Age-Related Cognitive Decline and Prevalence of Mild Cognitive Impairment in the Iwaki Health Promotion Project

Abstract

Background:

The Iwaki Health Promotion Project (IHPP) is a community-based study for the prevention of lifestyle-related diseases and improvement of quality of life.

Objective:

Between 2014 and 2017, a total of 4,442 Iwaki town residents from 19 to 93 years of age participated in annual surveys to clarify the natural course of age-related cognitive decline and mild cognitive impairment (MCI).

Methods:

Modified OLD and SED-11Q questionnaires, MMSE, Logical Memory II, educational history, and APOE genotypes were examined at the first screening. MCI and dementia were diagnosed at the second examination by detailed neurological examination, CDR, and MRI, and followed for 3 years. Spline regression analyses based on a linear mixed model was adopted for statistical analysis.

Results:

MMSE scores declined with age from 55 to 64 years. There was also interaction between levels of education and ages. At the second examination, 56 MCI and 5 dementia patients were identified. None of the MCI cases progressed to dementia during the 3 years. During follow-up examinations, 13 cases showed improved MMSE scores (0.95 point/year), 5 remained stable, and 7 deteriorated (–0.83 point/year). Five cases showed improved CDR-SOB scores (–0.28 point/year), 9 remained stable, and 6 deteriorated (0.3 point/year).

Conclusion:

IHPP revealed that age- and education-related cognitive decline began and advanced from 55 years of age. The prevalence of MCI and dementia was estimated to be 5.9%in the Iwaki town cohort over 60 yeas of age. About 30%of MCI cases showed progression of cognitive decline.

Keywords

INTRODUCTION

The number of patients with dementia in Japan is rapidly increasing owing to the increasingly elderly population. Cognitive impairment decreases life independence and impairs the quality of life. For patients with dementia, medical therapy along with home and social care interventions are necessary. These burdens are stressful for family members and caregivers and are major factors in the increasing national cost of medical care. Recent advances in dementia research have suggested that early detection of cognitive impairment and intervention are beneficial for dementia prevention, use of disease-modifying therapy, and development of social care systems for dementia patients [1 –3]. Large cohort studies including the Alzheimer’s Disease Neuroimaging Initiative (ADNI) [4] and the Dominantly Inherited Alzheimer Network (DIAN) have verified the efficacy of neuropsychiatric tests, cerebrospinal fluid (CSF) biomarkers, and neuroimaging including MRI, FDG-PET, and amyloid and/or tau PETs. Such studies have provided strong evidence for the signatures of Alzheimer’s disease (AD) in the brain [5] and have defined AD based on the NIA-AA framework criteria [6, 7] for developing new disease-modifying therapies. Registration networks for clinical trials such as the Trial-Ready Cohort for the Prevention of Alzheimer’s Dementia [8] and the European Prevention of Alzheimer’s Dementia project [9] are gaining attention and are growing.

The Iwaki Health Promotion Project (IHPP) is a community-based study started in 2005 and is designed to prevent lifestyle-related diseases and to improve quality of life. More than 1,100 inhabitants over 20 years of age in Iwaki town, Hirosaki city, Japan have participated in the annual health checkups. Clinical data consisted of a total of 18,000 participants over 10 years. In 2014, this project was expanded to survey big data sets of more than 600 items from every participant in order to identify risk factors of dementia. We previously analyzed data from IHPP participants in 2014 and showed that aging and APOE ɛ4 are defined factors of plasma Aβ₄₂ levels [10]. Here, we verified the feasibility of the annual survey of IHPP to detect age-related cognitive decline [6 , 12] and the prevalence of mild cognitive impairment (MCI) [11], with a prospective goal to collaborate with global registration networks [8, 9].

METHODS

Subjects

Between 2014 and 2017, a total of 4,442 residents in Iwaki town participated in the annual IHPP survey, which consisted of 1,148 surveys in 2014, 1,099 in 2015, 1,138 in 2016, and 1,057 in 2017. In terms of participation, 554 residents took part in all 4 surveys, 321 in 3, 381 in 2, and 563 took part once. The total population of Iwaki town in 2014 was 11,285, comprising 6,042 females and 5,243 males. An estimated 10%of residents aged over 20 years joined the study. The percentage of the Iwaki town population participated in the survey separately by age group; 16.3%in 60–69 year age group, 11.7%in 70–79 year age group, and 2.6%in 80 years and over age group. The mean age of participants was 54 years, with 43.8%of participants over 60 years; the mean age of this group was 69 years, and the male to female ratio was 0.56. Average education duration was 11 years. The mean Mini-Mental State Examination (MMSE) [13] score was 29, and the average score on the Logical memory II test (LMII) of the Wechsler Memory Scale-Revised (WMS-R) was 8.9 (Table 1, Fig. 1).

Table 1

Baseline characteristics of participants in the IHPP

Study year	2014	2015	2016	2017	Total
Total number of participants	1,148	1,099	1,138	1,057	4,442
Mean age, range (y)	54, 19∼91	54, 19∼91	54, 19∼91	54, 19∼93	54, 19∼93
Total female	715	675	686	624	2,700
APOE ɛ4 (%)	19.8	20.3	20.1	19.9	20
Elder participants (≧60 y)	515	493	485	456	1,949
Mean age (y)	69±6.5	68±6.3	69±6.7	69±6.3	69±6.5
Female	341	325	308	275	1,249
Education duration (y)	11±1.9	11±2	11±2.1	11±1.9	11±2.0
MMSE	29±1.7	29±1.9	29±1.8	29±1.7	29±1.8
LM II	7.4±4.1	8.7±4.2	9.5±4.7	10.1±4.1	8.9±4.5

Descriptions using mean±standard deviation; MMSE, Mini-Mental State Examination; LMII, Logical memory II tests from the Wechsler Memory Scale-Revised (WMS-R).

Fig. 1

Flowchart detailing the study design. The first screening of MCI candidates was conducted based on subjective and objective forgetfulness assessed by the modified OLD and SED-11Q questionnaires, less than 3 points on question 5 of the MMSE, and a decreased score on LMII (≦8 points for 16 years education, ≦4 points for between 10–15 years education, and ≦2 points for 9 years or less) in participants ≧60 years old. Physical and neurological examinations, MMSE, clinical dementia rating (CDR), WMS-R, neuropsychiatry tests for cortical functions, routine blood examinations, and 3T brain MRI were added for the second detailed medical evaluation. Clinical diagnosis of dementia, Alzheimer’s disease dementia (ADD), and MCI were based on the NIA-AA clinical criteria.

A total of 200 medical and paramedical staff examined participants between 6 : 30 and 13 : 00 over 10 days in Iwaki town culture center. Other study items including blood pressure, height, body weight, BMI, body fat ratio (BFR), common and special laboratory tests, physiological examinations, motor reaction studies, and plasma assay of Aβ₄₀ and Aβ₄₂ levels have already been published [10].

Cognition tests

After written informed consent, participants filled out a detailed questionnaire, which included sections on basic demographics, family structure, residence, educational record, medication and health history, smoking and drinking, sleep evaluation, labor, exercise, physical activity, meal evaluation, a 36-item short-form health survey (SF-36) for quality of life and ADL, oral and dental hygiene, lumbar and knee joint programs, mental health, and gynecological, urinary, and bowel problems. MMSE was conducted for all participants. LMII and questionnaire assessment of subjective and objective memory disturbances using a modified version of the Observation List for Possible Early Signs of Dementia (OLD) [14] and the Symptoms of Early Dementia-11 Questionnaire (SED-11Q) [15] were conducted for participants over 60 years of age (Table 2). The modified 15-item OLD and 15-item SED-11Q were used with participants and family members, respectively.

Table 2

Modified Old and SEQ-11 questionnaires on MCI and dementia

	2014∼2017	First Screening	Second Screening
	Total	MCI	CU	MCI	Dementia
Modified OLD 15 items for participants	2,286	191	40	56	5
1. Continuously forgets which day it is	153	25^**	3	10^**	1
2. Often forgets events that occurred a short time ago	534	69^***	13	23^**	3
3. Unable to repeat recently heard stories or information	226	34^**	3	12^*	1
4. Often repeats him-/herself without being aware of it	358	40	9	14	2
5. Regularly tells the same story during a conversation	91	20^***	2	5	1
6. Often unable to find certain words	1,252	102	26	31	2
7. Quickly loses the thread of story	105	20^**	3	5	1
8. From the answer can be concluded that the question was not understood	136	26^***	4	8^*	2^*
9. Has more difficulty understanding a conversation	69	12^*	1	5^*	0
10. Gets times muddled up	32	7^*	1	1	2^**
11. Makes up excuses	288	28	4	7	1
12. No specific item available	333	48^***	8	11	3^*
13. Loses familiar way	22	5	1	1	1^*
14. Forgets serious events or engagement	72	12^*	1	3	0
15. Mistakes in job, domestic duties and community activity	88	12	1	5	1
Modified 15SED-11Q for family members	1,708	191	40	56	5
1. Talk and Ask the same things repeatedly	216	9^***	1	6	1
2. Unable to understand the context of facts	30	8^*	1	3	0
3. Indifferent clothing and other concerns	75	9	3	0	1
4. Forgets to turn off the faucet, close the door or unable to clean up	71	16^*	2	6^*	1
5. Forget one procedure when doing 2 things at once	226	10^***	2	7	0
6. Unable to take medicine under proper management	19	14^****	4^**	3^*	1
7. Take longer time to work, household than before	141	13	2	7	1
8. Unable to make a plan	25	12	0	3	2
9. Unable to understand complex topics	115	8	0	4	1
10. Less interested and willing, and stopped hobbies	38	12^**	5^**	2	1
11. Irritable and suspicious than before	137	8	1	3	0
12. Social withdraw and refusal to take bath	14	17^****	4^***	6^****	0
13. Often misplace and search	265	12^***	1^*	9	0
14. Claims to have had a valuables stolen	3	1	0	0	0
15. Illusions, e.g., sees some persons or bugs that are not there	5	0	0	0	0

MCI, mild cognitive impairment; CU, cognitively unimpaired. ^*p < 0.05; ^**p < 0.01; ^***p < 0.001; ^****p < 0.0001.

APOE genotyping

DNA samples of 3,136 Iwaki residents were purified from peripheral whole blood using a QIAamp^® 96 DNA Blood Kit (QIAGEN, Hilden, Germany), and the presence of APOE ɛ4 was determined by Toshiba corporation using a Japonica Array consisting of population-specific SNP markers designed from the 1,070 whole genome reference panel [10].

Diagnosis of MCI and dementia

Screening of MCI candidates by secondary medical examination in Hirosaki University Hospital was conducted based on subjective and objective forgetfulness, which were assessed by questionnaires, less than 3 points on question 5 of the MMSE, and a decreased score on LMII (≦8 points for 16 years education, ≦4 points for between 10–15 years education, and ≦2 points for 9 years or less) [12]. Clinical diagnosis of dementia, Alzheimer dementia (ADD), and MCI were based on the NIA-AA clinical criteria [16, 17]. Physical and neurological examinations, MMSE, Clinical Dementia Rating (CDR), WMS-R, neuropsychiatry tests for cortical functions, routine blood examinations, and 3T brain MRI were added for final medical evaluation.

Statistical analysis

Since age-dependent cognitive decline is not linear [18] and MMSE scores are subject to a ceiling effect [19], a linear regression model is not available. Spline regression analyses based on a linear mixed model (LMM) have been indicated to fit the nonlinear longitudinal association between age and cognitive outcome score more precisely [18]. The roles of education history and the presence of APOE ɛ4 on age-dependent alterations of MMSE scores were analyzed. The participants were analyzed separately according to the presence of APOE ɛ4 and Japanese ADNI (J-ADNI) classification of education duration, which consisted of the following groups: 9 years or less, 10–15 years, and 16 years or more [12].

We built a linear mixed effect model in combination with splines and set MMSE as a dependent variable with the following fixed effects: 1) age groups divided using 3 knots into quarters every 10 years; 2) presence of APOE ɛ4 alleles (0, 1); 3) education record (≦9 years, 10–15 years, and ≧16 years); 4) random effects with individuality including intercept and/or slope. Optional positions of the 3 knots in the LMM were evaluated by moving the first knot from 40 to 65 years at an interval of 5 years. Each Akaike Information Criterion (AIC) of the LMM formula was compared using ANOVA to select a significant fit formula, and the LMM formula that fit best was identified by a stepwise method. Statistically significant differences were determined using ANOVA for continuous variables, the chi-squared test for categorical variables, and the Kruskal-Wallis test for skewed variable. p-values less than 0.05 were considered significant. These analyses were performed with R 4.04 version (Packages: stats, Ime4, splines, sjPlot, and lmerTest) [20 –23] and GraphPad Prism, version 9 (GraphPad Software, San Diego, CA).

RESULTS

Basic profiles and age-related cognitive decline

The basic characteristics of participants in the annual surveys were not significantly different (Table 1). A change in education history was seen from 50 years of age, with the percentage of participants with less than 9 years of education increasing from 1.3%for participants in their 20s to 53.7%for those in their 80s. These values decreased from 82%to 42.6%for the 10–15 years education duration group and from 16.7%to 3.7%for the ≧16 years education group (Fig. 2a). The mean frequency of APOE ɛ4 (heterozygotes and homozygotes) was 20%in respective annual surveys; however, the frequency of APOE ɛ4 varied with age. The mean percentage of non-carriers, heterozygotes, and homozygotes of APOE ɛ4 were 78.2%, 20.86%, and 0.94%, respectively, between 20 and 60 years of age. These frequencies were lower for participants in their 70s and 80s (Fig. 2b), with an overall decrease in APOE ɛ4 carriers seen over 60 years of age.

Fig. 2

Age-dependent education history and APOE ɛ4. Top) Education history. Red line: percentage of ≦9 years education history; Yellow line: 10–15 years education history; Blue line: ≧16 years education history. Bottom) APOE ɛ4 allele frequency. Blue line: APOE ɛ4 allele 0; Yellow line: APOE ɛ4 allele 1; Red line: APOE ɛ4 allele 2

We then statistically analyzed age-dependent cognitive function using spline regression analyses based on LMM with consideration of the education history and APOE ɛ4 frequencies of participants. The best-fit formula was selected based on the most relevant AIC (AIC = 12,140; BIC = 12,292; log-likelihood = –6,046.0; deviance = 12,092; p < 0.6005 0 < 2e-16^***), which was divided into 4 groups using 3 knots at 55, 65, and 75 years of age. Selected variables were determined a using stepwise method, with age, education duration, and the presence of APOE ɛ4 selected as interaction variables. Random effect variables were determined for the respective participants, with the slopes and intercepts measured.

MMSE scores declined in an age-dependent manner from 55 to 64 years of ages. With regard to educational duration, the order of the groups in terms of declining levels of MMSE scores was ≦9 years, 10–15 years, ≧16 years (Fig. 3a). Age-dependent fixed effects were longitudinally decreased for participants aged 55–64 (–1.165), 65–74 (–2.497), and over 75 (–4.78) years. There was interaction between educational records and ages. The rates of decrease of MMSE scores were significant for participants aged 55–64 years (p = 0.01139) and 65–74 years (p = 0.00867). However, no interaction was observed between education duration and ages over 75 years (p = 0.266).

Fig. 3

Age-related cognitive decline. a) Predicted value lines and 95%confidence interval of MMSE scores. Association between age and MMSE score. The line represents the splines with 3 knots at the ages 55, 65, and 75 years. Figure categorizes risk factors (education grade) that have a significant age-dependent association with the outcome measure. Red line: ≦9 years education history; Blue line: 10–15 years education history; Green line: ≧16 years education. b) Predicted value lines and 95%confidence interval of MMSE. Association of risk factors with MMSE. Figure categorizes risk factors (APOE ɛ4 allele frequency). Red line: APOE ɛ4 allele 0; Blue line: APOE ɛ4 alleles 1 and 2.

The presence of APOE ɛ4 correlated with a significant reduction of MMSE scores (p = 0.01579) for participants 55–64 years of ages. However, interaction between the presence of APOE ɛ4 and the divided age groups was excluded by the stepwise analytic method (Fig. 3b).

Modified 15-item OLD for participants and modified 15-item SED-11Q for family members

Significantly positive answers in the modified 15-items OLD questionnaire were seen for questions 1, 2, 3, 5, 7, 8, 9, 10, 12, and 14 in the first screening of MCI candidates compared with all participants over 60 years of age. Significantly positive answers in the second screening of the MCI groups were seen for 1, 2, 3, 8, and 9 compared with the cognitively unimpaired (CU) and dementia groups. These questions (1, 2, 3, 8, and 9) were useful for differentiating MCI from CU. Although questions 8, 10, 12, and 13 were positive in the dementia group, the subject number was too small to evaluate significance.

In the modified 15-item SED-11Q questionnaire, response rates of family members were quite low. Significantly positive answers were seen for questions 1, 2, 4, 5, 6, 10, 12, and 13 between the first screening of MCI candidates and all participants. In the second screening, only question 4 was significant in the MCI group compared with the CU and dementia groups. These findings suggest that family members did not pay close attention to cognitive impairment of the participants (Table 2).

Results of the first screening of MMSE and LMII

Participants were selected to be MCI candidates in the first screening: 55 patients (10.7%) in 2014, 51 (10.3%) in 2015, 47 (9.6%) in 2016, and 38 (8.3%) in 2017. Although there were variations among the 4 annual surveys, it was deemed necessary to screen 9.8%of participants in a second examination for precise diagnosis. The mean age of these candidates was 73 years. The male to female ratio (96/95; 1.01), mean education duration (11 years), and MMSE score (27 points) were consistent each year. There were variations in LMII scores among the surveyed years, but the mean total score was 2.4 points. Mean scores for MMSE and LMII declined by 2 and 7 points, respectively, compared with those of all participants over 60 years of age. APOE ɛ4 frequency increased from 20%to 22.5%(Table 3).

Table 3

Results of the first screening

Study year	2014	2015	2016	2017	Total
Number (%)	55 (10.7%)	51 (10.3%)	47 (9.6%)	38 (8.3%)	191 (9.8%)
Age	72±7.3	72±6.7	75±8.4	73±6.1	73±7.5
Female	25	29	22	19	95
Education duration	11±2.2	11±2.7	11±3.1	11±2.2	11±2.6
MMSE	27±2.4	27±3.0	27±2.5	27±2.7	27±2.6
LM II	2.1±1.5	2.7±2.0	2.2±1.9	2.3±1.8	2.4±1.8
APOE ɛ4 (%)	25.5	25.5	17	21.1	22.5

Descriptions using mean±standard deviation; MMSE, Mini-Mental State Examination; LMII, Logical memory II tests from the Wechsler Memory Scale-Revised (WMS-R).

Results of the second detailed examination

For the second evaluation, medical interviews, examination by neurologists, and MMSE, CDR, and brain MRI examinations were carried out. Out of the 191 candidates of the first screening, 39 (70.9%) in 2014, 25 (49%) in 2015, 25 (53.1%) in 2016, and 12 (31.6%) in 2017 participated. A total of 40 CU participants, 56 MCI subjects, and 5 dementia patients were identified. The male to female ratios were 24/16 (1.5), 26/30 (0.87), and 3/2 (1.5) in the CU, MCI, and dementia groups, respectively. The mean age increased from 71 to 75 and 78 years of age in accordance with the clinical stages of CU, MCI, and dementia. MMSE scores were 29, 27, and 21 for the CU, MCI, and dementia groups, respectively, and LMII scores were, 4, 2, and 0 for these groups. CDR-SOB scores were 0.4, 1.3, and 4.9 for the CU, MCI, and dementia groups, respectively. Three patients with ADD, 1 with Parkinson’s disease with dementia, and 1 with semantic dementia were diagnosed in the dementia group. The mean education durations were 13 years in the CU group and 11 years in both MCI and dementia groups. Comparing the first and the second examinations of the MCI subjects, scores for MMSE and LMII were identical (Table 4).

Table 4

Results of the second detailed examination

Study Year		2014	2015	2016	2017	Total
Cognitively	Number	9	10	12	9	40
Unimpaired (CU)	Female	4	4	5	3	16
	Age	70 (62∼83)	70 (62∼79)	72 (60∼85)	72 (62∼85)	71±6.9
	LM II	3±1.69	4±2.66	4±1.91	3±2.24	4±2.11
	MMSE	29±0.88	29±0.7	29±1.1	29±0.9	29±0.9
	CDR (CDR-SOB)	0 (0.3)	0 (0.4)	0 (0.3)	0 (0.5)	0 (0.4)
MCI	Number	28	13	12	3	56
	Female	16	7	6	0	30
	Age	72 (62∼85)	75 (65∼85)	81 (67∼87)	78 (75∼80)	75±7.0
	LM II	2±1.23	3±1.66	2±1.67	2±0.58	2±1.40
	MMSE	27±2.25	27±2.2	27±2.6	26±3.1	27±2.3
	CDR (CDR-SOB)	0.5 (1.2)	0.5 (1.6)	0.5 (1.4)	0.5 (1.2)	0.5 (1.3)
Dementia	Number	2	2	1	0	5
	Female	0	1	0	0	2
	Age	81, 87	66, 81	76	–	78±7.9
	Disease	ADD 2	ADD1, PDD 1	SD 1	–	ADD3, PDD1, SD1
	LM II	0	1±1.41	0	–	0.4±0.89
	MMSE	20, 20	20, 24	20	–	21±1.8
	CDR (CDR-SOB)	1 (6, 5)	1 (4, 5)	1 (4.5)	–	1 (4.9)

Descriptions using mean±standard deviation; age: years of age; MMSE, Mini-Mental State Examination; LMII, Logical memory II tests from the Wechsler Memory Scale-Revised (WMS-R); MCI, mild cognitive impairment; CU, cognitively unimpaired; CDR, Clinical Dementia Rating; CDR-SOB, CDR sum of box; ADD, Alzheimer’s disease dementia; PDD, Parkinson’s disease with dementia; SD, semantic dementia.

Annual follow-up of MCI subjects

For 3 years, 25 of the 56 MCI subjects who underwent the second examination were followed up on (Table 5). The status of all cases remained MCI, with none of the cases progressing to dementia. Thirteen patients (52%) showed improved MMSE scores (0.95 point/year), 5 (20%) remained stable, and 7 (28%) deteriorated (-0.83 point/year). Twenty patients accepted serial CDR follow-up examinations with 5 (25%) showing improved CDR-SOB scores (–0.28 point/year), 9 (45%) remaining stable, and 6 (30%) deteriorating (0.3 point/year). Consequently, we identified 191 MCI candidates (9.8%of participants over 60 years of age) in the first screening of the 4 annual surveys. Out of the 101 participants of the second examination, 40 were CU (40%), 56 had MCI (55%), and 5 had dementia (5%). Based on these percentages, the estimated prevalence of MCI and dementia in the IHPP cohort over 60 years of age were calculated as 5.4%and 0.5%, respectively.

Table 5

Annual follow-up of MCI subjects

Participants	Age	Sex	Education years	Follow years	Baseline MMSE	Final MMSE	ΔMMSE	ΔMMSE/year	Baseline CDR	Final CDR	Baseline CDR-SOB	Final CDR-SOB	ΔCDR-SOB	ΔCDR-SOB/year
1	69	F	10	3	30	24	6	2.00	0.5	0.5	0.5	2	1.5	0.50
2	66	F	9	3	27	25	2	0.67	0.5	0.5	1.5	1.5	0	0.00
3	87	M	8	3	29	27	2	0.67	0.5	–	1	–	–	–
4	81	M	12	3	28	26	2	0.67	0.5	0.5	1.5	2.5	1	0.33
5	73	F	12	3	26	25	1	0.33	0.5	0.5	1	1	0	0.00
6	71	M	16	3	30	30	0	0.00	0.5	0.5	1	1	0	0.00
7	75	M	12	3	29	29	0	0.00	0.5	–	1	–	–	–
8	65	M	12	3	29	29	0	0.00	0.5	–	0.5	–	–	–
9	80	M	9	3	30	30	0	0.00	0.5	0.5	1	2	1	0.33
10	83	F	9	3	26	27	–1	–0.33	0.5	0.5	1	1	0	0.00
11	88	M	12	3	27	28	–1	–0.33	0.5	0.5	0.5	2	1.5	0.50
12	80	F	9	3	25	26	–1	–0.33	0.5	0.5	2	1	–1	–0.33
13	70	F	12	3	28	30	–2	–0.67	0.5	0.5	1	1	0	0.00
14	69	F	12	3	28	30	–2	–0.67	0.5	0.5	0.5	1	0.5	0.17
15	75	F	14	3	28	30	–2	–0.67	0.5	0.5	1	0.5	–0.5	–0.17
16	79	F	12	3	26	30	–4	–1.33	0.5	0.5	1.5	1.5	0	0.00
17	74	M	9	3	22	28	–6	–2.00	0.5	0.5	2	1.5	–0.5	–0.17
18	68	M	12	2	30	28	2	1.00	0.5	–	1	–	–	–
19	76	F	16	2	30	29	1	0.50	0.5	–	2	–	–	–
20	78	M	12	2	26	26	0	0.00	0.5	0.5	1	1.5	0.5	0.25
21	81	M	12	2	25	27	–2	–1.00	0.5	0.5	2.5	2	–0.5	–0.25
22	74	M	9	2	26	29	–3	–1.50	0.5	0.5	1	1	0	0.00
23	76	M	9	2	25	28	–3	–1.50	0.5	0.5	1	1	0	0.00
24	86	M	12	1	22	23	–1	–1.00	0.5	0.5	2	2	0	0.00
25	85	F	11	1	24	25	–1	–1.00	0.5	0.5	2	1.5	–0.5	–0.50

Education and follow-up: years; ΔMMSE, MMSE score of baseline MMSE points –final MMSE points; ΔMMSE/year, change rate/year; ΔCDR-SOB: baseline CDR-SOB –final CDR-SOB; ΔCDR-SOB/year, change rate/year.

DISCUSSION

The basic profiles of participants from Iwaki town were similar for each year surveyed between 2014 and 2017. Each participant voluntarily took part in the annual surveys having given prior written consent under the revised Japanese ethical guidelines of the Ministry of Health, Labor, and Welfare for human genome analysis in 2013. As a result, the expanded IHPP has become one of the newest cohort studies of risk factors of dementia in Japan. Since this annual survey was not a comprehensive study of Iwaki town residents, the number of participants varied among the 4 years.

The education history of participants changed from 50 years of age, with the rate of participants with low levels of education appearing to increase in participants in their 60s. These marked changes seem to have been caused by reformation of the educational system and economic development after World War II. The same development of educational attainment has also been observed worldwide [24].

Genotypes of APOE in the IHPP cohort consisted of 0.09%of ɛ2/ɛ2, 6.69%of ɛ2/ɛ3, 0.89%of ɛ2/ɛ4, 71.05%of ɛ3/ɛ3, 20.43%of ɛ3/ɛ4, and 0.84%of ɛ4/ɛ4. Compared with the Framingham cohort [25] and the Rotterdam study [26], the frequencies of ɛ2/ɛ2 and ɛ2/ɛ3 were lower in the IHPP and that reported by the Japanese Genetic Study consortium for AD [27], whereas the frequencies of APOE ɛ4 genotypes were almost the same. The IHPP cohort included younger participants in their 20s, resulting in lower participation of APOE ɛ4 carriers over 60 years of age. In the J-ADNI, which showed a high conversion rate of 45.1%from late amnestic MCI to dementia, the percentages of APOE ɛ4 carriers were 24%in the normal cognition group, 52%in late amnestic MCI group, and 60%in the mild AD group [12]. The low prevalence of MCI and annual conversion rate from MCI to dementia in the IHPP cohort compared with other cohort studies may have been caused by the lower participation rate of older APOE ɛ4 carriers with a higher risk of incipient AD. This is a limitation of this study, and it will be important to clarify whether the onset of dementia or comorbidity in elderly residents will cause them dropout from IHPP.

The MMSE is a widely used neuropsychiatry screening test [13], with a cut off value of 23/24 discriminating moderate and severe cognitive impairment [28]. Scores are highly dependent on age and educational level [29 –32]. Population-based norms at each age level have been reported from 6 to over 85 years [33, 34], and a natural decline course of MMSE scores of 3–4 points per year is indicative of AD [35]. The MMSE is a basic variable and one of the end points for observational studies, such as ADNI [4], DIAN [5], and intervention studies. Nonparametric spline regression analyses by LMM showed that MMSE scores declined from ages 55–64 to 65–74 years and continued to decline over 75 years, indicating that cognitive decline in the IHPP cohort began from 55 to 65 years and sequentially advanced. The Longitudinal Aging Study Amsterdam (LASA) analyzed cognitively healthy subjects across the age range of 55–85 years and determined the best-fit spline model with 2 knots at 70 and 80 years of age. The decline of MMSE scores per year was –0.06 between 55 and 70 years of age, –0.18 between 70 and 80 years of age, and –0.25 over 80 years of age [18]. The IHPP cohort was fitted to a model with 3 knots at 55, 65, and 75 years of age, showing a steeper rate of decline per year of MMSE scores of –1.165 for 55–65 years of age, –2.497 for 65–74 years of age, and –4.78 for over 75 years of age. However, unlike the LASA cohort, the IHPP cohort consisted of participants below 55 and over 85 years of age. Furthermore, the LASA subjects comprised a cognitively healthy population with over 27 points on the MMSE. Thus, major differences exist between IHPP and LASA. However, both LASA and IHPP showed age-related decline of MMSE scores that commenced from the age of 55 to 65 years, with the rate of decline increasing with age. Taniguchi et al. reported that MMSE score trajectories of non-demented adults aged 65–90 years in community-dwelling Japanese showed 3 patterns that depended on the initial MMSE score, years of education, and daily activity, and that the decline of MMSE scores started from 65 years of age, with the lowest rate of decline seen in the low MMSE trajectory group [36]. The DIAN study has clarified the natural course of cognitive decline and biomarker changes over 40 years in inherited AD patients [37]. Recently, Luo et al. analyzed 3,284 cognitively normal individuals from 18 to 101 years of age and showed that changes in a detailed cognitive composite occurred at 62.41 years [38]. Together with these reports, our study shows that cognitive decline in healthy individuals is likely to begin in individuals aged in their early 60s.

As shown in Fig. 3a, the predicted values of MMSE scores in IHPP declined from the age of 20, and this decline was dependent on educational levels. Thus, education history was a definite risk factor for age-related cognitive decline in the IHPP cohort, corroborating with previous reports [29 –32]. However, no interaction was observed between education duration and ages over 75 years. Since more than half of the participants were in the group with low levels of education and only a small percentage of participants over 75 years of age had high levels of education, statistical analyses did not show a clear dependence on education.

The PROSPER study indicated that APOE ɛ4 is associated with more rapid cognitive decline [39]. Although the frequency of the APOE ɛ4 allele varied in 4 major cohort studies including the National Alzheimer’s Coordination Center (NACC), The Rotterdam study, the Framingham Heart study, and the Sacramento Area Latino Study (SALSA), APOE ɛ4 frequency, age, and low educational levels consistently increased the risk of MCI or dementia [40]. A harmonized longitudinal study of 14 cohorts showed accelerated decline in MMSE scores with age and APOE ɛ4 carriers [41]. The Australian Imaging Biomarkers and Lifestyle (AIBL) cohort study indicated that Aβ and APOE ɛ4 combined to influence memory decline in non-demented older adults and increasing age may further exacerbate these effects [42, 43]. APOE ɛ4 is associated with a consistently faster rate of memory decline from midlife to the early years of old age [44]. All these reports indicate that aging and APOE ɛ4 are risk variables for cognitive decline in healthy adults. As with these studies, the majority of participants of IHPP are healthy elderly residents. The lower participation rate of APOE ɛ4 carriers aged over 60 years may be a reason for the weak influence of APOE ɛ4 on age-related cognitive decline.

The criteria of MCI consist of cognitive compl-aints, not normal age, no dementia, cognitive decline, and essentially normal functional ability [11]. Clinically, MCI is defined by neuropsychiatric tests such as MMSE and a score of 0.5 in CDR. Since not all MCI is due to AD, further examination by neuroimaging and biomarkers is necessary to diagnose the causative disease. The estimated prevalence of MCI is between 15%and 20%in persons aged 60 years or older. The annual rate in which MCI progresses to dementia varies between 8%and 15%per year [11]. The first report from ADNI revealed that the subjects with MCI progressed to dementia in 12 months at a rate of 16.5%per year [4]. Comparison between Japanese and North American ADNI showed that the conversion rates from MCI to dementia were 28.8%versus 20.2%at 12 months and 45.1%versus 40.9%at 24 months. The patterns of decline and 3-year mean changes were -1.52 in MMSE and 1.18 in CDR-SOB scores in amyloid PET-positive MCI in J-ADNI [12]. A clinical review by Langa and Levine reported that the prevalence of MCI in adults aged over 65 years is 10–20%, with annual conversion rates from < 5%to 20%[45]. In the Framingham Heart Study, 34%of subjects over 60 years of age had MCI [46]. A follow-up study reported that 71.4%remained stable, 13.5%reverted from MCI to normal, 15%transitioned from normal to MCI, and 4.8%developed incident dementia after an average of 6.5 years [43]. A systematic review of 7 studies from 618 studies between 2007 and 2016 revealed that MCI incidence per 1,000 person-years were 22.5 for ages 75–79 years, 40.9 for ages 80–84 years, and 60.1 for ages 85 years and older [47]. The Japanese population-based study in Ama-cho, which covered 97%of the 924 residents over 65 years of age, revealed that the overall prevalence of MCI and dementia were 23.4%and 16.4%, respectively [48]. Compared with these studies, the estimated overall prevalence of MCI and dementia (5.4%and 0.5%) over 60 years of age in the IHPP cohort was low. The conversion rate from MCI to dementia in the IHPP cohort was low, and the rates of decline rate per year of MMSE score and CDR-SOB scores were also low in individuals with MCI. However, the dropout bias at the annual follow-up studies was not negligible. For this reason, we have shown all observation courses of the 25 MCI participants in Table 5 and have only described the annual change rates of the MMSE and CDR-SOB scores. Because the IHPP is not a complete enumeration-based survey but a voluntary survey, the IHPP participants likely have higher health awareness at baseline compared to non-participants. The percentage of the Iwaki town population participated in the survey was 2.6%in 80 years and over age group. Therefore, it is not hard to anticipate that the voluntary participants of the IHPP were cognitively healthier than non-participants, which can explain the extremely low prevalence of MCI and dementia in the IHPP. These results indicate that the IHPP cohort consists of cognitively healthy subjects, and that early MCI was in accordance with findings of APOE ɛ4 frequency. It is possible that the inclusion of residents cared for at home and at care facilities in the survey would increase the prevalence of MCI and dementia. However, the IHPP dementia risk survey is based on voluntarily participation, requiring prior written consent under the new Japanese ethical guidelines, and the recently defined criteria of MCI. This will allow for the validation of risk factors that are in line with global network registrations, with a view of future collaboration. Based on results of IHPP between 2014 to 2017, we hope to continue with and expand the annual survey in order to precisely predict and intervene effectively with age-related cognitive decline and MCI.

Footnotes

ACKNOWLEDGMENTS

We thank Sakiko Narita, Kaoru Sato, Sachiyo Ich-inohe, Sachiyo Kasai, and members of the Iwaki Health Promotion Project (IHPP) group for research assistance. IHPP was supported by Hirosaki University, Hirosaki City, Hirosaki City Medical Asso-ciation, and Aomori Prefecture. This study was supported by Scientific Research (C) (18K07385 MS, 19K07989 TK) from the Ministry of Education, Science, and Culture of Japan; Study of prevention for neurodegenerative diseases by new anti-aging methods from the Hirosaki University Institutional Research Grant, the Center of Innovation Science and Technology based Radical Innovation and Entrepreneurship Program from the Japan Science and Technology Agency and JST COI Grant Number JPMJCE1302. This study was approved by the ethical committee of Hirosaki University (2014-014, 2015-377; 2016-028; 2017-026). All participants provided written informed consent.

Authors’ disclosures available online ().

References

Livingston

, Sommerlad

, Orgeta

, Costafreda

, Huntley

, Ames

, Ballard

, Banerjee

, Burns

, Cohen-Mansfield

, Cooper

, Fox

, Gitlin

, Howard

, Kales

, Larson

, Ritchie

, Rockwood

, Sampson

, Samus

, Schneider

, Selbæk

, Teri

, Mukadam

(2017) Dementia prevention, intervention, and care. Lancet 390, 2673–2734.

Nagandu

, Lehtisalo

, Solomon

, Levälahti

, Ahtiluoto

, Antikainen

, Bäckman

, Hänninen

, Jula

, Laatikainen

, Lindström

, Mangialasche

, Paajanen

, Pajala

, Peltonen

, Rauramaa

, Stigsdotter-Neely

, Strandberg

, Tuomilehto

, Soininen

, Kivipelto

(2015) A 2 year multidomain intervention of diet, exercise, cognitive training, and vascular risk monitoring versus control to prevent cognitive decline in at-risk elderly people (FINGER): A randomized controlled trial. Lancet 385, 2255–2263.

WHO (2019) Risk reduction of cognitive decline and dementia. www.who.int/news-room/detail/14-05-2019-adopting-a-healthy-lifestyle-helps-reduce-the-risk-of-dementia.

Petersen

, Aisen

, Beckett

, Donohue

, Gamst

, Harvey

, Jack

Jr , Jagust

, Shaw

, Toga

, Trojanowski

, Weiner

(2010) Alzheimer’s Disease Neuroimaging Initiative (ADNI): Clinical characterization. Neurology 74, 201–209.

Bateman

, Xiong

, Benzinger

, Fagan

, Goate

, Fox

, Marcus

, Cairns

, Xie

, Blazey

, Holtzman

, Santacruz

, Buckles

, Oliver

, Moulder

, Aisen

, Ghetti

, Klunk

, McDade

, Martins

, Masters

, Mayeux

, Ringman

, Rossor

, Schofield

, Sperling

, Salloway

, Morris

; Dominantly Inherited Alzheimer Network (2012) Clinical and biomarker changes in dominantly inherited Alzheimer’s disease. N Engl J Med 367, 795–804.

Knopman

, Haeberlein

, Carrillo

, Hendrix

, Kerchner

, Margolin

, Maruff

, Miller

, Tong

, Tome

, Murray

, Nelson

, Sano

, Mattsson

, Sultzer

, Montine

, Jack

Jr , Kolb

, Petersen

, Vemuri

, Canniere

, Schneider

, Resnick

, Romano

, van Harten

, Wolk

, Bain

, Siemers

(2018) The National Institute on Aging and the Alzheimer’s Association Research Framework for Alzheimer’s disease: Perspectives from the Research Roundtable. Alzheimers Dement 14, 563–575.

Jack

Jr , Bennett

, Blennow

, Carrillo

, Dunn

, Haeberlein

, Holtzman

, Jagust

, Jessen

, Karlawish

, Liu

, Molinuevo

, Montine

, Phelps

, Rankin

, Rowe

, Scheltens

, Siemers

, Snyder

, Sperling

; Contributors (2018) NIA-AA Research Framework: Toward a biological definition of Alzheimer’s disease. Alzheimers Dement 14, 535–562.

Aisen

, Sperling

, Cummings

, Donohue

, Langford

, Jimenez-Maggiora

, Rissman

, Rafii

, Walter

, Clanton

, Raman

(2020) The Trial-Ready Cohort for Preclinical/Prodromal Alzheimer’s Disease (TRC-PAD) Project: An overview. J Prev Alzheimers Dis 7, 208–212.

Ritchie

, Molinuevo

, Truyen

, Satlin

, Van der Geyten

, Lovestone

; European Prevention of Alzheimer’s Dementia (EPAD) Consortium (2016) Development of interventions for the secondary prevention of Alzheimer’s dementia: The European Prevention of Alzheimer’s Dementia (EPAD) project. Lancet Psychiatry 3, 179–186.

10.

Nakamura

, Kawarabayashi

, Seino

, Hirohata

, Nakahata

, Narita

, Itoh

, Nakaji

, Shoji

(2018) Aging and APOE-ɛ4 are determinative factors of plasma Aβ42 levels. Ann Clin Transl Neurol 5, 1184–1191.

11.

Petersen

(2016) Mild cognitive impairment. Continuum (Minneap Minn) 22(2 Dementia), 404–418.

12.

Iwatsubo

, Iwata

, Suzuki

, Ihara

, Arai

, Ishii

, Senda

, Ito

, Ikeuchi

, Kuwano

, Matsuda

; Japanese Alzheimer’s Disease Neuroimaging Initiative, Sun

, Beckett

, Petersen

, Weiner

, Aisen

, Donohue

; Alzheimer’s Disease Neuroimaging Initiative (2018) Japanese and North American Alzheimer’s Disease Neuroimaging Initiative studies: Harmonization for international trials. Alzheimers Dement 14, 1077–1087.

13.

Folstein

, Folstein

, McHugh

(1975) “Mini-mental state”. A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 12, 189–198.

14.

Hopman-Rock

, Tak

, Staats

(2001) Development and validation of the Observation List for early signs of Dementia (OLD). Int J Geriatr Psychiatry 16, 406–414.

15.

Maki

, Yamaguchi

(2013) Symptoms of Early Dementia-11 Questionnaire (SED-11Q): A brief informant-operated screening for dementia. Dement Geriatr Cogn Dis Extra 3, 131–142.

16.

McKhann

, Knopman

, Chertkow

, Hyman

, Jack

Jr , Kawas

, Klunk

, Koroshetz

, Manly

, Mayeux

, Mohs

, Morris

, Rossor

, Scheltens

, Carrillo

, Thies

, Weintraub

, Phelps

(2011) The diagnosis of dementia due to Alzheimer’s disease: Recommendations from the National Institute on Aging-Alzheimer’s Association workgroups on diagnostic guidelines for Alzheimer’s disease. Alzheimers Dement 7, 263–269.

17.

Albert

, DeKosky

, Dickson

, Dubois

, Feldman

, Fox

, Gamst

, Holtzman

, Jagust

, Petersen

, Snyder

, Carrillo

, Thies

, Phelps

(2011) The diagnosis of mild cognitive impairment due to Alzheimer’s disease: Recommendations from the National Institute on Aging-Alzheimer’s Association workgroups on diagnostic guidelines for Alzheimer’s disease. Alzheimers Dement 7, 270–279.

18.

Legdeur

, Heymans

, Comijs

, Huisman

, Maier

, Visser

(2018) Age dependency of risk factors for cognitive decline. BMC Geriatr 18, 187.

19.

Ichii

, Nakamura

, Kawarabayashi

, Takatama

, Ohgami

, Ihara

, Shoji

(2020) CogEvo, a cognitive function balancer, is a sensitive and easy psychiatric test battery for age-related cognitive decline. Geriatr Gerontol Int 20, 248–255.

20.

R Core Team (2021) R:Alanguage and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. https://www.R-project.org/.

21.

Bates

, Maechler

, Bolker

, Walker

(2015) Fitting linear mixed-effects models using lme4. J Stat Softw 67, 1–48.

22.

Ludecke

(2018) sjPlot: Data Visualization for Statistics in Social Science. URL: http://doi.org/10.5281/zenodo.1308157), R package version 2.6.1, https://CRAN.R-project.org/package=sjPlot.

23.

Kuznetsova

, Brockhoff

, Christensen

RHB

(2017) lmerTest Package: Tests in linear mixed effects models. J Stat Softw 82, 1–26.

24.

Gakidou

, Cowling

, Lozano

, Murray

(2010) Increased educational attainment and its effect on child mortality in 175 countries between 1970 and 2009: A systematic analysis. Lancet 376, 959–974.

25.

Myers

, Schaefer

, Wilson

, D’Agostino

, Ordovas

, Espino

, Au

, White

, Knoefel

, Cobb

, McNulty

, Beiser

, Wolf

(1996) Apolipoprotein E epsilon4 association with dementia in a population-based study: The Framingham study. Neurology 46, 673–677.

26.

van der Lee

, Wolters

, Ikram

, Hofman

, Ikram

, Amin

, van Duijn

(2018) The effect of APOE and other common genetic variants on the onset of Alzheimer’s disease and dementia: A community-based cohort study. Lancet Neurol 17, 434–444.

27.

Kuwano

, Miyashita

, Arai

, Asada

, Imagawa

, Shoji

, Higuchi

, Urakami

, Kakita

, Takahashi

, Tsukie

, Toyabe

, Akazawa

, Kanazawa

, Ihara

; Japanese Genetic Study Consortium for Alzheimer’s Disease (2006) Dynamin-binding protein gene on chromosome 10q is associated with late-onset Alzheimer’s disease. Hum Mol Genet 15, 2170–2182.

28.

Creavin

, Wisniewski

, Noel-Storr

, Trevelyan

, Hampton

, Rayment

, Thom

, Nash

, Elhamoui

, Milligan

, Patel

, Tsivos

, Wing

, Phillips

, Kellman

, Shackleton

, Singleton

, Neale

, Watton

, Cullum

(2016) Mini-Mental State Examination (MMSE) for the detection of dementia in clinically unevaluated people aged 65 and over in community and primary care populations. Cochrane Database Syst Rev 13, CD011145.

29.

Anthony

, LeResche

, Niaz

, von Korff

, Folstein

(1982) Limits of the ‘Mini-Mental State’ as a screening test for dementia and delirium among hospital patients. Psychol Med 12, 397–408.

30.

Tangalos

, Smith

, Ivnik

, Petersen

, Kokmen

, Kurland

, Offord

, Parisi

(1996) The Mini-Mental State Examination in general medical practice: Clinical utility and acceptance. Mayo Clin Proc 71, 829–837.

31.

O’Bryant

, Humphreys

, Smith

, Ivnik

, Graff-Radford

, Petersen

, Lucas

(2008) Detecting dementia with the mini-mental state examination in highly educated individuals. Arch Neurol 65, 963–967.

32.

Spering

, Hobson

, Lucas

, Menon

, Hall

, O’Bryant

(2012) Diagnostic accuracy of the MMSE in detecting probable and possible Alzheimer’s disease in ethnically diverse highly educated individuals: An analysis of the NACC database. J Gerontol A Biol Sci Med Sci 67, 890–896.

33.

Crum

, Anthony

, Bassett

, Folstein

(1993) Population-based norms for the Mini-Mental State Examination by age and educational level. JAMA 269, 2386–2391.

34.

Shoji

, Fukushima

, Wakayama

, Shizuka-Ikeda

, Ikeda

, Kawakami

, Sakazume

, Ikeda

, Harigaya

, Matsubara

, Kawarabayashi

, Murakami

, Nagano

, Manabe

, Abe

(2002) Intellectual faculties in patients with Alzheimer’s disease regress to the level of a 4–5-year-old child. Geriatr Gerontol Int 2, 143–147.

35.

Knopman

and Selens

(2012) Neuropsychiatry of dementia. In: Clinical Neuropsychology, 5th edn., Heilman KM, Valenstein E, eds. Oxford University Press, Oxford, New York, pp. 582–636.

36.

Taniguchi

, Kitamura

, Murayama

, Amano

, Shinozaki

, Yokota

, Seino

, Nofuji

, Nishi

, Yokoyama

, Matsuyama

, Fujiwara

, Shinkai

(2017) Mini-Mental State Examination score trajectories and incident disabling dementia among community-dwelling older Japanese adults. Geriatr Gerontol Int 17, 1928–1935.

37.

McDade

, Wang

, Gordon

, Hassenstab

, Benzinger

TLS

, Buckles

, Fagan

, Holtzman

, Cairns

, Goate

, Marcus

, Morris

, Paumier

, Xiong

, Allegri

, Berman

, Klunk

, Noble

, Ringman

, Ghetti

, Farlow

, Sperling

, Chhatwal

, Salloway

, Graff-Radford

, Schofield

, Masters

, Rossor

, Fox

, Levin

, Jucker

, Bateman

; Dominantly Inherited Alzheimer Network (2018) Longitudinal cognitive and biomarker changes in dominantly inherited Alzheimer disease. Neurology 91, e1295–e1306.

38.

Luo

, Agboola

, Grant

, Masters

, Albert

, Johnson

, McDade

, Vöglein

, Fagan

, Benzinger

, Massoumzadeh

, Hassenstab

, Bateman

, Morris

, Perrin

, Chhatwal

, Jucker

, Ghetti

, Cruchaga

, Graff-Radford

, Schofield

, Mori

, Xiong

(2020) Sequence of Alzheimer disease biomarker changes in cognitively normal adults: A cross-sectional study. Neurology 95, e3104–e3116.

39.

Packard

, Westendorp

, Stott

, Caslake

, Murray

, Shepherd

, Blauw

, Murphy

, Bollen

, Buckley

, Cobbe

, Ford

, Gaw

, Hyland

, Jukema

, Kamper

, Macfarlane

, Jolles

, Perry

, Sweeney

, Twomey

, Prospective Study of Pravastatin in the Elderly at Risk Group (2007) Association between apolipoprotein E4 and cognitive decline in elderly adults. J Am Geriatr Soc 55, 1777–1785.

40.

Qian

, Wolters

, Beiser

, Haan

, Ikram

, Karlawish

, Langbaum

, Neuhaus

, Reiman

, Roberts

, Seshadri

, Tariot

, Woods

, Betensky

, Blacker

(2017) APOE-related risk of mild cognitive impairment and dementia for prevention trials: An analysis of four cohorts. PLoS Med 14, e1002254.

41.

Lipnicki

, Crawford

, Dutta

, Thalamuthu

, Kochan

, Andrews

, Lima-Costa

, Castro-Costa

, Brayne

, Matthews

, Stephan

, Lipton

, Katz

, Ritchie

, Scali

, Ancelin

, Scarmeas

, Yannakoulia

, Dardiotis

, Lam

, Wong

, Fung

, Guaita

, Vaccaro

, Davin

, Kim

, Han

, Kim

, Anstey

, Cherbuin

, Butterworth

, Scazufca

, Kumagai

, Chen

, Narazaki

, Ng

, Gao

, Reppermund

, Brodaty

, Lobo

, Lopez-Anton

, Santabárbara

, Sachdev

, Cohort Studies of Memory in an International Consortium (COSMIC) (2017) Age-related cognitive decline and associations with sex, education and apolipoprotein E genotype across ethnocultural groups and geographic regions: A collaborative cohort study. PLoS Med 14, e1002261.

42.

Albrecht

, Szoeke

, Maruff

, Savage

, Lautenschlager

, Ellis

, Taddei

, Martins

, Masters

, Ames

, Foster

; AIBL Research Group (2015) Longitudinal cognitive decline in the AIBL cohort: The role of APOE ɛ4 status. Neuropsychologia 75, 411–419.

43.

Lim

, Kalinowski

, Pietrzak

, Laws

, Burnham

, Ames

, Villemagne

, Fowler

, Rainey-Smith

, Martins

, Rowe

, Masters

, Maruff

(2018) Association of β-amyloid and apolipoprotein E ɛ4 with memory decline in preclinical Alzheimer disease. JAMA Neurol 75, 488–494.

44.

Rawle

, Davis

, Bendayan

, Wong

, Kuh

, Richards

(2018) Apolipoprotein-E (APOE) ɛ4 and cognitive decline over the adult life course. Transl Psychiatry 8, 18.

45.

Langa

, Levine

(2014) The diagnosis and management of mild cognitive impairment: A clinical review. JAMA 312, 2551–2561.

46.

Jak

, Preis

, Beiser

, Seshadri

, Wolf

, Bondi

, Au

(2016) Neuropsychological criteria for mild cognitive impairment and dementia risk in the Framingham Heart Study. J Int Neuropsychol Soc 22, 937–943.

47.

Gillis

, Mirzaei

, Potashman

, Ikram

, Maserejian

(2019) The incidence of mild cognitive impairment: A systematic review and data synthesis. Alzheimers Dement (Amst) 11, 248–256.

48.

Wada-Isoe

, Uemura

, Nakashita

, Yamawaki

, Tanaka

, Yamamoto

, Shimokata

, Nakashima

(2012) Prevalence of dementia and mild cognitive impairment in the rural island town of Ama-cho, Japan. Dement Geriatr Cogn Dis Extra 2, 190–199.