Abstract
Summary
Despite growing evidence on prodromal symptoms of Parkinson’s disease, it remains unknown whether they culminated in clinically significant events. We found elevated risks of hip fractures from 3 years before until up to 10 years after diagnosis of Parkinson’s disease, underscoring the need for early and continuous fracture risk management.
Purpose
Parkinson’s disease (PD) is associated with a higher risk of hip fracture. However, the risk of hip fracture at the prodromal stage of PD is unknown. Our study aimed to investigate the incidence of hip fractures in persons with and without PD in Finland from 5 years before to 10 years after the diagnosis of PD.
Methods
We included persons diagnosed with PD between 2000 and 2009 in Finland and 1:2 matched comparison cohort. Hazard ratio [HR] was computed to assess overall risk of incident hip fracture. The annual incidence rate per 1000 person-year [IR/1000PY] and incidence rate ratios [IRR] were calculated for each year of follow-up from 5 years before to 10 years after PD diagnosis.
Results
During the follow-up among 33,153 eligible persons, 13.4% persons with and 5.3% persons without PD had an incident hip fracture (HR 1.82, 95% confidence interval [CI] 1.47–2.26). Persons with PD had higher annual incidence rates of hip fracture starting from 3 years before diagnosis (with PD: IR/1000PY 2.83, 95% CI 1.93–4.02; without PD: IR/1000PY 1.64, 95% CI 1.15–2.28). The higher annual incidence rates continued until 10 years after diagnosis (with PD: IR/1000PY 15.7, 95% CI 11.7–20.5; without PD: IR/1000PY 4.53, 95% CI 3.08–6.44). Prevalence of fall-risk-increasing medications but not anti-osteoporosis medications was higher at hip fractures among persons with PD than without PD.
Conclusion
As the risk of hip fracture is elevated already years before the diagnosis of PD, it is important to manage fall risks concomitantly while applying the diagnostic procedures for PD.
Supplementary Information
The online version contains supplementary material available at 10.1007/s00198-025-07632-9.
Keywords: Antiresorptive medication, Fall-risk-increasing medications, Hip fracture, Parkinson’s disease
Introduction
Parkinson’s disease (PD) is the second most common neurodegenerative disease after Alzheimer’s disease [1]. More than 1% of people aged 60 years and older have PD [2]. The prevalence of PD more than doubled from 1990 to 2016, with a 21% increase in the age-standardized rate [3]. The disability-adjusted life years of PD is predicted to more than double from 2022 to 2050 [4].
PD is characterized by a mixture of motor and non-motor symptoms. Motor symptoms include tremors, bradykinesia, rigidity and postural instability; non-motor symptoms include constipation, postural hypotension, sleep problems, depressive and psychotic symptoms [5]. Furthermore, non-motor symptoms are also common during the prodromal period [5].
Compared to healthy controls, persons with PD have at least double the risk of fractures due to falls and osteoporosis [6, 7]. Hallmark symptoms of Parkinson’s disease, including gait and balance impairment, increase the risk of falls [8, 9]. A systematic review of 14 studies showed that persons with PD have lower bone mineral density [6]. Poor bone health may arise from the pathogenesis of PD, physical inactivity, lower body weight and vitamin D deficiency [10, 11]. The elevated risk is further evidenced by the inclusion of PD in different data-driven fracture prediction models [12, 13].
Among all fragility fractures, hip fractures are the most costly in terms of direct treatment, surgery and rehabilitation [14]. Half of the people with hip fractures have increased dependence and a quarter are deceased after 1 year [15, 16]. Quantifying the risk of hip fractures among persons with PD is important for better health policy and planning.
Previous studies reported a twofold to fourfold increase in risk of hip fractures among persons with PD compared to persons without [6, 7]. Geographical variation in the risk of hip fractures has been reported [17, 18], with the highest age-standardized rate of hip fracture reported among Nordic countries [17]. Apparent risk differences may also arise due to differences in follow-up duration because PD is a progressive disease [19–21]. Motor and non-motor prodromal symptoms may also result in increased use of fall-risk-increasing medications, thus impacting the risk of hip fracture before diagnoses of PD [22–24]. However, it remains uncertain how the risk of hip fracture changes across the course of the disease.
Our study aimed to investigate the incidence of hip fractures in persons with and without PD in Finland from 5 years before to 10 years after the diagnosis of Parkinson’s disease.
Methods
Data sources
Our study was conducted using data from Finland’s nationwide healthcare registers [25]. All residents in Finland have a unique personal identification number that allow linkage of information to other national health and social care registers. Diagnoses made during hospital admissions and specialized healthcare outpatient visits were retrieved from Care Register for Healthcare. All diagnoses were coded according to the International Classification of Diseases and Related Health Problems, 8th (ICD-8), 9th (ICD-9) and 10th Revision (ICD-10), during the years of 1972–1986, 1987–1995 and 1996–2019 respectively. Records of reimbursed medications dispensed in community pharmacies were retrieved from Prescription Register and death information was retrieved from Statistics Finland.
Study population
We analyzed data from a nationwide matched cohort of persons with and without PD in Finland (Finnish Study on Parkinson’s disease [FINPARK]) [26]. Persons with PD aged 35 years and above were identified for inclusion in the cohort based on eligibility for reimbursement of PD medications with ICD-10 code for PD (G20) recorded in the national Special Reimbursement Register (SRR) between 1996 and 2015. To be included in the SRR, each person’s PD diagnosis (ICD-10 code G20) is confirmed in neurological specialist settings. Diagnostic information for each person is reviewed and confirmed by the Social Insurance Institution. During the study period, diagnostic requirements were consistent with that of United Kingdom Parkinson’s Disease Society Brain Bank criteria [27].
To improve the specificity of identifying persons with PD, we excluded persons aged younger than 35 years old at diagnosis. In addition, persons diagnosed with other extrapyramidal and movement disorders, systemic atrophies affecting central nervous system and other cognitive disorders and neurodegenerative diseases within 2 years of index date were excluded. This resulted in a cohort of 22,189 persons with PD diagnosed between 1996 and 2015.
For each person with PD, the date of PD diagnosis was defined as the index date and two matched comparison persons without PD were identified with the following matching criteria: same sex, age ( 1 year) and region of residence at the index date. Inclusion, exclusion and matching criteria of the study population have been described in detail previously [24, 26].
The study plan for FINPARK study was approved by the national health and social data permit authority Finnish Social and Health Data Permit Authority (FINDATA). Ethics approval and informed consent were not required because the study was conducted under the National Act on Secondary Use of Health and Social data. This applied because study participants were not contacted and all data were pseudonymized before being provided to the research team.
Follow-up for hip fracture
In this study, follow-up for hip fractures started from 5 years before until 10 years after each person’s index date. Individuals were censored at hip fracture or other censoring events, including death and additionally, diagnosis of PD for comparison cohort. We restricted our cohort to persons diagnosed with PD between 2000 and 2009 to ensure 10-year follow-up after index date for each study participant. This resulted in 11,141 persons with PD diagnosed between 2000 and 2009 and 74,242 age- and sex-matched comparison persons.
Person who had a hip fracture more than 5 years before their index date and all their matched counterparts were also excluded. Hip fractures were identified from hospital discharges of the Care Register for Health Care using ICD-8 (820), ICD-9 (820) and ICD-10 (S72.0-S72.2) codes. Persons without matched counterparts or excess comparison persons (> 2) matched for the same persons with PD were further excluded. Details for the inclusion and exclusion are presented in a flow diagram (Fig. S1).
Characteristics at baseline and hip fracture
Baseline comorbidities were identified from the SRR since 1972 until the start of follow-up at 5 years before the index date. Data on baseline comorbidities recorded in the Hospital Discharge Register and Prescription Register within 5 years of the index date were also included (Table S1, Figure S2). We reported comorbidities common to persons with Parkinson’s disease and conditions that may increase the risk of falls or fractures (cardiovascular diseases, diabetes, stroke, renal failure, rheumatoid arthritis, asthma/chronic obstructive pulmonary disease (COPD), osteoporosis, previous fractures, substance abuse) [26, 28, 29]. Use of fall-risk-increasing medications (psychotropics [including antidepressants, antipsychotics, benzodiazepines and related drugs], opioids, anti-epileptics, urinary anticholinergics, alpha blockers for benigh prostate hyperplasia) and anti-osteoporosis medications (bisphosphonates, other drugs affecting bone structure and mineralization [including denosumab], calcitonin) was ascertained using dispensing information within 1 year before the start of follow-up (Table S2, Figure S2) [18, 28, 30]. For characteristics at the time of hip fracture, the timeframe for records identification was shifted to the date of hip fracture among persons who experienced a hip fracture (Figure S2).
Statistical analysis
Descriptive statistics for characteristics of cohorts were presented. Median and interquartile range (IQR) were reported for continuous variables. Pearson’s chi-squared test and Wilcoxon rank sum test were used to test for significant differences across cohorts for categorical and continuous variables respectively.
Fine-Gray subdistribution hazard model was used to estimate the hazard ratio (HR) of hip fracture over the 10-year follow-up period. The model was chosen to account for competing risks of other censoring events (death and diagnosis of Parkinson’s disease for comparison cohort). Additionally, the model was adjusted for age, sex and variation across time using interaction term between Parkinson’s disease and time.
Annual incidence rate (IR) of hip fracture per year of follow-up were estimated as the incidences of first hip fracture occurring per 1000 person-years (PY). Persons who experienced hip fracture before the start of each year and their matched counterparts were excluded. The 95% confidence intervals (CI) of annual IR were estimated using exact limits of means with Poisson distribution. Annual incidence rate ratios (IRR) per year of follow-up were estimated using generalized linear mixed model with adaptive Gaussian quadrature rule accounting for low event rates and matching between Parkison’s disease and comparison cohort.
To examine variations across sex and age, analyses in subgroups of sex and age (< 70 years, 70–79, 80 years old at index date) were conducted. Comparison persons were categorized into the same age group as their matched counterparts (age 1) to retain 1:2 matching. All analyses were performed using R (version 4.0.0).
Results
Our study included 11,051 persons with PD and 22,102 age- and sex-matched comparison persons. Among persons with PD, 54.3% (n = 5051) were males and 45.7% (n = 6000) were females (54.3%, n = 18,000) and had a median age of 72 (IQR 64–78) years at diagnosis (Table S3). At baseline, persons with PD had comparable rates of comorbidities and usage of anti-osteoporosis medications than persons without PD (Tables S3). However, fall-risk-increasing medications were more prevalent among persons with PD than persons without PD (Tables S3).
Overall risk of hip fracture
Among persons with PD, 13.4% (n = 1373) had an incident hip fracture during the follow-up period from 5 years before to 10 years after diagnosis of PD. In contrast, only 5.3% (n = 1164) of the comparison cohort experienced a hip fracture during follow-up. Across the 15-year follow-up period, 42.7% (n = 4964) of persons with PD and 29.7% (n = 6561) comparison persons died without experiencing hip fracture or the end of follow-up. Only < 0.001% (n = 15) of comparison persons were censored due to diagnosis of PD. After accounting for competing events, persons with PD had 1.82 times the risk of hip fractures than those without PD (HR 1.82, 95% CI 1.47–2.26). The hazard ratios were higher among females (HR 2.08, 95% CI 1.58–2.73) than males (HR 1.47, 95% CI 1.04–2.07). The highest hazard ratio was found among those aged 80 years at index date (HR 2.73, 95% CI 1.91–3.89), and no increased risk was found among those aged < 70 years (HR 1.01, 95% CI 0.55–1.83) (Figure S3, Table S4).
Annual incidence of hip fractures
The annual incidence rates of hip fractures increase with the number of years of follow-up for both persons with and without PD (Fig. 1, Table S5). The annual incidence rates of hip fracture were similar at 5 years before index date among persons with (IR/1000PY 0.81, 95% CI 0.37–1.54) and without (IR/1000PY 1.31, 95% CI 0.88–1.88) PD (IRR 0.61, 95% CI 0.29–1.28). The annual incidence rates increased and started to diverge from 3 years before index date, with a higher annual incidence rate among persons with (IR/1000PY 2.83, 95% CI 1.93–4.02) PD than persons without PD (IR/1000PY 1.64, 95% CI 1.15–2.28) (IRR 1.69, 95% CI 1.05–2.72). The annual incidence rate continued to increase and peaked at 8 years after index dates among persons with (IR/1000PY 19.0, 95% CI 15.0–23.9) and without (IR/1000PY 4.43, 95% CI 3.18–6.00) PD. The difference among annual incidence rates was highest at 5 years after index date (IRR 5.78, 95% CI 3.77–8.87) among persons with PD (IR/1000PY 16.1, 95% CI 13.0–19.7) compared to persons without PD (IR/1000PY 3.40, 95% CI 2.44–4.61) (Fig. 1, Table S5).
Fig. 1.
Annual incidence rates of hip fracture per 1000 person-years (PY) in persons with and without Parkinson’s disease (PD) from 5 years before to 10 years after PD diagnosis (index date) in entire cohort and by sex
Higher annual incidence rates were observed among females than males consistently at all years of follow-up, and a similar increasing trend was observed for both sexes across the follow-up period. The difference in annual incidence rates for males started to emerge from 1 year before the index date, with a higher annual incidence rate among persons with (IR/1000PY 4.22, 95% CI 2.73–6.23) and without (IR/1000PY 1.52, 95% CI 0.90–2.40) PD (IRR 2.66, 95% CI 1.46–4.84) (Fig. 1, Table S6).
When stratified by age groups, the difference in annual incidence rates only showed statistically significant differences from 1 year before index date for all age groups due to the smaller sample size resulting in less precise estimates. Older people had higher annual incidence rate ratios across the whole follow-up period. For persons aged 80, no statistically significant differences were found after the incidence rate difference peaked at 5 years after index date (at 5 year after index date: persons with PD (IR/1000PY 65.7, 95% CI 38.9–103.8) vs. persons without PD (IR/1000PY 11.9, 95% CI 4.78–24.5), IRR 10.7 [95% CI 2.79–40.8]). This was likely due to depletion of population at risk with more people censored by death and/or hip fractures at later years (Fig. 2, Table S7).
Fig. 2.
Annual incidence rates per 1000 person-years (PY) of hip fracture in persons with and without Parkinson’s disease (PD) from 5 years before to 10 years after PD diagnosis (index date) stratified by age
Characteristics of persons with and without hip fractures at baseline
Among persons with PD, only rheumatoid arthritis, asthma/COPD and previous fractures occurred more frequently in persons with hip fractures than persons without. In the comparison cohort, most comorbidities were more prevalent in persons with hip fractures than persons without. Among persons with PD, persons with and without hip fractures had no difference in the use of fall-risk-increasing medications and anti-osteoporosis medications at 5 years before PD diagnosis. In the comparison cohort, persons with hip fractures used more fall-risk-increasing medications and anti-osteoporosis medications at baseline (Table 1).
Table 1.
Characteristics of people with and without hip fractures among persons with and without Parkinson’s disease at 5 years before index date
| Parkinson’s disease | Comparison cohort | |||||
|---|---|---|---|---|---|---|
| Characteristic | Hip fracture N = 1667 |
No hip fracture N = 9384 |
p-value1 | Hip fracture N = 1613 |
No hip fracture N = 20,489 |
p-value1 |
| Sex, n (%) | < 0.001 | < 0.001 | ||||
| Female | 986 (59.1) | 4065 (43.3) | 1005 (62.3) | 9097 (44.4) | ||
| Male | 681 (40.9) | 5319 (56.7) | 608 (37.7) | 11,392 (55.6) | ||
| Age at index date, median (IQR) | 74 (8) | 70 (10) | < 0.001 | 77 (72, 81) | 71 (64, 77) | < 0.001 |
| Age category at index date | < 0.001 | < 0.001 | ||||
| < 70 | 484 (29.0) | 4197 (44.7) | 289 (17.9) | 9,108 (44.5) | ||
| 70–79 | 817 (49.0) | 3757 (40.0) | 830 (51.5) | 8296 (40.5) | ||
| 80 + | 366 (22.0) | 1,430 (15.2) | 494 (30.6) | 3,085 (15.1) | ||
| Comorbidities, n (%) | ||||||
| Cardiovascular diseases | 622 (37.3) | 3301 (35.2) | 0.093 | 639 (39.6) | 7141 (34.9) | < 0.001 |
| Diabetes | 115 (6.9) | 717 (7.6) | 0.3 | 149 (9.2) | 1466 (7.2) | 0.002 |
| Stroke | 24 (1.4) | 138 (1.5) | > 0.9 | 41 (2.5) | 289 (1.4) | < 0.001 |
| Renal failure | 3 (0.2) | 12 (0.1) | 0.5 | < 3 (< 0.1) | 30 (0.1) | 0.7 |
| Rheumatoid arthritis | 64 (3.8) | 252 (2.7) | 0.009 | 82 (5.1) | 620 (3.0) | < 0.001 |
| Asthma/COPD | 125 (7.5) | 569 (6.1) | 0.026 | 109 (6.8) | 1345 (6.6) | 0.8 |
| Osteoporosis | 33 (2.0) | 143 (1.5) | 0.2 | 42 (2.6) | 285 (1.4) | < 0.001 |
| Previous fractures | 48 (2.9) | 193 (2.1) | 0.034 | 58 (3.6) | 500 (2.4) | 0.004 |
| Substance abuse | 10 (0.6) | 73 (0.8) | 0.4 | 24 (1.5) | 169 (0.8) | 0.006 |
| Medications, n (%) | ||||||
| Fall-risk-increasing medications | ||||||
| Psychotropics | 337 (20.2) | 1894 (20.2) | > 0.9 | 407 (25.2) | 3377 (16.5) | < 0.001 |
| Antidepressants | 149 (8.9) | 832 (8.9) | > 0.9 | 135 (8.4) | 1,269 (6.2) | < 0.001 |
| Antipsychotics | 70 (4.2) | 422 (4.5) | 0.6 | 66 (4.1) | 501 (2.4) | < 0.001 |
| Benzodiazepines and related drugs | 254 (15.2) | 1,363 (14.5) | 0.4 | 317 (19.7) | 2,538 (12.4) | < 0.001 |
| Opioids | 39 (2.3) | 258 (2.7) | 0.3 | 64 (4.0) | 500 (2.4) | < 0.001 |
| Anti-epileptics | 39 (2.3) | 196 (2.1) | 0.5 | 32 (2.0) | 277 (1.4) | 0.037 |
| Urinary anticholinergics | 33 (2.0) | 149 (1.6) | 0.2 | 26 (1.6) | 214 (1.0) | 0.034 |
| Alpha blockers for BPH | 42 (2.5) | 306 (3.3) | 0.11 | 42 (2.6) | 486 (2.4) | 0.6 |
| Anti-osteoporosis medications | ||||||
| Bisphosphonates | 25 (1.5) | 101 (1.1) | 0.13 | 30 (1.9) | 220 (1.1) | 0.004 |
| Calcitonin | 14 (0.8) | 45 (0.5) | 0.063 | 16 (1.0) | 96 (0.5) | 0.004 |
1Pearson’s chi-squared test for categorical variables; Wilcoxon rank sum test for continuous variables
BPH benign prostate hyperplasia, COPD chronic obstructive pulmonary disease
Characteristics of persons with an incident hip fracture at the date of hip fracture
Among persons with an incident hip fracture, persons with PD were younger than the comparison cohort at the time of hip fracture (80 [IQR 75–84] vs. 83 [IQR 78–88] years old, p < 0.001) (Table 2). At the time of hip fracture, persons with PD had a lower prevalence of comorbidities than persons without PD, except for osteoporosis and previous fractures. Within 1 year prior to hip fractures, persons with PD had a higher prevalent use of antidepressants, antipsychotics, anti-epileptics and urinary anticholinergics than persons without PD. There were no differences in the use of anti-osteoporosis medications among persons with and without PD (Table 2).
Table 2.
Characteristics of persons with incident hip fracture during follow-up in the PD and comparison cohort at the date of hip fracture
| Characteristic | Parkinson.s disease N = 1667 |
Comparison cohort N = 1613 |
p-value1 |
|---|---|---|---|
| Sex, n (%) | 0.200 | ||
| Female | 986 (59.1) | 1005 (62.3) | |
| Male | 681 (40.9) | 608 (37.7) | |
| Age at hip fracture, median (IQR) | 80 (75, 84) | 83 (78, 88) | < 0.001 |
| Age at index date, median (IQR) | 75 (69, 79) | 77 (72, 81) | 0.607 |
| Age category at index date, n (%) | < 0.001 | ||
| < 70 | 484 (29.0) | 289 (17.9) | |
| 70–79 | 817 (49.0) | 830 (51.5) | |
| 80 + | 366 (22.0) | 494 (30.6) | |
| Time since index date (year), median (IQR) | 5.7 (2.6, 8.7) | 6.7 (2.4, 10.5) | < 0.001 |
| Comorbidities, n (%) | |||
| Cardiovascular diseases | 926 (55.5) | 1022 (63.4) | < 0.001 |
| Diabetes | 250 (15.0) | 327 (20.3) | < 0.001 |
| Stroke | 141 (8.5) | 217 (13.5) | < 0.001 |
| Renal failure | 23 (1.4) | 40 (2.5) | 0.019 |
| Rheumatoid arthritis | 79 (4.7) | 103 (6.4) | 0.033 |
| Asthma/chronic obstructive pulmonary disease | 176 (10.6) | 211 (13.1) | 0.023 |
| Osteoporosis | 265 (15.9) | 186 (11.5) | < 0.001 |
| Previous fractures | 303 (18.2) | 283 (17.5) | 0.708 |
| Substance abuse | 15 (0.9) | 32 (2.0) | 0.008 |
| Medications, n (%) | |||
| Fall-risk-increasing medications | |||
| Psychotropics | 918 (55.1) | 745 (46.2) | < 0.001 |
| Antidepressants | 530 (31.8) | 375 (23.2) | < 0.001 |
| Antipsychotics | 384 (23.0) | 210 (13.0) | < 0.001 |
| Benzodiazepines and related drugs | 500 (30.0) | 466 (28.9) | 0.487 |
| Opioids | 284 (17.0) | 292 (18.1) | 0.349 |
| Anti-epileptics | 174 (10.4) | 132 (8.2) | 0.038 |
| Urinary anticholinergics | 118 (7.1) | 54 (3.3) | < 0.001 |
| Alpha blockers for benign prostate hyperplasia | 162 (9.7) | 153 (9.5) | 0.965 |
| Anti-osteoporosis medications | |||
| Bisphosphonates | 119 (7.1) | 97 (6.0) | 0.232 |
| Other drugs affecting bone structure and mineralization, including denosumab | 14 (0.8) | 9 (0.6) | 0.454 |
| Calcitonin | 47 (2.8) | 37 (2.3) | 0.414 |
1Linear mixed model for continuous variables; generalized linear mixed model for categorical variables
Discussion
To our knowledge, this is the first study to assess the annual incidence of hip fracture in relation to PD diagnosis. This is also the first study that investigated the risk of hip fracture before the diagnosis of PD. Compared to persons without PD, we found that persons with PD had a higher annual incidence of hip fracture from 3 years before their PD diagnosis, which continued until at least 10 years after diagnosis. This adds to the current evidence of persons with PD having an elevated risk of hip fracture at diagnosis, and the varied risks reported following diagnosis [19–21, 31–34].
Our finding of higher hip fracture risk before PD diagnosis may be partly due to the higher prevalence of fall-risk-increasing medications among persons with PD compared to persons without PD at 5 years before PD diagnosis. We have also previously shown that the prevalences of antidepressants, antipsychotics and muscle relaxants were higher up to 7 years before the PD diagnosis in this cohort [22–24], possibly due to non-motor and motor prodromal symptoms of PD. The motor symptoms likely also contributed to the increased risk of hip fracture before PD diagnosis. A case–control study of 8166 persons with PD in the UK found significantly more presentations of dizziness and balance impairment at 5 years before diagnosis, compared with age- and sex-matched controls [35]. Subtle motor deficits have been reported to appear from 3 to 4 years before diagnosis and score of mobility deteriorating from 2 years before diagnosis [36, 37]. While previous studies have shown the presence of prediagnostic motor and non-motor symptoms in PD [36, 38], our study was the first to report an elevated risk of a clinically significant event, hip fracture, before diagnosis. Rate of late or misdiagnosis is high in PD, likely complicated by the unspecific prodromal symptoms [39, 40]; thus, it is important to assess and manage risk factors for falls and fractures among persons presenting with early symptoms not yet fulfilling the diagnostic criteria of PD. Early referral of persons with a combination of motor (e.g. tremor, rigidity, impaired balance) and non-motor prodromal symptoms (e.g. anxiety, depression, constipation) may facilitate diagnosis of PD and managing risks of falls and fractures.
In our study, the risk of hip fracture remained higher among persons with PD until the end of the 10-year follow-up. While some studies found the risk of hip fracture peaked shortly after PD diagnosis and attenuated afterwards due to restricted mobility [31, 32], our results were aligned with most previous studies that showed the increased risk of hip fracture sustaining till the end of follow-up [19–21, 33, 34]. This substantiates the need to continuously assess and manage the risks of falls and fractures as PD progresses.
Similar to the general population [41], hip fracture incidence was highest among females and increased with age in the PD cohort. However, persons with PD consistently had a higher risk of hip fracture across all age groups, and in both sexes, underscoring the importance of managing the risk of hip fracture in persons with PD of all ages and sexes.
Compared to the comparison cohort, we observed a higher prevalence of fall-risk increasing medications in the PD cohort at baseline and at the time of hip fracture, consistent with earlier observations on the use of antidepressants and antipsychotics among persons with PD in this study population [22, 23]. The fall-risk-increasing medications may have been used to manage non-motor symptoms of PD, e.g. anxiety, depression, psychotic symptoms, sleep disorder and urinary dysfunction, which may emerge up to 10 years before PD diagnosis [5, 22, 23, 42]. Fall-risk-increasing medications have been associated with an increased risk of hip fractures [43, 44]. The ‘World guidelines for falls prevention and management for older adults’ strongly recommend assessing history and risk of falls before prescribing fall-risk-increasing medications and using validated tools (e.g. STOPPFall) to identify and deprescribe fall-risk-increasing medications when performing medication review [28, 45]. The use of fall-risk-increasing medications may compound the already higher risk of falls and hip fractures among persons with PD; thus, a regular review of fall-risk-increasing medications should be integral in the management of PD. Future research on alternatives (e.g. non-pharmacological management) that can both alleviate symptoms and minimize the risk of falls and hip fractures is also warranted.
Despite the higher risk of hip fractures, persons with PD had similar prevalence of anti-osteoporosis medications within 1 year prior to baseline and hip fractures, compared with persons without PD. While the risk of hip fractures may be mainly driven by the higher risk of falls among persons with PD, persons with PD have also been shown to be at higher risk of osteoporosis [6]. The similar treatment rate may indicate that persons with PD were inadequately treated for osteoporosis. However, clinicians may recognize persons with PD are at a higher risk of osteoporosis as PD progresses. The higher risk of osteoporosis may result from the pathogenesis of PD, physical inactivity, lower body weight and vitamin D deficiency as PD progresses [6]. Further study on the longitudinal use of anti-osteoporosis medications is needed to assess whether persons with PD were appropriately treated with the progression of PD.
The main strength of our study was reporting the annual incidence rates of hip fractures from before diagnosis of PD, which provides a more comprehensive longitudinal understanding of the risk of hip fracture than previous population-based studies reporting overall hazard ratios or odds ratios [19, 21, 31–33]. Additionally, our study was associated with minimal selection bias because it was conducted using nationwide registers. Our study also has some limitations common to register-based studies. Our data did not include information on specific clinical features of PD, e.g. severity of motor and non-motor symptoms, balance, mobility or bone health, nor physiological information that may affect the risk of hip fracture, e.g. weight, height or stature. Our study was limited to incident hip fractures; therefore, future studies are needed to quantify the risk and prevention of recurrent hip fractures in persons with PD.
Conclusion
In our matched cohort of persons with and without PD in Finland, we found higher annual incidences of incident hip fractures among persons with PD, starting from 3 years before until up to 10 years after the diagnosis of PD, regardless of age and sex. Our results underscore the importance of early and regular assessment and management of fall and fracture risk among persons with PD of all age and sex including assessment of medication and deprescribing fall-risk-increasing medications.
Supplementary Information
Below is the link to the electronic supplementary material.
DOCX (359 KB)
Acknowledgements
MTYL is supported by Australian Government Research Training Scholarship.
Funding
Open Access funding enabled and organized by CAUL and its Member Institutions.
Data Availability
Restrictions apply to the availability of all data generated and analyzed during this study to preserve patient confidentiality. Data are not publicly available, however data are available from the authors upon reasonable request and with the permission of the register maintainers.
Declarations
Conflict of interest
None.
Footnotes
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Feigin VL, Nichols E, Alam T et al (2019) Global, regional, and national burden of neurological disorders, 1990–2016: a systematic analysis for the Global Burden of Disease Study 2016. Lancet Neurol 18:459–480. 10.1016/S1474-4422(18)30499-X [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Zhu J, Cui Y, Zhang J et al (2024) Temporal trends in the prevalence of Parkinson’s disease from 1980 to 2023: a systematic review and meta-analysis. Lancet Healthy Longev 5:e464–e479. 10.1016/S2666-7568(24)00094-1 [DOI] [PubMed] [Google Scholar]
- 3.Dorsey ER, Elbaz A, Nichols E et al (2018) Global, regional, and national burden of Parkinson’s disease, 1990–2016: a systematic analysis for the Global Burden of Disease Study 2016. Lancet Neurol 17:939–953. 10.1016/S1474-4422(18)30295-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Vollset SE, Ababneh HS, Abate YH et al (2024) Burden of disease scenarios for 204 countries and territories, 2022–2050: a forecasting analysis for the Global Burden of Disease Study 2021. Lancet 403:2204–2256. 10.1016/S0140-6736(24)00685-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Schapira AHV, Chaudhuri KR, Jenner P (2017) Non-motor features of Parkinson disease. Nat Rev Neurosci 18:435–450. 10.1038/nrn.2017.62 [DOI] [PubMed] [Google Scholar]
- 6.Torsney KM, Noyce AJ, Doherty KM et al (2014) Bone health in Parkinson’s disease: a systematic review and meta-analysis. J Neurol Neurosurg Psychiatry 85:1159–1166. 10.1136/jnnp-2013-307307 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Louvois M, Ferrero S, Barnetche T et al (2021) High risk of osteoporotic fracture in Parkinson’s disease: meta-analysis, pathophysiology and management. Revue Neurologique 177:660–669. 10.1016/j.neurol.2020.07.015 [DOI] [PubMed] [Google Scholar]
- 8.Murueta-Goyena A, Muiño O, Gómez-Esteban JC (2024) Prognostic factors for falls in Parkinson’s disease: a systematic review. Acta Neurol Belg 124:395–406. 10.1007/s13760-023-02428-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Camicioli R, Morris ME, Pieruccini-Faria F et al (2023) Prevention of falls in Parkinson’s disease: guidelines and gaps. Mov Disord Clin Pract 10:1459–1469. 10.1002/mdc3.13860 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Liu B, Chen G, Yu Z et al (2021) Bone mineral density and related scores in Parkinson’s disease: a systematic review and meta-analysis. World Neurosurg 146:e1202–e1218. 10.1016/j.wneu.2020.11.132 [DOI] [PubMed] [Google Scholar]
- 11.Carroll C, Clarke CE, Grosset D et al (2024) Addressing comorbidities in people with Parkinson’s disease: considerations from an expert panel. JPD 14:53–63. 10.3233/JPD-230168 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Heo K-N, Seok JY, Ah Y-M et al (2023) Development and validation of a machine learning-based fall-related injury risk prediction model using nationwide claims database in Korean community-dwelling older population. BMC Geriatr 23:830. 10.1186/s12877-023-04523-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Nordström P, Ahlqvist VH, Ballin M, Nordström A (2024) A novel clinical prediction model for hip fractures: a development and validation study in the total population of Sweden. eClinicalMedicine 77:102877. 10.1016/j.eclinm.2024.102877 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Tatangelo G, Watts J, Lim K et al (2019) The cost of osteoporosis, osteopenia, and associated fractures in Australia in 2017. J Bone Miner Res 34:616–625. 10.1002/jbmr.3640 [DOI] [PubMed] [Google Scholar]
- 15.Mitchell PJ, Chan D-CD, Lee J-K et al (2022) The global burden of fragility fractures – what are the differences, and where are the gaps. Best Pract Res Clin Rheumatol 36:101777. 10.1016/j.berh.2022.101777 [DOI] [PubMed]
- 16.Leung MTY, Marquina C, Turner JP et al (2023) Hip fracture incidence and post-fracture mortality in Victoria, Australia: a state-wide cohort study. Arch Osteoporos 18:56. 10.1007/s11657-023-01254-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Cauley JA, Chalhoub D, Kassem AM, Fuleihan GE-H (2014) Geographic and ethnic disparities in osteoporotic fractures. Nat Rev Endocrinol 10:338–351. 10.1038/nrendo.2014.51 [DOI] [PubMed] [Google Scholar]
- 18.Sing C, Lin T, Bartholomew S et al (2023) Global epidemiology of hip fractures: secular trends in incidence rate, post-fracture treatment, and all-cause mortality. J Bone Mineral Res 38:1064–1075. 10.1002/jbmr.4821 [DOI] [PubMed] [Google Scholar]
- 19.Koo HY, Cho EB, Kong SH et al (2023) Fracture risk in Parkinson’s disease according to its severity and duration. Osteoporos Int 34:81–89. 10.1007/s00198-022-06562-0 [DOI] [PubMed] [Google Scholar]
- 20.Kim T, Byun SJ, Seong MY et al (2022) Fracture risk and impact of osteoporosis in patients with Parkinson’s disease: a nationwide database study. J Bone Miner Metab 40:602–612. 10.1007/s00774-022-01322-w [DOI] [PubMed] [Google Scholar]
- 21.Chen Y-Y, Cheng P-Y, Wu S-L, Lai C-H (2012) Parkinson’s disease and risk of hip fracture: an 8-year follow-up study in Taiwan. Parkinsonism Relat Disord 18:506–509. 10.1016/j.parkreldis.2012.01.014 [DOI] [PubMed] [Google Scholar]
- 22.Lilja S, Tolppanen A-M, Koponen M et al (2022) Incidence of antipsychotic use among community dwellers with and without Parkinson’s disease. Parkinsonism Relat Disord 103:69–72. 10.1016/j.parkreldis.2022.08.024 [DOI] [PubMed] [Google Scholar]
- 23.Hentilä E, Tiihonen M, Taipale H et al (2021) Incidence of antidepressant use among community dwellers with and without Parkinson’s disease – a nationwide cohort study. BMC Geriatr 21:202. 10.1186/s12877-021-02145-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Paakinaho A, Karttunen N, Koponen M et al (2020) Incidence of muscle relaxant use in relation to diagnosis of Parkinson’s disease. Int J Clin Pharm 42:336–340. 10.1007/s11096-020-01002-7 [DOI] [PubMed] [Google Scholar]
- 25.Wettermark B, Zoëga H, Furu K et al (2013) The Nordic prescription databases as a resource for pharmacoepidemiological research—a literature review. Pharmacoepidemiol Drug 22:691–699. 10.1002/pds.3457 [DOI] [PubMed] [Google Scholar]
- 26.Ilmaniemi S, Tolppanen A, Herukka S et al (2023) Incidence and outcomes of head injuries in people with and without Parkinson disease. Euro J of Neurol 30:1648–1657. 10.1111/ene.15782 [DOI] [PubMed] [Google Scholar]
- 27.Hughes AJ, Daniel SE, Kilford L, Lees AJ (1992) Accuracy of clinical diagnosis of idiopathic Parkinson’s disease: a clinico-pathological study of 100 cases. J Neurol Neurosurg Psychiatry 55:181–184. 10.1136/jnnp.55.3.181 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Seppala LJ, Petrovic M, Ryg J et al (2021) STOPPFall (screening tool of older persons prescriptions in older adults with high fall risk): a Delphi study by the EuGMS task and finish group on fall-risk-increasing drugs. Age Ageing 50:1189–1199. 10.1093/ageing/afaa249 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Hommel ALAJ, Krijthe JH, Darweesh S, Bloem BR (2022) The association of comorbidity with Parkinson’s disease-related hospitalizations. Parkinsonism Relat Disord 104:123–128. 10.1016/j.parkreldis.2022.10.012 [DOI] [PubMed] [Google Scholar]
- 30.By the 2023 American Geriatrics Society Beers Criteria® Update Expert Panel (2023) American Geriatrics Society 2023 updated AGS Beers Criteria® for potentially inappropriate medication use in older adults. J Am Geriatri Society 71:2052–2081. 10.1111/jgs.18372 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.An SJ, Lee S-H, Lee S-Y et al (2017) Femur fractures in Parkinsonism: analysis of a national sample cohort in South Korea. J Clin Neurol 13:380. 10.3988/jcn.2017.13.4.380 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Vestergaard P, Rejnmark L, Mosekilde L (2007) Fracture risk associated with parkinsonism and anti-Parkinson drugs. Calcif Tissue Int 81:153–161. 10.1007/s00223-007-9065-6 [DOI] [PubMed] [Google Scholar]
- 33.Pouwels S, Bazelier MT, De Boer A et al (2013) Risk of fracture in patients with Parkinson’s disease. Osteoporos Int 24:2283–2290. 10.1007/s00198-013-2300-2 [DOI] [PubMed] [Google Scholar]
- 34.Sleeman I, Che ZC, Counsell C (2016) Risk of fracture amongst patients with Parkinson’s disease and other forms of parkinsonism. Parkinsonism Relat Disord 29:60–65. 10.1016/j.parkreldis.2016.05.026 [DOI] [PubMed] [Google Scholar]
- 35.Schrag A, Horsfall L, Walters K et al (2015) Prediagnostic presentations of Parkinson’s disease in primary care: a case-control study. The Lancet Neurology 14:57–64. 10.1016/S1474-4422(14)70287-X [DOI] [PubMed] [Google Scholar]
- 36.Maetzler W, Mirelman A, Pilotto A, Bhidayasiri R (2024) Identifying subtle motor deficits before Parkinson’s disease is diagnosed: what to look for? JPD 14:S287–S296. 10.3233/JPD-230350 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Pauwels A, Phan ALG, Ding C et al (2024) Rate of motor progression in Parkinson’s disease: a systematic review and meta-analysis. Front Neurol 15:1452741. 10.3389/fneur.2024.1452741 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Darweesh SKL, Verlinden VJA, Stricker BH et al (2017) Trajectories of prediagnostic functioning in Parkinson’s disease. Brain 140:429–441. 10.1093/brain/aww291 [DOI] [PubMed] [Google Scholar]
- 39.Tolosa E, Garrido A, Scholz SW, Poewe W (2021) Challenges in the diagnosis of Parkinson’s disease. The Lancet Neurology 20:385–397. 10.1016/S1474-4422(21)00030-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Cheng H, Ulane CM, Burke RE (2010) Clinical progression in Parkinson disease and the neurobiology of axons. Ann Neurol 67:715–725. 10.1002/ana.21995 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Feng J, Zhang C, Li B et al (2024) Global burden of hip fracture: the global burden of disease study. Osteoporos Int 35:41–52. 10.1007/s00198-023-06907-3 [DOI] [PubMed] [Google Scholar]
- 42.Burchill E, Watson CJ, Fanshawe JB et al (2024) The impact of psychiatric comorbidity on Parkinson’s disease outcomes: a systematic review and meta-analysis. The Lancet Regional Health - Europe 39:100870. 10.1016/j.lanepe.2024.100870 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Thorell K, Ranstad K, Midlöv P et al (2014) Is use of fall risk-increasing drugs in an elderly population associated with an increased risk of hip fracture, after adjustment for multimorbidity level: a cohort study. BMC Geriatr 14:131. 10.1186/1471-2318-14-131 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Donnelly K, Bracchi R, Hewitt J et al (2017) Benzodiazepines, Z-drugs and the risk of hip fracture: a systematic review and meta-analysis. PLoS ONE 12:e0174730. 10.1371/journal.pone.0174730 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.Montero-Odasso M, Van Der Velde N, Martin FC et al (2022) World guidelines for falls prevention and management for older adults: a global initiative. Age and Ageing 51:afac205. 10.1093/ageing/afac205 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
DOCX (359 KB)
Data Availability Statement
Restrictions apply to the availability of all data generated and analyzed during this study to preserve patient confidentiality. Data are not publicly available, however data are available from the authors upon reasonable request and with the permission of the register maintainers.