Longitudinal changes in the prevalence of dementia in a Japanese rural area

Authors


Dr Yosuke Wakutani MD, PhD, Department of Neurology, Institute of Neurological Sciences, Faculty of Medicine, Tottori University, 36-1 Nishi-cho, Yonago 683-8504, Japan. Email: wakutani@utoronto.ca

Abstract

Background:  The increasing number of patients with dementia in Japan, together with the rapid aging of society, is currently considered to have a substantial impact on Japan's medical, economic and sociological systems. Therefore, the longitudinal estimation of changes in the prevalence of dementia based on accurate diagnostic evaluation has important implications.

Methods:  We undertook three separate epidemiological studies on long-term changes, 10 years apart (1980, 1990 and 2000), in the prevalence of dementia in an elderly population using identical methods (DSM-III and Hachinski's ischemic score) for the same rural area in Japan (Daisen-cho).

Results:  The percentage of the population that was elderly (over 65 years of age) increased steadily from 16.0% in 1980 to 21.7% in 1990 and 27.1% in 2000. The prevalence of dementia (cases/100 people aged 65 years or older, adjusted to the population structure of 1980) in 1980, 1990 and 2000 was 4.4, 4.5 and 5.9, respectively, for all types of dementia, 1.9, 2.5 and 3.6, respectively, for Alzheimer-type dementia (DAT) and 2.0, 1.7 and 2.2, respectively, for vascular dementia (VaD).

Conclusions:  These findings of an increase in the number of cases and prevalence of DAT and VaD in a Japanese rural community have important implications for interventional medicine.

INTRODUCTION

One of the most important issues in the public health of Japan is the rapid aging of society. It is highly possible that the increasing number of patients with dementia may become a serious social problem, impacting on Japan medically, economically and sociologically. Therefore, longitudinal estimation of changes in the prevalence of dementia based on accurate diagnostic evaluation has important implications.

It has been reported previously that vascular dementia (VaD) is more predominant than dementia of the Alzheimer type (DAT) among the Japanese population.1 However, several recent reports have shown that the incidence of DAT is equal to or greater than that of VaD.2–5 At present, there are few reports that consider longitudinal changes in the prevalence of dementia in Japan.

Several clinical criteria have been developed to standardize the diagnosis of dementia, including DAT and VaD. Significant differences in patient classification have been reported, depending on the criteria used. In particular, recent studies have demonstrated that clinical criteria for VaD are not interchangeable.6,7 Thus, the use of identical clinical criteria is indispensable for the accurate estimation of changes in the prevalence of dementia. We have been conducting longitudinal prevalence studies of dementia, 10 years apart, in the elderly population using identical methods for the same area (Daisen-cho) in Japan.

METHODS

Epidemiological studies were repeated at 10 year intervals (1980, 1990 and 2000) for the entire population of Daisen-cho. Daisen-cho is located in a rural area of western Japan (Fig. 1). The population structure of Daisen-cho in 1980, 1990 and 2000 is shown in Fig. 2. The population was 7741 (3668 men and 4073 women) in 1980, 7749 (3674 men and 4075 women) in 1990 and 7020 (3354 men and 3666 women) in 2000. The number of elderly people over 65 years of age increased over two decades: 1236 (16.0%) in 1980, 1626 (21.0%) in 1990 and 1851 (26.4%) in 2000. The migration rate of the population was approximately 1% or less and is therefore considered very low, especially among the elderly population of Daisen-cho. We examined the prevalence rate of dementia in the elderly population over a 10 year period using methods detailed previously8–10 (Fig. 3).

Figure 1.

Map of Japan, showing the location of Daisen-cho.

Figure 2.

Population structure and the number and ratio of people above 65 years of age in Daisen-cho in 1980, 1990 and 2000.

Figure 3.

Methods used to investigate dementia.

First, we performed screening tests of the data obtained from the Daisen-cho questionnaire for all inhabitants over 20 years of age. The Daisen-cho questionnaire data consist of lifestyle items (including occupation and working hours), an abridged medical history (including information about hypertension, hyperlipidemia, diabetes mellitus, cerebrovascular disease, Parkinson's disease, DAT and cancer) and recent subjective symptoms focusing on neurological issues (including amnesia, headache, numbness, weakness and speech and gait disturbances). The total response rate of the Daisen-cho questionnaire in 2000 was 85.5% and the response rate for the elderly population (over 65 years of age) was 82%. We identified individuals who had cerebrovascular disease, DAT and Parkinson's disease based on their medical history and who may have had amnesiac episodes or other neurological signs based on subjective symptoms.

We conducted further documentary searches, including the population stroke screening record, National Health Insurance Medical Records and nursing care insurance records. Volunteer health officers operate in each small community in Daisen-choand we interviewed them to determine whether there are any individuals with neurological disabilities, including amnesiac symptoms, in their communities. There were 127, 167 and 186 possible cases of dementia in 1980, 1990 and 2000, respectively.

For supplementary evaluation of dementia, qualified neurologists visited the candidates and their family member(s) in their homes or met with them in the official day-care center of Daisen-cho. The supplementary evaluation consisted of assessment of these patients based on a thorough medical history, physical examination, including a drug inventory, neurological examination, comprehensive cognitive evaluation using the Mini-Mental State Examination,11 activity of daily life evaluation with the Barthel Index,12 psychosocial assessment of the patient's environment and routine laboratory tests. Patients who satisfied the DSM-III and those scoring 4 points or less on Hachinski's ischemic score were diagnosed as having DAT.13,14 Patients who satisfied the DSM-III and those scoring 7 points or more on Hachinski's ischemic score were diagnosed as having VaD. The degree of dementia (mild, moderate or severe) was assessed according to a functional assessment staging of Alzheimer's disease (FAST).15

RESULTS

The progressive aging of society was clearly evident in Daisen-cho. The percentage of individuals over 65 years of age was 16.0% in 1980, 21.0% in 1990 and 26.4% in 2000. The number of all types of dementia was 56 of 1236 people aged 65 years or more in 1980, 82 of 1626 persons in 1990 and 137 of 1823 persons in 2000. Therefore, the number of all types of dementia in 1990 and 2000 had increased approximately 1.5- and 2.4-fold, respectively, compared with that in 1980 (Fig. 4).

Figure 4.

Number of cases of dementia in people over 65 years of age in 1980 (1236; □), 1990 (1626; inline image) and 2000 (1823; ▪). DAT, Alzheimer-type dementia; VaD, vascular dementia.

Unadjusted prevalence rates for dementia in the elderly population were 4.4 per 100 population in 1980, 4.9 in 1990 and 7.4 in 2000. The age-adjusted prevalence rate in those aged 65 years or more compared with the 1980s population structure in Daisen-cho was 4.5 per 100 population in 1990 and 5.9 in 2000. The number of DAT cases was 24 in 1980, 41 in 1990 and 66 in 2000. The adjusted prevalence rates of DAT were 1.9 in 1980, 2.3 in 1990 and 2.8 in 2000. There were 26 cases of VaD in 1980, 31 cases in 1990 and 56 cases in 2000. The adjusted prevalence rates of VaD were 2.0 in 1980, 1.7 in 1990 and 2.2 in 2000 (Fig. 5). The ratio of VaD to DAT was 1.1 in 1980, 0.8 in 1990 and 0.8 in 2000, indicating that DAT had clearly become more prevalent than VaD over the two decades.

Figure 5.

Adjusted prevalence rates of dementia in people over 65 years of age in 1980 (□), 1990 (inline image) and 2000 (▪). DAT, Alzheimer-type dementia; VaD, vascular dementia.

Dividing the cases of dementia into two groups according to FAST severity, the ratio of mildly demented patients had increased over the two decades. In particular, the increase in the ratio of mildly demented DAT patients was obvious through the two decades, whereas the ratio of mildly demented VaD patients increased from 1980 to 1990 and decreased from 1990 to 2000 (Fig. 6).

Figure 6.

Ratio of mild and severe–moderate cases of dementia in people over 65 years of age in 1980, 1990 and 2000. DAT, Alzheimer-type dementia; VaD, vascular dementia.

DISCUSSION

The present study shows the longitudinal transition of the prevalence of dementia in the population over 65 years of age in a community (Daisen-cho) situated in a rural area of western Japan. Because Daisen-cho was an evidently stable population, it was suitable for investigations of longitudinal changes in the prevalence of dementia patients. Further, to avoid discrepancy of the longitudinal prevalence owing to differences in patient collection methods and diagnostic criteria, we used identical methods throughout the present study. We used DSM-III criteria for dementia evaluation and Hachinski's ischemic score to differentiate DAT and VaD.

The progressive aging of the population was shown to be significant in Daisen-cho. As predicted, the number of dementia patients increased steadily. Unadjusted prevalence rates for dementia in the elderly population aged 65 years or more were 4.4 per 100 population in 1980, 4.9 in 1990 and 7.5 in 2000, indicating that the progressive aging of the population has had an impact on the increased number of dementia patients. The unadjusted prevalence rate for dementia in Daisen-cho in 2000 substantially agrees with the recently developed epidemiological study of dementia in Japan.2–5 Furthermore, the age-adjusted prevalence of dementia obviously increased in 2000 compared with 1980 and 1990. Recent epidemiological studies in Japan have demonstrated that the prevalence of DAT exceeds that of VaD.2–5

Although it is predicted that the Japanese lifestyle (particularly dietary habits), even in rural areas, is closely associated with the increased ratio of DAT, the precise factors responsible are yet to be identified. The increased number and prevalence of VaD in Daisen-cho is consistent with the recent results of a computed tomography based study conducted in another rural area in Japan.4 Although the precise factor(s) explaining the increasing prevalence of dementia, DAT and VaD in Daisen-cho remains unknown, the increasing ratio of moderate or severe VaD may reflect reduced mortality from cerebrovascular diseases and the increase in disease duration in Japan. Moreover, owing to the therapeutic progress in Japan for aging-related diseases, such as infectious diseases (e.g. pneumonia), lifestyle-related diseases (e.g. hypertension, diabetes mellitus, hyperlipidemia, coronary heart diseases, chronic cardiac failure and cerebrovascular diseases), orthopedic diseases (e.g. bone fractures) and cancers, the number of elderly people having (or surviving) those diseases has increased in the Japanese population and this issue may lead to increased numbers of elderly people ‘at risk’ of developing dementia. Recent epidemiological studies have shown that hypertension, diabetes mellitus or other atherosclerosis-related factors (e.g. increased plasma levels of homocysteine) are important risk factors in the elderly population for the development of dementia, VaD and DAT.16–19 Assuming that not only vascular factors, but also other unidentified factors (e.g. alterations in hormonal homeostasis) based on these diseases are closely related to the pathogenesis of DAT, the decrease in acute and mortal vascular diseases (cardiovascular diseases or cerebrovascular diseases) as a result of effective therapies could be inversely associated with the increase in the prevalence of chronic brain diseases, especially DAT.

Conversely, in the severity analysis based on FAST staging, an increased ratio of mild dementia cases, in particular DAT cases, was observed. Although it may be predicted that recent developments in Japan in medical and social intervention for the aging-related diseases mentioned above could also have a beneficial impact on the progression of DAT, leading to an increased ratio of mild cases, these predicted aspects will need to be investigated in the future.

In conclusion, we have shown an increased prevalence of dementia, in particular DAT, in a Japanese rural area using clinical criteria. We did not have neuroradiological or pathological evidence of the dementia subtype in our patients. However, our data have important implications for future interventional medicine for dementia in Japan.

ACKNOWLEDGMENTS

The authors thank all the inhabitants of Daisen-cho for participation in the present study. The author also thank the Daisen-cho health officers, especially Akemi Matsunami, Yoko Fujita and Satoko Ishizashi, for their constant support. This study was supported, in part, by a Grant-in-Aid for scientific research from the Ministry of Education, Culture, Sports, Science and Technology, Government of Japan (TT and MK), Health and Labor Sciences Research Grants (TT and KN.)

Ancillary