Medication Adherence in Chinese Patients with Essential Tremor: A Real World Study

Introduction

Essential tremor (ET) is one of the most prevalent movement disorders, clinically characterized by symmetric action tremor of bilateral upper limbs, with or without tremor in other parts of the body such as the head, face, voice, or lower limbs [1]. ET affects approximately 0.9% of the global population. Epidemiological studies conducted in China have revealed significant regional variations in ET prevalence. For example, a study among individuals aged 55 and above in Beijing reported an overall ET prevalence of 3.29%, with varying rates across rural (4.29%), urban (2.85%), and mountainous (2.29%) areas [2]. However, a study conducted in a rural area of Shanghai reported a prevalence of 0.306% among residents aged 50 and older [3]. Given the obvious clinical heterogeneity of ET, the 2018 consensus statement on tremor from the International Parkinson and Movement Disorder Society proposed a new concept of ET plus, defined as ET in the presence of additional neurological signs of uncertain significance such as impaired tandem gait, questionable dystonic posturing, and mild memory impairment [4]. Nevertheless, the underlying mechanisms behind these differences remain unclear; it is ambiguous whether they signify distinct pathophysiological entities or merely represent a more advanced state along a disease continuum. Previous studies have identified shared familial aggregation and alcohol responsiveness between pure ET and ET plus, supporting the hypothesis that these conditions exist on a common disease spectrum [56]. Furthermore, electrophysiological evaluations have identified a similar tremor oscillator in both groups, providing additional evidence that they represent different manifestations within a spectrum of tremor disorders [7].

Despite its high prevalence, ET remains underdiagnosed and undertreated. Pharmacotherapy, including propranolol and primidone, is the first-line treatment for ET. However, long-term medication adherence rates remain suboptimal due to side effects and variable patient responses [8]. The concept of ET plus, a subtype encompassing additional neurological signs, has further complicated adherence research. ET plus patients often exhibit more severe symptoms and faster progression [59], which may theoretically increase adherence due to greater perceived need for treatment.

Medication adherence research holds profound implications across multiple dimensions. For patients, improved adherence enhances treatment efficacy, optimizes clinical outcomes, and ultimately elevates quality of life by ensuring consistent therapeutic benefits. At the societal level, understanding adherence patterns facilitates rational allocation of healthcare resources and informs evidence-based public health policy formulation. For healthcare systems, addressing non-adherence reduces avoidable medical expenditures, improves operational efficiency through targeted interventions, fosters multidisciplinary collaboration among clinicians, pharmacists, and behavioral specialists, and stimulates development of comprehensive adherence-enhancing strategies. Current research on medication adherence in ET remains substantially limited. Existing studies have primarily focused on Western populations [1011], while longitudinal follow-up data from Asian cohorts are particularly scarce. Notably, no studies have systematically compared medication adherence patterns between pure ET and ET plus in the international literature. Given the progressive nature of ET and the requirement for lifelong pharmacotherapy, this significant research gap urgently needs to be addressed.

This study aims to conduct a real world study on patients with pure ET and ET plus, in order to quantify medication adherence rates in a large Chinese ET cohort, compare adherence between pure ET and ET-plus subtypes, and identify clinical predictors of adherence. Our findings provide a comprehensive analysis of medication adherence in Chinese ET patients through a longitudinal follow-up investigation, offering insights for personalized interventions and highlighting unmet needs in ET management.

Methods

Results

Medication adherence in ET

87 ET patients were adherent to anti-tremor medications with a prevalence of 27.4%. ET plus patients had a higher odds of being adherent to medication compared to pure ET patients (32.2% vs 19.0%, χ² = 6.473, P = 0.011). As concerns the anti-tremor medications, 72 (22.6%) patients used arotinolol, 7 patients (2.2%) took propranolol, 6 patients (1.9%) took primidone, and 2 patients (0.6%) took trihexyphenidyl (shown in Figure 1). Among 231 ET patients with non-adherent medication, 97 patients (42.0%) discontinued treatment due to perceived inefficacy, 67 patients (29.0%) declined pharmacotherapy as they considered their symptoms mild and not significantly affecting daily life, 39 patients (16.9%) were influenced by other factors including economic constraints and adverse drug effects, and 28 patients (12.1%) reported difficulties in accessing medications due to transportation issues or residing in remote areas.

Figure 1

In comparison to patients with non-adherent medication, those with medication adherence were more likely to reside in urban areas (58.6% vs 44.6%, χ² = 4.982, P = 0.026) and had a higher prevalence of head tremor (42.5% vs 25.1%, χ² = 9.155, P = 0.002). Additionally, Mild cognitive impairment (40.2% vs 28.1%, χ² = 4.286, P = 0.038) was also significantly more prevalent in patients with adherent medication. Moreover, patients with adherent medication had higher scores in TETRAS-I (13 (8, 20) vs 15 (10, 22.5), Z = –1.990, P = 0.047), and TETRAS-II (17 (14, 21) vs 20 (15, 24.75), Z = –2.658, P = 0.008), as well as lower score in MoCA (26 (23, 27.5) vs 25 (21, 27), Z = –2.303, P = 0.021) than patients with non-adherent medication (Table 1).

Table 1

Comparison of demographic and clinical characteristics between the adherent and non-adherent patients to medication.

VARIABLES	NON-ADHERENT (n = 231)	ADHERENT (n = 87)	P VALUE
Gender (male, %)	113 (48.9%)	41 (47.1%)	0.776
Age (years)	62 (51.5, 69)	62 (54.5, 68)	0.646
Age at onset (years)	48 (35, 60)	52 (42.5, 60)	0.483
Duration (years)	10 (4, 20)	8 (4, 12)	0.239
Education level (years)	11 (6, 12)	9 (6, 12)	0.396
Area (urban/rural)	103/128	51/36	0.026
Tremor distribution
Head tremor (%)	58 (25.1%)	37 (42.5%)	0.002
Face tremor (%)	62 (26.8%)	25 (28.7%)	0.735
Voice tremor (%)	63 (27.3%)	27 (31.0%)	0.507
Lower limbs tremor (%)	57 (24.7%)	23 (26.4%)	0.747
ET plus (%)	137 (59.3%)	65 (74.7%)	0.011
Mild cognitive impairment	65 (28.1%)	35 (40.2%)	0.038
Rest tremor	34 (14.7%)	13 (14.9%)	0.960
Questionable dystonic posturing	50 (21.6%)	25 (28.7%)	0.184
Impaired tandem gait	42 (18.2%)	20 (23.0%)	0.335
TETRAS-I score	13 (8, 20)	15 (10, 22.5)	0.047
TETRAS-II score	17 (14, 21)	20 (15, 24.75)	0.008
MMSE score	28 (25, 29)	27 (24, 29)	0.094
MoCA score	26 (23, 27.5)	25 (21, 27)	0.021
NMSS score	11 (4, 20)	12 (6, 19)	0.607
Cardiovascular subscore	0 (0, 0)	0 (0, 0)	0.023
Sleep/fatigue subscore	3 (0, 6)	3 (0, 6)	0.541
Mood/cognition subscore	0 (0, 2)	0 (0, 3)	0.208
Perceptual problems subscore	0 (0, 0)	0 (0, 0)	0.302
Attention/memory subscore	3 (0, 4)	4 (0, 4)	0.140
Gastrointestinal subscore	0 (0, 1)	0 (0, 2)	0.255
Urinary subscore	0 (0, 2)	0 (0, 2)	0.456
Sexual function subscore	0 (0, 0)	0 (0, 0)	0.457
Miscellaneous subscore	0 (0, 2)	0 (0, 2)	0.927

[i] Note: TETRAS = Tremor Research Group Essential Tremor Rating Assessment Scale; MMSE, Mini-Mental State Examination; MoCA, Montreal Cognitive Assessment; NMSS = Non-motor Symptoms Scale.

Bold values represent statistically significant differences.

After adjusting for potential confounders in multivariable logistic regression, higher TETRAS-II scores (OR = 1.041, 95% CI = 1.001–1.082, P = 0.047), urban residence (OR = 1.775, 95% CI = 1.066–2.957, P = 0.028), and head tremor (OR = 1.936, 95% CI = 1.125–3.332, P = 0.017) were independently associated with improved adherence to anti-tremor medication (Table 2).

Table 2

Factors associated with adherence to anti-tremor medication.

INDEPENDENT VARIABLES	UNIVARIATE P–VALUE*	ODDS RATIO/95% CI	MULTIVARIATE P–VALUE
GENDER (MALE, %)	0.776	–	Not included
Age (years)	0.976	–	Not included
Age at onset (years)	0.303	–	Not included
Duration (years)	0.120	–	Not included
Education level (years)	0.979	–	Not included
Area (urban)	0.025	1.775/1.066–2.957	0.028
Head tremor	0.003	1.936/1.125–3.332	0.017
Face tremor	0.768	–	Not included
Voice tremor	0.507	–	Not included
Lower limbs tremor	0.747	–	Not included
Mild cognitive impairment	0.039	0.720/0.330–1.571	0.106
Rest tremor	0.960	–	Not included
Questionable dystonic posturing	0.186	–	Not included
Impaired tandem gait	0.336	–	Not included
TETRAS–I score	0.089		Not included
TETRAS–II score	0.007	1.041/1.001–1.082	0.047
MMSE score	0.125	–	Not included
MoCA score	0.048	0.978/0.891–1.074	0.144
NMSS score	0.792	–	Not included

[i] Note: TETRAS = Tremor Research Group Essential Tremor Rating Assessment Scale = MMSE, Mini-Mental State Examination; MoCA = Montreal Cognitive Assessment; NMSS = Non-motor Symptoms Scale.

Bold values represent statistically significant differences.

Discussion

In this real world study, we showed that medication adherence among ET patients is suboptimal. We reported a prevalence of adherence to anti-tremor medications to be 27.4%, which is lower than foreign studies [8101617]. This discrepancy may be partially explained by our extended follow-up duration of approximately 24 months, as sustained pharmacological treatment has been shown to progressively diminish medication adherence over time. Most ET patients exhibit poor adherence to medication, which may be attributed to limited effectiveness of these medications and the emergence of drug resistance [1819]. The leading causes of non-adherence in our study were perceived drug inefficacy, tolerable symptoms, access limitations, economic constraints, and adverse effect concerns. Furthermore, the available options for medications are relatively limited, and no new effective drugs have been developed in recent years.

The pharmacological management patterns observed in our Chinese ET cohort reveal several noteworthy findings regarding first-line treatment preferences. Arotinolol, propranolol, and primidone emerged as the most frequently prescribed medications, consistent with their status as first-line agents in Chinese treatment guidelines. However, the particularly high prescription rate of arotinolol (82.8% of adherent ET patients) warrants special consideration. This pattern stands in marked contrast to Western cohorts, where propranolol and primidone typically dominate first-line therapy [81120].

Several factors may explain the predominance of arotinolol in our cohort. A key reason is its superior tolerability compared to primidone. Although primidone is often effective, its use is frequently limited by a high incidence of dose-limiting adverse effects such as acute sedation, dizziness, and nausea. In contrast, β-blockers was associated with a more favorable side effect profile, suggesting that it may represent a more sustainable treatment option for a broader ET patient population. Beyond tolerability, drug accessibility serves as another critical factor influencing real-world treatment decisions. β-blockers are more readily accessible within the Chinese healthcare system than primidone. A notable advantage of arotinolol, in particular, lies in its dual blockade of α1- and β-adrenoceptors. This distinctive pharmacological characteristic enables partial mutual neutralization of adverse effects typically associated with selective receptor blockade. For example, its α1-blocking activity may counteract the peripheral vasoconstriction and heart rate reduction induced by beta-blockers, while also mitigating adverse effects on glucose and lipid metabolism. In addition, arotinolol demonstrates limited permeability across the blood-brain barrier, a characteristic that differentiates it from propranolol. Consequently, central nervous system related adverse events—such as dizziness, fatigue, and mental depression—occur significantly less frequently [2122]. Given these considerations, arotinolol has become the most commonly used medication for ET in China. Future large-scale, prospective studies are warranted to directly compare the long-term efficacy and adherence rates between these agents.

It is also noteworthy that while our cohort included a substantial number of patients with head tremor, the severity was generally mild and did not necessitate treatment with botulinum toxin injections or any other non-oral treatments (such as deep brain stimulation or magnetic resonance image guided focused ultrasound). Following our center’s clinical protocol, botulinum toxin therapy is typically reserved for patients with more severe, disabling, or medication-resistant head tremor. Since the patients in our study population predominantly had mild symptoms and were seeking first-line oral pharmacological treatment, botulinum toxin was not utilized.

Our study further highlights significant variations in adherence between ET subtypes, with ET plus patients demonstrating higher adherence (32.2%) compared to pure ET patients (19.0%). This discrepancy may be attributed to the more severe and multifaceted symptomatology of ET plus, which includes mild memory impairment, impaired tandem gait, and other neurological signs, thereby increasing patients’ perceived need for treatment. However, the high prevalence of mild cognitive impairment in ET plus patients may have a dual—and paradoxical—effect on adherence. On one hand, cognitive deficits can increase dependence on caregivers for medication management; on the other hand, they may compromise the patient’s own ability to adhere to treatment regimens [23]. Patients with mild cognitive impairment may be more likely to persist with treatment due to heavier disease burden or more consistent caregiver involvement. Indeed, patients with more pronounced cognitive decline often receive greater assistance and supervision with medications, which may contribute to higher adherence rates. Conversely, pure ET patients, who typically exhibit fewer motor and non-motor symptoms, may perceive their tremor as less debilitating, leading to lower adherence.

We revealed a significant relationship between tremor severity and medication adherence, with each 1-point increase in TETRAS-II scores associated with a 4.1% increase in the odds of medication adherence. This relationship suggests that patients with severe tremor symptoms are more likely to recognize the necessity of treatment due to significant impacts on daily living activities, whereas those with milder symptoms may underestimate the need for medication as their condition causes minimal disruption to daily functioning. This aligns with previous research demonstrating that disease severity is a key determinant of adherence in chronic neurological conditions. We further identified head tremor and urban residence as two critical determinants of medication adherence in ET patients, which may be mechanistically explained by the following observations. As a highly visible motor manifestation, head tremor may significantly affect patients’ physical appearance and social presentation. This “visible” symptom likely induces psychological distress (e.g., social anxiety or self-image concerns), thereby motivating patients to maintain medication adherence more rigorously to mitigate its cosmetic impact. Urban dwellers may benefit from structural advantages in healthcare accessibility, including: greater density of medical facilities reducing travel burden for prescription refills, and more robust health insurance coverage and medication supply systems. These systemic factors may collectively promote long-term treatment persistence among urban patients.

Our study did not identify an independent association between ET Plus and medication adherence, supporting the hypothesis that ET plus and pure ET may not represent distinct clinicopathological entities. This finding is consistent with recent comparative studies on thalamic deep brain stimulation, which have reported similar tremor-suppressing effects in patients with ET and those with ET plus, indicating comparable therapeutic efficacy between the two groups [242526]. These outcomes align with the disease staging theory proposed by Lenka and Louis [27], which posits that ET plus may correspond to a specific stage within the spectrum of disease progression.

This study provides critical insights into medication adherence among Chinese ET patients, revealing that patients were more likely to adhere to anti-tremor medication if they had a higher TETRAS-II score, resided in urban areas, or exhibited head tremor. The higher adherence in ET plus patients underscores the need for subtype-specific management strategies, while the predominance of arotinolol highlights regional disparities in treatment practices. In future, the development of novel therapies could significantly improve adherence and outcomes for ET patients.

This study possesses several notable strengths. Its prospective design and extended longitudinal follow-up period minimize recall bias and allow for a more robust assessment of clinical progression over time. Furthermore, our rigorous approach to differentiating classic ET from ET plus phenotypes enhances the diagnostic accuracy and provides valuable insights into the potential heterogeneity of the disorder. Despite these strengths, several limitations must be acknowledged. First, the single-center nature of our study may limit the generalizability of our findings. The participant population was recruited from a tertiary referral center, which may not be fully representative of the broader community-based ET population, potentially introducing selection bias. Second, medication adherence was assessed solely through self-report. The incorporation of objective measures, such as electronic pill monitors or drug level assays, in future studies would provide a more reliable assessment. Additionally, although we employed standardized rating scales, the assessment of tremor characteristics remains subject to a degree of inherent subjectivity. Future research should aim to replicate these findings in larger, multi-center cohorts to enhance external validity. Longitudinal studies incorporating objective adherence monitoring and advanced neuroimaging biomarkers are also warranted to further elucidate the underlying mechanisms of disease progression and treatment response in ET.

Conclusions

Medication adherence among Chinese ET patients remains suboptimal. Pure ET patients demonstrated markedly poorer adherence compared to ET plus patients. Disease severity, urban residence, and the presence of head tremor were identified as significant predictors of medication adherence in ET patients. Our study highlights the urgent need to develop personalized medication adherence intervention strategies.

Competing Interests

The authors have no competing interests to declare.