Self-Reported Dystonia and Dystonia on Neurological Examination: Prevalence and Concordance across Time in a Prospectively Followed Essential Tremor Cohort Open Access

Screening questions and questionnaires are widely used to assess the likelihood that an individual in a particular population has a disease [1]. In clinical settings, screening is the cornerstone of early diagnosis, improving treatment outcomes and population-level health benefits [1‒3]. In research, screening allows investigators to identify and refine the characteristics of specific target populations. For a screening method to achieve optimal results, the proportion of false positives and false negatives should be minimal [1].

In the context of movement disorders, and more specifically, essential tremor (ET), screening for other involuntary movements is similarly useful. Dystonic movements and postures are not uncommon in patients with ET, and in research studies, there is often a need to either stratify or exclude ET cases with these movements [4]. Self-report data collected during the interview can potentially help identify dystonia, as it provides information based on previous interactions with their treating physicians and the self-awareness of having the disease. Although a promising resource, a gap in knowledge is that no studies have yet assessed the validity of self-report of dystonia in ET patients.

Hereby, we present data from a large cohort of 274 carefully phenotyped ET cases. These cases were followed prospectively, thereby providing a rich source of longitudinal data. Our goals were to (1) report the prevalence of self-reported dystonia and dystonic postures and movements on neurological examination, (2) analyze the concordance between self-reported dystonia and dystonia on examination (DOE), and (3) provide an estimate of the validity of self-reported dystonia in ET cases. These data aim to guide clinical researchers and treating physicians working with ET populations regarding the validity of self-report data on dystonia.

Our initial sample comprised 379 ET cases enrolled in a currently ongoing, prospective, longitudinal study of cognition in elders with ET (COGNET: Clinical Pathological Study of Cognitive Impairment in Essential Tremor, National Institutes of Health Award #R01-NS086736). Nationwide enrollment began in July 2014; to date, cases reside in 39 states. Eligibility requirements are (1) a diagnosis of ET, (2) a baseline age of at least 55 years, and (3) enrollment as an eventual donor in the Essential Tremor Centralized Brain Repository. The Yale University, Columbia University, and University of Texas Southwestern Medical Center Institutional Review Boards approved the study protocol, approval number STU2020-0564. All cases provided written, informed consent.

Evaluations took place at baseline and again at 18-month intervals, with the initial intent to follow the cohort for 5 years. The follow-up was subsequently expanded to 10 years among those who wished to remain in the cohort. A trained research assistant administered evaluations during home visits. Each involved the completion of demographic and clinical questionnaires, participation in a videotaped neurological examination, and completion of a battery of neuropsychological tests. Evaluations typically lasted 4–5 h and were usually completed in multiple sessions on 1 day (i.e., morning, afternoon) or on two adjacent days.

Longitudinal data were collected from 379 cases in two waves. Each wave included three evaluations conducted approximately 18 months apart. Wave I took place from 2014 through 2019. Wave II began in 2019 and is still ongoing. Of our initial sample of 379 cases, 93 participated only in wave I, 148 in both waves, and 138 only in wave II (online suppl. Fig. 1; for all online suppl. material, see https://doi.org/10.1159/000546428).

From our initial sample of 379, we eliminated 105 cases for the following reasons: (1) tremor was not severe enough to assign an ET diagnosis (see below) or case was determined to have Parkinson’s disease (n = 15); (2) a complete baseline assessment of dystonia or additional critical data were not available (n = 39); or (3) case was assigned a diagnosis of amnestic mild cognitive impairment or dementia, potentially impacting the validity of self-reported data (see below) (n = 51). Exclusion of these cases yielded a final baseline sample of 274, M baseline age = 78.6 ± 8.4 years. Of the baseline sample, 69 completed all six evaluations, and 205 did not. Specifically, 47 (17.2%) cases died and 37 (13.5%) withdrew during study enrollment. The remaining cases who did not complete all six assessments are either still actively enrolled in the study but have yet to complete their participation, or declined to re-enroll in wave II of data collection (i.e., evaluations 4 through 6). Cases who completed all six evaluations were younger at baseline (M = 74.0 ± 8.0 years) than those who did not (M = 80.1 ± 8.0 years), p < 0.001. These two groups did not differ in sex, race, years of education, age of tremor onset, tremor duration, or tremor severity, all p’s ≥ 0.17.

Cases provided basic demographic information including age, sex, race, and education, as well as details of their tremor history, including age of tremor onset. We calculated the difference between baseline and onset age to yield a measure of tremor duration.

Based on a detailed neuropsychological assessment, cases were assigned a diagnosis of normal cognition, mild cognitive impairment, or dementia, as detailed elsewhere [5, 6].

A videotaped neurological examination was performed at each evaluation, as detailed elsewhere [7]. The severity of postural tremor (one examination item) and kinetic tremor (five examination items) was assessed in each arm on a 0–3 scale by an experienced movement disorders neurologist, yielding a total tremor score. Total tremor score values range from 0 (low severity) to 36 (high severity) [8].

Based on a review of the questionnaire and videotaped examination data, the neurologist assigned clinical diagnoses of ET based on criteria from the Washington Heights-Inwood Genetic Study of Essential Tremor (WHIGET) [9]. The criteria include gradations of possible, probable, and definite ET. At a minimum, possible ET required moderate or greater amplitude kinetic tremor during the examination of at least three activities in the absence of another known cause (e.g., medication-induced tremor and tremor from hyperthyroidism). These diagnostic criteria for ET were developed for a population-based genetic study and based on data from approximately 2,000 non-diseased controls, and the criteria carefully detail the specific examination maneuvers during which tremor should be present and the severity of tremor that should be evident during these maneuvers to distinguish normal from ET [10]. These criteria have been shown to be both reliable [11] and valid [12] and have been used by tremor investigators in the USA and worldwide. The presence of mild dystonic postures in the setting of long-standing and severe ET did not exclude the ET diagnosis [4].

As part of the questionnaire, each case was asked whether they had dystonia – “Has a doctor diagnosed you as having dystonia?” In addition to this self-report of dystonia, the neurologist assessed whether dystonic postures, movements, or tremor were present on the videotaped neurological examination (i.e., sustained or intermittent muscle contractions causing abnormal, often repetitive, movements, postures, or both). As such, detailed assessment for dystonia was performed through review of videotaped views of the face, neck, trunk, and extremities while seated, standing, and walking (including turning); with posture (arms extended in front of body); and with multiple tests of action (finger taps, hand-opening/closure, finger-to-nose, pouring/drinking/lifting water with a spoon, and alternating toe-heel taps). Handwriting was videotaped and reviewed. Audio recordings of sustained phonation and speech were also reviewed to assess for dysphonia and oral, buccal, lingual dystonia. Dystonia was noted to be present or absent, regardless of its severity (e.g., ranging from subtle dystonic postures in the fingers or hand [13, 14] to less subtle postures and movements). For each case for whom dystonic movements were identified as present, the neurologist further described the anatomical location as either the (1) neck, (2) limb, (3) trunk, (4) voice, (5) ocular region, (6) oral, buccal, or lingual, (7) other area, (8) spanning multiple locations, or (9) generalized in nature. Because a number of in-person home visits were replaced with video visits or telephone visits during COVID (i.e., early 2020 through 2021), and in limited instances, due to study subject frailty and/or fatigue, videotaped neurological exams on which evaluations of dystonia were based were not available for all cases at all evaluations.

The distributions of baseline sample characteristics appear in Table 1. At each evaluation, we calculated the proportion of cases reporting the presence of dystonia, as well as the proportion of cases in whom dystonia was present upon examination (DOE) (Table 2). The distribution of anatomical locations of DOE appears in Table 3. Differences in prevalence across evaluations for these measures were assessed via chi-square tests (Tables 2, 3).

Contingency tables reporting the sensitivity, defined as the percent of cases with DOE who self-reported dystonia; specificity, percent of cases without DOE who did not self-report dystonia; positive predictive value (PPV), the likelihood that in a case with self-reported dystonia, DOE was present; and negative predictive value (NPV), the likelihood that in a case without self-reported dystonia, DOE was absent (Table 4). False positives were the percentage of individuals with a positive self-report without DOE. False negatives were the percentage of individuals with negative self-reported dystonia with DOE. A heat map depicting the pattern of concordance between self-reported dystonia and DOE appears in Figure 2. Additional heat maps depict these patterns of concordance in (a) only those cases in which the DOE was located at the neck and (b) only those cases in which the DOE was located at the limbs (Fig. 2, 3, respectively).

A total of 274 cases were followed in the study. The mean baseline age of the sample was 78.6 ± 8.4 years. Cases were predominantly female (61.3%) and Caucasian (98.4%), with a mean of 16.1 ± 2.5 years of education. The mean age of tremor onset was 41.3 ± 21.6 years, the mean tremor duration at baseline was 37.1 ± 21.1 years, and the mean total tremor score at baseline was 22.9 ± 5.7 (Table 1).

The prevalence of self-reported dystonia was 1.5%, 2.5%, 3.1%, 4.3%, 0.0%, and 4.3% at evaluations 1–6, respectively, and differed significantly across time, p = 0.01 (Table 2). The prevalence of DOE was 8.8%, 10.4%, 12.3%, 17.5% 14.8%, and 8.6% at evaluations 1–6, respectively (Table 2). Again, these values differed significantly across time (p = 0.007).

DOE occurred most commonly in the neck and limb (Table 3). When present in the limb, it only involved the upper limb, and it manifested as dystonic postures of the fingers or wrist. DOE was also observed in multiple regions in some cases (Table 3).

Across all evaluations, self-reported dystonia had a sensitivity ranging from 0.0% to 40.0%, a specificity of 96.1%–100%, an PPV of 0.0–67.0% (median = 30%), and an NPV of 83.1%–94.4% (Table 4). The proportion of false positives ranged from 0.0% to 3.9%, and the proportion of false negatives ranged from 60.0% to 100%.

Figure 1 illustrates the concordance patterns between self-reported dystonia and DOE. Out of the 74 cases with either a self-report of dystonia and/or DOE; 3 (4.1%) had positive concordance at each of their evaluations. Eleven (14.9%) of 74 cases had a positive self-report of dystonia but no DOE. Sixty-two (82.4%) of 74 cases did not self-report dystonia, but they had DOE.

In 28 cases, DOE involved the neck during one or more evaluation. In only one (3.6%) case in this group, there was concordance between a positive self-report of dystonia and presence of DOE (Fig. 2). In 29 cases, the DOE involved the limb. In only three (10.3%) cases in this group, there was concordance between a positive self-report of dystonia and presence of DOE (Fig. 3).

In this study, we assessed the validity of self-reported dystonia and concordance with the presence of DOE. We did so in a longitudinal cohort of 274 ET cases. Across the six evaluations, self-report had an estimated sensitivity that ranged from 0% to 40%, meaning that the self-report had low validity in identifying ET cases with DOE. Conversely, a specificity of 96.1%–100% suggests that ET cases who do not report dystonia are unlikely to have DOE. Furthermore, we report a PPV of 0.0%–67% (median = 30%), meaning that when a person has a positive self-report, there is a low probability that the person has DOE. The NPV of 83.1%–94.4% demonstrates that when there is no dystonia by self-report, the likelihood of DOE is low.

We also assessed the location of the dystonia, with the idea being that torticollis might have been viewed by a treating neurologist as “dystonic,” whereas subtle limb postures might have been overlooked. We hypothesized that the sensitivity of the self-report of dystonia might be higher in this setting. However, our data showed only 1 of 28 cases (3.6%) with torticollis self-reported its presence.

We attribute the low validity to a variety of challenges associated with self-reporting of dystonia. First, it could be that cases are not as self-aware of their dystonic postures and movements because these are mild in presentation or transient and therefore did not report these to their neurologist. Second, these cases’ neurologists may not have diagnosed dystonia – subtle dystonic movements may be missed by neurologists who are not trained in their recognition. Third, we did not capture data on regularity of neurological care outside of our study; it is possible that greater regularity of visits may have resulted in higher validity. Finally, a small number of cases who reported dystonia had no DOE. The reasons are not clear but could include misdiagnosis outside of the study or transient presence of dystonia.

Additionally, we observed a self-reported dystonia prevalence ranging from 0% to 4.3% and DOE from 8.6% to 17.5%. Our findings add to the existing literature that has reported the prevalence of dystonia, which thus far has reported a wide range of prevalence values in ET (2.9%–47%) [15‒17]. Our findings align with studies reporting lower prevalence rates of dystonia in (2.9% [15], 10.8% [17]) in their ET cohorts.

Our study had several limitations. First, we acknowledge that self-report data are prone to recall bias, which can result in underreporting during the interview. Second, we recognize that our sample of cases represents one cohort with its attendant demographic and clinical characteristics and that our data may not generalize to all cohorts of ET cases. For example, our cases were relatively advanced in age (M = 78.6 years). Finally, in addition to assessing the frequency with which people provide accurate self-reports of dystonia, it would be valuable to identify qualities that distinguish patients with dystonia who accurately report their condition from those who do not. Our cohort did not include a sufficient number of individuals with dystonia who accurately reported their condition to address this. However, this is an important area for future research.

Our study has several strengths. First, we are not aware of prior studies assessing the validity of ET cases’ self-report of dystonia. Second, our cohort longitudinally followed hundreds of cases using standardized and detailed clinical assessments.

In conclusion, our study provides important insights into the validity of self-report data on dystonia in ET. Secondarily, we report the prevalence of dystonic movements and postures alongside the anatomical distribution of said cases. Our findings suggest that self-reported dystonia was not a valid means to identify DOE in ET. We encourage further research to focus on testing additional questions for their utility as screening tools to improve dystonia detection in ET populations.

Research complies with internationally accepted standards for research practice and reporting. Research with human subjects was performed with approval of appropriate Ethics Committees and appropriate written informed consent was required for participation in the study. The protocol was reviewed and approved by the Yale University, Columbia University, and University of Texas Southwestern Medical Center Institutional Review Boards approved the study protocol, Approval No. STU2020-0564.

Elan D. Louis was a member of the journal’s Editorial Board at the time of submission. No other authors have a relationship at the time of submission that could reasonably be perceived as a potential conflict of interest.

This work was supported by National Institutes of Health award #R01-NS086736. NIH played no role in the design, collection, analysis, or interpretation of these data or in the writing of this report.

T.O.G. and D.S.B.: data analysis and writing of first draft; V.D.S.: data coding; and E.D.L.: conceptualization and design and writing of first draft.

The data that support the findings of this study can be requested from Elan D. Louis, MD, MS (E-mail address: [email protected]).