Prevalence and risk factors of falls among older adults with Parkinson’s disease in a Malaysian tertiary centre

Abstract

Background

Falls are a major source of morbidity in Parkinson’s disease (PD) and contribute substantially to fractures, disability, and reduced quality of life. Although both motor and non-motor factors contribute to fall risk, data from Southeast Asian populations remain limited.

Objective

To determine the prevalence of falls and identify clinical factors associated with falls among older adults with Parkinson’s disease attending a tertiary hospital falls clinic in Malaysia.

Methods

This cross-sectional study included older adults with Parkinson’s disease attending a tertiary falls clinic. Participants were classified as fallers if they reported at least one fall within the preceding 12 months. Clinical assessments included demographic characteristics, disease severity, depressive symptoms, cognitive status, orthostatic blood pressure measurements, functional mobility, and frailty-related parameters including body mass index (BMI). Multivariable logistic regression analysis was used to identify factors independently associated with falls.

Results

Falls were reported by 67% of participants with Parkinson’s disease. In multivariable analysis, depressive symptoms, orthostatic hypotension, and lower BMI were independently associated with falls.

Conclusion

Falls were highly prevalent among older Malaysian adults with Parkinson’s disease. Non-motor features, including depressive symptoms and autonomic dysfunction, together with markers of physiological vulnerability such as low BMI, were independently associated with falls. Recognition of these potentially modifiable factors may inform targeted fall prevention strategies in Parkinson’s disease.

Introduction

Falls are a major public health concern among older adults and represent a leading cause of injury-related morbidity and disability worldwide. Approximately one-third of adults aged 65 years and older experience at least one fall annually, with substantial consequences including fractures, hospitalization, and loss of independence [1]. Individuals with Parkinson’s disease (PD) experience an even greater risk of falls due to the combined effects of motor impairment, postural instability, and progressive neurodegeneration.

Epidemiological studies suggest that 45–68% of individuals with Parkinson’s disease experience at least one fall each year, and recurrent falls occur in approximately one-third of patients [2–4]. Falls in PD are associated with significant adverse outcomes, including fractures, fear of falling, functional decline, and increased healthcare utilization [5]. While motor features such as postural instability, bradykinesia, and freezing of gait are well-established contributors to falls, growing evidence indicates that non-motor symptoms also play a critical role in fall vulnerability [6–8].

Autonomic dysfunction, particularly orthostatic hypotension, affects approximately 30–50% of individuals with Parkinson’s disease and may contribute to falls through dizziness, impaired cerebral perfusion, and reduced postural stability [9]. In addition, frailty and sarcopenia have emerged as important contributors to falls among older adults with neurodegenerative diseases by reducing muscle strength, balance control, and physiological reserve [10, 11]. Regional factors may also influence fall risk. Studies among older adults in Southeast Asia have reported substantial fall prevalence and multiple associated risk factors, highlighting the importance of examining fall risk within local contexts [12]. However, evidence examining fall prevalence and associated factors among individuals with Parkinson’s disease in Southeast Asian populations remains limited.

Therefore, this study aimed to determine the prevalence of falls and identify clinical factors associated with falls among older adults with Parkinson’s disease attending a tertiary hospital falls clinic in Malaysia. We hypothesized that non-motor factors including depressive symptoms, autonomic dysfunction, and markers of physiological vulnerability would be associated with increased fall risk.

Methods

Study design and setting

This cross-sectional study included older adults with Parkinson’s disease attending a tertiary hospital falls clinic from August to December 2022. Ethical approval was obtained from the Medical Research Ethics Committee (MREC), University of Malaya Medical Centre (Approval No: MREC 2020627-11343). Participants were recruited during routine clinical evaluation. Individuals were eligible if they had a clinical diagnosis of Parkinson’s disease and were able to participate in clinical assessments. Patients with marked cognitive impairment, which was considered likely to limit reliable history-taking and completion of study assessments, were excluded (Fig. 1).

Fig. 1.

Flow chart on patient recruitment and clinical assessment

Ethical approval was obtained from the institutional ethics review committee and the study was conducted in accordance with the Declaration of Helsinki. Written informed consent was obtained from all participants prior to study enrolment.

Participants

Inclusion criteria

• Age ≥ 65 years

• Idiopathic PD (UK Brain Bank criteria)

• Regular follow-up at University of Malaya Medical Centre (UMMC)

Exclusion

• Terminal illness

• Severe frailty (CFS = 9)

• Marked Cognitive Impairment

Total sample: 138 participants.

Falls assessment

Participants were classified as fallers if they reported at least one fall within the previous 12 months, in line with definitions commonly used in epidemiological studies of falls. Although recurrent falls, defined as two or more falls, represent a more severe phenotype in Parkinson’s disease, even a single fall in older adults is clinically significant and linked to a higher risk of injury and future falls. While recurrent falls (≥ 2 falls) indicate a more severe phenotype, this study focused on the occurrence of any fall due to limitations in retrospective fall reporting.

Clinical assessments

Clinical assessments were selected to capture domains previously associated with falls in Parkinson’s disease while remaining feasible within the routine clinic workflow.

Cognitive function

Cognitive status was evaluated by a physiotherapist and an occupational therapist. Participants with severe cognitive impairment were excluded because such impairment was deemed likely to hinder reliable clinical history-taking and the completion of structured clinical assessments necessary for the study. We did not employ this threshold to establish a formal clinical severity category.

Depressive symptoms

Depressive symptoms were assessed using the Geriatric Depression Scale-15 (GDS-15).

Functional status

Functional status was assessed using Activities of Daily Living (ADL), which evaluate an individual’s ability to perform basic self-care tasks including bathing, dressing, toileting, transferring, continence, and feeding.

Functional mobility

Functional mobility was evaluated using the Timed Up and Go (TUG) test, a validated measure of mobility and balance.

Orthostatic blood pressure

Orthostatic blood pressure was assessed during transition from a seated to standing position, in keeping with routine clinic workflow. This seated-to-standing approach was used to capture postural blood pressure change during routine evaluation, although it may underestimate orthostatic hypotension compared with standardized supine-to-standing protocols.

Frailty-related parameters

Body mass index (BMI) was recorded as a marker of nutritional status and potential sarcopenic vulnerability.

Statistical analysis

Univariate analysis used t-tests or χ² tests. Variables with p < 0.25 were entered into a multivariate logistic regression using backward likelihood ratio. Age and sex were forced covariates. Adjusted ORs with 95% CIs were reported.

Results

Participant characteristics

Baseline demographic and clinical characteristics of participants are presented in Table 1. Falls were reported by 67% of participants, indicating a high prevalence of falls among older adults with Parkinson’s disease attending the tertiary falls clinic. Mean age was 74.5 ± 6.0 years; 52.7% were female.

Table 1.

Demographics of fallers and non-fallers in Parkinson’s disease

Characteristic	Fallers (n = 93)	Non-Fallers (n = 45)	p-value
Age (years), mean ± SD	75.27 ± 6.0	72.8 ± 5.75	0.025*
Gender, n (%)			0.364
• Male	44 (47.3%)	25 (55.6%)
• Female	49 (52.7%)	20 (44.4%)
Ethnicity, n (%)			0.095
• Chinese	55 (59.1%)	26 (57.8%)
• Indian	19 (20.4%)	3 (6.7%)
• Malay	14 (15.1%)	12 (26.7%)
• Other	5 (5.4%)	4 (8.9%)
Residence, n (%)			0.636
• Home	85 (91.4%)	40 (88.9%)
• Nursing Home	8 (8.6%)	5 (11.1%)
Duration of PD, n (%)			<0.001*
• <5 years	14 (15.1%)	22 (48.9%)
• 5–10 years	51 (54.8%)	19 (42.2%)
• >10 years	28 (30.1%)	4 (8.9%)
Hoehn and Yahr Stage, n (%)			<0.001*
• Stage 1	1 (1.1%)	5 (11.1%)
• Stage 2	13 (14.0%)	30 (66.7%)
• Stage 3	47 (50.6%)	6 (13.3%)
• Stage 4	27 (29.0%)	3 (6.7%)
• Stage 5	5 (5.4%)	1 (2.2%)
BMI Category, n (%)			0.001*
• <18.5 kg/m²	30 (32.3%)	5 (11.1%)
• 18.5–22.9 kg/m²	57 (61.3%)	28 (62.2%)
• >23 kg/m²	6 (6.5%)	12 (26.7%)
Number of Medications, n (%)			0.015*
• <5 medications	34 (36.6%)	28 (62.2%)
• 5–7 medications	51 (54.8%)	16 (35.6%)
• >7 medications	8 (8.6%)	1 (2.2%)

^*Statistically significant at p < 0.05

Multivariable logistic regression analysis identifying independent predictors of falls is presented in Table 2 in multivariable logistic regression analysis, depressive symptoms, orthostatic hypotension, and lower BMI were independently associated with falls.

Table 2.

Clinical assessment of fallers and non-fallers

Clinical Variable	Fallers (n =93)	Non-Fallers (n =45)	p-value
Frailty Score, mean ± SD	3.96 ± 1.26	2.47 ± 1.26	<0.001*
Lowton ADL Score, mean ± SD	7.24 ± 3.60	11.4 ± 3.35	0.248
Geriatric Depression Scale (GDS), mean± SD	5.63 ± 2.56	3.20 ± 1.75	0.004*
Timed Up and Go (seconds), mean ± SD	18.9 ± 4.84	14.2 ± 3.37	0.110
Mini Mental State Examination (MMSE),mean ± SD	17.8 ± 5.69	24.3 ± 5.01	0.056
Postural Hypotension, n (%)	48 (51.6%)	8 (17.8%)	<0.001*

^*Statistically significant at p < 0.05

These findings indicate that autonomic dysfunction and markers of physiological vulnerability were associated with falls in this population.

As shown in Table 1 fallers were older (p = 0.025), had longer disease duration (p < 0.001), more advanced H&Y stage (p < 0.001), lower BMI (p = 0.001), and higher GDS scores (p = 0.004). Postural hypotension was significantly more common among fallers (51.6% vs. 17.8%, p < 0.001).

Frailty scores were significantly higher in fallers (p < 0.001). TUG showed non-significant trends.

Multivariate analysis

After adjustment, the results showed that:

• GDS independently predicted falls (OR 1.905, p < 0.001)

• Postural hypotension increased fall risk fourfold (OR 4.08, p = 0.034)

• Lower BMI independently predicted falls (OR 0.301, p = 0.039)

Age, sex, disease duration, H&Y stage, ADL, TUG, and polypharmacy were not independently associated.

Discussion

In this study, we observed a high prevalence of falls among older adults with Parkinson’s disease attending a tertiary falls clinic, with approximately two-thirds of participants reporting at least one fall in the preceding year. This finding is consistent with previous epidemiological studies indicating that 45–68% of individuals with Parkinson’s disease experience falls annually [2–4]. In comparison, fall prevalence among community-dwelling older adults without Parkinson’s disease is generally estimated at approximately 30%, highlighting the substantially increased fall risk associated with PD [1]. This elevated risk reflects the complex interaction between motor impairment, autonomic dysfunction, and reduced physiological reserve (Table 3).

Table 3.

Logistic regression analysis of risk factors for falls in Parkinson’s disease

Variable	Adjusted OR	95% CI	p-value
Age	0.988	0.881 – 1.108	0.835
Race (Chinese vs others)	1.954	0.820 – 4.658	0.131
Duration of PD	0.428	0.116 – 1.580	0.206
Hoehn and Yahr Stag	1.972	0.688 – 5.654	0.206
Geriatric Depression Scale (GDS)	1.905	1.316 – 2.757	<0.001*
Postural Hypotension	4.08	1.07 – 15.6	0.034*
BMI (kg/m²)	0.301	0.099 – 0.913	0.039*
Number of Medications	1.585	0.575 – 4.370	0.374
MMSE	0.805	0.646 – 1.002	0.052
Lowton ADL Score	0.930	0.640 – 1.350	0.702
Timed Up and Go (TUG)	0.991	0.759 – 1.293	0.946

^*Statistically significant at p < 0.05

Our findings support the concept that falls in Parkinson’s disease are multifactorial. Although motor features such as postural instability and gait dysfunction are well recognized contributors to falls, increasing evidence suggests that non-motor symptoms also play an important role in fall vulnerability [6–8]. In the present study, depressive symptoms were independently associated with falls. Depression is highly prevalent in Parkinson’s disease and has been linked to impaired attentional control, psychomotor slowing, and reduced balance confidence, which may adversely affect gait stability and postural control [6]. Additionally, depressive symptoms may contribute indirectly to fall risk through reduced physical activity and progressive deconditioning. Recent studies have further highlighted the role of neuropsychiatric symptoms as contributors to mobility impairment and fall risk in Parkinson’s disease [7].

Orthostatic hypotension was also independently associated with falls in our cohort. Autonomic dysfunction is a common non-motor manifestation of Parkinson’s disease and may affect up to half of individuals with PD [9]. Orthostatic hypotension may lead to transient cerebral hypoperfusion, dizziness, and impaired postural adaptation during position changes, thereby increasing susceptibility to falls. Longitudinal studies have demonstrated that orthostatic hypotension may predict future falls and mobility decline in Parkinson’s disease, underscoring the importance of recognizing autonomic dysfunction during fall risk assessment [13].

Lower body mass index (BMI) was another factor associated with falls in this study. Low BMI in older adults with Parkinson’s disease may reflect underlying frailty, sarcopenia, or nutritional vulnerability. Frailty represents a state of reduced physiological reserve and diminished resilience to stressors, while sarcopenia contributes to reduced muscle strength and impaired balance responses [10, 11]. In this context, reduced postural reserve refers to diminished neuromuscular capacity to maintain upright stability and respond effectively to internal or external balance disturbances.

Although gait impairment is widely recognized as a major contributor to falls in Parkinson’s disease, detailed gait analysis was not performed in this study. Functional mobility was assessed using the Timed Up and Go test, which provides a practical measure of mobility within routine clinical practice but does not capture specific gait abnormalities such as freezing of gait, gait variability, or impaired turning. Recent studies using wearable sensor technology have demonstrated that subtle gait abnormalities and increased gait variability are strongly associated with fall risk in Parkinson’s disease [14]. Future studies incorporating objective gait analysis may provide further insights into mechanisms underlying falls in this population.

The high prevalence of falls observed in this study also highlights the importance of examining fall risk in regional contexts. Data on falls among individuals with Parkinson’s disease in Southeast Asia remain limited. However, studies among older adults in the region have reported substantial fall prevalence and multiple associated risk factors [12]. Differences in environmental conditions, healthcare access, caregiving structures, and comorbidity patterns may influence fall risk across populations. Understanding these contextual factors is important for developing targeted fall prevention strategies tailored to local healthcare settings.

Taken together, our findings support the concept that falls in Parkinson’s disease arise from the interaction of motor impairment, non-motor symptoms, and reduced physiological reserve. Effective fall prevention strategies therefore require a multidimensional approach addressing not only mobility impairment but also autonomic dysfunction, mood disorders, and frailty-related vulnerability.

Strengths and limitations

This study offers insight into the prevalence and related factors of falls among older adults with Parkinson’s disease in a Southeast Asian population, where data remain relatively scarce. The study also included assessment of both motor and non-motor domains relevant to fall risk within a routine clinical setting. However, several limitations should be noted. The study did not distinguish between single and recurrent falls. Recurrent falls are an important clinical feature in Parkinson’s disease and should be explored in future prospective studies. The cross-sectional design restricts causal inference between identified factors and falls. Detailed gait analysis was not conducted, so specific gait abnormalities like freezing of gait could not be examined. Anxiety measures were also not collected, although anxiety might influence gait behaviour, balance confidence, and fear of falling. Additionally, orthostatic blood pressure was measured using a seated-to-standing protocol aligned with clinical workflow rather than a standardised supine-to-standing assessment, which may underestimate orthostatic hypotension. Excluding participants with severe cognitive impairment have resulted in under-representation of individuals with greater cognitive impairment and potentially higher fall risk, possibly affecting the reported prevalence and associated factors. Future longitudinal studies incorporating comprehensive gait assessment and autonomic testing may further reveal mechanisms underlying falls in Parkinson’s disease.

Conclusion

Falls were highly prevalent among older adults with Parkinson’s disease attending a tertiary falls clinic in Malaysia. Depressive symptoms, orthostatic hypotension, and lower BMI were independently associated with falls. These findings support the inclusion of non-motor symptoms and markers of physiological vulnerability in fall risk assessment among older adults with Parkinson’s disease. Multidimensional approaches addressing both motor and non-motor contributors may help inform fall prevention strategies in this population.

Authors’ contributions

NIS conceptualised the study and supervised data analysis. JQL and LS collected data and performed preliminary analyses. All authors contributed to manuscript drafting and approved the final version.

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

Approved by the Medical Research Ethics Committee (MREC), University of Malaya Medical Centre (MREC 2020627-11343). Conducted in accordance with the Declaration of Helsinki. Written informed consent was obtained from all participants.

Competing interests

The authors declare no competing interests.