The epidemiology of dizziness and vertigo

Abstract

This chapter gives an overview of the epidemiology of dizziness, vertigo, and imbalance, and of specific vestibular disorders. In the last decade, population-based epidemiologic studies have complemented previous publications from specialized settings and provided evidence for the high burden of dizziness and vertigo in the community. Dizziness (including vertigo) affects about 15% to over 20% of adults yearly in large population-based studies. Vestibular vertigo accounts for about a quarter of dizziness complaints and has a 12-month prevalence of 5% and an annual incidence of 1.4%. Its prevalence rises with age and is about two to three times higher in women than in men. Imbalance has been increasingly studied as a highly prevalent complaint particularly affecting healthy aging. Studies have documented the high prevalence of benign paroxysmal positional vertigo (BPPV) and vestibular migraine (VM), as well as of comorbid anxiety at the population level. BPPV and VM are largely underdiagnosed, while Menière's disease, which is about 10 times less frequent than BPPV, appears to be overdiagnosed. Risk factor research is only at its beginning, but has provided some interesting observations, such as the consistent association of vertigo and migraine, which has greatly contributed to the recognition of VM as a distinct vestibular syndrome.

Introduction

Well beyond counting cases, epidemiologic data on prevalence, incidence, risk factors, disease burden, and outcomes can help us understand the nature and impact of dizziness and vertigo and can be a valuable resource for evidence-based patient care. In clinical decision making, epidemiologic studies that systematically analyze patterns of disease in defined populations provide clinicians with probabilistic expectations on disease frequency (Lurie and Sox, 1999), as well as on outcome and prognosis. This chapter gives an overview of the epidemiology of dizziness, vertigo, and imbalance, and of specific vestibular disorders, thus updating previous reviews of the epidemiology of dizziness and vertigo (Neuhauser, 2007, Neuhauser, 2013).

Compared to cardiovascular or cancer epidemiology, the epidemiology of vertigo and imbalance is still a small and emerging field. However, its potential impact on patient care is rather large. For example, the awareness of vestibular migraine (VM) as a vestibular syndrome causally linked to migraine was promoted by epidemiologic observations indicating a more than chance association of migraine with vertigo and dizziness and not by pathophysiologic hypotheses (Kuritzky et al., 1981, Kayan and Hood, 1984, Neuhauser et al., 2001, Vukovic et al., 2007). Moreover, as robust data on the population-wide high prevalence of dizziness and vertigo and their specific underlying disorders accumulate, a need for improved recognition and therapy of these diseases beyond specialized dizziness clinics and neurotologic training programs becomes evident.

This chapter will focus on the frequency and distribution of dizziness, vertigo, and imbalance, and of selected vestibular disorders and will report recent findings on associated risk factors and personal and healthcare impact. A few definitions and comments on epidemiologic concepts may facilitate the reading. It is essential to bear in mind that the clinical value of most epidemiologic findings reported here is not dictated by their statistical significance, i.e., by their precision, but by minimization of bias in the study design, i.e., minimization of systematic error that may affect the validity of the study (the ability to measure the truth), the reliability (the ability to reproduce the results), and the generalizability (does this study result apply to my patient?).

Bias cannot be corrected no matter how statistically sophisticated the analysis. There are two main types of bias, both of which are common in epidemiologic studies on dizziness and vertigo. Selection bias occurs when study participants are systematically different from the group about which the study wants to make an inference. Examples of selection bias are prevalence estimates or prognostic studies from specialized dizziness clinics which may not apply to more unselected patients. For example, the relative frequency of Menière's disease (MD) of 5–11% in the specialized care setting (Neuhauser et al., 2001, Brandt, 2004, Guilemany et al., 2004) is almost certainly due to selection bias and considerably overestimates the prevalence in the community. The least informative are studies with little information on their sampling design, inclusion and exclusion criteria, and proportions of eligible patients who were actually included. Information bias due to misclassification of both symptoms and diagnoses is a particular concern for dizziness studies at two levels: misclassification by study participants, who are given options of describing their subjective symptoms and have to choose among them, and misclassification by investigators, who have to interpret standardized or nonstandardized symptom descriptions and assign medical terms and diagnoses based on insufficiently operationalized diagnostic criteria or criteria that have been modified for study purposes without validation. Patient descriptions may be unclear, inconsistent, and unreliable (Newman-Toker et al., 2007), and there are language-specific linguistic issues. Moreover, patients are more likely to misclassify their symptoms when they are not offered enough options that cover the entire range of specific subcategories and appear to be equally valued by the investigators.

On the other hand, even physicians do not agree on the meaning of the word “vertigo” (Stanton et al., 2007) and investigators tend to diagnose conditions that they know about or are interested in, while ignoring others (Sloane et al., 1989, Maarsingh et al., 2010b). Patients and many physicians tend to use the terms vertigo and dizziness interchangeably, while dizziness experts use vertigo as a vestibular symptom, defined as a sensation of self-motion when no self-motion is occurring (Committee on Hearing and Equilibrium, 1995, Bisdorff et al., 2009). As a general rule, unless the terms dizziness and vertigo and individual diagnoses have been explicitly defined and reported, they may be imprecise and not comparable among studies or even within studies (Maarsingh et al., 2010b).

The recently published classification of vestibular symptoms by the Committee for the Classification of Vestibular Disorders of the Bárány Society (Bisdorff et al., 2009) is a very valuable basis for future studies but has not been applied yet, so most of the findings reported in this chapter do not refer to the exact terms and definitions of this classification. In this chapter, the term vertigo denotes a vestibular symptom but the exact definition varies among studies. Measures of disease frequency in the population are incidence (proportion of newly developed – incident – disease over a specific period of time) and prevalence (proportion of an existing disease, either at one point in time – point prevalence – or during a given period, i.e., period prevalence, e.g., 1-year prevalence). Lifetime prevalence denotes the cumulative lifetime frequency of a disease to the present time, i.e., the proportion of people who have had the event at any time in the past.