dementia adjusted for the competing risk of death prior to dementia were analyzed.

Results:

During follow-up, 2,647 participants developed dementia. The smoking status–diabetes interaction on development of dementia was statistically significant (p≤0.001). Among those patients exposed to both factors, 17% of risk of development of dementia was attributable to the interaction of these factors. Current smokers with diabetes had significantly greater risks of development of dementia than never smokers without diabetes (reference): multivariable-adjusted risk of dementia in current smokers without diabetes (subdistribution hazard ratio [sHR], 1.25; 95% confidence interval [CI], 1.05–1.48); never smokers with diabetes (1.31, 1.16–1.47); and current smokers with diabetes (1.86, 1.39–2.48). However, no such association was noted for former smokers with and without diabetes.

Conclusions:

Current smoking, but not former smoking, was associated with increased risk of development of dementia in older adults with and without diabetes. Moreover, the synergistic effect of current smoking and diabetes on dementia was noted.

Keywords

Cigarette smoking dementia diabetes mellitus insurance long-term care

INTRODUCTION

Diabetes mellitus and cigarette smoking are significant risk factors for the development of dementia [1 –9]. The association between cigarette smoking and diabetes mellitus risk is well established [10, 11]; however, the increased risk of dementia in a current smoker with diabetes mellitus is yet to be clarified. In this study, we hypothesized that cigarette smoking enhanced the risk of development of dementia in older adults with diabetes mellitus and therefore evaluated our hypothesis using the data obtained from Komatsu City, Ishikawa Prefecture, Japan.

MATERIALS AND METHODS

Study population

In Japan, health checkups are annually carried out on residents with national health insurance. These are done to identify early signs of illness and promote health. However, participation in these checkups is not mandatory. The long-term care insurance (LTCI) is a mandatory social insurance to assist the disabled older adults with their activities of daily living [12]. The insured person aged ≥65 years is eligible for the LTCI insurance service.

The present study was conducted using the annual health checkup, LTCI certification, and death or emigration data of Komatsu City, Ishikawa Prefecture, Japan, between April 2006 and March 2017. The total population of Komatsu City in 2006 was 109,721 (53,237 men and 56,484 women), and the number of older people aged ≥65 years was 22,481 (9,335 men and 13,146 women).

The participants of this study included a total of 11,438 residents (4,375 men and 7,063 women) aged ≥65 years who had a health checkup in 2006 (participation rate, 50.8%) and who were followed up from April 1, 2007. Compared with the total population aged ≥65 years of Komatsu City (n = 22,481), the participants who were checked up on (n = 11,438) were mostly female (p < 0.001). We excluded the individuals whose LTCI information data were unavailable (n = 1,813) and who had already been certified as having dementia by the LTCI before follow-up (n = 213) and who had no blood data (n = 9). Thus, the data from 9,403 residents were analyzed for the purposes of this study.

We obtained a dataset from Komatsu City. Moreover, the health checkup, LTCI, and death or emigration data were linked at the individual participant level for analysis.

Ethical issues

This study was approved by the medical ethics review board of the Kanazawa University, Kanazawa, Japan (approval number 864). All data were transferred from the Komatsu City government under an agreement paper related to Academic Research and Privacy Protection. Consent for use of the dataset was granted by the municipal government of Komatsu City after a formal application, along with an explicit pledge to protect the confidentiality of the data supplied.

Follow-up and diagnosis of dementia

The primary outcome was incident dementia. The diagnosis of dementia was defined using the criteria of the LTCI system. According to the standardized physicians’ manual issued by the Ministry of Health, Labor and Welfare of Japan [13], the grading of the Dementia Scale is summarized in Supplementary Table 1. Each municipality’s Certification Committee for Long-Term Care Needs has assigned the levels of the Dementia Scale based on both the opinion of a primary physician and that of a home-visit interview. According to the LTCI system, dementia was defined as incident functional disability rank II or higher on the Dementia Scale (Degree of Independence in Daily Living for Elderly Patients with Dementia) [14]. The Dementia Scale is classified into eight ranks (0, I, IIa, IIb, IIIa, IIIb, IV, M), and a rank exceeding I is usually used as an outcome measure of incident dementia because individuals who have mild or moderate dementia are classified as rank IIa [14].

Risk factor measurements

During the health checkup in 2006, the participants completed a self-administered lifestyle questionnaire regarding cigarette smoking habit, alcohol consumption, and medical history. Cigarette smoking status was classified as never, former, or current, while alcohol consumption was classified as current use or not. Blood pressure was measured three times using a standard mercury sphygmomanometer in the sitting position after at least 5 min of rest, and the mean of three measurements was used for the analysis. Hypertension was defined as a blood pressure reading of 140/ 90 mmHg or greater or a medical history of hypertension such as the use of antihypertensive drugs. An electrocardiographic abnormality was defined as ST depression or arrhythmia. Body height and weight were measured, and body mass index (BMI) (kg/m²) was calculated as an indicator of obesity. In addition, blood samples were collected during the annual health checkup, and the hemoglobin A1c (HbA1c) and serum total cholesterol levels were determined for all participants. HbA1c was examined using a high-performance liquid chromatography, and its value was determined by the Japan Diabetes Society (JDS) method, which is 0.4% lower than the National Glycohemoglobin Standardization Program value [15]. Based on the Japan National Diabetes Survey, HbA1c ≥6.1% was used as a criterion for probable diabetes mellitus [15]. The serum total cholesterol levels were measured enzymatically. Moreover, hyperlipidemia was defined as a serum total cholesterol level of 240 mg/dL or greater or a medical history of hyperlipidemia such as the use of antihyperlipidemic drugs.

Statistical analyses

The person-years of follow-up for each participant were counted from April 2007 until the date of incident dementia, transfer from Komatsu City, or death or the end of the study period (March 2017), whichever occurred first.

The baseline characteristics were evaluated using the ANOVA and chi-square test for continuous and categorical variables, respectively. For the interaction analysis, the Fine-Gray subdistribution hazards model and binominal distribution, logistic regression model were conducted to analyze two main variables (smoking and diabetes mellitus) and their interaction to evaluate the interaction effect [16, 17]. The participants without diabetes mellitus who smoked but who did not do so currently were categorized as the reference. Then, the relationship between cigarette smoking status, diabetes mellitus diagnosis, and incident dementia which accounts for the competing risk of death prior to dementia was examined using the multivariable-adjusted model including age, sex, systolic blood pressure, total cholesterol, electrocardiogram abnormality, and alcohol consumption. Additionally, to investigate the effects of the difference between the former and never smokers on the development of dementia, we used the Fine-Gray subdistribution hazards model to calculate the subdistribution hazard ratios (sHRs) and 95% confidence interval (CI) of the incidence of dementia which accounts for the competing risk of death prior to dementia according to cigarette smoking status (never, former, current) and diabetes mellitus diagnosis. The participants without diabetes mellitus who never smoked were categorized as the reference. In addition, the relationship between cigarette smoking status, diabetes mellitus diagnosis, and dementia incidence was examined using the Fine-Gray subdistribution hazards model with adjusted potential confounders as mentioned earlier.

Data were reported as mean±standard deviation unless otherwise specified. p < 0.05 was considered statistically significant. Furthermore, all statistical analyses were performed with EZR (Saitama Medical Center, Jichi Medical University, Saitama, Japan), which is a graphical user interface for R (The R Foundation for Statistical Computing, Vienna, Austria). More precisely, it is a modified version of R commander designed to add statistical functions frequently used in biostatistics [18].

RESULTS

In the final analysis, compared with the population lost to follow-up (n = 1,813), the sociodemographic characteristics of the study population (n = 9,403) were the following: older age (p≤0.001); mostly female (p≤0.001); higher systolic blood pressure level (p≤0.001); and BMI (p≤0.001); more frequent hypertension (p≤0.001) and hyperlipidemia (p≤0.01); less electrocardiogram abnormality percentage (p≤0.001); lower serum total cholesterol level (p≤0.001); and fewer number of former and current smokers (p≤0.001) at baseline (Supplementary Table 2). Additionally, the clinical characteristics of the study population are summarized according to the smoking status and diabetes mellitus diagnosis in Table 1. Compared with never smokers, former smokers were significantly more likely to be male, consume alcohol, have higher diastolic blood pressure, higher HbA1c (JDS) levels, and higher BMI, higher electrocardiogram abnormality percentage, and lower serum total cholesterol level. Also, the current smokers were significantly more likely to be male, consume alcohol, and have a higher HbA1c (JDS) level and lower BMI and serum total cholesterol level. The mean values of the systolic blood pressure, BMI, hypertension and hyperlipidemia frequency, and electrocardiogram abnormality percentage were higher in the participants with probable diabetes mellitus than in those participants with HbA1c ≤6.0%. Furthermore, these participants were significantly more likely to be male, consume alcohol, and have lower diastolic blood pressure and serum total cholesterol levels.

Table 1

Clinical characteristics of the study population

	Smoking			Probable diabetes mellitus
Characteristic n	Never n = 7461	Former n = 1134	Current n = 808	No n = 8178	Yes n = 1225
Age	75.04 ± 6.04	74.84 ± 5.72	73.87 ± 5.52^a	75.89 ± 6.00	75.10 ± 5.75
Male (%)	23.6	91.4^a	86.3^a	35.8	46.0^d
Systolic blood pressure (mmHg)	133.62 ± 16.34	133.77 ± 17.14	133.90 ± 17.04	133.38 ± 16.47	135.55 ± 16.53^d
Diastolic blood pressure (mmHg)	76.17 ± 10.10	77.07 ± 10.45^c	75.99 ± 10.54	76.42 ± 10.14	75.19 ± 10.35^d
Hypertension (%)	45.2	47.6	40.2	44.3	50.3^d
Electrocardiogram abnormality (%)	35.1	47.7^a	39.4	36.2	42.1^d
HbA1c (JDS) (%)	5.43 ± 0.69	5.52 ± 0.78^a	5.57 ± 0.81^a	5.24 ± 0.36	6.88 ± 0.86^d
Probable diabetes mellitus (%)	11.8	17.2	18.2	0	100
Body mass index (kg/m²)	22.97 ± 3.19	23.23 ± 2.97^a	22.64 ± 3.11^a	22.85 ± 3.12	23.77 ± 3.33^d
Serum total cholesterol	202.29 ± 32.12	189.58 ± 32.94^a	188.51 ± 33.00^a	199.87 ± 32.59	197.58 ± 33.58^e
Hyperlipidemia (%)	44.0	35.2^a	36.5^a	41.9	45.2^f
Smoking status (former, %)	0	100	0	11.5	15.9
Smoking status (current, %)	0	0	100	8.1	12.0
Alcohol consumption (%)	25.2	62.9^a	65.5^a	32.8	36.2^f

^ap < 0.001; ^bp < 0.01; ^cp < 0.05 versus the never smokers; ^dp < 0.001; ^ep < 0.01; ^f p < 0.05 versus the diabetes mellitus (-) group.

During the follow-up of 9,403 participants for 10 years, 2,647 of participants developed dementia and 649 of participants died prior to dementia. The follow-up time was 8.67±2.31 years. According to Fine-Gray subdistribution hazards model, the smoking status–diabetes mellitus interaction on development of dementia was statistically significant (p≤0.001). The relative excess risk due to the interaction (RERI) effect between current smoking and diabetes mellitus was estimated as 1.85 − 1.31 − 1.23 + 1.00 = 0.30, the attributable proportion (AP) due to interaction was 0.30/1.85 = 0.17, and the synergy index (SI) was (1.85 − 1)/[(1.31 − 1)+(1.23 − 1)] = 1.56, implying that 17% of the risk among both current smoking and diabetes mellitus of development of dementia was caused by their interaction (Table 2). A significant synergistic interaction between the two factors affecting the development of dementia was noted, suggesting that the combined effect of current smoking and diabetes mellitus on development of dementia was greater than the sum of the individual effects of the two factors (p < 0.001, RERI > 0, AP > 0, SI > 1). Additionally, the interaction analysis using logistic regression model also showed that the smoking status–diabetes mellitus interaction on development of dementia was statistically significant (Wald test = 31.224 on 3df, p≤0.001) (Supplementary Table 3). The lower limit of the CI of current smokers with diabetes mellitus was lower than the upper limit of the CI of current smokers without diabetes mellitus and not currently smokers with diabetes mellitus; these overlaps of CI were recognized. The association between smoking status, diabetes mellitus diagnosis, and risk of development of dementia was estimated in Table 3. With regard to the incidence of dementia, the multivariable-adjusted sHRs were as follows: never smokers without diabetes mellitus, 1.00 (reference); former smokers without diabetes mellitus, 1.04 (95% CI, 0.89–1.21; p = 0.640); current smokers without diabetes mellitus, 1.25 (1.05–1.48; p≤0.001); never smokers with diabetes mellitus, 1.31 (1.16–1.47; p≤0.001); former smokers with diabetes mellitus, 1.26 (0.95–1.65; p = 0.100); and current smokers with diabetes mellitus, 1.86 (1.39–2.48; p≤0.001).

Table 2

Multivariable-adjusted risks of the development of dementia according to smoking status and diabetes mellitus diagnosis

	β	HR (95% CI)	p
Current smoking	0.21	1.23 (1.04–1.45)	0.001
Diabetes (+)	0.27	1.31 (1.17–1.46)	<0.001
Current smoking/diabetes (+)	0.61	1.85 (1.39–2.47)	<0.001
RERI	0.30
AP	0.17
SI	1.56

Multivariable adjustment was made for age, sex, systolic blood pressure, total cholesterol, electrocardiogram abnormality, and alcohol consumption. HR, hazard ratio; CI, confidence interval; RERI, relative excess risk due to the interaction; AP, attributable proportion due to interaction; SI, synergy index.

Table 3

Multivariate-adjusted risk of the development of dementia according to smoking status and diabetes mellitus diagnosis

	Person-years at risk	Dementia (n)	HR (95% CI)	p	Death prior to death (n)	HR (95% CI)	p
Never smoking/diabetes (–)	51,891	1,943	Reference		359	Reference
Former smoking/diabetes (–)	7,399	244	1.04 (0.89–1.21)	0.640	105	1.33 (1.05–1.71)	0.020
Current smoking/diabetes (–)	5,095	178	1.25 (1.05–1.48)	<0.01	77	1.56 (1.20–2.04)	<0.001
Never smoking/diabetes (+)	6,711	324	1.31 (1.16–1.47)	<0.001	66	1.28 (0.99–1.67)	0.057
Former smoking/diabetes (+)	1,504	56	1.26 (0.95–1.65)	0.100	24	1.49 (0.99–2.27)	0.057
Current smoking/diabetes (+)	1,117	52	1.86 (1.39–2.48)	<0.001	18	1.63 (1.01–2.63)	0.045

Multivariable adjustment was made for age, sex, systolic blood pressure, total cholesterol, electrocardiogram abnormality, and alcohol consumption. HR, hazard ratio; CI, confidence interval.

DISCUSSION

In this prospective longitudinal study, we demonstrated that smoking habit and diabetes mellitus were significant risk factors for development of dementia. Moreover, a significant synergistic interaction between the two factors affecting the development of dementia was noted, suggesting that the combined effect of current smoking and diabetes mellitus on development of dementia was greater than the sum of the individual effects of the two factors. In particular, the current smokers with diabetes mellitus were associated with a greater risk of dementia, whereas the sHRs regarding the incidence of dementia in the former smokers with diabetes mellitus were almost the same as those of the never smokers with diabetes mellitus. Furthermore, our results showed that no significant association was observed between dementia risk and the former smokers without diabetes mellitus. These findings suggested that abstaining from smoking reduces the risk of development of dementia in older adults with and without diabetes mellitus.

The findings of this study suggest that current smoking is an independent risk factor for development of dementia. Consistent with our results, most population-based prospective studies have reported that current smoking is a risk factor for the development of dementia, especially Alzheimer’s disease (AD) [5 –9]. Regarding vascular dementia (VaD), a few prospective studies have examined the association between smoking and VaD risk, but the findings were inconsistent [7–9 , 20]. In some studies, late life or midlife smoking was significantly associated with risk of VaD [7 , 20], whereas no such association was observed in other studies [8, 19]. The pathophysiological mechanisms underlying the association between smoking and increased dementia risk including AD and VaD remain unclear. A study using an animal model of AD reported that smoking increases the severity of amyloidogenesis, neuroinflammation, and tau phosphorylation [21]. Smoking may increase VaD because it is a major risk factor for cerebrovascular diseases. As a result of smoking, the state of diabetes mellitus may be exacerbated, which may increase the risks of dementia. Additionally, the gas of cigarette smoke has extremely high concentrations of free radical species causing oxidative stress such as reactive oxygen and nitrogen species [22]. Smoking-related oxidative stress may initiate and promote the AD pathophysiology and increased risk of AD in patients with diabetes mellitus.

In our study, we implied that diabetes mellitus was an independent risk factor for development of dementia, which was in agreement with the findings of previous studies [1 –4]. Diabetes mellitus increases the risks of AD and VaD. Four major mechanisms underlying diabetes mellitus-induced dementia have been proposed [23]: atherosclerosis; microvascular disease; glucose toxicity leading to the accumulation of advanced glycated proteins and increased oxidative stress; and change in the insulin metabolism resulting in an insulin-resistant state and distorted amyloid metabolism in the brain. The former two pathways are considered to be involved in the development of VaD, while the latter two pathways may mainly contribute to the development of AD [23].

Smoking is a major risk factor for the exacerbation of diabetes mellitus, although the exacerbation pathomechanisms of diabetes mellitus remain unclear. In the blood, smoking may cause a decrease in the concentration of adiponectin, which has various biological functions including insulin sensitization [24]. As such, decreased blood adiponectin levels have been consistently associated with incidence of diabetes mellitus. Blood adiponectin levels were reported to be inversely associated with incidence of both smoking and diabetes mellitus [25], and adiponectin significantly mediated the association between current smoking and incidence of diabetes mellitus [25]. In a number of case–control and longitudinal studies, inconsistent findings regarding the association between blood adiponectin levels and AD were noted: some studies reported that the higher blood adiponectin levels were associated with an increased risk of AD [26 –29], while in others, no such association was observed [30, 31]. Further, one case–control study has shown that the decreased blood concentrations of adiponectin are associated with mild cognitive impairment and AD [32]. In cerebrospinal fluid (CSF), the decreased adiponectin levels were reported to be associated with AD, despite the elevated serum adiponectin levels in patients with AD [28]. Interestingly, CSF adiponectin levels were significantly correlated with CSF amyloid-β protein₄₂ and phosphorylated tau protein and hippocampal atrophy, suggesting that the loss of function of adiponectin plays a role in the pathogenesis of AD [28]. On the other hand, the increased CSF adiponectin levels were reported to be associated with mild cognitive impairment, suggesting that the adiponectin levels in the plasma reflect the level adiponectin in the CSF [26]. Although the mechanisms are unclear, adiponectin may play some significant roles in the synergistic effect of current smoking and diabetes mellitus on development of dementia, which was observed in the present study. Unfortunately, the adiponectin levels in either blood or CSF were not investigated in this study. Therefore, further studies in a large population involving both blood and CSF adiponectin levels are required to reveal the mechanisms underlying the exacerbation of diabetes mellitus and dementia through cigarette smoking.

The present study has a few limitations. Firstly, the “Dementia Scale,” which was our procedure for the diagnosis of dementia, is used only in Japan, not generally worldwide. The validation study [33] of the Dementia Scale provided epidemiological evidence for association of the Dementia Scale of LTCI with dementia defined by the diagnostic criteria in the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition [34]. The study suggests that the selection of grades ≥IIa as the cutoff point would provide considerable specificity of 96%, however, the sensitivity of 54% is not high [33]. Secondly, we did not investigate the following: the types of dementia; number of cigarettes smoked; presence or absence of complications with diabetes mellitus; or treatment of diabetes mellitus. Thirdly, the 75 g oral glucose tolerance test was not performed to assess diabetes mellitus. Fourthly, among the source population of 22,481, the participation rate (50.8%; n = 11,438) was not high that might provide a selection bias. Nonetheless, the advantages of the current study are its longitudinal design, large dataset used, relatively long follow-up duration, and opportunity to adjust for possible confounding factors.

In conclusion, diabetes mellitus and current smoking were considered as potential risk factors for dementia. Current smoking synergistically enhances the risk of the development of dementia in older adults with diabetes mellitus, while former smoking does not. Therefore, we emphasize the significance of cessation of smoking for the prevention of dementia.

Footnotes

ACKNOWLEDGMENTS

The authors thank the staff of the Healthy Living Department, Healthcare Insurance Support Center, Health Division, and Graceful Aging Division of Komatsu City for their cooperation in this study.

This study was supported in part by Topic-Setting Program to Advance Cutting-Edge Humanities and Social Sciences Research (Responding to Real Society), Japan Society for the Promotion of Science. The sponsors had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; or preparation, review, or approval of the manuscript.

Authors’ disclosures available online ().

The supplementary material is available in the electronic version of this article: .

References

Leibson

, Rocca

, Hanson

, Cha

, Kokmen

, O’Brien

, Palumbo

(1997) Risk of dementia among persons with diabetes mellitus: A population-based cohort study. Am J Epidemiol 145, 301–308.

Ott

, Stolk

, van Harskamp

, Pols

HAP

, Hofman

, Breteler

MMB

(1999) Diabetes mellitus and the risk of dementia: The Rotterdam study. Neurology 53, 1937–1942.

Peila

, Rodriguez

, Launer

(2002) Type 2 diabetes, ApoE gene, and the risk for dementia and related pathologies. The Honolulu-Asia aging study. Diabetes 51, 1256–1262.

Ohara

, Doi

, Ninomiya

, Hirakawa

, Hata

, Iwaki

, Kanba

, Kiyohara

(2011) Glucose tolerance status and risk of dementia in the community. The Hisayama study. Neurology 77, 1126–1134.

Launer

, Andersen

, Dewey

, Letenneur

, Ott

, Amaducci

, Brayne

, Copeland

, Dartigues

, Kragh-Sorensen

, Lobo

, Martinez-Lage

, Stijnen

, Hofman

(1999) Rates and risk factors for dementia and Alzheimer’s diseases: Results from EURODEM pooled analyses. Neurology 52, 78–84.

Merchant

, Tang

M-X

, Albert

, Manly

, Stern

, Mayeux

(1999) The influence of smoking on the risk of Alzheimer’s disease. Neurology 52, 1408–1412.

Juan

, Zhou

DHD

, Li

, Wang

JYJ

, Gao

, Chen

(2004) A 2-year follow-up study of cigarette smoking and risk of dementia. Eur J Neurol 11, 277–282.

Reitz

, den Heijer

, van Duijn

, Hofman

, Breteler

MMB

(2007) Relation between smoking and risk of dementia and Alzheimer disease. The Rotterdam study. Neurology 69, 998–1005.

Ohara

, Ninomiya

, Hata

, Ozawa

, Yoshida

, Mukai

, Nagata

, Iwaki

, Kitazono

, Kanba

, Kiyohara

(2015) Midlife and late-life smoking and risk of dementia in the community: The Hisayama study. J Am Geriatr Soc 63, 2332–2339.

10.

Willi

, Bodenmann

, Ghali

, Faris

, Cornuz

(2007) Active smoking and the risk of type 2 diabetes. A systematic review and meta-analysis. JAMA 298, 2654–2664.

11.

Yeh

, Duncan

, Schmidt

, Wang

, Brancati

(2010) Smoking, smoking cessation, and the risk for type 2 diabetes mellitus: A cohort study. Ann Intern Med 152, 10–17.

12.

Tamiya

, Noguchi

, Nishi

, Reich

, Ikegami

, Hashimoto

, Shibuya

, Kawachi

, Campbell

(2011) Population ageing and wellbeing: lessons from Japan’s long-term care insurance policy. Lancet 378, 1183–1192.

13.

Health and Welfare Bureau for the Elderly, Ministry of Health, Labour and Welfare (1993) Standardized assessment manual for the grade of activity daily living assessment manual for the grade of activity daily living related to dementia among elderly with dementia. Tokyo, Japan (in Japanese)

14.

Ikegami

(1997) Public long-term care insurance in Japan. JAMA 278, 13130–1314.

15.

Kashiwagi

, Kasuga

, Araki

, Oka

, Hanafusa

, Ito

, Tominaga

, Oikawa

, Noda

, Kawamura

, Sanke

, Namba

, Hashiramoto

, Sasahara

, Nishio

, Kuwa

, Ueki

, Takei

, Umemoto

, Murakami

, Yamakado

, Yatomi

, Ohashi

; Committee on the Standardization of Diabetes Mellitus-Related Laboratory Testing of Japan Diabetes Society (2012) International clinical harmonization of glycated hemoglobin in Japan: From Japan diabetes society to national glycohemoglobin standardization program values. J Diabetes Investig 3, 39–40.

16.

Hallqvust

, Ahlbom

, Diderichsen

, Reuterwall

(1996) How to evaluate interaction between causes: a review of practices in cardiovascular epidemiology. J Intern Med 239, 377–382.

17.

Andersson

, Alfredsson

, Kallberg

, Zdravkovic

, Ahlbom

(2005) Calculating measures of biological interaction. Eur J Epidemiol 20, 575–579.

18.

Kanba

(2013) Investigation of the freely available easy-to-use software ‘EZR’ for medical statistics. Bone Marrow Transplant 48, 452–458.

19.

Tyas

, White

, Petrovitch

, Webster Ross

, Foley

, Heimovitz

, Launer

(2003) Mid-life smoking and late-life dementia: the Honolulu-Asia Aging Study. Neurobiol Aging 24, 589–596.

20.

Rusanen

, Kivipelto

, Quesenberry

Jr , Zhou

, Whitmer

(2011) Heavy smoking in midlife and long-term risk of Alzheimer disease and vascular dementia. Arch Intern Med 171, 333–339.

21.

Moreno-Gonzalez

, Estrada

, Sanchez-Mejias

, Soto

(2013) Smoking exacerbates amyloid pathology in a mouse model of Alzheimer’s disease. Nat Commun 4, 1495.

22.

Ambrose

, Barua

(2004) The pathophysiology of cigarette smoking and cardiovascular disease. J Am Coll Cardiol 43, 1731–1737.

23.

Biessels

, Staekenborg

, Brunner

, Brayne

, Scheltens

(2006) Risk of dementia in diabetes mellitus: a systematic review. Lancet Neurol 5, 64–74.

24.

Kadowaki

, Yamauchi

, Kubota

, Hara

, Ueki

, Tobe

(2006) Adiponectin and adiponectin receptors in insulin resistance, diabetes, and the metabolic syndrome. J Clin Invest 116, 1784–1792.

25.

Hilawe

, Yatsuya

, Li

, Uemura

, Wang

, Chiang

, Toyoshima

, Tamakoshi

, Zhang

, Kawazoe

, Aoyama

(2015) Smoking and diabetes: Is the association mediated by adiponectin, Leptin or C-reactive protein? J Epidemiol 25, 99–109.

26.

Une

, Takei

, Tomita

, Asanuma

, Ohrui

, Furukawa

, Arai

(2011) Adiponectin in plasma and cerebrospinal fluid in MCI and Alzheimer’s disease. Eur J Neurol 18, 1006–1009.

27.

van Himbergen

, Beiser

, Ai

, Seshadri

, Otokozawa

, Au

, Thongtang

, Wolf

, Schaefer

(2012) Biomarkers for insulin resistance and inflammation and the risk for all-cause dementia and Alzheimer’s disease: Results from the Framingham heart study. Arch Neurol 69, 594–600.

28.

Waragai

, Adame

, Trinh

, Sekiyama

, Takamatsu

, Une

, Masliah

, Hashimoto

(2016) Possible involvement of adiponectin, the anti-diabetes molecule, in the pathogenesis of Alzheimer’s disease. J Alzheimers Dis 52, 1453–1459.

29.

Bednarska-Makaruk

, Graban

, Wisniewska

, Lojkowska

, Bochynska

, Gugala-Iwaniuk

, Sławińska

, Ługowska

, Ryglewicz

, Wehr

(2017) Association of adiponectin, leptin and resitin with inflammatory markers and obesity in dementia. Biogerontology 18, 561–580.

30.

Bigalke

, Schreitmuller

, Sopova

, Paul

, Stransky

, Gawaz

, Stellos

, Laske

(2011) Adipocytokines and CD34⁺ progenitor cells in Alzheimer’s disease. PLoS One 6, e20286.

31.

Matthew

, Hynan

, Weiner

, Texas Alzheimer’s Research and Care Consortium (2012) Lipids and adipokines as risk factors for Alzheimer’s disease. J Alzheimers Dis 29, 151–157.

32.

Teixeira

, Diniz

, Campos

, Miranda

, Rocha

, Talib

, Gattaz

, Forlenza

(2013) Decreased levels of circulating adiponectin in mild cognitive impairment and Alzheimer’s disease. Neuromol Med 15, 115–121.

33.

Noda

, Yamagishi

, Ikeda

, Asada

, Iso

(2018) Identification of dementia using standard clinical assessments by primary care physicians in Japan. Geriatr Gerontol Int 18, 738–744.

34.

American Psychiatric Association (1994) Diagnostic and statistical manual of mental disorders, 4th ed. American Psychiatric Association, Washington, DC.