Epidemiology of Dizziness

Timothy C. Hain, MD • Page last modified: March 7, 2021

Although there are many publications concerning the topic, relatively little is known about the precise incidence and prevalence of dizziness in the general population. It is generally accepted to be common, although not so common as diabetes or heart disease, and to increase with age.

To deal with it rationally, one must break dizziness down into categories:

There is also considerable ambiguity between conditions defined by symptoms rather than objective signs. Often a clinician can choose between several symptom inventories --- in fact, each specialty (ear, internal medicine, neurology, psychiatry) tends to have it's own "jack of all trades" symptom inventory or even a collection of these, where patients are allocated.

Even with a breakdown, coming to a reasonable conclusion is very difficult due to differences in populations being studied, differences in definition, and problems with study technique, and variability according to age.

A core problem to epidemiology is that it depends on the population you are drawing from as to what you conclude. Parker et al (2018), showed this in a metanalysis of 42 case series and noted that ENT clinics were dominated by BPPV, psychogenic and Meniere's, while ED populations were dominated by "Other", cardiac and neurological. Or in other words, if you are sampling apples from a barrel that has mostly oranges, you might not get as many apples as you expect.

Another core problem is that most of the diagnostic categories are defined mainly by symptom inventories, which overlap, and as well the fidelity of the data (e.g. from insurance coding) is suspicious to say the least.

On this page, we attempt a synthesis of a large amount of data. Our bottom line is --

If more data becomes available, these estimates may change of course, but this is likely a reasonable estimate. The main source of error would seem to us to be that there might be an overestimate of otologic dizziness, and consequently an underestimate of medical or undiagosed. These numbers are not at all appropriate for extremes of age -- young or old.

Overall -- the proportion of the entire population that is dizzy.

A different question is -- what is the one year prevalence of dizziness ? Here we define "dizziness" as the sum of vertigo, imbalance and faintness. Considerably more data is available on the elderly, where there have been some large studies. Roughly, there seems to be about 10% of elderly with vertigo, 20% with imbalance, and about 10% with faintness. Younger people have little to no imbalance, and moderately less vertigo -- with 6-8% of children having vertigo being a reasonable estimate.

Measure Percentage Reference
Vertigo health insurance claims in Taiwan 3.13% Lai, Y. T., T. C. Wang, et al. (2011).
Dizziness in children in UK 5.7% Humphriss, R. L. and A. J. Hall (2011).
Cross sectional study of dizziness in 2547 persons 21% Mendel, B., J. Bergenius, et al. (2010).
Survey of 1287 persons (Germany) 15.8% Wiltink, J., R. Tschan, et al. (2009)
Cross section of 17,638 adults in Sweden with dizziness or faintness 28.7% Tamber, A. L. and D. Bruusgaard (2009).
Cross section of 2751 persons in Australia 36.2% Gopinath, B., C. M. McMahon, et al. (2009).
1801 elderly in germany over age of 65 29% Gassmann, K. G. and R. Rupprecht (2009).
2925 elderly in UK over the age of 65, dizziness 11.1% Stevens, K. N., I. A. Lang, et al. (2008).
2925 elderly in UK over the age of 65, imbalance 21.5% Stevens, K. N., I. A. Lang, et al. (2008).
Cross section of 4869 adults in Germany 22.9% Neuhauser, H. K., A. Radtke, et al. (2008).
938 children in Finland with vertigo 8% Niemensivu, R., I. Pyykko, et al. (2006).
2011 persons over age of 70 in Sweden, dizziness or imbalance 36% Jonsson, R., E. Sixt, et al. (2004).
Young adults, metanalysis 1.8% Sloane, P. D., R. R. Coeytaux, et al. (2001).
Elderly, metanalysis 30% Sloane, P. D., R. R. Coeytaux, et al. (2001).
Persons 65 or older, 6158 studied 9.6% Aggarwal, N. T., D. A. Bennett, et al. (2000).
Cross sectional study of adults, dizziness defined as imbalance 35.4% Agrawal, Carey, Della Santina, Schubert and Minor (2009); Agrawal, Ward and Minor; 2013, and 1 other.

As a rough estimate, the figure from Wiltink of 15% prevalence of dizziness in middle aged adults seems reasonably likely to be correct.

Outliers -- it's not as bad as some suggest.

The outliers are largely on the bigger side. Three analyses of National Health Survey data (i.e. other people's data) were published by Agrawal, and co-authored by Dr. Carey, Della-Santina, Ward and Minor. These papers suggested a very high prevalence (about 35%) of the US population aged 40 years or older had "balance dysfunction". The authors stated that this number was an estimate of "vestibular dysfunction". This is an amazing leap of logic, as they are suggesting that all balance dysfunction is vestibular dysfunction. Perhaps this leap was made easier for the authors because they are all otolaryngologists. Other specialties -- such as medicine and neurology for example, know from their patients that most imbalance is not vestibular in origin. In fact, more reasonable studies suggest that only about 1/4 of dizziness (which includles imbalance) is vestibular.

The problem is that balance is not a vestibular measure, any more than weight loss is a cancer measure. Imbalance can be caused by inner ear disturbances, just as weight loss can be caused by cancer. However, one cannot say that the two are equivalent. (Agrawal et al, 2009; Agrawal et al, 2010; Agrawal et al, 2013).

A less precise method of measuring dizziness is to just look at how many patients present to medical clinics. Dizziness is the primary complaint in 2.5% all primary care visits = 8 million/year visits (Sloane, 1989).

Another rough estimate of the overall impact of dizziness can be obtained from survey studies. The NHIS (US national health interviews survey) provided data. About 300,000 persons were affected in the United states over 1986-1988 per year with dizziness and vertigo. About 26% of them were unable to work, accounting for an economic impact of about 75,000 persons per year unable to work due to dizziness and vertigo. If we assume that this causes $30,000 of economic damage each year per person, this means that dizziness/vertigo costs the nation roughly 2.25 billion $ each year. Given that there are about 300 million people in the US, this means that roughly 1 in 1000 people in the US was disabled due to dizziness each year. This is only 0.1%.

Proportions of patients in each category: There are substantial otologic, neurological, general medical, and psychiatric/undiagnosed causes of vertigo (see below).

The proportion of patients in each category is known to some extent, from clinics that have reported differential diagnosis.





Dizzy Clinic4


Otologic (%)




















Unknown or Unlocalized

42.3 (both)


5.5, 40%5



number studied





Table 1 summarizes the results of several studies of diagnoses obtained in dizzy patients according to the setting where the patients were seen, and shows that dizziness is multifactorial. No specialty, be it otology, neurology, internal medicine or psychiatry, has a clear preponderance of dizzy patients.

Brandt et al (2004) provided an table of their diagnostic experience in Germany in their "neurological dizziness unit", and reported that BPPV, and then "PPV", which can be variously interpreted as psychogenic vertigo or "I can't figure it out". These prevalence numbers are very much setting dependent, as well as practioner dependent, and mean little.

Much more about dizziness in the ED can be found here.

OTOLOGIC (AKA Vestibular) dizziness/vertigo

According to Neuhauser and others (2005), vestibular vertigo (i.e. otologic vertigo) is common in the general population, affecting more than 5% of adults in 1 year. The frequency and health care impact of vestibular symptoms at the population level have been underestimated. We agree with this assessment.


Measure Percentage Reference Comment
Cross sectional study of dizziness in 2547 persons 10.5% of population Mendel, B., J. Bergenius, et al. (2010). Seems reasonable.
Metaanalysis of 12 articles, peripheral cause 44% of dizzy patients Kroenke, K., R. M. Hoffman, et al. (2000). This is a proportion of all dizziness, not prevalence.
Meniere's disease
Meniere's disease 0.19% population Harris, J. P. and T. H. Alexander (2010). This is likely ballpark correct.
Meniere's disease in German insurance records of 70 million people. 0.2% prevalence Hulse et al, 2018 This could easily be an artifact of the methodology that likely overcounts.
BPPV in 4869 adult Germans 1.5% one year prev. von Brevern, M., A. Radtke, et al. (2007). Probably right.
BPPV in ENT clinic in Japan 23-41% of dizzy Uno, A., M. Nagai, et al. (2001). This is a proportion, not prevalence.
BPPV in "elderly" 9% Prevalence. Oghalai, J. S., S. Manolidis, et al. (2000). Seems a bit high.
BPPV in German insurance records of 70 million people. 0.46% prevalence Hulse et al, 2018 Amazingly low.
Unspecific dizziness
Dizziness in Scottish children 15% Russell, G. and I. Abu-Arafeh (1999). Could be anything
"Vestibular vertigo" in 2751 Australians 10% Gopinath, B., C. M. McMahon, et al. (2009).  
"Vestibular vertigo" in 4869 adult Germans 4.9% Neuhauser, H. K., A. Radtke, et al. (2008).  
Vestibular, not BPPV in ENT clinic in Japan 22% Uno, A., M. Nagai, et al. (2001). Just a subset.
Unspecified cause of vertigo in German insurance records of 70 million people 4.8% Hulse et al, 2018 This could easily be an artifact of the methodology. Probably wrong.

Otologic vertigo estimates vary widely. The two largest studies are those of Mendel et al (2010) and Neuhauser et al (2008), which suggested percentage of the population having either dizziness or "vestibular vertigo" varies between 4.9% and 10.5%. If we accept the estimate that 15% of the population has dizziness of any cause, this would suggest somewhere between 1/3 and 2/3 of all dizzy patients have inner ear problems.

A more recent study, of Hulse et al (2018) used ICD-10 codes from German insurance companies. While this was an immense study (70 milliion insurance claims), in the author's experience as a practicing physician for 30 years, ICD-10 codes are not to be relied upon for anything important -- they are basically guesses about diagnoses. There is a "confirmation bias" with ICD-10 codes -- if a condition becomes popularized, it becomes more prevalent. If there is a compensation system for some diagnosis -- perhaps a disabling one -- it may have an increase in selection. As another example, in the Framingham study, it was found that there are 10 times as many people who say they have Meniere's disease (in a survey), than actually meet the criteria. Because we are dubious about the validity of the ICD-10 method, we do not rely at all on these statistics. We are also dubious that the age dependence of these diagnoses rises so greatly with age. Certainly as people get older, they develop more hearing loss, dizziness and tinnitus. We suspect that they also collect more wrong diagnoses on insurance claims.

Many studies have documented that BPPV accounts for roughly 20% of all dizziness. Nedzelski, in a primarily ENG setting, et al reported BPPV to compose 17.1% of their patients (1986). This is a reasonable estimate for most "dizzy" practices.




Measure Percentage Reference
MAV in midlife women 5% Hsu, L. C., S. J. Wang, et al. (2011).
Migraine vertigo in 4869 Germans 0.89 % 1 year Neuhauser, H. K., A. Radtke, et al. (2006).
Metaanalysis of 12 articles, central cause 11% of dizzy patients Kroenke, K., R. M. Hoffman, et al. (2000).
Metanalysis of 12 articles, stroke 6% of dizzy Kroenke, K., R. M. Hoffman, et al. (2000).
Metanalysis of 12 articles, Brain tumor 1% of dizzy Kroenke, K., R. M. Hoffman, et al. (2000).
117 elderly veterans in Neurology clinic, brainstem 22% of dizziness Davis, L. E. (1994).

Neurological dizziness is loosely defined here as a group of conditions causing dizziness that are usually diagnosed and treated by neurologists. In addition to dizziness associated with dysfunction of the central nervous system, it also includes dizziness or ataxia caused by sensory neuropathy and visual dysfunction.

Kroenke et al (2000) suggested that 11% of dizzy patients had central vertigo. The "600 pound gorilla" of central dizziness is Migraine. Roughly 1% of the population has migraine associated vertigo. If we accept the 15% overall prevalence of dizziness above, 1/15 is 6.6%, and 6.6/11.1 is roughly 50%. Thus migraine should reasonably cause roughly half of all central dizziness. There are huge age and gender effects as migraine is mainly a disorder of middle aged women, and fades with age. Women have far more migraine than men.

The incidence of neurological dizziness in the emergency setting is very low. The incidence in the primary care setting is unknown, but probably also low. Only neurologists report a substantial proportion of dizzy patients with neurological dizziness, and even in this setting, "neurological" diagnoses account for only about 1/3 of their cases (Drachman and Hart, 1972; Macrae, 1960).

Vertigo in the neurology setting is dominated by vascular causes, combined with a larger number of rarer miscellaneous causes. A variety of miscellaneous causes of dizziness make up the remainder. Vertigo attributed to "miscellaneous", includes disorders of CSF drainage, multiple sclerosis, the Chiari malformation, cervical vertigo, and conditions accompanied by central positional nystagmus or tremor.



Measure Percentage Reference
Orthostatic dizziness in Germany 10.9 (12 month) Radtke, A., T. Lempert, et al. (2011).
Cross-sectional study of elderly in Netherlands, all dizzy 8.3% Maarsingh, O. R., J. Dros, et al. (2010B)
Cross-sectional study of elderly in Netherlands, cardiovascular 57% of dizzy Maarsingh, O. R., J. Dros, et al. (2010A)
Dizzy clinic in Japan 19% of dizzy Uno, A., M. Nagai, et al. (2001).
Cardiac Arrythmia 1.5% of dizzy Kroenke, K., R. M. Hoffman, et al. (2000).

The main cause of "medical" dizziness is low blood pressure, which is very common, but often called something other than dizzy such as "light headed", or "presyncope". Here the reported data varies hugely. The statistic of 10.9% of the population seems unduly high to us, considering our choice of 15% of the entire population having dizziness overall.

General medical diagnoses are assigned to about 33% of cases of dizziness seen in the emergency setting. These include disorders which alter blood pressure, blood sugar, or which are attributed to an infection or medication. In the primary care setting, medical causes account for a similar fraction of diagnoses. Hypertension, disturbances of blood glucose, and coronary atherosclerosis are the most common diagnoses (Sloane 1989). General medical diagnoses are very unusual in specialty settings, associated with only 2% of diagnoses in the otology setting and 5% in the neurotology setting.


Measure Percentage Reference
Cross-sectional study of elderly in Netherlands, unknown 39% of all dizzy Maarsingh, O. R., J. Dros, et al. (2010B)
Vertiginous patients in ENT clinic 26.1% Isaradisaikul, S., N. Navacharoen, et al. (2010).
Epidemiological study of 2169 patients with vertigo (Japan) 26.8% of all dizzy Yin, M., K. Ishikawa, et al. (2009).
Patients referred to neurotology clinic with normal examination 10.6% Odman, M. and R. Maire (2008).
"Phobic postural vertigo" in outpatient dizziness unit 22-26% of all dizzy Strupp, M., M. Glaser, et al. (2003).
Metaanalysis of 12 articles, psychiatric cause 16% of dizzy patients Kroenke, K., R. M. Hoffman, et al. (2000).
Metanalysis of 12 articles, unknown cause 13% of dizzy Kroenke, K., R. M. Hoffman, et al. (2000).

It is very clear that unlocalized dizziness is extremely common. We think it is best to "lump" dizziness attributed to psychiatric origin with undiagnosed dizziness. With this in mind, it appears that the median for the studies above is roughly 26%. We will use 25% as it is a more convenient number.

Unlocalized vertigo includes patients whose symptoms are attributed to a patients whose vertigo is attributed to an event without further definition (such as head trauma) and patients with vertigo and dizziness of unclear origin. Common variants of unlocalized variants include "hyperventilation syndrome", "post-traumatic vertigo," and "nonspecific" dizziness. There are also some recent acronyms - -"PPV", and "PPPD", attributed to psychogenic causes, but again are essentially undiagnosed patients who have particular combinations of symptoms.