The DSP is a "smart" case history form to aid in the diagnosis and management of patients presenting with dizziness.
By Gary P. Jacobson, Ph.D., FASHA
One of the most important parts of a vestibular function assessment is the case history as reported by the patient. This is, in part, because the diagnosis of many dizziness diseases are often based on the patient’s symptoms. For example, there is no Meniere’s disease test. However, if a patient reports having recurring spells of rotatory vertigo that are accompanied by tinnitus, aural fullness and hearing loss in one ear, then Meniere’s disease would be one of a short listing of possible diseases in the differential diagnosis. Unfortunately, the conventional case history interview requires a great deal of time, patience, and training on the part of the clinician. Even though the clinician may attempt to guide the patient through the case history, valuable time can be lost in the process of mining for those valuable bits and pieces (i.e. the gold nuggets) of the history that contribute to the diagnosis and management of the dizziness disorder.
About 3 years ago we embarked on a project that was an attempt to develop a “smart” case history form. It was our objective to develop a case history device that would not require verbal responses to a long series of questions. Instead, the questions were converted into a series of statements that patients respond to using something called a Likert scale. Now, we all have had the experience of using a Likert scale. These scales usually consist of 5 choices that range from “strongly disagree,” to “disagree,” to “neutral” to “agree” and then lastly to “strongly agree”. The content of the questions was based on accepted diagnostic criteria for each of the most common dizziness diseases (e.g. Meniere’s disease, benign paroxysmal positional vertigo-BPPV, vestibular neuritis, vestibular migraine, persistent, postural perceptual dizziness-3PD). We wanted the device to be brief enough that it could be completed by a patient in 10 minutes or less. It was our objective for the scoring of the device to be automatic. That is, the device would automatically generate a list of possible dizziness diagnoses for the clinician to consider in their differential. In this way we hoped that such a device would be attractive to the primary caregiver, general otolaryngologist, audiologist, and other clinicians assessing dizzy patients. In the future, this questionnaire might be adapted for use on a tablet PC or IPad.
The questionnaire we developed is called the Dizziness Symptom Profile (DSP). It was our hope that, for example, if the result of the DSP “pointed” to BPPV then the clinician might focus on conducting both Dix-Hallpike and head-roll maneuvers. If either test was positive, then the clinician would perform the appropriate repositioning maneuver and “cure” the patient. If the result of the DSP pointed to vestibular migraine then the clinician might place the patient on a strict migraine diet and re-assess in 30 days. If the result of the DSP suggested that the patient’s dizziness was initiated partially or entirely as a result of anxiety (i.e. Persistent Postural Perceptual Dizziness-PPPD) the clinician might manage the patient with medication in an effort to reduce or eliminate the anxiety-related dizziness. Alternately, where the result of the DSP suggested that the dizziness/vertigo might be occurring due to Meniere’s syndrome, vestibular neuritis or superior semicircular canal dehiscence-SCD then the patient would be forwarded directly to a specialty provider (e.g. an otoneurologist, neurotologist or otologist) for assessment and management.
There were 3 phases to the development of the DSP. Our study protocol was approved by our local Institutional Review Board (IRB). The first 2 phases were focused on the development of the DSP questionnaire. Contrary to popular belief, the development of a validated and reliable scale is a long, arduous process. In the first phase the investigators created 64-items covering a range of common and a few less common dizziness diseases. We administered the 64-item questionnaire to 162 patients and conducted a statistic called an exploratory factor analysis (EFA). This is a statistic that is sensitive to groups of items that subgroups of patients answer in a similar manner. The analysis also made it possible for us to delete items that did not contribute statistically to the questionnaire. After the first analysis we were able to reduce the 64-item DSP to 35 items (i.e. much shorter). We randomized the remaining 35 items and re-administered the questionnaire to an additional 352 patients. As we did before, the responses were analyzed again with a factor analysis (a confirmatory factor analysis-CFA this time) and we were able to confirm the groups of items and reduce the number. This process yielded the final version of the DSP, which is 31-items in length and fits nicely on the front and back of a 8 ½” x 11” sheet of paper.
For the third phase of the study we wanted to know how often the result of the DSP agreed with the ear specialist’s differential diagnosis after they had completed only a case history and a physical examination. In order to conduct this project we had to develop a scoring system. Since the total points for each subscale (i.e. each diagnosis) varied from 12 points to 20 points, we decided to “normalize” the scoring so that the maximum score for each subscale was 100 points. A subscale score of 60 points or more meant that diagnosis should be considered in the differential for that patient. We conducted this third phase of the study with a group of 195 consecutive dizzy patients. Patients completed the DSP prior to their vestibular function studies in the Division of Audiology. The patient then met with their ear specialist (in the Division of Otology/Neurotology) and underwent a case history and physical examination. We scored the DSP and obtained a rank-ordering of those diagnosis subscales that exceeded 60%. Separately the otologist/neurotologist rank-ordered their own differential diagnosis based on the patient history and physical examination. We then calculated what percent of the time the ear specialist’s differential diagnosis was in agreement with the differential diagnosis suggested by the DSP.
We found that, overall, the DSP agreed with the ear specialist ~70% of the time. The best agreement occurred for the diagnoses BPPV, vestibular migraine and Meniere’s syndrome. We need to continue collecting data in an effort to increase our likelihood of encountering patients with diagnoses that occur less frequently such as superior semicircular canal dehiscence (SCD).
This study has been published in the journal Ear and Hearing (zdoi; 10.1097/AUD.0000000000000628). Providers who have an interest in obtaining the DSP may contact me at [email protected].
Figure 1: Table that is generated by the Dizziness Symptom Profile (DSP) when the 31-item data was entered into an Excel spreadsheet for a patient whose history is suggestive of Meniere’s disease.