The prevalence and characteristics of dizziness in an elderly community

The prevalence and characteristics of dizziness in an elderly community

Nicola R. Colledge


A postal questionnaire was sent to 1000 subjects aged over 65 years randomly selected from the age/sex register of five group practices, 90% of subjects returning adequate information. Thirty per cent of responders reported dizziness; 27% of these had symptoms more than once per month and 37% had symptoms which lasted longer than 1 minute. Dizziness was most commonly provoked by postural change and head and neck movement.

The prevalence of dizziness increased with age and was higher in women but these differences were not statistically significant. The prevalence of symptoms occurring more than once per month was significantly greater with increasing age (p=0.0003). Dizziness was significantly associated with angina and previous myocardial infarction (p<0.001) and antihypertensive therapy (p<0.05) but not with current smoking, diabetes mellitus or previous stroke.


Dizziness is one of the commonest problems afflicting elderly people. It was the most frequent symptom in people aged over 75 years who visited their family physician in the United States in 1985 (1). It is associated with functional disability, and 10-20% of sufferers fall because of their symptoms (2). Rotatory dizziness has been associated with an increased risk of subsequent stroke (3). The aims of the present study were to measure the prevalence of dizziness in people aged 65 years and over living at home in South West Edinburgh; to define the characteristics of reported dizziness, and to investigate the relationship between dizziness and cardiovascular risk factors.


A sample of 1000 people aged over 65 years was selected from the age/sex registers of five group practices in South West Edinburgh by stratified random sampling. Four strata were used, defined by two age groups (65-74 and 75 + years) and sex. The age-sex distribution of the sample was matched to that of the total over-65-year-old population of the practices. Those people known to suffer from dementia or who lived in institutional care were excluded from the study. Each subject was sent a previously piloted questionnaire with a covering letter from his/her general practitioner and a stamped addressed envelope for return to the researchers. The questionnaire comprised ten questions. The first two covered the presence of dizziness and cardiovascular risk factors. Specific enquiries were made regarding angina or previous myocardial infarction, current antihypertensive therapy, previous stroke, smoking habit and the presence of diabetes mellitus. If a subject reported dizziness he/she was asked to complete a further eight questions regarding its characteristics. A number of possible responses were given, and the subject was asked to tick the one which best described his/her symptoms.

Those that did not respond within 1 month were sent a reminder letter and, if there was still no response, attempts were made to telephone or visit the subject.

The associations between prevalence of dizziness, age and sex were examined using logistic regression with age used as a continuous variable. The frequency of cardiovascular risk factors reported by dizzy and non-dizzy responders was compared using [x.sup.2] tests.


Eight hundred and ninety subjects returned completed questionnaires of which 14 were inadequate. Further information was obtained by telephone from 17 non-responders; 14 individuals could not be contacted and 79 questionnaires were returned as the addressee was either not known at that address or had died. Adequate information was therefore obtained from 90% of the total sample. The response rate did not vary with age, sex or particular group practice.

Thirty per cent (267/893) of subjects reported dizziness (95% confidence interval 26.5-33.0%). Fifty-one per cent were aged 65-75 years, and 49% were over 75 years. Eighty per cent reported that they had had symptoms for more than 6 months and only 5% reported that they had had them for less than 1 month. Table I shows the proportion of men and women in each age group who complained of dizziness. As the sample contained more elderly women than men, the odds ratio quoted for age in Table II has been adjusted for sex, and the odds ratio quoted for sex has been adjusted for age. There is thus a 10% increase in the odds of dizziness for every 5 years of increasing age, and 30% greater odds of dizziness in women than in men. The confidence intervals are wide and these differences are not statistically significant. However, in those with the most severe symptoms (present for 6 months or more, with dizziness occurring at least monthly), there was a 29% increase in the odds of dizziness with every 5 years of increasing age, and this was highly significant.

Table I. Dizziness and demographic features

Men Women

Age No.(%) No.(%)

group Total with Total with

(years) no. dizziness no. dizziness

65-69 110 24(22) 161 45(28)

70-74 90 25(28) 114 41(36)

75-79 76 21(28) 138 42(30)

>80 63 19(30) 141 50(35)

Total 339 89(26) 554 178(32)

Table III shows the characteristics of dizzy subjects’ symptoms. Symptoms were most commonly described as a sensation of movement or vertigo. Six per cent of subjects reported constant dizziness, but the vast majority had episodic symptoms. Dizzy spells lasted only a few seconds in the majority, with few lasting more than several minutes. The most common precipitant of dizziness was postural change, but head and neck movements were also frequently reported. Ten per cent of subjects had fallen while dizzy and 5% had lost consciousness. Other reported symptoms included tinnitus, breathlessness, headache and nausea. Dizzy responders reported significantly more ischaemic heart disease and current anti-hypertensive therapy, but there were no significant differences in reported smoking diabetes mellitus or previous stroke (Table IV).

Table III. Characteristics of dizzy subjects’ symptoms

No.(%) of dizzy


Description of dizziness

Lightheadedness 55(21)

Unsteadiness 63(24)

Lightheaded and unsteady 47(18)

Vertigo 85(32)

No description 17(6)

Total duration of symptoms

<1 month 14(5)

1-6 months 37(14)

6 months-2 years 93(36)

2-10 years 72(28)

>10 years 40(16)

No information 11(4)

Frequency of dizziness

Constant 16(6)

Daily 33(12)

Weekly 24(9)

Monthly 20(8)

Less than monthly 57(21)

Once or twice/year 85(32)

No information 32(12)

Duration of dizzy spells

A few seconds 126(47)

Up to a minute 42(16)

Several minutes 59(22)

Up to an hour 20(8)

More than an hour 15(6)

No response 5(2)

Provoking factors

Bending over 112(42)

Rising from lying to standing 113(42)

Turning head 78(29)

Looking up 78(29)

Walking 39(15)

Associated symptoms

Chest pain 10(4)

Breathlessness 52(20)

Nausea 44(17)

Headache 48(18)

Tinnitus 58(22)

Loss of consciousness 13(5)

Fall 27(10)

Table IV. Dizziness and cardiovascular risk factors

No.(%) of responders

Dizzy Non-dizzy p value

Current smoker 35(13) 105(17) 0.15

History of angina or MI 75(30) 100(16) <0.001

History of stroke 18(8) 30(5) 0.22

Anti-hypertensive treatment 72(29) 135(22) 0.05

Diabetes mellitus 16(6) 33(5) 0.77


This survey has established that there is a 30% prevalence rate of dizziness in individuals aged over 65 years living in South West Edinburgh. This may be an underestimate as those suffering from dementia and those living in institutional care were excluded. Recent population surveys have identified similar rates (1)(2)(3). Of 758 representative 75-year-olds in Goteborg interviewed in 1987, 40% of women and 30% of men complained of postural disturbance (2), while an interview survey of 1622 over-60-year-olds living in the community in North Carolina found a 33.5% prevalence of dizziness (1). British studies dating from 1948 (4) and 1953 (5) found much higher prevalence rates, but a more recent study from Newcastle identified similar rates of 33.5% and 40% in men and women, respectively (3). The studies from Goteborg and Newcastle are particularly comparable with the present one in that neither used severity criteria in defining dizziness (2)(3).

In keeping with the American and Newcastle studies (1, 3) there was only a marginal increase in the prevalence of dizziness with age, but it is of note that there was a strong association with age in those with the most frequent symptoms. We found no significant association between the prevalence of dizziness and sex after controlling for age. The American study found a significant association on bivariate analysis, but this was lost when multiple variables were included. The Newcastle study found only non-rotatory dizziness to be more common in women but no multivariate analysis was performed. The Swedish study did not include any comparable analyses (2).

There is a multitude of potential causes for dizziness in this age group and previous reports from specialized clinics have suggested that benign positional vertigo (6)(7)(8), cerebrovascular disorders (7)(8)(9), multiple neurosensory impairments (8)(10) and psychological factors (11) are the most frequent. Drugs are also frequently cited as a cause of dizziness (12)(13)(14), but a previous study found that although cardiovascular medications were associated with dizziness, apparent associations with other drugs were abolished by multivariate analysis (1). In the interests of simplicity and hence an adequate response, our questionnaire did not include a question regarding specific drug therapy. Belal and Glorig suggested that no specific cause for dizziness could be found in up to 79% of elderly subjects. They coined the term ‘presbyastasis’ which they defined as dysequilibrium of ageing (15). However, many subjects in their study did not undergo detailed investigation, and doubts must remain regarding the validity of such a label.

Around one-half of the dizzy subjects in this study had attacks precipitated by changes in posture, suggesting impaired cardiovascular responses (8)(14), and a third had symptoms associated with head or neck movement, suggesting cervical spine (12)(16) or vestibular abnormalities (12)(14). Further investigation would be required to confirm this but it is possible that a different range of causes of dizziness may be identified in a community-derived patient sample to a sample derived from secondary or tertiary referrals to a specialized clinic.

Most of our subjects had short-lived infrequent symptoms and further assessment of these would probably not be fruitful. However, in one-third, dizzy spells were more frequent and troublesome and this group might benefit from further investigation and specific treatment. Such patients could be identified by general practitioners in their annual assessment of people aged over 75 years.

Despite the common belief that dizziness is a significant cause of falling (2)(10)(12), only 10% of dizzy subjects had fallen as a result of their symptoms. This may be because dizziness alerts the subject to sit down or hold on to a support (14). Although it is difficult to evaluate reporting of risk factors in a cross-sectional survey, the study supports the previous finding that there is an association between cardiovascular risk factors and dizziness (1)(3). The association between anti-hypertensive therapy and dizziness is open to interpretation as symptoms could be drug-related rather than due to hypertension itself. None the less, vascular disease seems important in the pathogenesis of dizziness in old age, and any investigation should include a full assessment of cardiovascular risk status.

Dizziness is highly prevalent amongst people aged over 65 years, with approximately 10% having symptoms of sufficient frequency and severity to justify further investigation. Although a symptom appears clinically important, this does not necessarily mean that further investigation will be of value and this is being explored in an ongoing study. Further research is also required to establish whether there is a different pattern of causes of dizziness in a community-derived sample of patients compared with that found in patients referred to specialized clinics, and to allow the development of improved clinical diagnostic criteria and a cost-effective investigative algorithm.


This study was funded by the Health Service Research Committee of the Chief Scientist’s Office (Scotland) (Grant K/OPR/15/3/1/F8).

We wish to thank the general practitioners who allowed us access to their patients.


1. Sloane P, Blazer D, George LK. Dizziness in a community elderly population. J Am Geriatr Soc 1989;37:101-8. 2. Sixt E, Landahl S. Postural disturbances in a 75-year-old population: I. Prevalence and functional consequences. Age Ageing 1987;16:393-8. 3. Grimley Evans J. Transient neurological dysfunction and risk of stroke in an elderly English population: the different significance of vertigo and non-rotatory dizziness. Age Ageing 1990;19:43-9. 4. Sheldon JH. The social medicine of old age. London: Oxford University Press, 1948. 5. Droller H, Pemberton J. Vertigo in a random sample of elderly people living in their own homes. J Laryngol Otol 1953;67:689-95. 6. Bloom J, Katsarkas A. Paroxysmal positional vertigo in the elderly. J Otolaryngol 1989;18:96-8. 7. McLure J. Vertigo and imbalance in the elderly. J Otolaryngol 1986;15:248-51. 8. Sloane PD, Baloh RW. Persistent dizziness in geriatric patients. J Am Geriatr Soc 1989;37:1031-8. 9. Baloh RW. Stroke and vertigo. Cerebrovasc Dis 1992; 2:3-10. 10. Drachman DA, Hart CW. An approach to the dizzy patient. Neurology 1972;22:323-34. 11. Brashear RE. Hyperventilation syndrome: managing elderly patients. Geriatrics 1984;39(7):114-25. 12. Luxon LM. Disturbances of balance in the elderly. Br J Hosp Med 1991;45:22-6. 13. Venna N. Dizziness, falling and fainting: differential diagnosis in the elderly. Geriatrics 1986;41(6): 30-42. 14. Baloh RW. Dizziness in older people. J Am Geriatr Soc 1992;40:713-21. 15. Belal A, Glorig A. Dysequilibrium of ageing (prebyastasis). J Laryngol Otol 1986;100:1037-41. 16. de Jong JMBV, Bles W. Cervical dizziness and ataxia. In: Bles W, Brandt T, eds. Disorders of posture and gait. Amsterdam: Elsevier, 1986;185-206.

Authors’ addresses

N. R. Colledge, W. J. MacLennan Geriatric Medicine Unit, University of Edinburgh, City Hospital, Greenbank Drive, Edinburgh EH10 5SB

J. A. Wilson Department of Otolaryngology, Head and Neck Surgery, Royal Infirmary, Glasgow

C. C. A. Macintyre Medical Statistics Unit, University of Edinburgh, Medical School, Teviot Place, Edinburgh

Received in revised form 25 May 1993

COPYRIGHT 1994 Oxford University Press

COPYRIGHT 2004 Gale Group