alzhub.org
Get help now — free helplines

You can call any of these for free. You do not need a diagnosis, an appointment, or a crisis to call — questions and plain exhaustion are reasons enough.

  • Alzheimer's Association 24/7 Helpline 800-272-3900 24 hours a day, 365 days a year Free, confidential support and information for people with dementia, caregivers, families and the public. Master's-level clinicians. Also Spanish.
  • 988 Suicide & Crisis Lifeline 988 24/7 Call or text 988, or chat at 988lifeline.org. For anyone in emotional distress or a mental-health crisis, including caregivers.
  • Alzheimer's Foundation of America (AFA) Helpline 866-232-8484 9:00 am - 9:00 pm ET, seven days a week Licensed social workers providing support, info and referrals. Also text message and web chat. Text: 646-586-5283

In a medical emergency, call 911. More numbers, and what each line does, on the help page.

Home › Research › Trends

Alzheimer's over time, 1999-2024

Every other headline number on this site is a single frozen figure from an unstated year. This page is the first to show change over a full 26-year time series — and, because a change over time is easy to over-read, it puts the reason to be skeptical right next to the numbers, not in a footnote at the bottom.

+69%

change in the national median rate, 1999 → 2024

CDC WONDER / NCHS, age-adjusted
50 of 51

states with a higher age-adjusted rate in 2024 than 1999

CDC WONDER / NCHS, age-adjusted
×4.3

gap between the highest (Mississippi) and lowest (District of Columbia) state rate in 2024 — both age-adjusted

CDC WONDER / NCHS, 2024
-15%

change in the national median rate from its 2017 peak to 2024

CDC WONDER / NCHS
2017

the only year all 51 states report the Alzheimer's/vascular/unspecified/other split

CDC WONDER / NCHS

Read this before the state chart below: age-adjusted still doesn't mean comparable

Every rate on this page is already age-adjusted — standardized to the 2000 US population, specifically so that a state with more older residents doesn't automatically show a higher rate. That rules out "the population got older" as the explanation for what follows.

Which means the pattern below — a more than four-fold gap between the highest- and lowest-rate state in the same year, and 50 of 51 states showing a higher rate in 2024 than in 1999 — is much more likely to reflect how death certificates get filled out and coded from state to state (which physician or medical examiner signs it, how consistently Alzheimer's is listed as the underlying cause versus a contributing one, and a well-documented national shift in coding practice toward Alzheimer's-specific codes since around 2010) than a real four-fold difference in who develops the disease. Nothing on this page can fully separate real disease change from certification-practice change — that separation would need clinical diagnostic data this file doesn't have — but the direction of the effect is well established enough that reading this chart as a map of where Alzheimer's is "worse" would be a mistake.

The national line

The national rate, 1999-2024

Median of all 51 states' age-adjusted rates each year (17.0 in 1999 → 28.7 in 2024, peak 33.6 in 2017).

US age-adjusted Alzheimer's death rate, 1999-2024Median across 51 states/DC, from 17.0 per 100,000 in 1999 to 28.7 in 2024, peaking at 33.6 in 2017.010203040199920042009201420192024deaths per 100,000, age-adjusted1999: first ICD-10 yearmedian across states (1999-2024)population-weighted (2000-2024 — no state population for 1999)

This chart is wider than your screen — drag it sideways, or open the table underneath.

Source: CDC WONDER / NCHS Stats of the States. “National” here is the median of the 51 state rates each year, not a population-weighted average, because state population is only in this file from 2000 onward (see provenance below) — 1999 predates the Census Bureau's intercensal population series, so that single year would go unweighted while every other year in the file does not, mixing two different kinds of number silently. Where both can be computed (2000-2024) they are shown together and they do not agree exactly: the weighted line sits below the median because the most populous states (e.g. New York, Florida) have below-median rates, so weighting by population pulls the national figure down. 1999 is the first year the US used ICD-10 to code cause of death; the jump from 1999 to 2000 (+15.9%) is the single largest year-over-year move in the series, consistent with a coding-system transition rather than one year of the disease itself changing.
Every year in this chart, as a table
YearMedian rate (per 100k)States reportingPopulation-weighted rate (2000-2024; 1999 has no state population)
199917.051—
200019.75118.3
200120.55119.5
200222.35121.2
200322.65122.4
200423.35123.0
200524.65124.4
200625.25124.2
200725.15124.3
200827.45126.3
200925.55124.7
201026.85125.6
201126.25125.3
201225.15124.3
201323.65124.0
201426.65126.1
201530.05130.3
201631.75131.2
201733.65132.0
201832.75131.5
201932.35130.8
202033.25133.5
202132.75132.0
202231.45129.9
202330.45128.7
202428.75127.9

Which states rose, which fell, and by how much

All 51 states and DC, not just the extremes — Mississippi's rate roughly tripled over 26 years and Maine's is the only one that fell, but the honest picture is the whole ranked list, not the two ends of it.

Every state, 1999 → 2024

Ranked by percent change in the age-adjusted rate. Median change +59%, middle half of states between +34% and +94%. 50 of 51 states rose; only Maine fell.

Change in each state's age-adjusted Alzheimer's death rate, 1999 to 2024Mississippi +253%; South Dakota +158%; Arkansas +147%; Missouri +119%; Oklahoma +118%; Michigan +118%; Utah +117%; Louisiana +112%; California +108%; Alabama +107%; Hawaii +106%; Georgia +96%; Tennessee +94%; North Dakota +94%; Texas +88%; Alaska +87%; Nevada +81%; Idaho +76%; Ohio +75%; Connecticut +73%; Nebraska +69%; Rhode Island +69%; South Carolina +65%; New York +61%; Vermont +61%; Iowa +59%; New Jersey +53%; Minnesota +53%; Virginia +52%; North Carolina +49%; West Virginia +47%; Delaware +47%; Kentucky +47%; Wisconsin +44%; Indiana +43%; Illinois +39%; Pennsylvania +38%; Oregon +35%; New Mexico +34%; Kansas +33%; Washington +22%; Colorado +20%; Arizona +20%; Massachusetts +16%; District of Columbia +15%; Wyoming +10%; Montana +10%; New Hampshire +10%; Florida +9%; Maryland +4%; Maine -16%.no changeMississippi+253%South Dakota+158%Arkansas+147%Missouri+119%Oklahoma+118%Michigan+118%Utah+117%Louisiana+112%California+108%Alabama+107%Hawaii+106%Georgia+96%Tennessee+94%North Dakota+94%Texas+88%Alaska+87%Nevada+81%Idaho+76%Ohio+75%Connecticut+73%Nebraska+69%Rhode Island+69%South Carolina+65%New York+61%Vermont+61%Iowa+59%New Jersey+53%Minnesota+53%Virginia+52%North Carolina+49%West Virginia+47%Delaware+47%Kentucky+47%Wisconsin+44%Indiana+43%Illinois+39%Pennsylvania+38%Oregon+35%New Mexico+34%Kansas+33%Washington+22%Colorado+20%Arizona+20%Massachusetts+16%District of Columbia+15%Wyoming+10%Montana+10%New Hampshire+10%Florida+9%Maryland+4%Maine-16%rose (50 states)fell (1 state)

This chart is wider than your screen — drag it sideways, or open the table underneath.

n = 51 states + DC, each with both a 1999 and a 2024 value. Mississippi rose the most (+253%, 13.3→46.9 per 100k); Maine is the only state with a lower rate in 2024 than 1999 (-16%). Note this is change in the median of individual state changes (+59%), which is not the same number as the change in the national median rate itself — median is not linear, so those two ways of aggregating give different answers on purpose.
Every state's numbers, as a table
State1999 rate2024 rateChange (per 100k)Change (%)
Mississippi13.346.9+33.6+253%
South Dakota16.342.0+25.7+158%
Arkansas14.836.5+21.7+147%
Missouri15.032.8+17.8+119%
Oklahoma15.433.6+18.2+118%
Michigan15.433.5+18.1+118%
Utah17.337.6+20.3+117%
Louisiana17.938.0+20.1+112%
California16.634.5+17.9+108%
Alabama17.836.9+19.1+107%
Hawaii9.419.4+10+106%
Georgia18.836.8+18+96%
Tennessee17.934.8+16.9+94%
North Dakota18.135.1+17+94%
Texas18.534.7+16.2+88%
Alaska11.922.3+10.4+87%
Nevada13.624.6+11+81%
Idaho21.437.6+16.2+76%
Ohio18.231.9+13.7+75%
Connecticut11.419.7+8.3+73%
Nebraska16.327.6+11.3+69%
Rhode Island17.028.7+11.7+69%
South Carolina20.533.9+13.4+65%
New York7.011.3+4.3+61%
Vermont20.533.0+12.5+61%
Iowa18.229.0+10.8+59%
New Jersey12.018.4+6.4+53%
Minnesota21.132.2+11.1+53%
Virginia15.924.2+8.3+52%
North Carolina20.830.9+10.1+49%
West Virginia15.022.1+7.1+47%
Delaware15.022.0+7+47%
Kentucky19.328.3+9+47%
Wisconsin19.928.7+8.8+44%
Indiana18.927.1+8.2+43%
Illinois15.922.1+6.2+39%
Pennsylvania14.419.9+5.5+38%
Oregon24.132.5+8.4+35%
New Mexico16.422.0+5.6+34%
Kansas16.622.1+5.5+33%
Washington29.836.3+6.5+22%
Colorado24.529.4+4.9+20%
Arizona20.824.9+4.1+20%
Massachusetts16.519.2+2.7+16%
District of Columbia9.510.9+1.4+15%
Wyoming23.926.3+2.4+10%
Montana21.423.5+2.1+10%
New Hampshire23.225.4+2.2+10%
Florida14.315.6+1.3+9%
Maryland15.416.0+0.6+4%
Maine29.625.0-4.6-16%

What dementia deaths get coded as

Alzheimer's disease (G30) is one of four ICD-10 categories a dementia death can be coded under, alongside vascular dementia (F01), unspecified dementia (F03), and other degenerative disease (G31). 2017 is the only year in this file where every state reports that split — 1999-2016 and 2018-2024 are blank for these four columns, and nothing below is estimated to fill that gap.

What dementia deaths are coded as — 2017 only

2017 is the only year in this file where all 51 states report the G30/F01/F03/G31 breakdown. This is a single snapshot, not a trend — every other year in 1999-2024 is blank for these columns and is not estimated here.

Dementia deaths by underlying-cause subtype, 2017Alzheimer disease (G30) 46.4%; Vascular dementia (F01) 6.2%; Unspecified dementia (F03) 38.9%; Other degenerative (G31) 8.5%.46.4%38.9%8.5%Alzheimer disease (G30)Vascular dementia (F01)Unspecified dementia (F03)Other degenerative (G31)n = 51 states, 261,914 dementia deaths, 2017 only
Source: CDC WONDER Underlying Cause of Death, 2017, 51 states, 261,914 total dementia deaths. Shares are death-count-weighted across states, not a simple average of each state's percentage. Dementia cause-of-death coding shifted markedly toward G30 (Alzheimer disease) after roughly 2010, per this dataset's own documentation — a rising G30 share in any year this file doesn't cover could reflect that coding shift rather than a change in who has which disease.
Every number in this chart, as a table
SubtypeShare of dementia deaths, 2017
Alzheimer disease (G30)46.4%
Vascular dementia (F01)6.2%
Unspecified dementia (F03)38.9%
Other degenerative (G31)8.5%

Unpaid caregiving over time

An under-covered part of this story: how much unpaid care Alzheimer's and dementia caregiving actually demands, self-reported by caregivers themselves, and whether that differs for people caring for someone at midlife versus later life. Measured in the sense that it's a direct survey response, not a model — but it comes from an optional state survey module that not every state runs every year, so the states behind each point change from year to year.

Heavy caregiving (20+ hours/week)

Self-reported, BRFSS Caregiving module. States rotate this optional module, so the number of states behind each point varies year to year (shown in orange where fewer than 15 states reported) — this is not a fixed 51-state panel.

Heavy caregiving (20+ hours/week), by age band, 2015-20222015: 50-64 27.3%, 65+ 31.8%, n=24 states; 2016: 50-64 28.3%, 65+ 35.0%, n=20 states; 2017: 50-64 28.8%, 65+ 30.7%, n=12 states; 2018: 50-64 30.8%, 65+ 36.7%, n=5 states; 2019: 50-64 31.5%, 65+ 31.3%, n=10 states; 2020: 50-64 32.2%, 65+ 34.6%, n=8 states; 2021: 50-64 32.7%, 65+ 34.2%, n=47 states; 2022: 50-64 29.4%, 65+ 38.8%, n=14 states.0%10%20%30%40%20152016201720182019202020212022n=12n=5n=10n=8n=14ages 50-64ages 65+

This chart is wider than your screen — drag it sideways, or open the table underneath.

Source: CDC Alzheimer's Disease and Healthy Aging Data (BRFSS). Each point is the unweighted mean across whichever states asked this question that year. Because the set of reporting states changes every year, a move from one year to the next can reflect which states answered as much as any real change in caregiving — read the 2018 point (5 states) with particular caution.
Every year in this chart, as a table
YearAges 50-64Ages 65+States reporting
201527.3%31.8%24
201628.3%35.0%20
201728.8%30.7%12
201830.8%36.7%5
201931.5%31.3%10
202032.2%34.6%8
202132.7%34.2%47
202229.4%38.8%14

A related measure, the share of caregivers specifically reporting they care for someone with dementia (rather than any long-term condition), nearly doubled in this data between 2018 and 2019 — from about 13% to about 30%. That jump is larger than any other year-over-year move in the whole series, and it lands on a year where the number of reporting states also changed (5 states in 2018, 10 in 2019). A shift that size across a change in which states are even answering is more likely a composition effect than a real doubling of dementia caregiving in twelve months, so it is noted here rather than charted as a trend line.

For context: it's not just the US

This is the clearest version of the certification-practice story, so it's included even though it isn't US data. Read the warning printed directly on the chart before the numbers next to it.

For context: Western Europe's much larger rise

Not a US number — shown because the same certification-practice explanation applies here even more strongly than to the US state spread above.

Modelled estimate
Age-standardized dementia mortality, five Western European countries, 2000-2021United Kingdom: 20.36 in 2000 to 131.81 in 2021, ×6.5; Germany: 7.79 in 2000 to 79.01 in 2021, ×10.1; Netherlands: 36.33 in 2000 to 89.21 in 2021, ×2.5; Spain: 42.29 in 2000 to 74.54 in 2021, ×1.8; France: 37.96 in 2000 to 65.68 in 2021, ×1.7. Labelled on the chart: this rise is very unlikely to be biological given how much it varies between neighbouring countries; certification-practice change is the leading explanation in the mortality-statistics literature.02040608010012014020002005201020152020deaths per 100,000, age-standardized (modelled, IHME GBD)Netherlands ×2.5Germany ×10.1Spain ×1.8France ×1.7This is a rise in age-standardized mortality, so ageing populations are alreadyruled out as the explanation. A 1.7x-10x range between neighbouring countriesover the same 21 years is not plausible as a true difference in disease risk —researchers attribute most of it to countries changing how they certify andcode dementia deaths, not a real multi-fold rise in who develops it.

This chart is wider than your screen — drag it sideways, or open the table underneath.

Source: IHME Global Burden of Disease via Our World in Data. The death rate per 100,000 here is age-standardized by the publisher, and it is a model estimate, not a counted total. Five countries shown because they have complete 2000-2021 coverage in this file; not selected for the size of their rise. The multiplier ranges from ×1.7 (Spain) to ×10.1 (Germany) across countries with broadly similar demographics, which is the tell: a real biological rise would not vary this much between neighbours over the same 21 years.
Every number in this chart, as a table
Country2000 rate2021 rateMultiple
United Kingdom20.36131.81×6.5
Germany7.7979.01×10.1
Netherlands36.3389.21×2.5
Spain42.2974.54×1.8
France37.9665.68×1.7

Where these numbers come from

Records shown
of the 4472 in the data behind this page
Distinct sources
5
Years covered
1999–2024
Basis, all 4472
1326 measured · 3146 modelled · 0 not stated
Data collected
2026-09-27
Source files
panel_state_timeseries.csv (96 KB, 1326 rows), panel_country_timeseries.csv (164 KB, 3146 rows)

Most-cited sources on this page

What could make these numbers misleading

The US state series (1999-2024) is directly counted vital-records mortality data, not modelled, but 1999 is the first year of ICD-10 coding and its jump to 2000 is the largest year-over-year move in the series. State population, needed to weight state rates into a true national average, exists in this file from 2000 onward -- 1999 predates the Census Bureau's intercensal population series and has no published estimate, so it stays blank -- so the 1999-2024 national line is a median of states, not a weighted mean, while the population-weighted line runs 2000-2024. The dementia-subtype breakdown (G30/F01/F03/G31) is only complete for all 51 states in 2017; every other year is blank and is not estimated. Caregiving figures come from an optional state survey module that different states run in different years, so the reporting panel size changes year to year (as low as 5 states). The Western European country series is a modelled Global Burden of Disease estimate, not a counted death total, and its own multi-fold rise varies too much between neighbouring countries to be read as a real biological trend.

Every figure on this page is read straight out of a file in our data folder; nothing is typed in by hand. You can download the raw data and check any row yourself. This is a summary of published research, not medical advice.