You are here

Article Text

Article menu

Download PDFPDF

Movement disorders
Research paper
Hospitalisation and comorbidities in Parkinson's disease: a large Australian retrospective study
  1. Michal Lubomski1,2,
  2. R Louise Rushworth1,
  3. Stephen Tisch1,2
  1. 1The University of Notre Dame Australia, School of Medicine, Sydney, New South Wales, Australia
  2. 2Department of Neurology, St Vincent's Hospital, Sydney, New South Wales, Australia
  1. Correspondence to Dr Michal Lubomski, The University of Notre Dame Australia, School of Medicine, 160 Oxford St, Darlinghurst, Sydney, NSW 2010, Australia; michal.lubomski{at}nd.edu.au

Abstract

Objectives Patients with Parkinson's disease (PD) require higher levels of care during hospitalisation. Management of comorbidities in these patients aims to optimise function while minimising complications. The objective of this study was to examine patterns of hospitalisation of patients with PD in NSW with regards to sociodemographic factors, comorbidities and aspects of clinical management.

Methods A retrospective study of all patients with idiopathic PD and a control group of non-PD patients admitted for acute care to NSW hospitals between 2008 and 2012.

Results The study group comprised 5637 cases and 8143 controls. The mean PD patient age was 75.0 years (±10.9). Patients with PD had a significantly longer hospital stay (median 7 days, IQR 3–13 vs 1 day, IQR 1–7, p<0.001) than control patients. Patients with PD were five times more likely to be treated for delirium, three times more likely to experience an adverse drug event and syncope, more than twice as likely to require management of falls/fractures, dementia, gastrointestinal complications, genitourinary infections, reduced mobility and other trauma but half as likely to require hospitalisation for chronic airways disease and neoplasia, including melanoma, than the control group (all p<0.001).

Conclusions Patients with PD are more likely to suffer serious health problems, including delirium, adverse drug reactions, syncope, falls and fractures than controls. These findings highlight PD as a multisystem neuropsychiatric disorder in which motor and non-motor features contribute to morbidity. Increased awareness of the added risk PD poses in acute hospitalised patients can be used to inform strategies to improve patient outcomes.

  • Parkinson's Disease

https://doi.org/10.1136/jnnp-2014-307822

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    Introduction

    Parkinson's disease (PD), which affects approximately 3.4% of adults over 55 years,1 is a common progressive and disabling neurological disorder characterised by the degeneration of several different neuronal populations important for movement, autonomic function and cognition. There is a worldwide paucity of information on the causes and related comorbidities that lead to hospitalisation in patients with PD.2 Rising life expectancy is increasing the prevalence of chronic diseases such as PD, and it is estimated that there will be between 8.7 and 9.3 million people over the age of 50 years living with PD around the world by 2030.3 In Australia, the incidence of PD is approximately 8900 cases per year, and it is slightly more common in men than women.4–6 In 2009–2010, there were 3179 patients treated in hospital for PD in Australia.7

    Many challenging clinical features and comorbidities are encountered in managing the hospitalised patient with PD. Early recognition and management of these issues aims to optimise the potential benefits of an inpatient stay, along with decreasing hospitalisation-related complications.8 ,9 Although complications including falls, fractures and reduced mobility are believed to be a significant cause of PD-related admissions, medical comorbidities including pneumonia, cardiac diseases and non-motor PD symptoms including cognitive impairment, psychiatric complaints and adverse drug events are likely to result in an increasing need for hospitalisation in patients with PD.2 ,8–14 In Australia in 2009–2010, there were an estimated 2220 hospital admissions for accidental falls and 2138 admissions for pneumonia in the context of PD-related complications.7

    Patients with PD are admitted to hospitals at higher rates and frequently have longer stays than the general population.9 ,11 ,12 ,15 Admission for aspiration pneumonia, trauma (inclusive of fractures), psychosis and sepsis are higher in patients with PD than controls.9 Further, patients with PD, on average, have a longer length of stay (LOS) of between 2 and 14 days than controls, with 7–28% of patients with PD requiring yearly hospitalisation.16 Encouraging evidence supports improving access to outpatient clinics as an essential strategy to minimising inpatient hospitalisations.11 The objective of this study was to examine patterns of acute care hospitalisation of patients with PD in NSW with regards to sociodemographic factors, comorbidities and aspects relating to clinical management.

    Methods

    Study settings and subjects

    The NSW Ministry of Health maintains a database, the Admitted Patient Data Collection (APDC) on all separations (episodes of hospitalisation) from public and private hospitals in NSW. Information stored in the APDC includes patient age, gender, principal diagnosis and all secondary diagnoses, any surgical procedures, hospital type, mode of admission (eg, through the emergency department or planned admissions) and separation (eg, discharge home, death, transfer to another hospital) and LOS.

    For the purposes of this analysis, all patients with either a principal diagnosis or a secondary (comorbid) diagnosis recorded as Idiopathic Parkinson's disease (ICD-10-CM G20) and who had a separation (episode of care) from a NSW hospital (public or private) between the years 2008 and 2012 were included in the study. Exclusions included other causes of Parkinsonism: secondary Parkinson's disease, atypical Parkinson's disease, vascular Parkinsonism or Parkinson's Plus Syndromes. Admissions for inpatient rehabilitation, dialysis or to an inpatient psychiatric facility were excluded from this analysis. Ethics approval was received from the University of Notre Dame Australia Human Research Ethics Committee.

    Control group

    A random sample of patients admitted to NSW hospitals during 2008 without a principal or secondary diagnosis of PD was selected as the control group for this study. As public and private hospital data are stored separately, the selection of the control group was first weighted according to the relative proportion of beds available in these two sectors in NSW (74% and 26%, respectively).17 The control subjects were then selected using weights based on the age and gender distribution of the most recent Australian PD prevalence estimates, corresponding to a ratio of 1.10 males to 1 female and an age distribution of 6% (<55 year olds), 12% (55–64 year olds), 29% (64–74 year olds), 33% (75–84 year olds) and 20% (>85 year olds).7 Patients were excluded from the control group if they were admitted for rehabilitation, dialysis or to an inpatient psychiatric facility.

    Statistical analysis

    All admission diagnoses were grouped according to the ICD-10 Chapter headings. (International Classification of Diseases, Tenth Revision).18 Within these categories, diagnoses were grouped according to the clinical conditions related to PD. Cardiac diseases included myocardial infarction (I21) and conduction disorders (I44). Orthostatic hypotension (I95) and syncope and collapse (R55) were classified individually due to their specific relevance to PD. DBS insertion was represented by procedure codes: 92036 (insertion of intracranial electrode via burr holes) and 39138 (insertion of intracranial electrode via craniotomy).

    Two-sample, independent t tests were used to analyse differences between the groups for continuous variables. As LOS was highly skewed, a Mann–Whitney U test was used to compare this variable between the groups. χ2 tests were used to compare differences between categorical variables. Logistic regression models were constructed to evaluate differences in the prevalence of various health problems between the total PD and control groups after controlling for age, gender, marital status, hospital type and LOS. A 5% level of significance was used. Data analysis was performed using SPSS for Windows, V.20 (SPSS Inc, Chicago, Illinois, USA).

    Results

    Demographic characteristics

    There were 5637 patients receiving inpatient care for PD over the 5-year study period and these patients comprised the study subjects. These episodes of care for patients with PD represented approximately 0.04% of all the separations from NSW hospitals over that time and corresponded to a rate of 1127 separations/year from an estimated NSW PD patient population of 20 597 patients (54.7/1000 patients with PD per year).7 ,19 Of these, 1574 (27.9%) separations were for PD as a principal diagnosis and 4063 (72.1%) as a secondary diagnosis. The control group consisted of 8143 patients who received treatment in a NSW hospital and who did not have either a principal or secondary diagnosis of PD. The demographic and hospitalisation characteristics of the study and control groups are shown in table 1.

    View this table:
    Table 1

    Clinical and demographic characteristics during acute care admission of PD and control patients, NSW 2008–2012

    The mean age of all patients with PD was 75.0 years (±10.9) (range 30–100 years), which was slightly higher than that of the control group (73.1 years (±13.9)). There were proportionally more men in the PD than the control group (62.8% and 52.9%, respectively, χ2=131.4, df=1, p<0.001, table 1) and marital status differed significantly between the two groups (χ2=235.0, df=3, p<0.001) with the PD group having a greater proportion of married or de facto couples than the control sample (62.8% vs 50.6%, respectively). Three quarters of the PD group were treated in a private hospital compared with only 30% of the control group (75.7% vs 30.3%, χ2=2752.1, df=1, p<0.001).

    Within the total PD patient sample, patients with PD as a principal diagnosis were younger (68.2±11.1 years vs 77.6±9.5 years, t=−31.7, df=5635, p<0.001, table 2); were more likely to be married or in a de facto relationship (73.6% vs 58.6%, χ2=205.5, df=3, p<0.001) and be admitted to a private hospital (90.4% vs 70.1%, χ2=253.7, df=1, p<0.001) than those with PD as a secondary diagnosis.

    View this table:
    Table 2

    Clinical and demographic characteristics during acute care admission of PD patients, NSW 2008–2012 by principal PD diagnosis or secondary diagnosis

    Clinical characteristics

    Patients with PD had a significantly longer LOS than the control group (median LOS 7 vs 1 day, U test=13 705 361.0, df=1, p<0.001, table 1). There were also differences between the two groups with respect to the mode of admission (χ2=199.4, df=2, p<0.001) with more patients with PD having a planned admission than controls (58.4% vs 51.3%). In addition, there were significant differences between the two groups in their mode of separation (discharge outcome) (χ2=141.3, df=3, p<0.001). Fewer patients with PD than controls were discharged to their usual place of residence (69.4% vs 76.3%), and more patients with PD were transferred to other hospitals (13.5% vs 10.3%) or to nursing homes (5.8% vs 4.4%). There were 209 deaths in the total PD group, corresponding to an in-hospital mortality rate of 37.1/1000 separations, which was not significantly different from that in the control group (309 deaths or 38.0/1000, χ2=0.1, df=1, p=0.722).

    Within the sample of all patients with PD, those with a secondary diagnosis of PD had a longer LOS (median LOS 7 vs 6 days, U test=3 411 066.0, df=1, p<0.001, table 2). Differences were noted with respect to the mode of admission (χ2=528.9, df=2, p<0.001) with more patients with a secondary diagnosis of PD being admitted through the emergency department than those with a principal diagnosis of PD (42.2% vs 12.0%). Further, there were significant differences between the two PD groups in their mode of separation (χ2=215.5, df=3, p<0.001). Fewer patients with a secondary diagnosis than a principal diagnosis of PD were discharged to their usual place of residence (63.9% vs 83.7%); however, they were more likely to be transferred to other hospitals (16.2% vs 6.4%) or to nursing homes (6.7% vs 3.5%). There were 182 deaths in the group with a secondary diagnosis of PD, corresponding to an in-hospital mortality rate of 44.7/1000 separations, which was significantly higher than for those with a principal diagnosis of PD (27 deaths or 17.2/1000, χ2=25.5, df=1, p<0.001).

    Health problems during admission

    The frequencies of a number of health problems recorded for patients with PD and the control group are shown in table 3. Among the patients with PD, the most common health problems were falls/fractures (19.8%); cardiac diseases (ischemic heart disease/heart failure/arrhythmia) (16.7%); complications arising from dementia (14.4%); deep brain stimulation management (DBS) (14.1%) and psychiatric illness (13.6%). There were 794 admissions for DBS, corresponding to a rate of 28.2/1000 patients with PD per year. Of these, (50.6%) were for insertion and management of the device, (15.6%) were for removal, while remaining admissions (33.8%) were for programming and management. Patients who had in-hospital DBS management were significantly younger (t=30.4, df=5636, p<0.001), were more frequently in a married or in a de facto relationship (χ2=246.5, df=3, p<0.001) and were much more likely than patients in the control group to be treated in a private hospital (χ2=251.8, df=1, p<0.001). In fact, 98.1% of all DBS procedures were performed in the private sector. There were no patients in the control group who had DBS.

    View this table:
    Table 3

    Health problems during acute care admission of PD and control patients, NSW 2008–2012

    Overall patients with PD were approximately twice as likely than controls to have a hospital admission for falls/fractures (19.8% vs 8.6%), dementia (14.4% vs 5.1%), psychiatric illness (13.6% vs 7.9%), gastrointestinal complications (diverticular disease/constipation/dysphagia) (13.3% vs 6.5%), syncope/orthostatic hypotension (11.5% vs 3.8%), genitourinary infections (10.6% vs 4.6%), encephalopathy/delirium (10.3% vs 1.8%), reduced mobility/motor fluctuations (10.1% vs 4.5%), pneumonia (9.8% vs 5.5%), spinal pain (5.5% vs 3.0%), adverse drug events (5.1% vs 1.7%), other trauma (4.3% vs 2.0%) (all p<0.001), sleep disorders/restless legs syndrome (1.8% vs 1.4%, χ2=4.9, df=1, p=0.026), venous thromboembolism (1.4% vs 0.9%, χ2=7.0, df=1, p=0.008) and electroconvulsive therapy (0.8% vs 0.4%, χ2=11.4, df=1, p=0.001, table 3). By comparison, the patients with PD had proportionally fewer admissions in which there was a record of cardiac disease (16.7% vs 24.0%), neoplasia (7.3% vs 18.4%) and chronic airways disease (2.9% vs 7.2%) (all p<0.001). There was no difference in the proportions of each group who were recorded as having management of anaemia and stroke/TIA (table 3). Within the patient subgroup with a record of any neoplasm, there were fewer patients with melanoma in the PD group than the control group (0.25% vs 0.51%, χ2=5.9, df=1, p=0.015).

    Patients with PD as a principal diagnosis were approximately ten times more likely to have DBS recorded during their admission than patients with a secondary diagnosis of PD (41.1% vs 3.6%, table 4). By comparison, patients with a secondary diagnosis of PD had higher levels of falls/fractures (24.1% vs 8.6%), cardiac disease (21.7% vs 3.8%), dementia (16.8% vs 8.1%), gastrointestinal complications (15.4% vs 7.9%), syncope (13.5% vs 6.3%), pneumonia (12.5% vs 2.8%) and neoplasia (9.6% vs 1.3%) (table 4). Significantly more patients with a secondary diagnosis of PD had electroconvulsive therapy during their admission than those with a principal PD diagnosis (1.1% vs 0.1%, χ2=14.5, df=1, p<0.001). Interestingly, reduced mobility/motor fluctuations were recorded as a diagnosis in approximately 10% of the patients in both these groups (table 4).

    View this table:
    Table 4

    Health problems during acute care admission of PD patients, NSW 2008–2012 by principal PD diagnosis or secondary diagnosis

    Logistic regression models were developed to further evaluate the significance of differences between the total PD and control groups for specific health problems during admission. Statistical significance persisted after controlling for age, gender, marital status, hospital type and LOS for differences in encephalopathy/delirium (Wald χ2=310.0, df=5, p<0.001), adverse drug events (Wald χ2=114.4, df=5, p<0.001), falls/fractures (Wald χ2=431.5, df=5, p<0.001), dementia (Wald χ2=410.1, df=5, p<0.001), pneumonia (Wald χ2=111.3, df=5, p<0.001), syncope/orthostatic hypotension (Wald χ2=265.8, df=5, p<0.001), neoplasia (Wald χ2=315.2, df=5, p<0.001) and reduced mobility (Wald χ2=201.7, df=5, p<0.001).

    Discussion

    In our large population-based study, we identified many clinically important issues that are experienced more frequently by patients with PD than other patients. Hospitalisations for patients with PD in NSW occurred more often in the private sector, more frequently in men and those in a married or de facto relationship and were associated with a longer hospital stay than patients in the control group. In this study, we found that a greater proportion of the PD patients were treated in hospital following a planned admission and more of these patients were transferred to other hospitals or nursing homes postdischarge. Patients with PD receiving treatment in hospital had higher levels of reported health problems than controls for falls/fractures, dementia, DBS, psychiatric illness, gastrointestinal complications, syncope/orthostatic hypotension, genitourinary infections, encephalopathy/delirium, reduced mobility/motor fluctuation, pneumonia, spinal pain, adverse drug events, other trauma, sleep disorders/restless legs syndrome, venous thromboembolism and electroconvulsive therapy. However, patients with PD had a lower proportion of cardiac diseases, neoplasia and chronic airways disease than controls and the frequency of anaemia and stroke/TIA was comparable between the two groups.

    The results of the present study are similar to those from other international studies, which found that admissions for falls, mobility complications, pneumonia, psychiatric problems, genitourinary infections and trauma were more prevalent among patients with PD than controls.9 ,12 ,15 ,16 ,20 Previously identified common reasons for hospitalisation of patients with PD worldwide were also identified in our study.8–12 ,16 ,20–24 From an Australian perspective, we found higher levels of morbidity in patients with PD with regards to admission for falls/fractures, cardiac disease, syncope, gastrointestinal complications, urinary disorders, dementia and encephalopathy than previously described.10 Our sample not only reflected the experience of patients in the larger cities but also included patients hospitalised across regional and rural areas who tend to be older and are diagnosed at a later age than patients from metropolitan areas.25 These characteristics may be reflected in higher levels of age-related complications experienced during hospitalisation by patients in this sample and this, in turn, may have influenced their higher levels of morbidity. Further, the patients in our study had a shorter hospital stay than reported in other groups.9 ,23 This may be a reflection of the more integrated approach to hospital care in Australia for patients with PD, which aims to incorporate the early use of allied health and acute medical units.26 ,27 Alternatively, it may be a reflection of the slightly higher proportion of planned admissions in our PD group.

    We found that there were a number of health problems in this patient group that have been previously unrecognised in other surveys of patients with PD. In our sample, patients with PD were five time more likely to be treated for delirium, three times more likely to experience an adverse drug event and syncope, more than twice as likely to require management of falls/fractures, dementia, gastrointestinal complications, genitourinary infections, reduced mobility and other trauma than compared with a control group weighted for the age and gender distribution of patients with PD in Australia. These problems should be considered major drivers for hospitalisation and healthcare in PD. We suggest that clinicians should focus on addressing the complexity of presenting problems and the complications associated with PD hospitalisations and that additional health resources should be allocated to assist them. Such efforts should include increased access to specialists by outpatients, which has been shown elsewhere to prevent unnecessary hospitalisations in PD,11 as well as increasing the number of dedicated inpatient facilities akin to stroke units, in order to minimise adverse health outcomes and potentially reduce inpatient LOS. Further, it has been suggested that the quality of care during hospitalisation and patient motor outcomes could be improved by addressing preventable medication errors and allowing selected patients to take control of their own PD medication.28 This is because non-neurology admitting teams, including nursing staff without PD training, may not be as familiar with the time criticality of PD medications, especially in patients with motor fluctuations.

    Interestingly, we were able to report that patients with PD were half as likely to require hospitalisation for chronic airways disease and neoplasia, with significantly lower rates of in-hospital management of melanoma than in controls. The differences between the two groups with regards to neoplasia treatment persisted even after controlling for age and gender differences. Smoking prevalence has been shown to be lower in patients with PD,29 likely limiting the risk of developing smoking-related cancers.30 ,31 The lower rates of admission of patients with PD for treatment of chronic airways disease found in this study may be a reflection of lower smoking rates in this PD population. Overall cancer risks have been shown to be lower in PD compared with control populations,30 ,32 despite evidence that suggests that patients with PD are more likely to develop melanoma.30 ,33 Past case reports have speculated that there may be a positive association between levodopa use and melanoma risk,34 but these associations are largely unverified and remain controversial. The patients with PD in this study had lower levels of cardiac disease than the control group, which is consistent with the results from previous analyses.15 ,20

    In our PD patient sample, we identified that 14.1% of the patients received hospital treatment for DBS management, of which more than half were new insertion procedures. Patients were significantly more likely to receive DBS if they were younger, married and were treated in a private hospital. DBS is well regarded as an important treatment option for patients with advanced PD.35 ,36 Across tertiary medical centres, DBS surgery is usually considered a leading cause for elective hospitalisation in patients with PD.9 Surgical, as well as hardware-related complications, are also known to occur in this group of patients, in addition to the recognised complications of medical therapy during a hospitalisation.37

    We used hospital administrative data to examine patterns of morbidity in patients with PD. While these data provide very valuable information, they have some limitations. These include the possibility that there may be misclassification of patients with other causes of PD into the group classified as idiopathic PD used in this study. In addition, the data are in the form of unmatched patient records, and, therefore, it is not possible to identify multiple admissions for individual patients in this analysis. Administrative underreporting of patient comorbidities at the time of patient discharge may have underestimated the number of associated diagnoses attributed to an admission. However, this problem may be minimised by the activity-based funding used in the Australian healthcare system. In addition, the large sample sizes in this study may have contributed to the high levels of statistical significance identified in many of the comparisons. For this reason, we have chosen to focus on the clinical importance of the observed differences. In addition, we used regression models to explore the relationships between admissions after controlling for demographic differences and comorbidities between the sample groups. Lastly, the reduced LOS in the control group compared with the PD group may reflect a different distribution of health problems and their treatments, where patients with PD may use fewer short stay preventive or therapeutic procedures, such as colonoscopies and cataract surgery, compared with non-PD patients.

    Conclusion

    Our study identified differences in the demographic characteristics, health problems and treatments in hospital between patients with PD and a control group of patients without PD. Patients with PD had a considerably higher proportion of hospitalisations with problems, including delirium, adverse drug events, syncope, falls/fractures, dementia, gastrointestinal complications, genitourinary infections, reduced mobility and other trauma. These findings highlight the need for prehospital management and prevention of these issues through specialist clinics as well as primary care. We suggest that clinicians in hospitals should identify those patients at risk of complications early and work with multidisciplinary teams to ensure complications are minimised. Gait and balance assessments, adopting falls prevention strategies, early swallowing and speech therapy reviews, nutritional and dietary support, education on the common adverse drug events in PD, as well as timely administration of medication would be of benefit to the hospitalised PD patient. Further studies investigating risk factors for hospitalisation such as those by Hassan et al, identifying the effects of disease severity, patient comorbidities and DBS on hospital encounters,38 are vital for informing clinicians about strategies to avoid preventable admissions.

    Acknowledgments

    We acknowledge the support of Jithendra Uppalapati and John Agland from the Health System Information and Performance Reporting Branch of the NSW Ministry of Health for their support with data extraction. Further, we would like to acknowledge Mihovil Matic, senior analyst, for his support with analysis.

    References

    1. ↵
      1. Chan DK,
      2. Cordato D,
      3. Karr M,
      4. et al
      . Prevalence of Parkinson's disease in Sydney. Acta Neurol Scand 2005;111:7–11.
      OpenUrlPubMed
    2. ↵
      1. Chou KL,
      2. Zamudio J,
      3. Schmidt P,
      4. et al
      . Hospitalization in Parkinson disease: a survey of National Parkinson foundation centers. Parkinsonism Relat Disord 2011;17:440–5.
      OpenUrlCrossRefPubMed
    3. ↵
      1. Dorsey ER,
      2. Constantinescu R,
      3. Thompson JP,
      4. et al
      . Projected number of people with Parkinson disease in the most populous nations, 2005 through 2030. Neurology 2007;68:384–6.
      OpenUrlCrossRef
    4. ↵
      Access Economics. Living with Parkinson's Disease ‘Challenges and Positive Steps for the Future’. Report for Parkinson's Australia, June 2007.
    5. ↵
      1. Bucks RS,
      2. Cruise KE,
      3. Skinner TC,
      4. et al
      . Coping processes and health-related quality of life in Parkinson's disease. Int J Geriatr Psychiatry 2010;26:247–55.
      OpenUrlPubMed
    6. ↵
      1. Lubomski M,
      2. Rushworth RL,
      3. Lee W,
      4. et al
      . Sex differences in Parkinson's disease: a cross sectional analysis. J Clin Neurosci 2014. doi.org/10.1016/j.jocn.2013.12.016.
    7. ↵
      Access Economics. Living with Parkinson's Disease—update ‘Challenges and Positive Steps for the Future’. Report for Parkinson's Australia. October 2011.
    8. ↵
      1. Martignoni E,
      2. Godi L,
      3. Citterio A,
      4. et al
      . Comorbid disorders and hospitalisation in Parkinson's disease: a prospective study. Neurol Sci 2004;25:66–71.
      OpenUrlPubMedWeb of Science
    9. ↵
      1. Aminoff MJ,
      2. Christine CW,
      3. Friedman JH,
      4. et al
      . Management of the hospitalized patient with Parkinson's disease: current state of the field and need for guidelines. Parkinsonism Relat Disord 2011;17:139–45.
      OpenUrlCrossRefPubMed
    10. ↵
      1. Temlett JA,
      2. Thompson PD
      . Reasons for admission to hospital for Parkinson's disease. Intern Med J 2006;36:524–6.
      OpenUrlCrossRefPubMedWeb of Science
    11. ↵
      1. Klein C,
      2. Prokhorov T,
      3. Miniovitz A,
      4. et al
      . Admission of Parkinsonian patients to a neurological ward in a community hospital. J Neural Transm 2009;116:1509–12.
      OpenUrlCrossRefPubMed
    12. ↵
      1. Woodford H,
      2. Walker R
      . Emergency hospital admissions in idiopathic Parkinson's disease. Mov Disord 2005;20:1104–8.
      OpenUrlCrossRefPubMedWeb of Science
    13. ↵
      1. Hely MA,
      2. Reid WG,
      3. Adena MA,
      4. et al
      . The Sydney multicenter study of Parkinson's disease: the inevitability of dementia at 20 years. Mov Disord 2008;23:837–44.
      OpenUrlCrossRefPubMedWeb of Science
    14. ↵
      1. Hely MA,
      2. Morris JG,
      3. Traficante R,
      4. et al
      . The Sydney multicentre study of Parkinson's disease: progression and mortality at 10 years. J Neurol Neurosurg Psychiatry 1999;67:300–7.
      OpenUrlAbstract/FREE Full Text
    15. ↵
      1. Guttman M,
      2. Slaughter PM,
      3. Theriault ME,
      4. et al
      . Parkinsonism in Ontario: comorbidity associated with hospitalization in a large cohort. Mov Disord 2004;19:49–53.
      OpenUrlCrossRefPubMedWeb of Science
    16. ↵
      1. Gerlach OH,
      2. Winogrodzka A,
      3. Weber WE
      . Clinical problems in the hospitalized Parkinson's disease patient: systematic review. Mov Disord 2011;26:197–208.
      OpenUrlCrossRefPubMed
    17. ↵
      AIHW. Australian hospital statistics 2009–10. Health services series no. 40. Cat. no. HSE 107. Canberra: AIHW, 2011. Viewed 16 November 2013 http://www.aihw.gov.au/publication-detail/?id=10737418863.
    18. ↵
      WHO. International Statistical Classification of Diseases and Related Health Problems 10th Revision. 2010. Viewed 21 November 2013. http://apps.who.int/classifications/icd10/browse/2010/en.
    19. ↵
      Australian Bureau of Statistics. Themes: Data & analysis. Viewed 06 December 2013. 2013. http://www.abs.gov.au/websitedbs/censushome.nsf/home/data?opendocument&navpos=200.
    20. ↵
      1. Vossius C,
      2. Nilsen OB,
      3. Larsen JP
      . Parkinson's disease and hospital admissions: frequencies, diagnoses and costs. Acta Neurol Scand 2010;121:38–43.
      OpenUrlCrossRefPubMed
    21. ↵
      1. Tan LC,
      2. Tan AK,
      3. Tjia HT
      . The profile of hospitalised patients with Parkinson's disease. Ann Acad Med Singapore 1998;27:808–12.
      OpenUrlPubMed
    22. ↵
      1. Cosentino M,
      2. Martignoni E,
      3. Michielotto D,
      4. et al
      . Medical healthcare use in Parkinson's disease: survey in a cohort of ambulatory patients in Italy. BMC Health Services Research 2005;5:26.
      OpenUrlCrossRefPubMed
    23. ↵
      1. Oguh O,
      2. Videnovic A
      . Inpatient management of Parkinson disease: current challenges and future directions. Neurohospitalist 2012;2:28–35.
      OpenUrlCrossRefPubMed
    24. ↵
      1. Guneysel O,
      2. Onultan O,
      3. Onur O
      . Parkinson's disease and the frequent reasons for emergency admission. Neuropsychiatr Dis Treat 2008;4:711–14.
      OpenUrlPubMed
    25. ↵
      1. Lubomski M,
      2. Rushworth RL,
      3. Lee W,
      4. et al
      . A cross-sectional study of clinical management, and provision of health services and their utilisation, by patients with Parkinson's disease in urban and regional Victoria. J Clin Neurosci 2013;20:102–6.
      OpenUrlCrossRefPubMed
    26. ↵
      1. Scott I,
      2. Vaughan L,
      3. Bell D
      . Effectiveness of acute medical units in hospitals: a systematic review. Int J Qual Health Care 2009;21:397–407.
      OpenUrlAbstract/FREE Full Text
    27. ↵
      1. Daly S,
      2. Campbell DA,
      3. Cameron PA
      . Short-stay units and observation medicine: a systematic review. Med J Aust 2003;178:559–63.
      OpenUrlPubMedWeb of Science
    28. ↵
      1. Gerlach OH,
      2. Broen MP,
      3. Weber WE
      . Motor outcomes during hospitalization in Parkinson's disease patients: a prospective study. Parkinsonism Relat Disord 2013;19:737–41.
      OpenUrlCrossRefPubMed
    29. ↵
      1. Nefzger MD,
      2. Quadfasel FA,
      3. Karl VC
      . A retrospective study of smoking in Parkinson's disease. Am J Epidemiol 1968;88:149–58.
      OpenUrlAbstract/FREE Full Text
    30. ↵
      1. Becker C,
      2. Brobert GP,
      3. Johansson S,
      4. et al
      . Cancer risk in association with Parkinson disease: a population-based study. Parkinsonism Relat Disord 2010;16:186–90.
      OpenUrlCrossRefPubMed
    31. ↵
      1. Allam MF,
      2. Campbell MJ,
      3. Hofman A,
      4. et al
      . Smoking and Parkinson's disease: systematic review of prospective studies. Mov Disord 2004;19:614–21.
      OpenUrlCrossRefPubMedWeb of Science
    32. ↵
      1. Bajaj A,
      2. Driver JA,
      3. Schernhammer ES
      . Parkinson's disease and cancer risk: a systematic review and meta-analysis. Cancer Causes Control 2010;21:697–707.
      OpenUrlCrossRefPubMedWeb of Science
    33. ↵
      1. Zanetti R,
      2. Loria D,
      3. Rosso S
      . Melanoma, Parkinson's disease and levodopa: causal or spurious link? A review of the literature. Melanoma Res 2006;16:201–6.
      OpenUrlCrossRefPubMedWeb of Science
    34. ↵
      1. Fiala KH,
      2. Whetteckey J,
      3. Manyam BV
      . Malignant melanoma and levodopa in Parkinson's disease: causality or coincidence? Parkinsonism Relat Disord 2003;9:321–7.
      OpenUrlCrossRefPubMedWeb of Science
    35. ↵
      1. Gregory R
      . Surgery for movement disorders. J Neurol Neurosurg Psychiatry 2002;72(Suppl 1):I32–5.
      OpenUrlPubMed
    36. ↵
      1. Ngoga D,
      2. Mitchell R,
      3. Kausar J,
      4. et al
      . Deep brain stimulation improves survival in severe Parkinson's disease. J Neurol Neurosurg Psychiatry 2013;85:17–22.
      OpenUrlPubMed
    37. ↵
      1. Morishita T,
      2. Foote KD,
      3. Burdick AP,
      4. et al
      . Identification and management of deep brain stimulation intra- and postoperative urgencies and emergencies. Parkinsonism Relat Disord 2010;16:153–62.
      OpenUrlCrossRefPubMed
    38. ↵
      1. Hassan A,
      2. Wu SS,
      3. Schmidt P,
      4. et al
      . High rates and the risk factors for emergency room visits and hospitalization in Parkinson's disease. Parkinsonism Relat Disord 2013;19:949–54.
      OpenUrlCrossRefPubMed

    Footnotes

    • Competing interests None.

    • Ethics approval The University of Notre Dame Australia, Human Research and Ethics Committee.

    • Provenance and peer review Not commissioned; externally peer reviewed.