Home   •   Alzheimer's Disease Project

US Death Rates due to Alzheimer’s Disease:   Some Regional and Geologic Correlates

Wayne P. London MD, Alexander P. Gutterman MA, Susanna Hope Glendenning, Paul Eagle, Deborah M. O'Brien from the Brattleboro Research Institute, 139 Main St., Suite 706A, Brattleboro, VT 05301

Key words:  Alzheimer’s disease, death rates, regional, geography, geology, seismic.

(version of 15 Jun 2011)



State-by-state death rates due to Alzheimer’s disease (AD) in the US show regional patterns that are not readily explainable by social, cultural or racial factors but appear to correlate with geologic features.

The relevant geological factors that correlate with high death rates due to AD are (1) zones of seismic activity either from intraplate dynamics or active plate margins; higher death rates appear to correlate with subduction as opposed to strike-slip plate margins. (2) geographic zones with abundant felsic rocks containing minerals such as quartz and feldspar, some of which have conductive and piezoelectric properties.

The range in the state-by-state and the regional death rates is more than 3-fold, which is higher than most known risk factors for AD.

These preliminary findings are discussed in the context of the hypothesis of chromosomal missegregation in AD, and they have implications for understanding and preventing AD.



Alzheimer’s disease (AD) is a compelling world wide health problem with enormous social and economic consequences. The best case solution is prevention, which requires an understanding of both biological mechanisms and relevant environmental factors that produce the disease.

Most cases of AD arise de novo or sporadically, which implies an environmental factor or factors that trigger the disease process.

This brief communication suggests that geologic factors - some types of seismic activity and the presence of felsic minerals - might be environmental influences that affect the development of AD.

The Analysis by Geographic Region

We examined the number of deaths due to AD per 100.000 population state-by-state for 2006 (Figure 1, ref 1). These age-adjested death rates are very similar to those for 2005 (ref 2, available at http://www.alz.org/alzheimers_disease_facts_figures).

Figure 1   (available at: http://www.statehealthfacts.org/comparemaptable.jsp?ind=63&cat=2&sort=a&gsa=2 )

Independent of any analysis, the range in the state death rates was more than 4 fold – from New York (9.1) to Washington (38.3). This means that, living, for example, in Washington (38.3) or Tennessee (34.6) - the two states with the highest death rates - carries at least a three-fold increased death rate due to AD as compared with living in New York (9.1) or Hawaii (12.2) - the two states with the lowest death rates. This roughly three-fold risk factor is greater than most known risk factors in AD (Ref. 3).

Because we observed the pattern that many states that were contiguous had similar death rates, we grouped much of the state-by-state data into five regions that are shown in Figure 2 and in the following tables.

Figure 2.

Southern New England and Mid Atlantic (SNEMA) Region This region of nine contiguous states contains 9 of the 14 states having the lowest death rates – all well below the US median death rate of 24.1.

Rank in the 51 states
(lowest to highest rate)


Death Rate


New York






New Jersey






District of Columbia









Rhode Island






Central Region The death rates in this region of 10 contiguous states are all below or near the median US death rate of 24.1.

Rank in the 51 states
(lowest to highest rate)


Death Rate































Northern New England Region This small region of three contiguous states is included for eventual comparison with the adjoining SNEMA and Northwest regions and with the prevalence rates in adjoining Canada (not shown).

Rank in the 51 states
(lowest to highest rate)


Death Rate





New Hampshire


42. (9th highest)




Southeast Region The death rates in these eight contiguous states are substantially above the USA median of 24.1 and include some of the highest death rates.

Rank in the 51 states
(lowest to highest rate)


Death Rate








North Carolina





45. (7th highest)



46. (6th highest)



48. (4th highest)

South Carolina


50. (2nd highest)




Northwest Region The three contiguous states in this region have very high death rates.

Rank in the 51 states
(lowest to highest rate)


Death Rate

41. (11th highest)



42. ( 9th highest)



51. (highest)




The range and medians of the five regions and of the USA are shown in the following table.



Number of States



Southern New England/Mid Atlantic (SNEMA)


9.1 – 20.4

18.3 (lowest)



20.6 – 26.3

23.7 (intermediate)



9.1 - 38.3


Northern New England


26.2 – 29.0

26.5 (higher)



25.0 – 34.6

26.7 (higher)



28.1 – 38.3

29.0 (highest)

As discussed in the geology section, we were not able to define regions of contiguous states with similar death rates in the Rocky Mountain area.


Results of the Regional Analysis

The analysis showed that regions of contiguous states have similar death rates due to AD. This finding implies an environmental factor(s) that correlates or occurs regionally. It also means that living, for example, in the contiguous states of either the Northwest or Southeast regions carries a substantially increased death rate due to AD as compared to living in the contiguous states of the Southern New England/Mid Atlantic region.

But the state-by-state and regional data still showed many unexplained features:

The above considerations cast doubt that social, cultural or geographic factors are adequate in explaining the state-by-state or regional death rates due to AD. The analysis requires going beyond discussing similar death rates in contiguous states; for example, it should explain the low death rates in such noncontiguous and culturally distinct states as New York, Florida and Hawaii. These and related considerations prompted a geologic approach.

Geologic Correlates

The following table summarizes some relevant geology of the five regions.

Brief Summary of the Relevant Geology of the five Regions




Relevant Geology

Southern New
England-Mid Atlantic (SNEMA)

 9.1 – 20.4

18.3 (lowest)

Seismically inactive
Mafic minerals in some highly populated coastal areas


20.6 – 26.3

23.7 (intermediate)

Seismically inactive
No mountains
Limited felsic minerals


9.1 – 38.3



New England

26.2 – 29.0

26.5 (higher)

Seismically inactive, but a
gradient of felsic minerals that correlates with death rates: VT < NH < ME


25.0 – 34.6

26.7  (higher)

Three active intraplate seismic zones:

  • New Madrid
  • East Tennessee
  • South Carolina

Also felsic minerals.


19.6 – 38.3

29.0 (highest)

Very active seismic zone of
subduction (only one in US)
Abundant felsic minerals


Seismic Activity

The two regions with the highest AD death rates, the Northwest and the Southeast, are also the two most seismically active areas in the United States. The three regions with lower death rates, SNEMA, Central and Northern New England, are not seismically active.

Simply put, the map of death rates due to AD by region (Figure 2) basically superimposes (Figure 3) onto the map of seismic hazards in the US (.

Figure 3.  (in-depth data available at: http://earthquake.usgs.gov/earthquakes/states/seismicity/ )

In the Northwest region (Washington, Oregon and Idaho), the seismic activity is due to subduction of the Juan de Fuca plate under the North American plate. In addition, this region has abundant felsic minerals. In two adjoining states that have relatively low death rates – Montana (12th lowest death rate) and Wyoming (18th lowest rate) the seismic activity occurs only at the western edges of both states (Figure 3).

The Southeast region presents high death rates and very high seismic activity from three seismic zones –New Madrid, East Tennessee and South Carolina.

As shown on the seismic hazard map (Figure 3), Tennessee – the state with the 2nd highest death rate due to AD in the US - has two very active seismic zones - the New Madrid zone in West Tennessee and the East Tennessee Seismic Zone (ETSZ) in East Tennessee. Both zones are intraplate and are the most active and second most active areas for earthquake activity in the eastern U.S. respectively (Geology.com, Eastern Tennessee Earthquake Research Center)

Also as shown on the seismic hazard map (Figure 3), South Carolina - the state with the 4th highest death rate due to AD in the US - has high seismic activity, whereas the two adjoining states, North Carolina and Georgia have both lower death rates and lower seismic activity.

Alabama (7th highest in the US) is under the influence of three active seismic zones: the New Madrid, East Tennessee and South Carolina seismic zones (ref geology of AL)

Of all of the states in the eastern US, these states in the Southeast region are both the most seismically active and have the highest death rates due to AD.

Correlation with Type of Seismic Activity

The data suggest a correlation of the death rates and the type of seismic activity. As just described, the very high death rates due to AD correlated with both the subduction plate margin in the Northwest region and the intraplate activity in the Southeast region.

In contrast, in Alaska (rank 29th), the seismic activity is both strike-slip and subduction (e.g. volcanoes), and the death rate is just above the US median. In California (rank 25th), the seismic activity is strike-slip, and the death rate is the same as the US median and slightly below that of Alaska.

Hawaii has modest but steady seismic activity related to volcanism but not active plate margins; the state has the 2nd lowest death rate due to AD.
Thus, of five instances of seismic activity in the US, the high death rates due to AD appear to correlate with subduction at active plate margins and with intraplate dynamics.

Correlation with Felsic Minerals

In addition to seismic activity, high death rates due to AD also appear to correlate with the presence of felsic minerals (continental crust) and low death rates with mafic minerals (oceanic crust).

The SNEMA region, which has the lowest regional death rates, is seismically inactive and mafic and not felsic minerals are found in the highly populated coastal areas of southern New York, northern New Jersey, Connecticut and Rhode Island.

Hawaii (2nd lowest death rate) is seismically active but has mafic and no felsic minerals. We cannot distinguish if the very low death rate in Hawaii is because of the type of seismic activity or the lack of felsic minerals, or both.

Florida (6th lowest death rate) is seismically inactive and is a coral reef with abundant mafic limestone minerals and no felsic minerals.

Thus, parts of the SNEMA region, Hawaii and Florida, which are not contiguous, have some of the lowest death rates due to AD,  and they share a common geological factor of a predominance of mafic but not felsic minerals.

Furthermore, Florida (6th lowest death rate) is culturally similar to both the Southeast region (very high death rate) and to Arizona (5th highest death rate); but the differences in the death rates correlate, not with the cultural similarities, but with the geologic differences.

The Central Region has intermediate death rates near the US median, is seismically inactive and has limited felsic minerals and few mountains.

The Northern New England Region (VT, NH, ME), with its relatively high death rates, is also seismically inactive, but it has large amounts of felsic rock.

This region also has a gradient of felsic rocks that appears to correlate with the death rates due to AD. Vermont, which has the lowest death rate of the region has the least amount of felsic rocks that are actually limited to the northeast corner of the state, which is the least densely populated. New Hampshire - the Granite State - has both an intermediate death rate and amount of felsic minerals. Maine has both the highest death rate in the region - 9th highest in the US - and the most abundant felsic minerals.

The prevalence rates of AD by gender and age group in adjoining regions of Canada (Ref. 4) were consistent with the higher death rates in the Northern New England region. In that study of five regions in Canada, the prevalence of AD in the Atlantic Island and Quebec regions, which adjoin and have the same geology as the Northern New England region, were higher than in the Ontario region, which adjoins and has similar geology as the Central region in the US. (Prevalence rates of AD were also lower in the Prairies and British Columbia regions.)

The Rocky Mountain States
Our attempts to correlate death rates due to AD with regional and geologic factors were least successful in the Rocky Mountain states.  As shown in Figure 1, some states in this area with high death rates are adjacent to states with low rates, and we found no regions of contiguous states with similar death rates.

Also, our geologic analysis does not explain three states east of the Rockies that have relatively high death rates: North Dakota (3rd highest), adjoining South Dakota (8th highest), and Arizona (5th highest).  The failure of our approach might be due to the unusual orogeny of the Rocky Mountains, which is poorly understood, and most likely the result of an unusual subducting slab (Ref. 5). Regardless, the subtleties of both the geology and the orogeny of the Rocky Mountains are beyond the scope of this brief communication.


Discussion and Conclusion

The geologic factors and analysis in this brief communication are admittedly simplistic. A more sophisticated approach would undoubtedly consider more subtle geologic factors such as crustal thickening that can double at active plate margins (Ref. 6).

A more systematic study of death rates due to AD and geology also requires data of death rates within states that can be compared with more local geology, for example, by county or geographic area. A prime candidate for this more focused analysis is the state of Tennessee that has the 2nd highest death rate due to AD and is divided into three geographic and geologic regions – East (Blue Ridge Mountains and Tennessee Valley), Middle (the Cumberland Plateau) and West (Gulf Coastal Plain). County-by-county data in four states – Tennessee, South Carolina, Washington and New York – are consistent with the hypothesis of this brief communication [see www.brattleborohealing.org/ad].

Our findings suggest an effect of geology on the biology of AD.

Some seismic activity produces intense ULF (ultra low frequency) and ELF (extremely low frequency) electromagnetic waves by a piezoelectric effect (Ref. 7). ELF radiation is also a known risk factor in AD (Refs. 8,9) and is associated with increased production of amyloid beta and decreased production of melatonin (Ref. 9). Thus, suggesting that some types of seismic activity and felsic minerals that have conductive and piezoelectric properties would be risk factors for AD is merely identifying a natural source of intense ULF or ELF radiation.

As to an effect on biology, perhaps the most plausible biological mechanism of AD is improper chromosomal segregation as proposed by Potter in, for example, the paper "Down's Syndrome and Alzheimer's Disease: Two Sides of the Same Coin" (Ref. 10).  According to this well documented hypothesis, a trisomy of chromosome 21 during mitosis in the adult nervous system is a sufficient condition for the development of AD.

An intriguing possibility is that geologic factors might affect the movement of chromosome 21 during mitosis in the adult nervous system. Such a biological effect might be mediated by the piezoelectric effect of rocks such as granite that contain the felsic mineral quartz. These rocks would transduce seismic pressure into an electromagnetic field that could affect cell division. This putative effect of geology on the biology of AD might be most operative in the absence of sex hormones such as estrogen, which have a protective effect in AD (Ref. 11).

Regardless of mechanism, geologic factors might be one environmental factor that account, in part, for the sporadic nature of AD. However, geologic factors, like the frequency of the APOE4 allele, both of which are virtually constant over a several decade period, cannot explain the dramatic increase in the incidence of AD during the last 50 years.

In conclusion, perhaps the main point of this brief communication is that physical characteristics of place - geography and geology - appear relevant to understanding and preventing AD, and they require further systematic study.

We thank Jared Smith for invaluable geologic consultation.


1. The Centers for Disease Control and Prevention (CDC), National Center for Health Statistics, Division of Vital Statistics, National Vital Statistics Report Volume 57, Number 14, April 2009, Table 29. Available at http://www.cdc.gov/nchs/data/nvsr/nvsr57/nvsr57_14.pdf.

2. Kung HC, Hoyert DK, Xu JQ, Murphy SL: "Deaths: Final Data for 2005" National vital statistics reports. Vol 56, no 10, Hyattsville, MD, National Center for Health Statistics; 2008.

3. Caselli RJ, et al. The degenerative dementias: Alzheimer's disease. In: Goetz CG, et al. Textbook of Clinical Neurology. 3rd ed. Philadelphia, Pa.: Saunders Elsevier; 2007.

4. Canadian study of health and aging: study methods and prevalence of dementia. CMAJ. 1994 March 15; 150(6): 899–913.

5. United States Geological Survey document "Geologic Provinces of the United States: Rocky Mountains" 2004.

6. Tarbuck EJ, Lutgens FK, Tasa D: Earth: An Introduction to Physical Geology, 9th edition Prentice Hall, Inc, Upper Saddle River NJ, 2008

7. Hayakawa M, Hattori K, Olita K. Monitoring of ULF (ultra-low-frequency) Geomagnetic Variations Associated with Earthquakes. Sensors. 2007; 7: 1108-1122.

8. Garcia AM, Hoyos SA. Occupational exposure to extremely low frequency electric and magnetic fields and Alzheimer's disease: a meta-analysis. Int J Epidemiol. 2008 Apr; 37(2):341-3.

9. Davanipour Z, Sobel E: Long-term exposure to magnetic fields and the risks of Alzheimer's disease and breast cancer: Further biological research. Pathophysiology 2009 Aug;16(2-3):149-56.

10. Potter H: Down's Syndrome and Alzheimer's Disease: Two Sides of the Same Coin. Future Neurology. 2008; 3(1):29-37.

11. Pike CJ, Carroll JC, Rosano ER, Barron AM: Protective actions of sex steroid hormones in Alzheimer's disease. Frontiers in Neuroendocrinology. 2009; 30 (2), 239-258.