Abstract
Using the condensed list of causes of death in the Vital Statistics (yearly for 1995–2024, and by 5-year age group for 2005, 2016, and 2024), the rise in deaths from vascular and unspecified dementia and from Alzheimer's disease was divided into three parts and lined up: the age composition of the population, the age-specific rates, and the 2017 change in how deaths are counted. It claims no new mathematical theorem and no new law.
First, the tables are consistent. In every year, total = men + women and the sum of 5-year groups + unknown = total, and the totals in Volume 2 Table 2 for 2005, 2016, and 2024 match the figures for the same year in Volume 1 Table 5.13 (Section 3).
Second, from 1995 to 2024, deaths from vascular and unspecified dementia went from 2,697 to 24,666 (9.1-fold), and deaths from Alzheimer's disease from 511 to 25,595 (50.1-fold). Among the year-to-year ratios for 2014–2024, the largest for both is 2016 to 2017 (1.643 and 1.440); the other years lie within 0.973–1.090 and 1.006–1.135 (Section 4).
Third, this is the core of the paper. Deaths from Alzheimer's disease at ages 40+ rose 14.12-fold from 2005 to 2024; keeping the rates at 2005 and moving only the age composition to 2024 gives 1.93-fold, and keeping the population at 2005 and moving only the age-specific rates to 2024 gives 6.65-fold. For vascular and unspecified dementia, 7.40-fold divides into 2.28-fold and 3.42-fold. For both causes and all three periods, the rate part is larger than the age-composition part (Section 5).
Fourth, from 2005 to 2016, the ratio of Alzheimer's disease rates grows with age, from 3.38 at 80–84 to 9.48 at 100+. For vascular and unspecified dementia it stays within 2.01–2.29 over the same ages (Section 6).
Fifth, dividing the 2016-to-2017 ratio by the ratio from recoding the same death records under the new and old classifications (the new/old ratio of M041) gives 1.643 ÷ 1.509 = 1.089 for vascular and unspecified dementia and 1.440 ÷ 1.296 = 1.111 for Alzheimer's disease, both within the range of the other years. Assuming the new/old ratio acted equally at all ages, the rate part from 2016 to 2024 becomes 1.05 for vascular and unspecified dementia and 1.27 for Alzheimer's disease (Section 7).
Sixth, what separates them is the age composition of the population, the age-specific rates, or the 2017 change in how deaths are counted. The rise in dementia deaths contains all three, and the two causes divide them differently.
Scope note: This paper claims no new mathematical theorem and no new law — the death counts and death rates are in public tables, and the method of dividing into age composition and rates is long established. This paper applied it to two dementia causes and lined up the results. This paper is a structural interpretation of the shape of numbers in public aggregate data; it is not medical advice, not a diagnosis, and not a treatment recommendation. The author is not a physician. Decisions about individual health should be discussed with a physician or a medical institution. This is a preprint that has not been peer reviewed. It does not go into the diagnosis, treatment, prevention, or care of dementia. The cause of death is the underlying cause based on death certificates, and this does not blame those who wrote them. These tables count only the underlying cause, so when a person with dementia dies and the underlying cause is another disease, the death is not counted as dementia. Whether the rise of the rate at the same age is a change in how certificates are written or a change in how people actually die cannot be separated by these aggregate numbers. Considering that families who cared for someone at the end of life, or who are caring for someone now, may read it, it is not written as an outlook for any one person. About the data: the Vital Statistics (e-Stat) are used under the Government Standard Terms of Use (Version 2.0), and the ratios in this paper were created by processing them. What this paper adds: dividing the rise in deaths from two dementia causes into the age composition, the age-specific rates, and the 2017 change in how deaths are counted, and showing that the rise in the Alzheimer's disease rate grows with age, that the 2017 jump is nearly accounted for by the recoding ratio, and that once it is removed the two causes' rates move apart after 2016.
1. Notation and question
- The two causes: code 05100 "vascular and unspecified dementia" in the condensed list of causes of death (in the 2005 table the same code carries an older Japanese word for dementia), and code 06400 "Alzheimer's disease." In the text the former is also written "vascular etc. dementia."
- Rate: the death rate per 100,000 population. The population was obtained from Volume 1 Table 5.15 as all-cause deaths ÷ death rate × 100,000 (for each 5-year group).
- Age-composition part, rate part: over the 13 five-year groups from 40 up (40–44 … 100+), the deaths with rates kept at the earlier year and only the population moved to the later year, and the deaths with the population kept at the earlier year and only the rates moved to the later year, each divided by the deaths in the earlier year.
- New/old ratio: the ratio (new ÷ old) that M041 obtained from the material in which the same 2016 death records were counted under both the new and the old classification.
Question: Did dementia deaths rise because there are more old people, because the rate at the same age rose, or because the way of counting changed in 2017?
Observation 1.1: M055 divided the rise in deaths from senility into the age composition, the age-specific rates, and the swapping of causes within the same age. M041 showed, with the new/old ratio, that the partial revision of the cause-of-death classification in 2017 moved deaths from pneumonia to dementia, Alzheimer's disease, and others. This paper divides the rise of the two causes to which those deaths moved.
2. Survey and design
To be honest: the design was decided after seeing the tables.
| Content | |
|---|---|
| Tables | Volume 1 Table 5.13 (condensed causes × year), Volume 2 Table 2 (condensed causes × sex × 5-year age for 2005, 2016, and 2024), Volume 1 Table 5.15 (back-calculating the population) |
| Noticed in the survey | Among the yearly ratios in Table 5.13, only 2016 to 2017 jumped sharply for both causes |
| Years compared | 2005 and 2024. Also 2005 to 2016, before the 2017 partial revision, and 2016 to 2024, across it |
| Measures | The age-composition part and rate part, the ratio of age-specific rates at 80+, and the 2016-to-2017 ratio ÷ the new/old ratio of M041 |
3. Data and checks
Calculation 3.1 The 5 retrieved files were fixed by SHA-256, and the following was checked.
- The title of Table 5.13 is "Volume 1, Deaths, Table 5.13: deaths and death rates (per 100,000) by sex and year, by cause of death (condensed list)." The years are the 15 of 1995, 2000, 2005, 2010, and 2014–2024; rows for 05100 and 06400 exist for both sexes, men, and women, and total = men + women in every year.
- In Volume 2 Table 2 for 2005, 2016, and 2024, the age headings are total, each single year 0–4, 21 five-year groups, and unknown; for both causes, the sum of 5-year groups + unknown = total and men + women = total.
- In all three years, the total in Volume 2 Table 2 matches the figure for the same year in Table 5.13 (both causes; both sexes, men, and women).
4. Yearly deaths and the 2017 jump
Calculation 4.1 Deaths (Table 5.13, both sexes)
| Year | Vascular and unspecified dementia | Alzheimer's disease |
|---|---|---|
| 1995 | 2,697 | 511 |
| 2005 | 3,334 | 1,814 |
| 2016 | 11,898 | 11,972 |
| 2017 | 19,551 | 17,239 |
| 2024 | 24,666 | 25,595 |
| 1995 to 2024 | 9.1-fold | 50.1-fold |
| 2016 to 2017 | 1.643-fold | 1.440-fold |
| Other consecutive years (2014–2024) | 0.973–1.090-fold | 1.006–1.135-fold |
Observation 4.2: For both causes, the largest of the year-to-year ratios for 2014–2024 is 2016 to 2017. Vascular and unspecified dementia rose 1.643-fold, outside the other years' 0.973–1.090, and Alzheimer's disease 1.440-fold, outside the other years' 1.006–1.135. From 2017 on, both years are under the new classification, and 2017 to 2024 is 1.262-fold and 1.485-fold.
Figure 1: Deaths from vascular and unspecified dementia (navy) and from Alzheimer's disease (gold), 1995–2024. The vertical dotted line is 2017.
5. Core — the age-composition part and the rate part
Calculation 5.1 Dividing the rise in deaths at ages 40+ (13 five-year groups)
| Cause, period | Deaths | Ratio | Age-composition part | Rate part | Product of the two |
|---|---|---|---|---|---|
| Vascular etc., 2005 to 2016 | 3,333 → 11,897 | 3.57 | 1.70 | 2.10 | 3.57 |
| Vascular etc., 2016 to 2024 | 11,897 → 24,666 | 2.07 | 1.34 | 1.58 | 2.12 |
| Vascular etc., 2005 to 2024 | 3,333 → 24,666 | 7.40 | 2.28 | 3.42 | 7.79 |
| Alzheimer's, 2005 to 2016 | 1,813 → 11,972 | 6.60 | 1.55 | 4.02 | 6.22 |
| Alzheimer's, 2016 to 2024 | 11,972 → 25,594 | 2.14 | 1.31 | 1.64 | 2.15 |
| Alzheimer's, 2005 to 2024 | 1,813 → 25,594 | 14.12 | 1.93 | 6.65 | 12.83 |
Observation 5.2: The 14.12-fold rise in Alzheimer's disease from 2005 to 2024 divides into an age-composition part of 1.93-fold and a rate part of 6.65-fold, and the 7.40-fold rise in vascular and unspecified dementia into 2.28-fold and 3.42-fold. The relative difference between the product of the two and the actual ratio is within 0.091 in all six. For both causes and all three periods, the rate part is larger than the age-composition part. On top of the same growth in the age composition, the growth in the Alzheimer's disease rate sits one step larger.
6. Age-specific rates — Alzheimer's disease rose more at older ages
Calculation 6.1 Rates at 80+ (per 100,000; 2005 → 2016 → 2024) and their ratios
| Age | Vascular etc. rate | Ratio 2005 to 2016 | Alzheimer's rate | Ratio 2005 to 2016 | Ratio 2016 to 2024 |
|---|---|---|---|---|---|
| 80–84 | 13.9 → 31.3 → 60.6 | 2.25 | 11.5 → 38.7 → 65.5 | 3.38 | 1.69 |
| 85–89 | 46.0 → 93.2 → 155.8 | 2.03 | 23.0 → 103.7 → 168.4 | 4.51 | 1.62 |
| 90–94 | 120.6 → 242.3 → 328.9 | 2.01 | 39.5 → 221.4 → 347.4 | 5.61 | 1.57 |
| 95–99 | 240.4 → 551.6 → 687.2 | 2.29 | 52.5 → 409.4 → 621.0 | 7.80 | 1.52 |
| 100+ | 465.3 → 961.5 → 1240.2 | 2.07 | 51.3 → 486.2 → 972.4 | 9.48 | 2.00 |
Observation 6.2: From 2005 to 2016, for both causes, the rate rose in every 5-year group 80+. The ratio for Alzheimer's disease grows with age, from 3.38 at 80–84 to 9.48 at 100+. For vascular and unspecified dementia, the ratio stays within 2.01–2.29 over the same ages and hardly changes with age. In 2005 the Alzheimer's disease rate was about the same height at 95–99 (52.5) and 100+ (51.3), but in 2016 it was 409.4 and 486.2. From 2016 to 2024, the ratios are 1.52–2.00 for Alzheimer's disease and 1.25–1.93 for vascular and unspecified dementia.
Figure 2: For each 5-year age group at 80+, the 2016 rate ÷ the 2005 rate. Navy is vascular and unspecified dementia, gold is Alzheimer's disease. The dotted line is 1 (no change).
7. Removing the 2017 change in how deaths are counted
Calculation 7.1 The 2016-to-2017 ratio and the new/old ratio of M041 (read from the output of M041's checker, after confirming that its 2016 and 2017 deaths match the Table 5.13 figures used here)
| Cause | 2016-to-2017 ratio | New/old ratio | Ratio ÷ new/old ratio | Other consecutive years |
|---|---|---|---|---|
| Vascular and unspecified dementia | 1.643 | 1.509 | 1.089 | 0.973–1.090 |
| Alzheimer's disease | 1.440 | 1.296 | 1.111 | 1.006–1.135 |
Observation 7.2: Dividing the 2016-to-2017 ratio by the new/old ratio puts both causes within the range of the other years. The 2017 jump is nearly accounted for by recounting the same death records under the new classification.
Calculation 7.3 Assuming the new/old ratio acted equally at all ages, removing it from the rate part
| Cause, period | Rate part | ÷ new/old ratio |
|---|---|---|
| Vascular etc., 2016 to 2024 | 1.58 | 1.05 |
| Vascular etc., 2005 to 2024 | 3.42 | 2.26 |
| Alzheimer's, 2016 to 2024 | 1.64 | 1.27 |
| Alzheimer's, 2005 to 2024 | 6.65 | 5.13 |
Observation 7.4: Under that assumption, the rate part from 2016 to 2024 becomes 1.05 for vascular and unspecified dementia and 1.27 for Alzheimer's disease. Much of the rate growth after 2016 is the counting change for vascular and unspecified dementia, while for Alzheimer's disease some remains after removing it. But "equal at all ages" is an assumption this paper made; the new/old ratio by age is not in that material.
8. Limitations
- The cause of death is the underlying cause, and which cause name it is counted under follows the death certificate and the classification rules. A change in how certificates are written and a change in how people die cannot be separated. These are not the deaths of people with dementia.
- The effect of the 2017 partial revision was estimated with M041's new/old ratio (2016 death records, all ages). There is no new/old ratio by age, so the adjusted rate parts in Section 7 are numbers resting on an assumption.
- The population was back-calculated from deaths ÷ death rate. Deaths under 40 and of unknown age are not in the division, so in some years the deaths in Section 5 differ slightly from those in Section 4.
- National yearly totals; prefectures and facilities are not compared. It is not an outlook for any one person.
Conclusion
Deaths from vascular and unspecified dementia and from Alzheimer's disease rose 9.1-fold and 50.1-fold from 1995 to 2024. The 14.12-fold rise in Alzheimer's disease at ages 40+ from 2005 to 2024 divided into an age-composition part of 1.93-fold and an age-specific-rate part of 6.65-fold, and from 2005 to 2016 the ratio of rates grew with age, from 3.38 at 80–84 to 9.48 at 100+. The 2017 jumps (1.643 and 1.440), divided by the recoding ratios (1.509 and 1.296), fall within the range of the other years.
Three things separate them — the age composition of the population, the age-specific rates, or the 2017 change in how deaths are counted.
On hearing "more people are dying of dementia," check together whether there are more old people, whether the rate counted under a dementia name at the same age rose, and whether the way of counting changed that year.
Data and reproduction
| Data | Vital Statistics 2024 (final figures), Volume 1 Tables 5.13 and 5.15; Vital Statistics 2005, 2016, and 2024 (final figures), Volume 2 Table 2 (Ministry of Health, Labour and Welfare; e-Stat) (retrieved 26 September 2026) |
| Terms of use | e-Stat follows the Government Standard Terms of Use (Version 2.0). The ratios in this paper were created by processing it and are not figures produced by the government |
| Scope | Japan, 1995–2024 |
| Checker | 検算/M064.py (--zu also draws the figures). Every number in the text is taken from 検算/M064_output.txt. The new/old ratio is read from the output of M041's checker |
| Raw data | The retrieved files and SOURCES.md in data/ (URL, retrieval date, SHA-256, terms of use, definitions and how calculations are set up) |
| Figures | 図/M064_zu1, 図/M064_zu2 (png, svg, English versions) |
References
- Ministry of Health, Labour and Welfare. Vital Statistics 2024 (final figures), Volume 1 Tables 5.13 and 5.15, Volume 2 Table 2 (in Japanese). e-Stat (accessed 26 September 2026). https://www.e-stat.go.jp/
- Ministry of Health, Labour and Welfare. Vital Statistics 2005, Volume 2 Table 2; Vital Statistics 2016, Volume 2 Table 2 (in Japanese). e-Stat (accessed 26 September 2026). https://www.e-stat.go.jp/
- Yamagishi, Y. (2026). The Fall in Pneumonia Deaths in 2017 Is Nearly Predicted by the Ratio from Recoding the Same Death Records under the New and Old Classifications — for Pneumonia, an Actual Step of 0.826 Against a New/Old Ratio of 0.810, a Correlation of 0.969 Across 65 Causes, and the Lost Deaths Moved to Dementia, Alzheimer's Disease, Heart Failure, COPD, and Others [M041]. Zenodo. https://doi.org/10.5281/zenodo.22974374
- Yamagishi, Y. (2026). Of the 7.85-Fold Rise in Deaths from Senility from 2005 to 2024, the Age Composition Accounts for 2.51-Fold and the Age-Specific Rates for 3.25-Fold — At Every Age 80+, the Senility Rate Rose 2.66–4.36-Fold While the Sum of Pneumonia, Heart Disease, and Cerebrovascular Disease Rates Fell to 0.50–0.59 [M055]. Zenodo. https://doi.org/10.5281/zenodo.22975275
Reference
Yamagishi, Y. (2026). Of the 14.12-Fold Rise in Deaths from Alzheimer's Disease from 2005 to 2024, the Age Composition Accounts for 1.93-Fold and the Age-Specific Rates for 6.65-Fold [M064]. Zenodo.
DOI: 10.5281/zenodo.22976780
YOI: https://www.yamagishiyuuki.com/yoi/M064
ORCID: 0009-0006-8587-3450
PDF (English) · PDF (Japanese) · Date proof of the English PDF (OpenTimestamps .ots) · SHA256SUMS
The record page for this paper (in Japanese, with the PDFs in each language and the date proof): YOI M064. This paper is a preprint and has not been peer reviewed.