SEARCH

SEARCH BY CITATION

Keywords:

  • dementia;
  • Long-Term Care Insurance;
  • MRI;
  • old-old;
  • prevalence

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGMENTS
  8. REFERENCES

Background:  There have been no reports on the prevalence of dementia among the old-old people in Japan.

Methods:  We studied the old-old population in Kurihara, northern Japan. Analysis 1 of Participants 1 (n = 590) was performed to evaluate the prevalence of dementia and dementing diseases by intensive evaluation including MRI. Analysis 2 aimed to determine a good indicator for detecting ‘suspected dementia condition’ based on the Long-Term Care Insurance index. Analysis 3 of Participants 2 (n = 3915) aimed to estimate the prevalence of ‘suspected dementia condition’.

Results:  In Analysis 1, 73 people (12.4%) were diagnosed with dementia. The most common cause was Alzheimer's disease with cerebrovascular disease. In Analysis 2, level I of the Impairment Level of Dementia was found to be a good indicator of ‘suspected dementia condition’. In Analysis 3, the overall estimated prevalence of ‘suspected dementia condition’ was 23.6%. In men, the ratio increased gradually from 75 to 87 years old to about 20%, increased to 40% at the age of 88 and became stable thereafter. In contrast, in women, the ratio increased from 75 to 95+ years old, reaching about 70%.

Conclusions:  The prevalence was higher than that reported previously. There was a difference between the sexes: an ‘age-related’ increase occurred in men and an ‘ageing-related’ increase in women. Alzheimer's disease with cerebrovascular disease was the most common cause, which coincided with the previous findings of individuals aged 65 years and older; however, the ratio of mixed dementia was greater.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGMENTS
  8. REFERENCES

For countries with increasingly large populations of older adults, knowledge of the prevalence of dementia is necessary for health policy planning. We reviewed recent articles published over the past 10 years regarding the prevalence of dementia among adults aged 65 years and older. The articles clearly described the diagnostic criteria; a PubMed search looked for the terms ‘dementia’, ‘prevalence’, ‘DSM’ and ‘Alzheimer’ appearing in articles between 23 February 2000 and 19 February 2010. The results included 10 articles mainly from the newly developing countries, as well as from Sweden, Germany and Spain.1–10 They reported various prevalence rates: from 3.4% (India) to 8.2% (Korea) in those aged 65 years and older,4,7 and 15.8% (Italy) in those aged 75 years and older.3 For dementing diseases, the most common cause was Alzheimer's disease (AD), followed by depression and vascular dementia (VaD). For the AD/VaD ratio, the findings ranged from 1.4 (India, China) to 5.0 (Sri Lanka).2,5,7 Despite the fact that the data varied, aaaall of these studies showed an age-associated increase likely due to the different backgrounds of national populations (e.g. high mortality rate) and the varying methodology, including diagnostic criteria, used.

In Japan, there have been six systematic studies on the prevalence of dementia;11–16 the data varied from 3.8% to 11.0% in those aged 65 years and older, and the AD/VaD ratio varied from 0.7 to 4.1.12,16 We previously reported the prevalence to be 8.5% in those aged 65 years and older in Tajiri, Miyagi Prefecture, in northern Japan,13 but the most common cause of dementing diseases was AD with cerebrovascular disease (CVD), not VaD, as some studies had previously found; the AD/VaD ratio was 3.3. For the old-old population, defined here as aged 75 years and older, there have been no previous reports in Japan.

Therefore, our first aim was to examine the prevalence of dementia diagnosed with DSM-IV and the Clinical Dementia Rating (CDR), together with dementing diseases based upon laboratory tests and magnetic resonance imaging (MRI). A low participation ratio of old-old population was a potential limitation, as it was not always possible for this population to receive a medical diagnosis of dementia using the DSM-IV with the CDR, laboratory tests and MRI. Using the Long-Term Care Insurance (LTCI) was a possible alternative way to confirm medical diagnosis. Thus, our second aim was to determine an LTCI index that could act as a good marker for indicating ‘suspected dementia condition’ and that would enable us to estimate the age-specific prevalence of ‘suspected dementia condition’.

METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGMENTS
  8. REFERENCES

Kurihara

The Kurihara Project (2008–2010) was a community-based stroke, dementia and bed-confinement prevention programme for old-old adults, defined here as aged 75 years or older, in Kurihara, an agricultural town in Miyagi Prefecture, northern Japan. The total population of the city was approximately 76 708 during this period, and the old-old population was approximately 14 579 (17.9%; November 2010).

Study protocol

We performed three analyses (Analyses 1–3) to determine the prevalence of dementia. Analysis 1 of Participants 1 (see below) was performed to evaluate the prevalence of dementia according to the DSM-IV and the CDR stage 1 or greater. Dementing diseases were diagnosed by medical diagnosis according to each diagnostic criteria (see below), with reference to MRI and laboratory tests.

Analysis 2 of Participants 1 aimed to confirm that an LTCI index could be a good alternative indicator of ‘suspected dementia condition’.

Analysis 3 of Participants 2 aimed to estimate the prevalence of ‘suspected dementia condition’ with the LTCI index, as decided by Analysis 2.

Figure 1 illustrates the study protocol.

image

Figure 1. Analysis 1 of Participants 1 was performed to evaluate the prevalence of dementia and dementing diseases by medical diagnosis. Analysis 3 of Participants 2 aimed to estimate the prevalence of ‘suspected dementia condition’ according to the LTCI index determined by Analysis 2. LTCI, Long-Term Care Insurance.

Download figure to PowerPoint

Participants 1

From 255 communities in Kurihara, 19 were selected by city officials and were asked to participate in the project to help us reach our target population of 1254. For Analysis 1 (see below), these populations underwent the following examinations: (i) CDR assessment; (ii) neuropsychological tests; (iii) blood and urine tests; and (iv) MRI scans.

For Analysis 2 (see below), LTCI indices were analyzed with the CDR.

Analysis 1: prevalence of dementia (CDR 1+) and the ratio of dementing diseases

CDR assessments

A clinical team of physicians and public health nurses determined the CDR.17,18 Initially, public health nurses visited participants' homes to evaluate their daily activities. Observations by family members regarding the participants' lives were then described in a semi-structured questionnaire. The participants were interviewed by public health nurses and physicians to assess factors such episodic memory, orientation and judgement. Finally, with reference to the information provided by family members, the participants' CDR levels were determined at a joint meeting of the physicians and public health nurses. Detailed methodology of this process has been described previously.13

Laboratory tests

Vitamins B1, B6, and B12, thyroid hormones and haemoglobin A1c were determined for their utility in diagnosing dementing diseases.

MRI

MRI examinations were performed with a 1.5T-MRI at the Kurihara Central Hospital. MRI scans were used to assess brain atrophy and vascular changes and to enable the diagnosis of the dementing diseases.

Dementia diagnosis and cognitive assessment

After an individual rated a CDR 1+, dementia was diagnosed based on the DSM-IV criteria. For dementing diseases, the following ‘probable’ diagnoses (see below) were made by the physicians team, which included two neurologists, a psychiatrist and a physician.

  • 1
    AD: People who met the following criteria were considered to have AD (i.e. ‘pure’ AD without CVD): the National Institute of Neurological and Communicative Disorders and Stroke and the Alzheimer's Disease and Related Disorders Association criteria for probable AD and no CVD as indicated by MRI.19 On MRI images, signal changes, which showed low signal intensity on T1-weighted images, high signal intensity on T2-weighted images and high signal intensity surrounding the low signal intensity areas on fluid-attenuated inversion recovery images, were considered to be CVD.
  • 2
    AD with CVD: People who met the following criteria were considered to have AD with CVD: the National Institute of Neurological and Communicative Disorders and Stroke and the Alzheimer's Disease and Related Disorders Association criteria for probable AD and the presence of CVD as indicated shown by MRI. However, the CVD was considered to be a concomitant lesion, and the CVD areas were not deemed responsible for the cognitive deterioration.
  • 3
    VaD: People were considered to have VaD if they met the probable VaD criteria, as per the National Institute of Neurological Disorders and Stroke and Association Internationale pour la Recherché et l'Enseignement en Neurosciences.20
  • 4
    Subcortical VaD (SVD): People were thought to have SVD if they met Erkinjuntti et al.'s criteria for SVD.21
  • 5
    Mixed dementia: People who met the criteria for AD (National Institute of Neurological and Communicative Disorders and Stroke and the Alzheimer's Disease and Related Disorders Association) as well as the criteria for VaD (NINDS-AIREN) were considered to have mixed dementia.19,20
  • 6
    Others: Idiopathic normal pressure hydrocephalus, dementia with Lewy bodies (DLB) and frontotemporal lobar degeneration were diagnosed according to each of the consensus criteria.22–24

Analysis 2: determine an LTCI index of ‘suspected dementia condition’

LTCI indices, defined here as the Care Level, the Impairment Level of Dementia (0, I–IV), and the Disability Level (0, J1, J2, A1, A2, B1, B2), were analyzed as to whether they were good indicators of ‘suspected dementia condition’. Sensitivity and specificity were evaluated for detecting dementia.

Participants 2

The old-old population aged 75 years and older to whom the LTCI index applied in Kurihara (n = 3915) was analyzed (Analysis 3).

Analysis 3: estimated prevalence of ‘suspected dementia condition’

With the LTCI index that was found to be a good indicator of ‘suspected dementia condition’, the age-specific prevalence of ‘dementia state’ was evaluated for each sex. Namely, the number of residents who met the LTCI index according to Analysis 2 in each age group was divided by the total number of residents in each age group for both sexes.

Ethics

Written informed consent was obtained from all participants and their families. The study was approved by the ethics committees of Tohoku University Graduate School of Medicine, the city of Kurihara and Kurihara Central Hospital.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGMENTS
  8. REFERENCES

Analysis 1: prevalence of dementia (CDR 1+) and the ratio of dementing diseases

In Analysis 1, there were 590 participants, which was a 47% response rate. The reasons for refusal to participate were mainly ‘psychological’ (30.1%) and ‘physical’ (18.8%). Each age group achieved a statistically sufficient number for a 95% confidence interval, and the prevalence of dementia was 10%.

Of the 590 participants, 223 people (37.8%) were diagnosed as CDR 0 (healthy), 294 (49.8%) were CDR 0.5 (questionable dementia), and 73 (12.4%) were diagnosed with CDR 1+ (dementia).

Figure 2 illustrates the age-specific prevalence of dementia for both sexes. A greater prevalence was noted among the old-old population. There was no remarkable difference between the sexes for those aged 75–79 years or those aged above 85 years, but women exhibited greater prevalence than men in the 80–84 age group.

image

Figure 2. Age- and gender-specific prevalence of dementia in Participants 1.

Download figure to PowerPoint

Figure 3 presents the ratio of dementing diseases in the CDR 1+ state. The most common cause was AD with CVD, followed by AD, mixed dementia, VaD, SVD and idiopathic normal pressure hydrocephalus. Other dementing diseases included subdural hematoma (n = 2), frontotemporal dementia (n = 1), brain tumour (n = 1), diabetic metabolic dementia (n = 1) and depressive pseudodementia (n = 1).

image

Figure 3. Dementing diseases in Participants 1. AD, Alzheimer's disease; CVD, cerebrovascular disease, iNPH, idiopathic normal pressure hydrocephalus; SVD, subcortical vascular dementia; VaD, vascular dementia.

Download figure to PowerPoint

Analysis 2: determine a LTCI index of ‘suspected dementia condition’

The relationships between each CDR and LTCI index (i.e. Care Level, Impairment Level of Dementia (0, I–IV), and Disability Level (0, J1, J2, A1, A2, B1, B2)) are illustrated in Figure 4. The Care Level was blindly decided by public health nurses in almost half of CDR 1 adults, although a few CDR 0 and CDR 0.5 cases were decided as well. The I level of Impairment Level of Dementia was found to be a good indicator of ‘suspected dementia condition’, with both the sensitivity and specificity being 0.88, respectively.

image

Figure 4. Long-Term Care Insurance indices for each CDR group. The symbol legend indicates each stage of (a) Care Level, (b) Impairment Level of Dementia and (c) Disability Level. The Care Level includes NA (not applied), NS1 (Needed Support 1), NS2 (Needed Support 2), CL1 (Care Level 1), CL2 (Care Level 2), CL3+ (Care Level 3 or greater) and ND (no data). The Impairment Level of Dementia includes Indep (independence), and various levels (I, IIa, IIb, IIIa, IIIb, and IV). The Disability Level includes Indep (independence) and various levels (J1, J2, A1, A2, B1, and B2). CDR, Clinical Dementia Rating; MRI, magnetic resonance imaging.

Download figure to PowerPoint

Analysis 3: estimated prevalence of ‘suspected dementia condition’

Of the old-old population in Kurihara, 26.8%, or 3915 residents, received LTCI. As a whole, the estimated prevalence of ‘suspected dementia condition’ was 16.7% in men, 27.7% in women and 23.6% overall. This was calculated by dividing the number of residents who were level I or higher according to the Impairment Level of Dementia by the number of residents.

Figure 5 illustrates the estimated age-specific prevalence of ‘suspected dementia condition’ in both sexes. In men, the ratio increased gradually from 20% for ages 75–87 years to 40% at 88 years and then subsequently became stable. By contrast, in women, the ratio increased from 75 years to 95+ years oldto approximately 70%.

image

Figure 5. Age-specific LTCI application rates and estimated prevalence of ‘suspected dementia condition’ in both sexes. The open circles show the LTCI application rates in women, and the closed circles indicate level I of the Impairment Level of Dementia in women. The open squares and closed squares represent the same respective values in men. Level I+/Total is the number of residents who were level I or higher for the Impairment Level of Dementia divided by the number of residents in each age gorup. LTCI%, Long-Term Care Insurance application rates.

Download figure to PowerPoint

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGMENTS
  8. REFERENCES

Methodological issues

During this study, some methodological issues were encountered. For Analysis 1, although medical diagnosis was performed and the number of participants was sufficient for a 95% confidence interval, the response rate was low. As some of refusals to participate were because of ‘psychological’ reasons, we felt that further enquiries of older adults might be culturally inappropriate. More than 50% of the refusals were for ‘physical’ reasons; most of these individuals already received the LTCI services, and they considered participating in the survey to be unnecessary. Therefore, the low prevalence of VaD might have resulted from the refusal of families of the most severely affected residents to participate. For Analysis 3, we used the LTCI index as a marker of ‘suspected dementia condition’. Table 1 shows the demographics and LTCI index information between residents who agreed to participate (Participants 1) and those who disagreed for Analysis 1. For Participants 1, the LTCI index applied to 15.4%, and of those, 41.8% had ‘suspected dementia condition’ per the I level of Impaired Level of Dementia and well supported by the CDR. In contrast, the LTCI index did not apply to 84.6% of Participants 1, but 17.8% met the criteria of ‘suspected dementia condition’ per the index, though not well supported by the CDR. This indicated a high sensitivity (36/38 = 94.7%) of ‘suspected dementia condition’ to dementia (CDR 1+) in the residents who met the LTCI index criteria compared with those who did not (38/89 = 0.43). Regarding residents who declined to participate (disagreed), the LTCI index applied to 31.1%, and 91.7% of them met the criteria of ‘suspected dementia condition’. Primary care doctors described their diagnoses as ‘senile dementia’ in 177 out of the 189 cases (93.7%), which was a similar result to the 94.7% (36/38), although the diagnoses needed further confirmation. No data were available for the residents who did not use LTCI, so we could not speculate on the prevalence of ‘suspected dementia condition’ for all residents.

Table 1. Analysis 1: demographics and LTCI information between residents who agreed to participate (Participants 1) and those who disagreed
 Agreed (Participants 1) (n= 590)Disagreed (n= 662)
  1. There were no significant differences between two groups for age (t-test) and gender ratios (χ2 test).

  2. CDR, Clinical Dementia Rating; LTCI, Long-Term Care Insurance; ND, no data.

Age80.1 (4.2)82.2 (5.7)
Sex (men/women)225/365217/445
 AppliedNot appliedAppliedNot applied
LTCI application91 (15.4%)499 (84.6%)206 (31.1%)456 (68.9%)
Level I+ of impaired level of dementia38 (41.8%)89 (17.8%)189 (91.7%)ND
CDR 1+36 (39.6%)38 (7.6%)NDND

Despite these limitations, we believe that the results yielded information relevant to dementia in Japan.

Validity of LTCI indices

We should note a high specificity of the Care Level for dementia. With level I of the Impairment Level of Dementia as a cut-off, dementia diagnoses were validated. Previously level II was considered to be an indicator of ‘suspected dementia condition’ for health policy planning (Japanese Ministry of Health, Labour and Welfare, 2003, http://www.mhlw.go.jp/topics/kaigo/kentou/15kourei/3c.html). The current results suggest that the previous data might underestimate the prevalence of dementia.

Prevalence of dementia

We reported the estimated prevalence of ‘suspected dementia condition’ to be 23.6% (Analysis 3) and the prevalence of dementia to be 12.5% (Analysis 1) in those 75 years or older. The discrepancy in prevalence data was probably due to the low participation ratio (47.0%) of Participants 1. Because level I of the Impairment Level of Dementia was validated by the CDR rating, the value of 23.6% should be considered as accurate.

We previously reported the prevalence of dementia to be 8.5% among adults 65 year and older in Tajiri, Japan.13 Tajiri and Kurihara are similar areas in Miyagi Prefecture, and a reconsideration of the Osaki-Tajiri Project data can provide us with information: 12.5% of the old-old population in Tajiri was found to have dementia, which is similar to this study's findings.

We reviewed articles published over the past 10 years regarding the prevalence of dementia. Figure 6 illustrates the age-specific plotting from those articles, the Osaki-Tajiri Project and the results of this study. Our current data fell in the middle among these studies.

image

Figure 6. Previous results from studies on age-specific prevalence of dementia together with the current data.

Download figure to PowerPoint

Is dementia age-related or ageing-related?

Whether dementia is an ‘ageing-related’ or ‘age-related’ disorder, has been discussed previously.25 The former concept suggests that the incidence of dementia increases with age, while the latter concept suggests that dementia incidence is higher but stable in old age and does not always increase with age. A previous meta-analysis of epidemiological studies concluded that dementia is better conceptualized as an ‘age-related’ rather than an ‘ageing-related’ disorder.25 Previous results of neuropsychological investigations of healthy older adults and longitudinal studies of screening test performances over a 5-year period suggest that dementia is age-related.26,27

However, the results in this study showed a difference between the sexes: an ‘age-related’ increase in men and an ‘ageing-related’ increase in women. A previous study indicated that the ratio of pathologically proven AD or AD with CVD increased with age, up to age 90 in women, although there was a stable but not age-specific increase in men. In contrast, for VaD, the ratio was stable with age in women and decreased with age in men.28 We can speculate that the prognosis after dementia onset was poorer in men. Indeed, the prognosis for those with VaD was poor, despite the very mild stage.29 Another possibility was that few families used LTCI for older men who were 87 years old or younger; LTCI rates of use suddenly increased for men 88 years of age or older. The rates of LTCI use support the hypothesis (see Fig. 5).

Dementing diseases in the old-old in Japan

Our results showed that AD with CVD was the most common dementing disease, which coincided with the previous results of individuals aged 65 years and older.13 The ratio between AD with CVD and mixed dementia was greater than in the Osaki-Tajiri Project. As a matter of course, the old-old population was more likely to suffer from CVD than those in the younger population. People who had at least one form of CVD, which could cause cognitive deterioration, were assessed as having mixed dementia.

Surprisingly, we found no patients with DLB during the 3-year course of this study. We ensured that DLB patients with parkinsonism or visual hallucinations were not misinterpreted by family as having an ‘ageing-related’ phenomenon rather than a disease. Also, we considered that DLB patients may have been referred to physicians earlier or have already been institutionalized. A previous study reported that the lifespan after DLB diagnosis was relatively poor compared to other dementing diseases. Thus, DLB patients may have already died before reaching age 75 years, or their activities of daily living status had declined and they were unable to provide informed consent to participate in the study.

ACKNOWLEDGMENTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGMENTS
  8. REFERENCES

We are grateful to the staff of the city of Kurihara. This study was supported by grants from the Kurihara, Miyagi, Japan (2008–2010) and partly by the Japanese Ministry of Health, Labour and Welfare (2009–2010).

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. ACKNOWLEDGMENTS
  8. REFERENCES