BRIEF COMMUNICATION

Translation, cross-cultural adaptation and validationof the Bulgarian version of the Dizziness Handicap Inventory

Spaska Georgieva-Zhostova • Ognyan I. Kolev •

Katerina Stambolieva

Accepted: 20 February 2014

� Springer International Publishing Switzerland 2014

Abstract

Purpose The aim of the present study was the translation,

cross-cultural adaptation and validation of the Dizziness

Handicap Inventory in Bulgarian language (DHI-BG).

Methods Ninety-seven vestibular patients (19 men and 78

women, mean age 45.08 ± 13.85 years) took part in the

investigation. All participants were asked to fill in the DHI-

BG. Internal consistency was estimated using Cronbach’s

alpha and item-total correlation, reproducibility by calcu-

lating Bland–Altman’s limits of agreement and intraclass

correlation coefficients (ICCs). Associations were esti-

mated by Spearman’s correlation coefficients.

Results The Cronbach’s alpha for the total score, func-

tional, physical and emotional subscales of DHI-BG were

0.88, 0.75, 0.72 and 0.81. The floor and ceiling effects of the

DHI-BG total scale were evaluated with respect to the limits

of agreement which were ±9.4–14.53 points. Intraclass

correlation coefficients (ICCs) for all scale and subscales

were higher than the recommended value of 0.75 and

determined good test–retest reliability. The range of items

correlation for DHI-BG was from 0.27 (item 12) to 0.72

(item 3). No significant differences were observed in the

Cronbach’s alpha coefficients between the DHI-BG and the

original version, the German and Italian versions of the

questionnaire. The most significant difference was observed

in comparison with the German version of DHI. Construct

validity presented a moderate correlation between Romberg

coefficients and DHI-BG scores and strong correlation

between all scores of DHI and the self-perceived disability.

The results suggest that DHI-BG scores show a good dis-

criminative validity between groups with different levels of

self-assessed disability.

Conclusion The Bulgarian version of the DHI is a reli-

able and valid tool in assessing the impact of dizziness on

the quality of life in Bulgarian vestibular patients.

Keywords Dizziness Handicap Inventory (DHI) �Dizziness � Disability

Introduction

Dizziness is the second most common symptom after

headache, with prevalence up to 25 % in the general

population [1, 2]. Vestibular dizziness accounts for a

quarter of all reports of dizziness [3]. The vestibular dis-

orders affect the postural stability and have a significant

negative impact on patients’ activities with physical and

emotional consequences.

There are many generic and disease-specific instruments

for measuring the quality of life (QoL) in patient with

dizziness, most of them in English. Various questionnaires

such as Dizziness Handicap Inventory (DHI) [4], the

University of California Los Angeles Dizziness Question-

naire (UCLA-DQ) [5], etc. are used for measuring the QoL

of patients with dizziness. The most popular is the DHI. It

has several translations and cross-cultural adaptations—

Swedish [6], German [7], Italian [8], etc. but so far, there is

not a Bulgarian version.

The DHI is a reliable, comprehensively validated and

clinically useful tool to measure self-perceived handicap

S. Georgieva-Zhostova (&) � O. I. Kolev

University Hospital of Neurology and Psychiatry ‘‘St. Naum’’,

4-th km, Tzarigradsko Shosse Boulevard, 1113 Sofia, Bulgaria

e-mail: spacegeorgieva@abv.bg

K. Stambolieva

Institute of Neurobiology, Bulgarian Academy of Science, Sofia,

Bulgaria

123

Qual Life Res

DOI 10.1007/s11136-014-0660-5

associated with symptoms of dizziness and unsteadiness

from a variety of causes.

The aim of the present study was the translation, cross-

cultural adaptation and validation of the DHI in Bulgarian,

thus providing a reliable instrument for measuring QoL of

patients with dizziness and vertigo.

Subjects and methods

Ninety-seven vestibular patients (19 men and 78 women,

mean age 45.08 ± 13.85 years) took part in the investi-

gation at their first examination and 52 of them (10 men

and 42 women, mean age 45.65 ± 12.92 years) underwent

a second examination after 1 week.

All participants gave their written informed consent to

take part in the study, which was approved by the Ethics

Committee of Medical University—Sofia, conformity with

Declaration of Helsinki. The patients were diagnosed after

neuro-otological examination and had not a history of

psychiatric or somatic illness. The diagnostic groups of

patients were as follows: benign paroxysmal positional

vertigo—20 patients at the first (11 at the second exami-

nation); vestibular neuritis—19 [12]; uncompensated ves-

tibular hypo-function—18 [10]; labyrinthopathy—9 [5];

kinetosis—18 [7]; other vestibular dysfunctions—13 [7].

Patients with duration of dizziness up to 6 months were 46

(for retest 28), and 51 patients were with complaints for

more than 6 months.

The original DHI scale consists of 25 questions, orga-

nized in three subscales: physical (7 items), functional (9

items) and emotional (9 items). The response scale used in

the DHI is ‘‘yes/sometimes/no’’ scored as ‘‘4/2/0,’’

respectively [4]. After obtaining permission from the

authors [4], the English version of DHI was translated in

Bulgarian by three independent medical experts with very

good knowledge in English and retranslated in English by

two independent translators that were uninformed about the

original DHI version. The translated versions were dis-

cussed and edited according to the rules of the Bulgarian

language. The Bulgarian version of DHI (DHI-BG) was

first presented to ten patients with complaints of dizziness.

All questions were well accepted. We added examples

giving additional description only in item F12. The patients

filled in the final questionnaire in the hospital in our pre-

sence. They understood all the items and did not find any

difficulty in answering the questions (they did not ask for

additional explanations).

A static posturographic system (Synapsis posturography

system, France) was used to measure the postural stability

during upright stance with open (OE) and closed eyes (CE).

The Romberg quotient describing the ratio CE/OE mea-

sures was calculated [9].

Statistical analysis

Descriptive statistics were used for demographic, rating

scales and questionnaire data. The frequency distribution of

each item was evaluated for investigating the possible

ceiling and floor effects. Cronbach’s alpha and item-total

correlation (Spearman’s rank correlation coefficients) were

used to determine the internal consistency of DHI-BG

dimensions. In general, Cronbach’s alpha[0.7 (Nunnally’s

criterion) indicates high levels of internal consistency [10].

The Spearman’s rank correlation coefficient values \0.25

was considered to be weak, and values C0.76 were con-

sidered to indicate a strong relationship [11]. Test–retest

reliability was assessed by calculating the intraclass cor-

relation coefficient (ICC) and by calculating Bland–Alt-

man’s limits of agreement. The ICC values above 0.75 are

indicative of good reliability [12]. The Mann–Whitney

U test was used for estimating the difference between

correlation coefficients of items in Bulgarian and the other

language scales, where the correlation coefficients were

transformed by Fisher’s r to z transformation. Convergent

validity between DHI-BG scores and Romberg coefficients

was determined by Spearman’s rank correlation. The dis-

criminative validity was investigated by Kruskal–Wallis

ANOVA. It evaluated the differences between DHI-BG

scores and self-reported level of disability of the patients:

mild—10, moderate—37 and severe—50 patients.. A

p level \0.05 was considered as statistically significant.

The analyses were made using the SPSS version 16.0

computer software.

Results

No significant differences were observed in the Cronbach’s

alpha coefficients between the DHI-BG and the original

version [4], the German [7] and Italian [8] versions of the

questionnaire. All values of Cronbach’s alpha coefficients

for DHI-BG are in the range 0.72–0.88 (Table 1). A strong

Table 1 Reliability values (Cronbach’s alpha coefficients) for DHI-

BG compared with other language versions of DHI

Internal consistency Cronbach’s alpha

DHI-BG DHI DHI-G DHI-I

DHI-BG 0.88 0.89 0.90 0.92

Functional subscale 0.75 0.85 0.80 0.82

Physical subscale 0.72 0.78 0.71 0.75

Emotional subscale 0.81 0.72 0.82 0.84

DHI-BG—Bulgarian version; DHI—original US version; DHI-G—

German version; DHI-I—Italian version

Qual Life Res

123

positive correlation between the total DHI-BG score and

subscales was observed (rs [ 0.75).

The floor and ceiling effects of the DHI-BG total scale

were evaluated with respect to the limits of agreement

which were ±9.4–14.53 points. Out of 97 patients, 1 % had

a score of \9 points and 1 % a score of [91 points. Only

5.1 % had a score[85 points and 2.1 % had a score of\15

points. The results demonstrated no obvious floor and

ceiling effects (Table 2).

The ICCs, which evaluate the test–retest reliability, were

higher than the recommended value of 0.75 [10] and

showed no significant difference between the scores

(Table 2).

The range of items correlation (CI-TCs) for DHI-BG

was from 0.27 (item 12) to 0.72 (item 3) (Table 3). The CI-

TCs for DHI-BG were significantly higher for item 2 and

significantly lower for item 17 compared to the original

DHI version (p \ 0.05) and similar to the Italian version of

DHI. The most significant difference was observed in

comparison with the German version (items: 1, 2, 6, 10, 14,

21, 22 and 24) (Table 3).

Construct validity presented a moderate correlation

between Romberg coefficients and DHI-BG scores: for the

total scale rs = 0.38 (p \ 0.01), for functional subscale

rs = 0.37 (p \ 0.01), for physical subscale rs = 0.3

(p \ 0.01) and for emotional subscale rs = 0.29

(p \ 0.01).

The Kruskal–Wallis ANOVA with the three levels

(mild, moderate and severe degree) showed significant

effect of self-reported disability on the DHI-BG total score

(H [2] = 64.002, p \ 0.001) and the three subscales

(functional: H [2] = 57.626; p \ 0.001; physical:

H [2] = 45.833; p \ 0.001; emotional: H [2] = 35.216;

p \ 0.001). A post hoc Mann–Whitney U test showed

significant difference in total DHI-BG scores and subscale

scores between patients with mild and moderate and

between patients with moderate and severe disability

(p \ 0.001). The results suggest that DHI-BG scores show

a good discriminative validity between groups with dif-

ferent levels of self-assessment for disability, which

reflects on the QoL of patients with vestibular dysfunction.

Discussion

The results show that the questions in DHI-BG are well

understood, well accepted and easy to answer by patients.

Generally, the patients’ demographic and clinical charac-

teristics of this study do not differ from those used in the

other language versions of the scale [4, 7, 8]. The DHI-BG

shows high internal consistency. The total reliability of

DHI-BG is 0.88, and the Cronbach’s alpha values of the

different language versions are similar. These results con-

firm the validity of DHI-BG. Probable explanations for this

similarity are that the questions in DHI are clear and well-

defined. CI-TC results correlate well with results of the

original version of DHI [4], except slight difference in item

2 and item 17. Our results are close to the results obtained

in the Italian version. However, we found some differences

compared to the German version of DHI, mostly in

‘‘Emotional subscale.’’ That can be explained by differ-

ences in the lifestyle, temperament, communicability and

culture between Bulgarians and Germans. For example, the

sensation of embarrassment is typical for some Bulgarians,

mostly aged 50 years and over. The social area and

financial position are also important for the QoL of

patients.

The repeatability coefficients show that 92.5–100 % of

the differences in test–retest scores of the DHI-BG total

scale, as well as the three subscales, lay between two SDs

[6]. ICC values exceed the recommended value of 0.75

[12] and are similar with the other DHI versions [4, 7, 8].

The strong positive correlation between the DHI scores

and Romberg coefficients, as an objective measurement of

Table 2 Distribution of the scores in the test (DHI-1) and retest (DHI-2) investigation of subgroup of 52 patients

Reproducibility n DHI-BG-1 DHI-BG-2 Mean differences and limits of agreement ICC 2/1

Median

(range)

Mean

(SD)

Median

(range)

Mean

(SD)

Mean

(SD)

Lower limit

(95 % CI)

Upper limit

(95 % CI)

(95 % CI)

DHI-BG 52 69.00 60.05 66.00 57.50 2.55 -12 12 0.94

(6–88) (20.62) (12–82) (17.58) (5.99) (-14.53 to -9.4) (9.4–14.53) (0.88–0.97)

Functional subscale 52 26.00 23.25 24.00 22.40 0.85 -5.5 5.5 0.93

(4–36) (8.20) (4–36) (7.22) (2.75) (-6.35 to -4.65) (4.65–6.35) (0.87–0.96)

Physical subscale 52 21.00 19.65 19.00 18.75 0.9 -4.9 4.9 0.90

(0–28) (6.31) (6–28) (5.16) (2.44) (-5.77 to -3.97) (3.97–5.77) (0.81–0.95)

Emotional subscale 52 16.00 17.15 15.00 16.35 0.8 -5.2 5.2 0.95

(2–34) (9.26) (2–34) (8.01) (2.59) (-5.98 to -4.38) (4.38–5.98) (0.91–0.98)

SD standard deviation, CI confidence interval, ICCs intraclass correlations

Qual Life Res

123

the vestibular dysfunction, and between the DHI scores and

self-assessment of disability of the vestibular patients

confirms the validity and sensitivity of the scale.

It will be of benefit to measure responsiveness of change

in the DHI-BG, in order to assess the effectiveness of

treatment and vestibular rehabilitation on patients’ QoL.

Conclusion

The results of this study show high internal consistency and

good test–retest reliability of the DHI-BG. DHI-BG is a

reliable and valid tool in assessing the impact of dizziness

on the QoL in Bulgarian vestibular patients.

Acknowledgments The authors wish to thank professor Jacobson

who his permission and assistance in the process of validating the

questionnaire.

References

1. Kroenke, K., Hoffman, R. M., & Einstadter, D. (2000). How

common are various causes of dizziness? A critical review.

Southern Medical Journal, 93, 160–167.

Table 3 Corrected item-total correlation (CI-TC) coefficients in DHI-BG, DHI-I, DHI, DHI-G

No. Item CI-TC DHI-BG

n = 97

DHI

n = 106

DHI-G

n = 127

DHI-I

n = 50

P1 Does looking up increase your problem? 0.66 0.57 0.54 0.32* 0.43

E2 Because of your problem, do you feel frustrated? 0.64 0.69 0.34* 0.51* 0.64

F3 Because of your problem, do you restrict your travel for business or

recreation?

0.69 0.72 0.76 0.61 0.65

P4 Does walking down the aisle of a supermarket increase your problems? 0.26 0.46 0.39 0.48 0.31

F5 Because of your problem, do you have difficulty getting into or out of bed? 0.43 0.45 0.50 0.41 0.45

F6 Does your problem significantly restrict your participation in social activities,

such as going out to dinner, going to the movies, dancing or going to

parties?

0.86 0.53 0.69 0.72* 0.65

F7 Because of your problem, do you have difficulty reading? 0.48 0.41 0.44 0.36 0.58

P8 Does performing more ambitious activities such as sports, dancing and

household chores (sweeping or putting dishes away) increase your

problems?

0.57 0.62 0.54 0.67 0.60

E9 Because of your problem, are you afraid to leave your home without having

someone accompany you?

0.54 0.59 0.43 0.49 0.67

E10 Because of your problem, have you been embarrassed in front of others? 0.51 0.49 0.46 0.27* 0.60

P11 Do quick movements of your head increase your problem? 0.48 0.33 0.51 0.41 0.54

F12 Because of your problem, do you avoid heights? 0.20 0.27 0.49 0.42 0.48

P13 Does turning over in bed increase your problem? 0.46 0.42 0.43 0.27 0.40

F14 Because of your problem, is it difficult for you to do strenuous homework or

yard work?

0.56 0.49 0.58 0.69* 0.64

E15 Because of your problem, are you afraid people may think you are

intoxicated?

0.42 0.38 0.30 0.48 0.47

F16 Because of your problem, is it difficult for you to go for a walk by yourself? 0.35 0.51 0.62 0.57 0.62

P17 Does walking down a sidewalk increase your problem? 0.21 0.36 0.58* 0.46 0.41

E18 Because of your problem, is it difficult for you to concentrate? 0.42 0.46 0.49 0.51 0.57

F19 Because of your problem, is it difficult for you to walk around your house in

the dark?

0.36 0.39 0.48 0.32 0.50

E20 Because of your problem are you afraid to stay home alone? 0.39 0.40 0.27 0.37 0.61

E21 Because of your problem. do you feel handicapped? 0.43 0.31 0.41 0.71* 0.58

E22 Has the problem placed stress on your relationships with members of your

family or friends?

0.43 0.34 0.46 0.60* 0.29

E23 Because of your problem, are you depressed? 0.64 0.57 0.41 0.63 0.59

F24 Does your problem interfere with your job or household responsibilities? 0.43 0.43 0.56 0.66* 0.63

P25 Does bending over increase your problem? 0.58 0.39 0.57 0.32 0.54

DHI-BG—Bulgarian version, DHI—original US version, DHI-G—German version, DHI-I—Italian version; * significant differences between

DHI-BG and other languages (Man–Whitney U test); significant level p \ 0.05

Qual Life Res

123

2. Yardley, L., Owen, N., Nazareth, I., et al. (1998). Prevalence and

presentation of dizziness in a general practice community sample

of working people. British Journal of General Practice, 48,

1131–1135.

3. Neuhauser, H. K., von Brevern, M., Radtke, A., et al. (2005).

Epidemiology of vestibular vertigo: A neurotologic survey of the

general population. Neurology, 65, 898–904.

4. Jacobson, G. P., & Newmann, C. W. (1990). The development of

the Dizziness Handicap Inventory. Archives of Otolaryngology:

Head and Neck Surgery, 116(4), 424–427.

5. Honrubia, V., Bell, T. S., Harris, M. R., et al. (1996). Quantitative

evaluation of dizziness characteristics and impact on quality of

life. American Journal of Otology, 17, 595–602.

6. Jarlsater, S., & Mattsson, E. (2003). Test of reliability of the

Dizziness Handicap Inventory and the activities-specific Balance

Confidence Scale for use in Sweden. Advances in Physiotherapy,

5, 137–144.

7. Kurre, A., Van Gool, C. J. A. W., Bastianen, C. H. G., et al.

(2009). Translation, cross-cultural adaptation and reliability of

the German version of the Dizziness Handicap Inventory. Otol-

ogy & Neurotology, 30, 359–367.

8. Nola, G., Mastardini, C., Salvi, V., et al. (2010). 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, 30, 190–197.

9. Hufschmidt, A., Dichgans, J., Mauritz, K. H., & Hufschmidt, M.

(1980). Some methods and parameters of body sway quantifica-

tion and their neurological applications. Archiv fur Psychiatrie

und Nervenkrankheiten, 228, 135–150.

10. Cronbach, L. J. (1951). Coefficient alpha and the internal struc-

ture of tests. Psychometrika, 16, 297–334.

11. Gill-Body, K. M., Beniato, M., & Krebs, D. E. (2000). Rela-

tionship among balance impairments, functional performance,

and disability in people with peripheral vestibular hypofunction.

Physical Therapy, 80, 748–758.

12. Shrout, P. E., & Fleiss, J. L. (1979). Intraclass correlations: Uses

in assessing rater reliability. Psychological Bulletin, 86, 420–428.

Qual Life Res

123