Multiple Sclerosis and Related Disorders (2013) 2, 21–28

Available online at www.sciencedirect.com

journal homepage: www.elsevier.com/locate/msard

Substantial burden of dizziness in multiple sclerosis Ruth Ann Marriea,n, Gary R. Cutterb, Tuula Tyryc a

Departments of Internal Medicine and Community Health Sciences, University of Manitoba, Winnipeg, CAN Department of Biostatistics, University of Alabama at Birmingham, Birmingham, USA c Division of Neurology, Barrow Neurological Institute, Phoenix, USA b

Received 1 April 2012; received in revised form 1 July 2012; accepted 14 August 2012

KEYWORDS Dizziness; Vertigo; Multiple sclerosis; Dizziness Handicap Inventory; Quality of life; Validation

Abstract Objective: Dizziness affects 49–59% of persons with MS but few prior studies examined the impact of dizziness in persons with multiple sclerosis (MS). Using the North American Research Committee on Multiple Sclerosis (NARCOMS) Registry, we examined the measurement properties of the Dizziness Handicap Inventory (DHI) in persons with MS, and the association between the DHI and health-related quality of life (HRQOL). Materials and methods: In 2004, NARCOMS participants completed the DHI. We scored the DHI for each participant and evaluated the association of the DHI scores and the participant demographic and clinical characteristics of interest using t-tests and Kruskal–Wallis tests as appropriate. We examined the association between the DHI scores and HRQOL (PCS-12, MCS-12) and disability status using Spearman rank correlations and generalized linear models. Results: 8123 participants completed the questionnaire, of whom 74.5% were women and 96.6% were white. Their mean (SD) age of symptom onset was 30.3 (9.7) years and median (IQR) score on Patient Determined Disease Steps (PDDS) was 4 (1–6). The median (interquartile range) DHI score was 20 (0–44). The impact of dizziness could be classified as mild in over 60% of participants (n = 4660), moderate in 30.9% (n = 2340) and severe in 7.55% (n= 572). The odds of reporting moderate or severe dizziness were reduced with higher socioeconomic status (postgraduate degree vs. less than high school education odds ratio [OR] 0.38; 95% confidence interval [CI]: 0.28–0.52) and increased with higher levels of disability (severe vs. mild disability OR 1.38; 95% CI: 1.13–1.68). Physical and mental HRQOL decreased as the impact of dizziness increased (DHI-PCS-12 r= 0.34, DHI-MCS-12 r = 0.37). Conclusions: Dizziness is common in MS and adversely impacts quality of life. More effective management of this symptom may improve HRQOL and deserves more attention. & 2012 Elsevier B.V. All rights reserved.

n Corresponding author: at. Health Sciences Center, GF-533, 820 Sherbrook Street, Winnipeg, MB R3A 1R9. Tel.: +1 204 787 4951; fax: +1 204 787 1486. E-mail address: [email protected] (R.A. Marrie).

2211-0348/$ - see front matter & 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.msard.2012.08.004

1.

Introduction

Dizziness is a symptomatic complaint which encompasses faintness or lightheadedness, disequilibrium and vertigo.

22

R.A. Marrie et al.

Vertigo is characterized by the illusion of movement when no movement is present. Dizziness is one of the most common symptomatic complaints in primary care practice, and affects 23% of the general population (Kroenke and Price, 1993; Neuhauser et al., 2008). The clinical manifestations of multiple sclerosis (MS) are diverse, and include dizziness and its subtype, vertigo. Dizziness, including postural intolerance, has variously been reported to affect 49–59% of persons with MS (Flachenecker et al., 1999; RaeGrant et al., 1999; Vita et al., 1993). Vertigo is the presenting manifestation of MS in up to 5% of persons (Herrera, 1990). Few of the prior studies of dizziness or vertigo have evaluated the impact of these symptoms on persons with MS. In one study, 5% of persons with MS surveyed ranked dizziness as one of their worst symptoms (Rae-Grant et al., 1999). However, the impact of dizziness has not been evaluated. Using a large socio-demographically diverse population, we examined the measurement properties of the Dizziness Handicap Inventory (DHI) in persons with MS, and the association between the DHI and health-related quality of life (HRQOL). We hypothesized that vertigo, as measured by the DHI, would have an independent negative impact on HRQOL.

2. 2.1.

Materials and methods NARCOMS Registry

The North American Research Committee on Multiple Sclerosis (NARCOMS) Registry is a voluntary self-report registry for persons with MS, (Consortium of Multiple Sclerosis Centers, 2002) developed by the Consortium of MS Centers. The Registry is approved by the Institutional Review Board at the University of Alabama at Birmingham, and participants agree to the use of their data for research purposes. Diagnoses of MS have been validated in a randomly selected sample of participants (Marrie et al., 2007). Participants enroll by mailing in a questionnaire or by completing a questionnaire online, according to their preference (Consortium of Multiple Sclerosis Centers, 2002). Subsequently, participants complete semi-annual surveys on paper or online. Participants report sociodemographic and clinical information, including disability status and HRQOL. Disability status is reported using validated instruments, Patient Determined Disease Steps (PDDS) and Performance Scales (Hohol et al., 1999; Marrie and Goldman, 2007). PDDS is a validated self-report measure which correlates well with a physician-scored Expanded Disability Status Scale (EDSS) (Hohol et al., 1999); NARCOMS using a slightly modified version of the scale which also correlates well with the EDSS and the Multiple Sclerosis Functional Composite (Marrie and Goldman, 2007). It is scored ordinally from 0 to 8, where a score of 0 approximates an EDSS score of 0, a score of 3 represents early gait disability without needing an assistive device and approximates an EDSS score of 4.0 to 4.5; and PDDS scores of 4, 5, and 6 represent EDSS scores of 6 to 6.5. Performance Scales uses a single question to assess each of eight domains: mobility, bowel/bladder, fatigue, sensory, vision, cognition, spasticity, and hand function. All of the subscales are scored ordinally as follows, 0 (normal),

1 (minimal), 2 (mild), 3 (moderate), 4 (severe), or 5 (total disability), except mobility which is scored from 0 to 6. HRQOL is reported using the Short-Form 12 (SF-12) version 2.0, a shortened version of the Short-Form 36 (SF-36). The SF-12 has 12 items, each capturing an aspect of one of the eight subscales in the SF-36. It is recommended for use in large surveys, and has been used in MS populations (Nortvedt et al., 2000). An aggregate Physical Component Scale (PCS-12) score summarizes physical HRQOL, and an aggregate Mental Component Scale (MCS-12) summarizes mental HRQOL; scores for both scales range from 0 (worst) to 100 (best). The PCS-12 and MCS-12 are standardized to reflect a general population mean of 50 and a standard deviation of 10.

2.2.

Dizziness

In the spring 2004 update questionnaire, we asked participants to complete the Dizziness Handicap Inventory (DHI) (Jacobson and Newman, 1990). The DHI is a 25-item instrument developed to assess the self-reported handicap related to dizziness, for the purposes of improving assessments of response to treatment of dizziness. The general instructions for the scale are as follows: ‘‘ The purpose of this scale is to identify difficulties that you may be experiencing because of your dizziness or vertigo. Dizziness is defined as a feeling of unsteadiness, faintness, or lightheadedness. Vertigo is a specific type of dizziness which includes the sensation that you or your environment is spinning. Please circle ‘‘Yes,’’ ‘‘No,’’ or ‘‘Sometimes’’ for each question. ‘‘Answer each question as it pertains to your dizziness or vertigo only.’’ The 25 items may be grouped into three subscales, functional (DHI-F, 9 items), emotional (DHIE, 9 items), and physical (DHI-P, 7 items). Each item has three potential responses, scored as 4 points for ‘‘yes’’, 2 points for ‘‘sometimes’’ and 0 points for ‘‘no’’. Total scores for the DHI range from 0 (no handicap) to 100 (significant self-perceived handicap), while total scores for the subscales range from 0 to 28 for the DHI-P, 0 to 36 for the DHI-F, and 0 to 36 for the DHI-E. Based on total DHI scores, the impact of dizziness can be classified as mild (0– 30 points), moderate (31–60 points), and severe (61–100 points) (Dros et al., 2011). The DHI has good internal consistency and high test–retest reliability (Tamber et al., 2009); however, since concerns have been raised about the appropriateness of the three-factor structure we evaluated it in our sample (Tamber et al., 2009).

2.3.

Analysis

First, we evaluated the measurement properties of the DHI. The internal consistency of the DHI and each of its subscales was measured using Cronbach’s alpha (Cronbach, 1951). We used principal components analysis (PCA) with oblique rotation to assess the factor structure of the DHI. Factors represent a linear combination of the original variables. We did not specify the number of factors to see if the three subscales would emerge as factors, as assessed by loading of the items from these subscales. If the three subscales were psychometrically valid, we would expect all or most of the items from each subscale to load on a common factor. We

Dizziness in MS evaluated item loadings as recommended elsewhere, where the minimum acceptable loading of an item was 0.32 and loadings of 0.32 or greater on two or more factors were considered to indicate cross-loadings (Costello and Osborne, 2005). To assess construct validity we correlated the DHI scale with Performance Scales using Spearman rank correlations. We expected moderate correlations between the DHI and fatigue (convergent validity) (Hebert et al., 2011), and HRQOL (Neuhauser et al., 2008). Second, we applied the DHI to examine the impact of dizziness in our population. We summarized continuous variables using means and standard deviations (SD) or medians and interquartile ranges (IQR). Categorical variables were summarized using frequencies and percents. For each participant we scored the DHI and each of its subscales, and report the median and interquartile range (IQR) for each. We evaluated the association of the DHI scores and the participant demographic and clinical characteristics of interest using t-tests and Kruskal–Wallis tests as appropriate. We examined the association between the DHI scores and HRQOL including the PCS-12 and MCS-12 using Spearman rank correlations. Following the univariate analyses we conducted multivariable analyses. In order to evaluate who was most likely to report a moderate or severe impact of dizziness we dichotomized the DHI as 0 (mild impact) or 1 (moderate or severe) (Dros et al., 2011), and conducted a logistic regression analysis considering as potential covariates age; sex; race; socioeconomic status as measured by education, annual household income, and health insurance status; marital status; disability status; year of symptom onset; and treatment status. We assessed multivariable associations between DHI and the PCS-12 using a general linear model. Because of non-linearity between the DHI and PCS12 the DHI was included as a categorical variable (mild, moderate, severe) as described above. We used the linear model to calculate mean SF-12 scores, adjusting for the following potential confounders derived from existing literature (Turpin et al., 2007): age; sex; race; socioeconomic status; marital status; disability status; year of symptom onset; and treatment status. We then repeated this analysis for MCS-12 scores. We calculated effect sizes for the categorized DHI on HRQOL by calculating the difference in mean scores in the affected and unaffected groups, and dividing by the scale’s construction standard deviation of 10 (Wu et al., 2007). Effect sizes are measures of the magnitude of effect in terms of standard deviation units in order to facilitate comparisons between populations (Samsa et al., 1999). We defined effect sizes of 0.2 as small, 0.5 as moderate, and 0.8 as large (Samsa et al., 1999). Race was categorized as White (reference group) and Other. Education was categorized as o12 years (reference group), high school diploma, Associate’s Degree or Technical Degree, Bachelor’s Degree, post-graduate degree. Annual household income was categorized as o$15,000 (reference group), $15,000–30,000, $30,000–50,000, $50,000–100,000, 4$100,000. Insurance status was categorized as private, public only (reference group), or none. Marital status was dichotomized as married/co-habiting versus single/ divorced/widowed/living alone. Age was included as a continuous variable in the linear model and as quintiles in the logistic model. Using PDDS, participants were classified

23 as having mild (PDDSr3, reference group), moderate (PDDS 4–5), or severe (PDDSZ6) disability (Marrie et al., 2006). Year of symptom onset was categorized aso1986, 1986– 1995,Z1996 because of non-linearity with the outcome measures. Treatment in the preceding 6 months with any of the approved disease-modifying therapies available in 2004 (IFNß-1a (Avonex), IFNß-1b (Betaseron), IFNß-1a (Rebif), glatiramer acetate (Copaxone)), was dichotomized as yes or no. All statistical analyses used SAS V9.2 (SAS Institute Inc., Cary, NC).

3.

Results

Of 16880 eligible participants, 8123 (48.1%) completed the spring 2004 questionnaire. The demographic and clinical characteristics of the responders are summarized in Table 1, and are consistent with other prevalent MS populations (Jacobs et al., 1999). As compared to responders, non-responders had a lower level of education (po0.0001) and lower annual income (po0.0001). The mean age of symptom onset (p=0.0002) was Table 1 Demographic and clinical characteristics of responders to the NARCOMS Spring 2004 Questionnaire. Characteristic

Responders

Sex, n (%) Female Male

6053 (74.5) 2067 (25.5)

Race, n (%) White Other

7636 (96.3) 292 (3.8)

Education, n (%) o12 yrs High school diploma Associate’s or Technical degree Bachelor’s degree Post-graduate degree

192 2911 1349 2102 1518

(2.4) (36.1) (16.7) (26.0) (18.8)

Annual income, n (%) o$15,000 $15,000–30,000 $30,000–50,000 $50,000–100,000 4$100,000 I do not wish to answer

925 1247 1968 2294 772 398

(11.4) (15.4) (24.2) (28.2) (9.5) (4.9)

Health insurance, n (%) Private 4203 Public 2666 None 19 Current age (years), mean (SD) 58.0 Age of symptom onset (years), mean (SD) 30.3 Age of diagnosis (years), mean (SD) 37.7

(61.2) (38.5) (0.3) (10.4) (9.7) (9.6)

Patient Determined Disease Steps (categorized), n (%) Mild 2637 (35.8) Moderate 3422 (46.5) Severe 1303 (17.6)

24

R.A. Marrie et al.

slightly younger among non-responders, but differed from responders by less than year, thus these differences were unlikely to be clinically significant. Non-responders were also more likely to report severe disability at enrollment (22.3% vs. 16.8%, po0.0001). The median (IQR) total DHI scale score was 20 (0–44) with 26.2% of participants reporting scores of zero (n=1984), indicating no concerns related to dizziness equivalently 73.8% reported at least some dizziness. Only 0.20% of participants had the maximum score. This suggests that there were no floor or ceiling effects. The DHI correlated moderately with fatigue (r= 0.42; 95% CI: 0.40–0.44), the PCS-12 (r= 0.34; 95% CI: 0.36 to 0.32) and the MCS-12 (r= 0.37; 95% CI: 0.39, 0.36). The internal consistency of the DHI was high (Cronbach’s a =0.95), as was the internal consistency of the DHI-P (a =0.87), DHI-F (a =0.89) and DHI-E (a =0.84). The Kaiser– Meyer–Olkin Measure of Sampling Adequacy (KMO) was 0.97, and Bartlett’s Test of Sphericity indicated the sample was suitable for PCA. Three factors had eigenvalues greater than one, and the scree plot also supported a two or three component solution, capturing 55.7% of the total variance before rotation. Each item loaded moderately or strongly on one or more factors; however, some items cross-loaded (Table 2). Further, some items did not load most strongly on

Table 2

their intended subscales. Therefore subsequent analyses focused on the DHI and the subscales were dropped. The impact of dizziness could be classified as mild in over 60% of participants (n= 4678), moderate in 25.9% (n =1964) and severe in 12.3% (n= 930). The top three questions to which respondents most commonly answered ‘yes’ or ‘sometimes’ were those querying whether symptoms worsened with head movements (56.0%), bending over (54.5%), and looking up (47.5%) (Table 3). On univariate analysis the median scores on the DHI did not differ according to sex or race (Table 4). The median scores on the DHI decreased with increasing levels of education, annual income. Persons with private health insurance also had lower DHI scores than those with public health insurance only. Thus the impact of dizziness was lower in persons with higher socioeconomic status. The impact of dizziness was higher in persons with moderate disability than in persons with mild or severe disability. Using a multivariable logistic regression model, several characteristics were associated with decreased odds of reporting a moderate or severe impact of dizziness (Table 5). Increasing socioeconomic status including education, income and insurance status were associated with decreased odds of moderate or severe dizziness. Age was inconsistently associated with impact.

Principal components analysis: factors and item loadings using oblique rotation.a

Dizziness Handicap Inventory (Items)

Factor 1 Factor 2 Factor 3

P—Does looking up increase your problem? P—Does walking down the aisle of a supermarket increase your problem? P—Does performing more ambitious activities like sports, dancing, household chores such as sweeping or putting dishes away increase your problem? P—Do quick movements of your head increase your problem? P—Does turning over in bed increase your problem? P—Does walking down a sidewalk increase your problem? P—Does bending over increase your problem? F—Because of your problem, do you restrict your travel for business or recreation? F—Because of your problem, do you have difficulty getting into or out of bed? F—Does your problem significantly restrict your participation in social activities such as going out to dinner, going to movies, dancing, or to parties? F—Because of your problem, do you avoid heights? F—Because of your problem, is it difficult for you to do strenuous housework or yard work? F—Because of your problem, do you have difficulty reading? F—Because of your problem, is it difficult for you to go for a walk around your house in the dark? F—Because of your problem, is it difficult for you to go for a walk by yourself? F—Does your problem interfere with your job or household responsibilities? E—Because of your problem, do you feel frustrated? E—Because of your problem, are you afraid to leave home without having someone with you? E—Because of your problem, have you been embarrassed in front of others? E—Because of your problem, are you afraid people may think you are intoxicated? E—Because of your problem, is it difficult for you to concentrate? E—Because of your problem, are you afraid to stay home alone? E—Because of your problem, do you feel handicapped? E—Has your problem placed stress on your relationship with members of your family or friends? E—Because of your problem, are you depressed?

0.384 0.548 0.688

0.797 0.469 0.646

0.235 0.464 0.359

0.422 0.270 0.668 0.542 0.704 0.497 0.746

0.833 0.377 0.438 0.782 0.406 0.371 0.404

0.285 0.740 0.470 0.366 0.461 0.711 0.498

0.522 0.756 0.558 0.606 0.729 0.819 0.672 0.671 0.706 0.599 0.662 0.460 0.798 0.754 0.755

0.610 0.599 0.515 0.566 0.411 0.513 0.611 0.251 0.361 0.531 0.549 0.059 0.369 0.259 0.339

0.282 0.387 0.413 0.358 0.486 0.402 0.289 0.600 0.202 0.073 0.306 0.554 0.359 0.271 0.224

a Factors generated by principal components analysis show that some items do not load most strongly on their intended subscales as indicated by the following letters: P—Physical subscale item; F—Functional subscale item; E—Emotional subscale item.

Dizziness in MS

Table 3

25

Responses to the Dizziness Handicap Inventory in the NARCOMS population.

Dizziness Handicap Inventory

Yes, n (%)

Sometimes, No, n (%) n (%)

P—Does looking up increase your problem? P—Does walking down the aisle of a supermarket increase your problem? P—Does performing more ambitious activities like sports, dancing, household chores such as sweeping or putting dishes away increase your problem? P—Do quick movements of your head increase your problem? P—Does turning over in bed increase your problem? P—Does walking down a sidewalk increase your problem? P—Does bending over increase your problem? F—Because of your problem, do you restrict your travel for business or recreation? F—Because of your problem, do you have difficulty getting into or out of bed? F—Does your problem significantly restrict your participation in social activities such as going out to dinner, going to movies, dancing, or to parties? F—Because of your problem, do you have difficulty reading? F—Because of your problem, do you avoid heights? F—Because of your problem, is it difficult for you to do strenuous housework or yard work? F—Because of your problem, is it difficult for you to go for a walk by yourself? F—Because of your problem, is it difficult for you to go for a walk around your house in the dark? F—Does your problem interfere with your job or household responsibilities? E—Because of your problem, are you afraid people may think you are intoxicated? E—Because of your problem, do you feel frustrated? E—Because of your problem, are you afraid to leave home without having someone with you? E—Because of your problem, have you been embarrassed in front of others? E—Because of your problem, is it difficult for you to concentrate? E—Because of your problem, are you afraid to stay home alone? E—Because of your problem, do you feel handicapped? E—Has your problem placed stress on your relationship with members of your family or friends? E—Because of your problem, are you depressed?

1890 (23.7) 1089 (13.9) 2173 (27.6)

1897 (23.8) 1068 (13.6) 1447 (18.4)

4193 (52.5) 5671 (72.5) 4246 (54.0)

2862 582 1192 2575 1378 863 1210

1586 953 1137 1739 808 1335 1130

3492 6377 5464 3600 5777 5746 5610

(36.0) (7.4) (15.30) (32.5) (17.2) ((10.9) (15.2)

(20.0) (12.0) (14.6) (22.0) (10.2) (16.8) (14.2)

(44.0) (80.6) (70.1) (45.5) (72.6) (72.3) (70.6)

1191 (15.0) 2283 (28.9) 2529 (32.2)

1389 (17.4) 767 (9.7) 1152 (14.7)

5390 (67.6) 4855 (61.4) 4178 (53.1)

1604 (20.4) 2178 (27.8)

1454 (18.5) 1099 (14.0)

4807 (61.1) 4561 (58.2)

1748 1543 1837 795

(22.1) (19.4) (23.0) (10.0)

1525 1063 1571 889

(19.3) (13.4) (19.7) (11.2)

4625 5328 4569 6242

(58.6) (67.2) (57.3) (78.8)

755 1255 210 1485 1092

(9.5) (15.8) (2.65) (18.7) (13.8)

1134 1787 386 1184 836

(14.3) (22.6) (4.87) (14.9) (10.5)

6046 4881 7333 5261 6003

(76.2) (61.6) (92.5) (66.4) (75.7)

897 (11.3)

1489 (18.8)

5550 (69.9)

P—Physical domain item; F—Functional domain item; E—Emotional domain item.

As compared to respondents reporting mild disability, those reporting moderate or severe disability based on the PDDS reported higher odds of moderate or severe dizziness, but with those reporting moderate disability having the highest odds. The mean (SD) PCS-12 was 36.3 (11.8). The mean PCS-12 decreased as the impact of dizziness increased. Thus it was 39.8 (12.1) in participants reporting a mild impact of dizziness, 32.7 (9.4) in those reporting a moderate impact of dizziness, and 28.7 (7.9) in those reporting a severe impact of dizziness. After adjustment for age, sex, socioeconomic status, year of symptom onset, disability status as measured by PDDS, and treatment status, participants reporting that dizziness had a severe (30.7; 95% CI: 29.8– 31.5) or moderate (31.7; 95% CI: 31.1–32.4) impact still had a lower PCS-12 than those reporting that dizziness had a mild impact (36.0; 95% CI: 35.4–36.6). This produced an effect size for severe as compared to mild impact of dizziness of 0.53, and for moderate as compared to mild impact of dizziness of 0.43. The mean (SD) MCS-12 was 45.7 (11.5). The mean MCS-12 decreased as the impact of dizziness increased. Thus it was 48.9 (10.5) in participants reporting mild dizziness, 43.2

(10.6) in those reporting moderate dizziness, and 36.2 (10.3) in those reporting severe dizziness. After adjustment for age, sex, socioeconomic status, year of symptom onset, disability status, and treatment status, participants reporting that dizziness had a severe (35.9; 95% CI: 35.3–36.5) or moderate (43.8; 95% CI: 43.0–44.6) impact still had a lower MCS-12 than those reporting that dizziness had a mild impact (48.9; 95% CI: 48.2–49.5). This produced a large effect size for severe as compared to mild impact of dizziness of 1.3, and for moderate as compared to mild impact of dizziness of 0.51.

4.

Discussion

Dizziness is a common symptomatic complaint in the general population, particularly in older persons (Dros et al., 2011; Kroenke and Price, 1993; Neuhauser et al., 2005). In a nationally representative sample of 4869 adults living in Germany, 22.9% had moderate or severe dizziness (Neuhauser et al., 2008). In a multi-community based sample of 13538 persons, 23.2% had dizziness of unspecified severity (Kroenke and Price, 1993). In the MS population

26

R.A. Marrie et al.

Table 4 Median (interquartile range) scores for the DHI according to participant characteristics. Characteristic

DHI

p-Value

Sex Female Male

20 (0–46) 22 (0–44)

0.76

Race White Non-white

20 (0–44) 16 (0–44)

0.49

Education o12 yrs 34 (12–62) o0.0001 High school diploma 28 (6–52) Associate’s or Technical degree 22 (2–46) Bachelor’s degree 16 (0–38) Post-graduate degree 10 (0–32) Annual income o$15,000 $15,000–30,000 $30,000–50,000 $50,000–100,000 4$100,000

40 (10–60) o0.0001 28 (5–52) 22 (2–46) 14 (0–36) 8 (0–26)

Health insurance Private Public None

16 (0–40) 34 (8–58) –

o0.0001

Patient Determined Disease Steps (categorized), n (%) Mild 12 (0–32) o0.0001 Moderate 36 (12–56) Severe 18 (0–52)

dizziness is also a frequent concern and appears to be more frequent than in the general population. We found that nearly 75% of respondents to our survey reported at least some difficulty with dizziness, and 38.2% were moderately or severely impacted by it. This is a considerable burden, substantiated by the significant impact on HRQOL as evaluated using effect sizes. Even after accounting for potential confounders, HRQOL was lower in respondents reporting a moderate or severe impact of dizziness than in respondents reporting a mild impact. For physical HRQOL severe dizziness had a moderate effect size of 0.53 and moderate dizziness had a small effect size of 0.43. For mental HRQOL severe dizziness had a large effect size of 1.3 while moderate dizziness had a moderate effect size of 0.51. Dizziness is also recognized to adversely affect HRQOL in the general population (Neuhauser et al., 2008; Nola et al., 2010), but we were unable to identify studies evaluating this in the MS population. In our population, lower socioeconomic status was associated with a greater impact of dizziness, as was moderate disability. Lower socioeconomic status is reported to be associated with greater physical disability, and disparities in the diagnosis and management of symptoms such as depression and urinary symptoms in persons with MS. Thus the observed association may reflect socioeconomic disparities

in the management of vertigo. It is possible that some of the inverse trends with socioeconomic status could be due to differential selection biases that entered due to the differences in response rates with healthier individuals in lower socioeconomic strata choosing not to respond. While this is a possibility given our unusually low response rate (usual response rates approach 70% in current surveys), even if the gradient is reduced or even eliminated, the levels of dizziness and lowered level of HRQOL still points to the impact that dizziness has within the MS population. Moderate disability was associated with higher odds of perceiving a moderate or severe impact of dizziness than severe disability. Several of the questions on the DHI assess the impact of dizziness on function, and some of these functions may already be so limited by MS-related physical disability that dizziness is not perceived as having as much impact. Longitudinal evaluation of the relationship between dizziness and physical disability would be required to elucidate this issue. The DHI was developed to assess the impact of dizziness. Initially thirty-seven items were created empirically based on symptoms reported by patients (Jacobson and Newman, 1990). This pool of items was reduced to 25 which were grouped into three subscales, physical, functional and emotional. Other investigators have questioned the validity of the three component structure of the DHI (Tamber et al., 2009), and our findings support this concern. However, the internal consistency reliability of the DHI was high and it did not exhibit floor or ceiling effects. It also had adequate construct validity. In a group of 51 persons with MS the DHI discriminated between fallers and non-fallers, and correlated strongly with the Activities-specific Balance Confidence test (r= 0.70) and more modestly with the Timed Up and Go test (r=0.35) and the Dynamic Gait Index (r= 0.39) (Cattaneo et al., 2006). In 25 persons with MS the DHI had a high test–retest reliability (intraclass correlation coefficient =0.90; 95% CI: 0.77–0.96). Collectively this suggests that it has reasonable psychometric properties in the MS population, although test–retest reliability still needs to be established. The etiology of dizziness in MS has received little attention. A study of patients with acute vertigo suggested that the most common etiology was benign paroxysmal positional vertigo, affecting 52% of individuals, followed by demyelinating brainstem lesions (32%); however this study included only 25 patients (Frohman et al., 2000). Postural intolerance is also reported to be a common cause of dizziness, but comprehensive evaluations of all etiologies of dizziness have not been performed (Flachenecker et al., 1999). The DHI was not designed to distinguish between the various forms of dizziness and thus we were unable to describe the types of dizziness experienced or determine whether the impact differs depending on the etiology of the symptoms. This should be the subject of future work. This study has limitations. Our response rate was lower than desirable. Although we currently send questionnaires by first class mail, for this study the paper questionnaire was sent using bulk mail and we do not know how many participants actually received the questionnaire as changes in addresses and deaths are often identified only after a participant failed to respond. Online response rates were affected by the unexpected emergence of SPAM filters. As

Dizziness in MS

27

Table 5 Odds ratios and 95% confidence intervals for adjusted associations between moderate or severe impact of dizziness and demographic and clinical characteristics of NARCOMS participants. Characteristic

Adjusted OR

Sex Male Female

b

Race White Other

b

Age Ager42 AgeZ43 tor48 AgeZ49 tor53 Age454 tor59 AgeZ60

95% CI

1.0 0.95 1.18 0.97 0.58

0.75, 0.92, 0.75, 0.44,

1.0 0.76 0.63

– 0.48, 1.22 0.50, 0.81

0.49 0.38

0.38, 0.63 0.28, 0.52

Annual income o$15,000 $15,000–30,000 $30,000–50,000 $50,000–100,000 4$100,000

1.0 0.73 0.63 0.49 0.38

0.57, 0.50, 0.38, 0.28,

Health insurance Public Private None

1.0 0.62 a, b

Education oHigh school High school diploma Associate’s or Technical degree Bachelor’s degree Post-graduate degree

1.21 1.49 1.24 0.76

0.93 0.81 0.63 0.52

0.51, 0.75

Marital status Single/divorced/ widowed Married

1.21

1.01, 1.44

Year of onset o1986 1986–1995 Z1996

1.0 0.75 0.61

0.63, 0.89 0.47, 0.78

Disability Mild Moderate Severe

1.0 2.81 1.38

2.35, 3.35 1.13, 1.68

Treatment No DMT DMT

1.0

b

a—all participants in the model had health insurance; b—not retained in the model.

compared to responders, non-responders had lower socioeconomic status and more disability thus these groups are likely underrepresented in our population. The NARCOMS Registry is a volunteer registry which does not fully represent the MS population in the United States. However, the NARCOMS population is large, socio-demographically diverse and its characteristics are similar to those reported for other MS populations (Jacobs et al., 1999; Marrie et al., 2003). Also, we cannot exclude the possibility that some participants mistakenly reported causes of unsteadiness, such as lower extremity weakness, other than those of interest. However, the specific instruction that ‘‘you or your environment is spinning’’ mitigate against this being a major explanation for our findings. Strengths of our study include evaluation of the psychometric properties of the DHI and the large population studied. Dizziness is common in persons with MS and adversely impacts daily activities and quality of life. The DHI is a potentially useful instrument to evaluate the impact of dizziness in the MS population. Given the impact of dizziness and the relative paucity of work in this area in persons with MS, further study to evaluate and mitigate the impact of this symptom is needed.

Role of the funding source The NARCOMS Registry is supported by the Consortium of Multiple Sclerosis Centers (CMSC). The CMSC had no role in the study design; in the collection, analysis and interpretation of the data; in the writing of the report; or in the decision to submit the article for publication.

Disclosures Ruth Ann Marrie receives research funding from: Canadian Institutes of Health Research, Public Health Agency of Canada, Manitoba Health Research Council, Health Sciences Center Foundation, Multiple Sclerosis Society of Canada, Multiple Sclerosis Scientific Research Foundation, Rx & D Health Research Foundation, and has conducted clinical trials funded by Bayer Inc. and Sanofi-Aventis. Gary Cutter has served on scientific advisory boards for and/or received funding for travel from Alexion, Allozyne, Bayer, Celgene, Consortium of MS Centers, Coronado Biosciences, Diogenix, Klein-Buendel Incorporated, Merck, Novartis, Nuron Biotech, Receptos, Somnus Pharmaceuticals, Spinifex Pharmaceuticals, St. Louis University, Teva pharmaceuticals; receives royalties from publishing Evaluation of Health Promotion and Disease Prevention (The McGraw Hill Companies, 1984); has received honoraria from GlaxoSmithKline, Novartis, Advanced Health Media Inc., Biogen Idec, EMD Serono Inc., EDJ Associates, Inc., the National Heart, Lung, and Blood Institute, National Institute of Neurological Diseases and Stroke, National Marrow Donor Program, Consortium of Multiple Sclerosis Centers; serves as a consultant to, Novartis, National Industrial Sand Association, Bayer Pharmaceuticals, and Teva Pharmaceuticals Industries Ltd.; has served on independent data and safety monitoring committees for Apotek, Biogen,Cleveland Clinic, EliLilly,Glaxo Smith Klein Pharmaceuticals, Medivation,

28 Modigenetech, NHLBI, NINDS, NMSS, Ono Pharmaceuticals, Prolor, Sanofi-Aventis, Teva.

Conflict of interest Tuula Tyry has no conflicts of interest to declare.

References Cattaneo D, Regola A, Meotii M. Validity of six balance disorders scales in persons with multiple sclerosis. Disability & Rehabilitation 2006;28:789–95. Consortium of Multiple Sclerosis Centers. NARCOMS Multiple Sclerosis Registry. 2002; 2008. Costello A, Osborne J. Best practices in exploratory factor analysis: four recommmendations for getting the most from your analysis. Practical Assessment, Research and Evaluation 2005;10:1–9. Cronbach LJ. Coefficient alpha and the internal structure of tests. Psychometrika 1951;16:297–334. Dros J, Maarsingh OR, Beem L, van der Horst HE, Gt Riet, Schellevis FG, et al. Impact of dizziness on everyday life in older primary care patients: a cross-sectional study. Health & Quality of Life Outcomes 2011;9:44–50. Flachenecker P, Wolf A, Krauser M, Hartung H-P, Reiners K. Cardiovascular autonomic dysfunction in multiple sclerosis: correlation with orthostatic intolerance. Journal of Neurology 1999;246:578–86. Frohman EM, Zhang H, Dewey RB, Hawker KS, Racke MK, Frohman TC. Vertigo in MS: utility of positional and particle repositioning maneuvers. Neurology 2000;55:1566–9. Hebert JR, Corboy JR, Manago MM, Schenkman M. Effects of vestibular rehabilitation on multiple sclerosis-related fatigue and upright postural control: a randomized controlled trial. Physical Therapy 2011;91:1166–83. Herrera W. Vestibular and other balance disorders in multiple sclerosis. Differential diagnosis of disequilibrium and topognostic localization. Neurologic Clinics 1990;8:407–20. Hohol MJ, Orav EJ, Weiner HL. Disease Steps in multiple sclerosis: a longitudinal study comparing disease steps and EDSS to evaluate disease progression. Multiple Sclerosis 1999;5:349–54. Jacobs LD, Wende KE, Brownscheidle CM, Apatoff B, Coyle PK, Goodman A, et al. A profile of multiple sclerosis: The New York State Multiple Sclerosis Consortium. Multiple Sclerosis 1999;5: 369–76. Jacobson GP, Newman CW. The development of the Dizziness Handicap Inventory. Archives of Otolaryngology Head & Neck Surgery 1990;116:424–7.

R.A. Marrie et al. Kroenke K, Price RK. Symptoms in the community: prevalence, classification, and psychiatric comorbidity. Archives of Internal Medicine 1993;153:2474–80. Marrie RA, Cutter G, Tyry T, Campagnolo D, Vollmer T. Validation of the NARCOMS Registry: diagnosis. Multiple Sclerosis 2007;13: 770–5. Marrie RA, Cutter G, Tyry T, Vollmer T, Campagnolo D. Does multiple sclerosis-associated disability differ between races? Neurology 2006;66:1235–40. Marrie RA, Goldman MD. Validity of Performance Scales for disability assessment in multiple sclerosis. Multiple Sclerosis 2007;13:1176–82. Marrie RA, Hadjimichael O, Vollmer T. Predictors of alternative medicine use by multiple sclerosis patients. Multiple Sclerosis 2003;9:461–6. Neuhauser HK, Radtke A, von Brevern M, Lezius F, Feldmann M, Lempert T. Burden of dizziness and vertigo in the community. Archives of Internal Medicine 2008;168:2118–24. Neuhauser HK, von Brevern M, Radtke A, Lezius F, Feldmann M, Ziese T, et al. Epidemiology of vestibular vertigo: a neurotologic survey of the general population. Neurology 2005;65:898–904. Nola G, Mostardini C, Salvi C, Erlocani AP, Ralli G. Validity of Italian adaptation of the Dizziness Handicap Inventory (DHI) and evaluation of the quality of life in patients with acute dizziness. Acta Otorhinolaryngologica Italica 2010;30:190. Nortvedt MW, Riise T, Myhr KM, Nyland HI. Performance of the SF36, SF-12, and RAND-36 summary scales in a multiple sclerosis population. Medical Care 2000;38:1022–8. Rae-Grant AD, Eckert NJ, Bartz S, Reed JF. Sensory symptoms of multiple sclerosis: a hidden reservoir of morbidity. Multiple Sclerosis 1999;5:179–83. Samsa G, Edelman D, Rothman ML, Williams GR, Lipscomb J, Matchar D. Determining clinically important differences in health status measures: a general approach with illustration to the Health Utilities Index Mark II. Pharmacoeconomics 1999;15:141–55. Tamber A, Wilhelmsen K, Strand L. Measurement properties of the Dizziness Handicap Inventory by cross-sectional and longitudinal designs. Health & Quality of Life Outcomes 2009;7:101–16. Turpin KV, Carroll LJ, Cassidy JD, Hader WJ. Deterioration in the health-related quality of life of persons with multiple sclerosis: the possible warning signs. Multiple Sclerosis 2007;13:1038–45. Vita G, Fazio M, Milone S, Blandino A, Salvi L, Messina C. Cardiovascular autonomic dysfunction in multiple sclerosis is likely related to brainstem lesions. Journal of Neurological Sciences 1993;120:82–6. Wu N, Minden SL, Hoaglin DC, Hadden L, Frankel D. Quality of life in people with multiple sclerosis: data from the Sonya Slifka Longitudinal Multiple Sclerosis Study. Journal of Health and Human Services Administration 2007;30:233–67.