Anxiety Scales in Lewy Body Disease
Background
Anxiety disorders and symptoms are frequent in Lewy Body Disease (LBD) and many scales are available to assess its severity. Yet, there is little consensus on which scale best evaluates anxiety across LBD patients.
Objective
Comparing the convergent validity of commonly used anxiety scales across synuclein-based diseases.
Lewy Body Disease & Anxiety:
Lewy body dementia (LBD) is an umbrella term for synucleinopathies that include Parkinson’s disease (PD), PD dementia (PDD), and dementia with Lewy bodies (DLB). DLB is a complex, heterogeneous disorder, clinically and pathologically related to PD. Besides cognitive fluctuation, depression, constipation, anosmia, orthostatic hypotension, anxiety is a common and early manifestation in DLB.
Ricci et al. reported that anxiety is more prevalent in patients with DLB than in patients with Alzheimer’s disease (AD), and this was later supported by Minna et al. A few studies found that the severity of anxiety was stable across stages of dementia, except for a slight decrease at the terminal stage. Chiu et al. reported that about 15 – 23% of patients with isolated REM Sleep Behaviour Disorder (iRBD) have mood disturbances including depression and anxiety.
Thus, frequent screening of anxiety symptoms in the early stages and the course of LBD is important for adequate and early-onset therapeutic management. However, it can be a challenge to select the right anxiety questionnaire/scale given the variety of options. Different rating scales do not capture the full spectrum of anxiety in patients with PD, LBD, or iRBD. In addition, some of the rating scales have only been validated in PD rather than DLB or prodromal LBD.
Therefore, this study aimed to evaluate the convergent validity of commonly used anxiety scales and to prepare the way for quantifying anxiety across a spectrum of LBD patients in both clinical and research settings.
Methods
With consent from the ethics committee of the University of Sydney Human Research, the database of Parkinson’s Disease Research Clinic at the Brain and Mind Centre, University of Sydney was screened and recruited 57 participants – 23 with isolated REM Sleep Behavior Disorder (iRBD), 17 with Parkinson’s Disease (PD), and 17 with Dementia with Lewy Bodies (DLB).
All participants underwent a neurological assessment by a neurologist and movement disorder specialist (SJGL). iRBD patients were confirmed using diagnostic polysomnography and did not satisfy diagnostic criteria for PD, DLB, or MSA.
Anxiety was assessed in all participants using the Hospital Anxiety and Depression Scale (HADS-A), State-Trait Anxiety Inventory (STAI), MDS-UPDRS Anxiety item, and the Parkinson Anxiety Scale (PAS).
The convergent validity amongst the anxiety rating scales across the different patient groups was analyzed using Spearman’s rank correlations.
Results
- Across all patient groups, the total score on the PAS measured by the STAI-Y2 form was significantly correlated with trait anxiety (iRBD: r = 0.80, p < 0.001; PD: r = 0.68, p = 0.003; DLB: r = 0.58, p = 0.018).
- Trait anxiety was correlated with the persistent anxiety subscore on the PAS (iRBD: r = 0.73, p < 0.001; PD: r = 0.58, p = 0.015; DLB: r = 0.61, p = 0.012).
- Trait anxiety was also correlated with episodic anxiety subscore on the PAS in the iRBD cohort (r = 0.53, p = 0.011).
- HADS-A total was correlated with the PAS-total in iRBD (r = 0.73, p < 0.001), PD (r = 0.58, p = 0.016), but not in DLB (r = 0.47, p = 0.09). HADS-A total was also correlated with persistent anxiety PAS sub score in the iRBD and PD groups (iRBD: r = 0.66, p = 0.001; PD: r = 0.54, p = 0.024) and with the episodic anxiety PAS sub score in all groups (iRBD: r = 0.65, p = 0.001; PD: r = 0.49, p = 0.044; DLB: r = 0.55, p = 0.043). HADS-A total was also correlated with Trait anxiety in the iRBD (r = 0.67, p = 0.001) and both trait and state anxiety in the PD cohort (trait: r = 0.67, p = 0.003; state: r = 0.6, p = 0.015).
- The avoidant PAS subscore was not related to any other anxiety scale measures and was consistent across all groups.
- There were no correlations between ratings on the UPDRS anxiety item, the PAS score (or subscores), and HADS-A total across all patient groups.
Discussion
Research has made some gains lately in sifting out the fact that a history of depression and anxiety might increase the risk of developing Lewy body dementia. However, there have been very few studies about anxiety in PD, DLB, and iRDB. Also, there is a dearth of research on the clinical scales that best evaluate anxiety across PD, DLB, and iRBD patients.
This study examined the convergent validity across common anxiety scales separately in PD, DLB, and iRBD patient groups. The most notable observation made was MDS-UPDRS anxiety item’s lack of correlation with any of the anxiety scales used. And, this could be considered one of the many reasons why anxiety remains underdiagnosed and undertreated in the patient populations. One possible explanation is that since item 1.4 in the MDS-UPDRS was the only scale that was completed by the examiner, it might have led the patients to under-report anxiety symptoms and the effect on their daily life compared to when questionnaires were self-administered. Another explanation is that the range of the UPDRS item was less than other anxiety scales and may have reduced the possibility for a meaningful correlation. In contrast, the PAS total was found to be the most suitable clinical scale to assess anxiety across the spectrum of LBD.
The greatest convergence across PAS, HADS-A, STAI was found when employed in the PD and iRBD cohort, whereas little convergence was found in DLB. In iRBD and PD cohorts, HADS-A was correlated with STAI-trait anxiety, PAS-total, as well as PAS-persistent and PAS-episodic. That means, in PD and iRBD, HADS-A has high convergent validity, as does PAS provided it was also associated with scores on the STAI-trait and HADS-A. In contrast, the only correlation between scales for the DLB cohort was between PAS-total and STAI-trait anxiety, indicating that HADS-A may not be as useful for assessing anxiety in DLB.
Previous studies reported that the psychometric properties of the HADS are satisfactory, although it has been found to have good interrater and test-retest reliability for detecting anxiety in patients with Parkinson’s disease. However, in this study, HADS-A showed weaker to moderate correlations to HADS-A, PAS -total (r = 0.47), and trait anxiety (r = 0.22), especially in DLB patients. This suggests a significant difference in anxiety levels in the patients which HADS-A failed to capture. The anxiety ratings were higher and more variable in BLB patients compared to other groups.
Limitations
This study has several limitations. First, the sample size is small. Second, it was limited to anxiety symptoms recorded in cross-sectional clinical settings, which is not always representative of the symptoms experienced at home settings. So, future studies might consider at-home monitoring using mobile apps to effectively track the severity of anxiety and its effect on a patient’s daily life. Third, there’s overlap amongst a small set of clinical scales in PD, DLB, and iRBD. Therefore, other clinical scales such as the Beck Anxiety Inventory, Hamilton Anxiety Scale, and the Generalized Anxiety Scale should be considered in future research.
Conclusion
PAS and STAI-Y2 are the most suitable anxiety rating scales to assess anxiety in synuclein-based diseases.
HADS-A showed strong convergent validity in PD and iRBD cohorts; it had weaker convergent validity in DLB patients.
The UPDRS anxiety item did not correlate with any of the other anxiety measures, and thus may not be effective at detecting anxiety symptoms.
Oncology Related Tools
Other
Latest Research
