The final form of the Vestibular Disorders Activities of Daily Living Scale.
Scores on the final version of the Vestibular Disorders Activities of Daily Living Scale for the Total Scale, functional subscale, ambulation subscale, and instrumental subscale.
Cohen HS, Kimball KT. Development of the Vestibular Disorders Activities of Daily Living Scale. Arch Otolaryngol Head Neck Surg. 2000;126(7):881-887. doi:10.1001/archotol.126.7.881
Existing scales of self-perceived disablement in patients with vestibular impairment either are too broad to detect the subtle deficits in this population or omit some important daily life tasks.
To develop a scale to assess self-perceived disablement in patients with vestibular impairment and to describe the development and initial testing of this new assessment tool.
A list of items was developed, sent to a panel of expert therapists for review, and then revised to yield the preliminary 31-item, 10-point scale that was administered to subjects. The scale was revised again, yielding the final 28-item scale, which was administered to a new group of subjects.
Data were collected from patients in an outpatient clinic of a tertiary care center.
Patients diagnosed as having benign paroxysmal positional vertigo and patients diagnosed as having chronic vestibulopathy, excluding Ménière disease, postsurgical vertigo, and postconcussion vertigo.
The final scale has 3 subscales: functional, ambulation, and instrumental. It has good face validity, high internal consistency (α≥.90), and high test-retest reliability (rc≥0.87). Scale ratings ranged from 1 (independent) to 10 (ceasing to participate in the activity), but median scores for most subjects were 4 or less. From 41% (39/94) to 44% (41/94) of subjects considered themselves to be independent on those tasks.
This scale has good face validity, high internal consistency, and high test-retest reliability. It may be useful for evaluating functional limitation and perceived handicap or disability before and after intervention and for helping patients become more realistic in understanding their own capabilities.
PEOPLE WITH vestibular disorders complain of vertigo, disequilibrium, and other symptoms. These complaints and the underlying impairments can lead to functional limitations or deficits in performing routine daily life tasks known as activities of daily living (ADL).1,2 Previous work3 has shown that people with vestibular impairments have deficits in these skills.
Different authors use different words to describe the various levels of physiological and performance problems. Cohen2 has described the history of this language discussion and has illustrated the application of these terms to otolaryngology. The operational definitions described by Cohen are used in this article: impairment refers to the loss of function at the organ level; functional limitation refers to reduced ADL performance; disability refers to reduced performance within the social environment; handicap is sometimes used synonymously with disability; and disablement, by contrast, includes all of these effects.
Activities of daily living can be categorized in many ways. Reed4 divided them into self-maintenance, productivity, and leisure. Self-maintenance comprises activities performed to maintain the person's health and well-being, including self-care, communication, and home management. Productivity comprises activities with economic value, which are performed to enable individuals to support themselves, their families, and society. These activities include paid employment, volunteer work, and hobbies. Leisure comprises activities performed for the intrinsic reward of enjoyment and psychological renewal; such occupations are important but are not required for physical self-maintenance and have no direct economic value.
Unlike Reed,4 Gresham and Dittmar5 divided ADLs into 2 categories: basic, including self-care and locomotion; and instrumental,6 including community living skills, work-related tasks, and other occupational role tasks. Early in the 20th century, Hall7 and Meyer8 each wrote influential articles suggesting that essential ADLs include purposeful occupations. Those comments have been followed by many other erudite discussions in the literature about daily life activity as a type of occupation, ie, goal-directed activity or that which occupies one's time (M. Frances Baxter, MA, OTR, oral communication, 1999). Some writers consider locomotion to be a separate activity; others consider it to be merely a performance component. (Hill9 provides an overview of the arguments, which are beyond the scope of this article.) As a practical matter, however, many clinicians approach mobility skills as a separate category from self-care.
The literature has many assessments of ADL performance (eg, the Functional Independence Measures, the Klein-Bell Scale, and the Barthel Index10- 12), but most instruments are designed for the type of patients usually referred to rehabilitation centers, ie, people with head injuries, spinal cord injuries, other severe neurologic problems, or significant orthopedic disorders. Using any of these assessments, the functional limitations of patients with a wide range of diagnoses can be compared. The rating scales of these assessments, however, are not refined enough to detect the subtle problems of vestibularly impaired patients.
Cohen3 reported on the ADL performance of vestibularly impaired patients using a scale similar to those used in the clinic by many occupational therapists working with other patient populations. Using a 5-point qualitative scale, patients rated their performance on daily life tasks ranging from bed mobility tasks, such as rolling over, to instrumental tasks, such as grocery shopping, before and after participating in a vestibular rehabilitation program. Subjects reported having decrements in all performance areas. Later studies13 with modified versions of the scale suggested that patients with Ménière disease have significant functional limitations while having Ménière attacks, and patients in the acute phase of recovery from acoustic neuroma resection also have significant limitations.14 Unfortunately, the psychometric characteristics of the scale were not documented, limiting its usefulness for future research by investigators and for treatment planning by clinicians.
Two normed scales of self-perceived functional limitation have been developed for patients with vestibular and balance disorders. The Dizziness Handicap Inventory was originally developed to assess the functional problems of patients with Ménière disease but has also been used with people who have other vestibular impairments.15 The Activities-specific Balance Confidence Scale was developed for elderly people with disequilibrium.16 Both self-administered scales measure individuals' perceptions of their own abilities. Powell and Myers16 use the term self-efficacy to describe this perception of one's own capabilities within a particular domain of activities. They suggest that self-efficacy affects the individual's decision of whether to participate in particular activities. Other factors, such as social constraints, also influence the extent of an individual's participation in various ADLs. Nevertheless, self-efficacy is an important variable and should be measured.
The Dizziness Handicap Inventory was the first scale of ADL self-efficacy developed for the vestibularly impaired population, and it has filled an important niche in the evaluation battery. It addresses a wide range of needs and behaviors, from self-care to psychosocial interactions. The 25-item, 3-point qualitative scale sums the scores to obtain subscores and total scores. Unfortunately, the use of a 3-point scale limits the possible sensitivity of the middle rating; indeed, many patients have commented that the definition of level 2, "sometimes," is variable. Also, because the domain of the scale is broad, it gives an overview rather than dealing with particular domains in depth.3 In contrast, the 16-item, 10-point qualitative Activities-specific Balance Confidence Scale has items more specific to balance.
Many functional assessments use multilevel scales. The Activities-specific Balance Confidence Scale uses an 11-level scale from 0% (no confidence) to 100% (complete confidence) in 10% gradations.16 The Canadian Occupational Performance Measure, a well-documented scale that has received accolades from therapists, has the client use 10-point scales to rate self-efficacy, satisfaction with self-efficacy, and level of importance of the task to the individual's life.17,18 The disability scale in the International Classification of Impairments, Disabilities, and Handicaps is also a 10-point rating, which is supplemented by a 10-point scale of outlook, ie, prognosis for recovery.19 The levels of the scales in the Activities-specific Balance Confidence Scale and the Canadian Occupational Performance Measure are merely qualitative and not defined, and indicate self-efficacy, whereas the International Classification of Impairments, Disabilities, and Handicaps has specific definitions for each level of disability within each specific domain, and the ratings are given by the evaluator. The operational definitions of the levels of the scales differ across domains. The 10-point scale used in the study described herein is based on the precedents set by these previous scales.
Since publication of the Dizziness Handicap Inventory, other scales have become available to assess the effect of vestibular impairments on psychosocial function, such as those developed by Yardley and colleagues.20- 22 Lacking from the literature, however, has been an assessment more directed to specific basic and instrumental activities. Such a scale should use a multilevel rating and focus on ADL. The goal of this study was to develop such an assessment.
The scale developed in this study, known as the Vestibular Disorders Activities of Daily Living Scale, was based on Cohen's 1992 ADL scale.3 This scale was used in a previous study of the effectiveness of some vestibular rehabilitation techniques.23 Subjects in the previous study stated that the scale was not sufficiently detailed to describe their problems accurately. Some subjects suggested other levels of the scale that should have been included. Therefore, the 1992 scale was expanded by breaking down some self-care and mobility tasks further and by eliminating items that were shown to be at normal levels in vestibularly impaired patients. The resulting list of 30 questions was sent to a panel of occupational therapists and physical therapists (n=20; mean ± SD experience, 20.3±5.8 years). They determined if each question was important for an ADL scale and suggested the placement of each question in a list of possible subscales. Based on their answers, the scale was modified: some questions were eliminated, some were added, and questions were organized by subscales. The new list of items was sent to a second panel of experienced occupational therapists and physical therapists (n=11; mean±SD experience, 21.5 ±7.3 years) who also rated each item on its value in the scale and its placement in a list of possible subscales. Items that were included by at least 2 therapists were retained. This procedure yielded an initial list of 30 items. They were divided into 3 subscales: functional, ambulation, and instrumental. The functional subscale included items about self-care and intimate activities. The ambulation subscale included items about walking and stair climbing. The instrumental subscale included items about home management, productivity, and leisure activities.
This process expanded the original 5-point rating scale to a 10-point scale plus a category of not applicable. The individual levels were similar to the standard ratings in existing ADL scales, but were expanded for finer distinctions based on comments of participants in the 1992 study and a subsequent study.23 Each level of the scale was defined in words and with a number to reduce possible ambiguities in interpretation. The rating scale was written on the scale vertically and was repeated on a separate page of text for ease of reading. The final form of the scale is given in Figure 1.
This assessment was administered to 94 subjects with peripheral vestibular disorders, recruited from among patients referred for vestibular rehabilitation after a diagnosis was made by an otolaryngologist or neurologist. Of these subjects, 66 had chronic vertigo (mean±SD age, 55±16.6 years; range, 25-92 years), the diagnosis of which was based on clinical examination and a diagnostic battery of dynamic posturography, bithermal caloric testing, and low-frequency sinusoidal testing in the rotatory chair.24 All subjects in this group with chronic vertigo had vertigo for at least 3 months; this group excluded subjects with Ménière disease,25,26 benign paroxysmal positional vertigo, residual vertigo after resection of acoustic neuroma, head trauma, and other neurologic disorders. They fit the profile for the diagnosis of chronic or recurrent vestibulopathy.27,28 In the other 27 subjects, who had posterior canal benign paroxysmal positional vertigo (mean±SD age, 52±11.7 years; range, 34-73 years), the diagnosis was made based on the finding of a positive response to the Dix-Hallpike maneuver29 and no other positive findings on diagnostic testing.30 These patients typically complained of vertigo during tasks involving pitch rotations of the head or when rolling over in bed.
After the initial data collection and analyses, described later in the article, the instrument was revised. The final version, shown in Figure 1, was administered to a second group of 28 vestibularly impaired subjects (mean±SD age, 54.2±14.1 years). This group was composed of 22 women and 6 men, of whom 25 had been diagnosed as having unilateral benign paroxysmal positional vertigo of the posterior semicircular canal and 3 had been diagnosed as having chronic vestibulopathies.
Before the scale was administered, the senior investigator (H.S.C.) explained the purpose of the scale and the levels of the rating scale, and instructed the subject to select only one level of the scale for each question. Subjects were instructed to rate their current performance on each task (their performance since having developed the vestibular disorder) compared with their premorbid performance (their performance before having developed the vestibular disorder). They were given as much time as they needed to complete the form.
Informed consent was obtained before test administration. The research protocol was approved by the Institutional Review Board for Human Subject Research.
The preliminary version of the scale included 30 items. Preliminary analyses showed that most subjects responded "not applicable" to a question about riding the bus or train, probably because in this geographic area the bus system is not widely used and no trains are available. Therefore, the transportation question was expanded to include buses or cars. This question was tested with 34 additional subjects who took the preliminary version of the scale.
This scale uses the median to summarize the total and subscale scores, which avoids the bias that can be introduced into a sum if a subject omits an answer or uses the not applicable rating. Unlike the mean, the median is not unduly influenced by extreme answers that do not agree with the remainder of the subject's assessment. Also, the median gave summary scores on the individual assessment that were more similar to an occupational therapist's qualitative evaluation than the mean score.
The Cronbach α coefficient was calculated to determine the reliability based on internal consistency.31 Item-rest correlations were used to detect items that fit poorly with the rest of the items on the scale or subscale. Several additional questions were eliminated, and the subscales were reorganized as a result of these analyses. The final scale had 28 items. Internal consistency on the final scale is high on the Total Scale and on the subscales: Total Scale α, .97; functional subscale α, .92; ambulation subscale α, .96; and instrumental subscale α, .91. Table 1, Table 2, Table 3, and Table 4 list the item-rest correlations and α coefficients for the Total Scale and for each subscale for the final scale. The highest item-rest correlations in each table indicate the items that best represent scores on that portion of the scale. For the Total Scale, those items are F-11 (meal preparation), A-15 (going up steps), and A-17 (walking in narrow spaces). For the functional subscale, those items are F-4 (dressing the lower body), F-5 (putting on socks or stockings), and F-10 (reaching down). For the ambulation subscale, those items are A-14 (walking on uneven surfaces), A-17 (walking in narrow spaces), and A-19 (walking in crowds). For the instrumental subscale, those items are I-24 (light household chores) and I-25 (heavy household chores).
Scores on the final version of the scale ranged from 1 to 8 on the Total Scale, from 1 to 5 on the functional subscale, from 1 to 8 on the ambulation subscale, and from 1 to 10 on the instrumental subscale. The distribution of subjects' scores on each scale showed that most subjects scored from 1 to 4 (Figure 2). Many subjects scored 1 on the Total Scale and the subscales. The median score on each subscale was 1. For the Total Scale, 41% (39/94) of the subjects had a median score of 1; functional subscale, 41% (39/94); ambulation subscale, 44% (41/94); and instrumental subscale, 38% (36/94).
To determine test-retest reliability during evaluation of the original version of the scale, 17 subjects with chronic vertigo who were participants in an ongoing study of vestibular rehabilitation took the test 2 hours apart, before and after testing, on the same day. The Lin concordance coefficient32 and the limits-of-agreement procedures of Bland and Altman33 were used to measure agreement between the test and retest results of the subjects. The Lin coefficient measures the accuracy (the concordance coefficient) and precision (the confidence interval [CI]) of agreement between the observed measurements and the line of perfect concordance (a line with an intercept of 0 and a slope of 1). The limits-of-agreement procedure of Bland and Altman complements the Lin concordance coefficient in that it reports the average difference and the 95% CI.
Concordance correlations were high for the Total Scale score (rc=1.00; 95% CI, 0.99 to 1.00) and for each subscale: functional (rc=0.87; 95% CI, 0.67 to 0.95), ambulation (rc=0.95; 95% CI, 0.87 to 0.98), and instrumental (rc=0.97; 95% CI, 0.92 to 0.99). Likewise, the actual differences were small and not statistically significant for the Total Scale score (rc=−0.03; 95% CI, −0.09 to 0.03) and for each subscale: functional (rc=−0.12; 95% CI, −0.28 to 0.52), ambulation (rc=0.15; 95% CI, −0.14 to 0.43), and instrumental (rc=0.12; 95% CI, −0.16 to 0.40).
This 28-item, 10-point, self-administered scale of self-perceived independence in ADLs has excellent internal consistency and test-retest reliability. It addresses 3 domains of occupation: basic self-maintenance tasks, mobility skills, and higher-level or more socially complex tasks within and outside of the home. The scale has good face validity as determined by a panel of experts, among whose professional instrumental ADLs is the task of evaluating patients' independence in ADLs.
As described earlier, the use of a multilevel scale is well-grounded in precedents in the literature. The not applicable rating was included because some subjects preferred not to answer all questions or thought that some questions were not relevant to their lifestyles. For example, some subjects said that they did not engage in meal preparation or heavy household chores because their spouses or adult children with whom they lived performed those tasks. Some subjects said they did not engage in active recreational activities although they did engage in quiet recreation. Also, some subjects preferred not to answer the question about intimate activities or were widowed and said that such activities were no longer part of their lives. Nevertheless, all of these activities are important ADLs, and the panel of therapists who were surveyed to obtain information about face validity said that these activities should be included.
The Vestibular Disorders Activities of Daily Living Scale has excellent internal consistency within each subscale and within the Total Scale. The measured parameter on the subscales and the Total Scale is the median score. Use of the median provides a valid measure of central tendency for this type of qualitative data. In this way, if the subject fails to answer a question, the score is not affected significantly.
The distribution of subjects' scores suggests that most patients believe their functional skills are not affected or are minimally affected, while a few report noticeable effects. Differences could be affected by diagnosis, personality type, constraints of the individual's lifestyle, and many other issues. Nevertheless, the finding that many patients use safety guarding for many daily life tasks has clinical significance for treatment planning and for understanding the effects of vestibular disorders on the quality of life. Performance on instrumental ADLs is more variable than on the other domains, probably because of the wider variety of tasks and roles in this category. The functional and ambulation tasks, such as bathing, donning clothes, and moving from place to place, are almost all universally essential tasks. The class of instrumental tasks, however, is more variable.
The scale also has excellent test-retest reliability within the 2-hour period used. Subjects could have remembered their original answers, but within the context of this study this issue could not be resolved further. All participants had already been diagnosed as having vestibular disorders and were referred to one of us (H.S.C.) for therapy. All subjects expected to be and were treated for their vestibular disorders before the end of the visit on the day they participated in the study, and some participants also underwent vestibular rehabilitation home programs. Therefore, to have asked subjects to have returned for subsequent testing on another day was not possible because their answers would have been confounded by having received therapy.
To meet the requirements for making a test meet the standards of the American Psychological Association,34 further research must be done on this instrument. Future work should address several issues, including test-retest reliability over a longer period, other tests of validity, the relation of the test to objective diagnostic tests and to existing ADL scales, and several other problems beyond the scope of this initial article.
A self-administered scale has some limitations. Self-efficacy, or self-perceived ability, is not the same as actual ability. Patients' perceptions may not correspond well to their actual performance abilities,35 although those perceptions undoubtedly color their willingness to attempt some tasks. Self-efficacy is also confounded by personality factors, past experience with illness or disability, psychological needs, and the demands of significant others for care, attention, or performance. For example, an elderly woman with moderate vertigo who is the primary caretaker for her chronically ill husband may perceive herself as inconvenienced but capable because she must be able to meet her husband's needs. In contrast, a young woman with mild vertigo who stopped working to be a full-time mother, whose husband is busy with his career, and who employs a maid to care for the house and a nanny to care for the baby may perceive herself as less capable, although objective diagnostic testing suggests that she has the lesser vestibular impairment. Therefore, this scale is best used in the context of an evaluation that includes a thorough history and, if possible, an objective assessment of functional skill.
This scale has many potential uses. It should be particularly useful for treatment planning, especially for vestibular rehabilitation. It should also be useful for assessing progress in response to treatment, with medication, surgery, or rehabilitation. Many subjects reported that filling out the scale was thought provoking because considering each item individually made them realize that they were either more or less independent than they had considered themselves. For example, several subjects who began the session by stating that the vertigo was "ruining my life" or similar comments discovered that they were more functional and less helpless than they had considered themselves to be. This change in self-image was probably therapeutic. The items also provoked discussion with significant others (eg, spouses) and provided a basis for discussion of practical issues to be addressed in therapy, discussed with the physician, or mentioned to family members and coworkers.
Accepted for publication January 26, 2000.
This study was supported by the Clayton Foundation for Research, Houston, Tex; and grants DC02412 and DC03602 from the National Institutes of Health, Bethesda, Md.
We thank Sharon Congdon, BS, Elizabeth Elizalde, BS, and Melody Fregia, BS, for technical assistance.
Reprints: Helen S. Cohen, EdD, OTR, Bobby R. Alford Department of Otorhinolaryngology and Communicative Sciences, Baylor College of Medicine, One Baylor Plaza, Houston, TX 77030 (e-mail: firstname.lastname@example.org).