Potentially inappropriate medications among older patients with Parkinson’s disease: a cross-sectional analysis of a national health insurance database in China

Parkinson’s disease (PD) is the one of the most common age-related neurodegenerative disorders [1, 2], the incidence of which sharply increases after 60 years of age [3]. While the progress of motor impairment can be taken under control with antiparkinsonian drugs, cognitive impairment, such as cognitive abnormalities, autonomic dysfunction, sleep disorders, and mood disorders, affecting 75% of PD patients who had a PD disease course of more than ten years, is deemed as one of the most distressing complications of PD for patients and their caregivers [4‐8]. Alleviating cognitive symptoms using pharmaceutical treatment plays an extremely significant role to promote the quality of life of PD patients [7, 8]. However, prescribing medications for patients with PD needs extra deliberations as adverse drug-drug and drug-disease interactions can exacerbate PD symptoms and aggravate healthcare burden [9]. Hence, it is of great importance to evaluate the appropriateness of medication use among PD patients. The American Geriatrics Society (AGS) Beers criteria list potentially inappropriate medications (PIMs) that should be avoided or used with caution among older patients [10‐12]. Previous studies observed that 35.8% of older patients with PD were exposed to PIMs in the US [13] and patients using PIMs had longer duration of hospital stay [14].

China has a rapidly ageing population [15] and 1.7% of older adults aged 65 years and above were diagnosed with PD [16]. It is estimated that the number of PD patients in China will reach 4.94 million by 2030, accounting for half of the global figure [17]. Though studies have evaluated PIM prescribing for the general older population in China [16, 17], older adults living with PD experience even heightened health risks due to their complex medical needs and higher susceptibility to PIM use. However, there is scarce literature investigating PIM use in this population in China and studies from abroad, mainly conducted in the United States, were unlikely to be generalizable to the Chinese context. Our study aimed to quantify the prevalence of PIM use and identify its risk factors among the older individuals with PD in China.

We conducted a cross-sectional analysis of PIM use among older PD patients in China between 2015 and 2017 based on a national representative insurance database. We observed that 57.8% of older adults with PD were exposed to at least one PIM, among which 17.1% and 42.9% were exposed to PIMs that could exacerbate their motor and cognitive impairments. Medications for the central nervous system and anticholinergics were most commonly used PIMs for older PD patients. Increasing age, female sex, being insured by URRBMI, and residing in the eastern or central regions were positively associated with the prevalence of PIM use.

The prevalence of PIM use among older PD patients was 57.8% in our study, which was much higher than that in the US (35.8%) [13], suggesting higher risks entailed in the pharmaceutical treatment for older PD patients in China. The prevalence of PIM use also differed between PD patients and the general population in China, with the latter being 29% as reported by a previous meta-analysis [28]. A total of 17.1% of sample patients had at least one PIM prescription that could aggravate their PD-related motor symptoms, which is double the estimate seen in a previous study in the US (8.7%) [13]. Our findings also revealed that 42.9% of older PD patients were exposed to at least one prescription containing PIMs that could exacerbate their cognitive impairment, also much higher than that in the US (29.0%) [13]. The potential cause of these discrepancies might be that pharmaceutical treatments for common cognitive symptoms of PD (e.g., antipsychotics for psychosis and benzodiazepines for insomnia) usually are PIMs that entail high risks of motor and/or cognitive impairment.

Most of the motor-impairing PIMs identified in our study are antipsychotic agents [29], which might also impair patients’ cognition. This is in line with the high prevalence of psychosis, with up to 60% of PD patients suffering from it [30]. Quetiapine and clozapine were the only two antipsychotics recommended for PDP by the AGS Beers criteria [12]. However, the most commonly used atypical antipsychotics in PD patients were olanzapine and risperidone in our study. As a previous study showed that olanzapine and risperidone were the top two consumed antipsychotics in China [31], this may indicate that clinicians did not differentiate between treatments of psychosis in the general population and PD patients. Besides, trihexyphenidyl, an anticholinergic anti-PD drug, was also a commonly used PIM, though it should be avoided in patients aged 60 and above for its possible cognitive impairment, as suggested by both the AGS Beers criteria and the Chinese guidelines for PD management [9, 32, 33]. Nevertheless, 16.5% of older adults with PD used trihexyphenidyl in our study, which was even higher than that among patients aged under 50 and with young-onset PD in China (14%) [34].

Our study also revealed that female PD patients were at greater risks of receiving PIMs than males, which was consistent with previous studies conducted in PD patients and in the general population [35, 36]. Epidemiological evidence indicated that affective disorders were more prevalent as comorbidities in female PD patients [37]. Therefore, antidepressants and benzodiazepines (BZDs) were more commonly used among females [38], which might partially explain the greater odds of receiving motor- and cognition-impairing PIMs of females. Besides, pharmaceutical treatment for older adults should be more cautious as aging is inextricably linked to the alterations in pharmacokinetics and pharmacodynamics, which may exacerbate drug-related problems [39, 40]. In this study, the overall prevalence of PIM use increased with age, however, the prevalence of motor- and cognition-impairing PIMs showed an inverse association with age. This may indicate that clinicians took more caution when prescribing medications likely to cause motor or cognitive impairments for older persons with PD.

Our results imply that medications’ potential harms or contra-indications for older adults with PD could to some extent be easily overlooked. Efforts shall be made to assess the appropriateness of medications prescribed for this already vulnerable population. For example, educational programs can keep clinicians updated about PIMs and competent in deciding the most appropriate mediations for older adults with PD. In other hand, prescribers are often well intentioned in their treatment but sometimes patients have symptoms that require treatment even if it comes with some degree of risks. For example, patients with behavioral issues that risk self-harm may require antipsychotics even if it will blunt the effects of their PD therapy. Pharmacists should be proactively involved in assessing medications prescribed as pharmaceutical therapy plays a cardinal role in PD treatment. Moreover, an expert guidance or consensus on medications to protect cognitive function of PD patients can also help clinicians better prescribe medications, which would in turn alleviate the burden of disease and health care expenditures [41‐43]. For example, quetiapine and clozapine, which have priority over other antipsychotics for patients with PD, could be recommended as the first choice for controlling their psychosis symptoms.

To our best knowledge, this is the first study to analyze the prevalence and risk factors of PIM use among older adults with PD in China using a nationally representative health insurance database. The present study focused not only on the prevalence of general PIMs that pose potential risks for all older individuals but also on PD disease-specific PIMs. It’s worth noting that PIMs posing motor-impairment and cognition-impairment risks contributed greatly to the overall PIM use among older PD patients, which is in urgent need for clinical intervention. Several limitations of this study should be noted. First, due to the lack of information about the patients’ creatinine clearance, drug doses, and duration in our database, we excluded some PIMs listed in the 2015 AGS Beers criteria (Table S1). Therefore, the overall prevalence of PIM use may be underestimated. Second, owing to the limitations of the database, many factors that might have affected clinicians’ prescribing and drug choice, such as patients’ medical history, cognitive status, economic status, and long-term treatment patterns were not measured. Therefore, we were unable to identify an exhaustive list of predictors of PIM use or and clarify their mechanisms of action. Third, we did not include Traditional Chinese Medicines (TCMs) in our study because the Beers criteria did not contain recommendations for TCMs.