Article Text
PDF
Abstract
Objective To evaluate the prevalence of psychosis associated with Parkinson's disease (PSY-PD) in its early stages, its incidence over a 24 month follow-up period and the association with motor and non-motor clinical features.
Methods PRIAMO is a 2 year longitudinal observational study that has enrolled patients with parkinsonism in 55 Italian centres. A cohort of 495 patients with early disease stage PD (baseline Hoehn and Yahr score ≤2, disease's duration (median) 3.4 years) were followed for 2 years. PSY-PD was evaluated by means of a clinician rated questionnaire and defined as the presence of at least one of the following symptoms occurring for at least 1 month: illusions, hallucinations, jealousy ideas and persecutory ideas. Patients with and without PSY-PD were compared on several clinical variables, encompassing motor and non-motor features.
Results The prevalence of PSY-PD at baseline was 3%; the incidences at 12 and 24 months were 5.2% and 7.7%, respectively. Longer disease duration and prescription of dopamine agonists at baseline were associated with the development of PSY-PD over the 24 month period. At this follow-up time, worse disease severity, decline in cognitive performances, presence of depressive symptoms and anxiety were more frequently observed in PSY-PD.
Conclusions Psychotic type symptoms may occur in the early stages of PD although less frequently than in later stages. Beyond dopaminergic treatment, there are disease related factors, such as disease severity and the occurrence of cognitive and depressive symptoms, which may underlie the onset of psychotic type symptoms from the earliest stages.
Statistics from Altmetric.com
Introduction
Psychotic symptoms, mainly visual hallucinations, affect 30–40% of Parkinson's disease (PD) patients, occurring in the later stages of the disease and are associated with poor prognosis, greater caregiver stress, higher admission rates to nursing homes and increased mortality.1 2 As well as visual hallucinations, other psychotic symptoms, such as paranoid delusions (often pathological jealousy), can develop in PD.3 4 Along with dopaminergic treatment, there are disease related factors, other than duration and severity,5 6 which play a pivotal role in the induction of PD psychosis.7 Indeed, cognitive impairment,3 5 6 sleep–wake disturbances8 and depression9 are now recognised as risk factors for hallucinations in PD patients with advanced disease.
Although psychosis typically tends to occur later in the course of the disease,10 it was recently reported that a small proportion of PD patients may develop hallucinations in the earlier stages.6 11 12 However, little is known about the prevalence of psychotic type symptoms in early stage PD and the factors associated with their occurrence. Specifically, the contribution of non-motor symptoms (NMS), such as depression and cognitive impairment, in early disease is unclear. The PRIAMO study is a large Italian multicentre observational study, including a cross sectional and longitudinal prospective 24 month phase, which was designed to assess the prevalence and evolution of NMS in patients affected with different parkinsonian syndromes.13 14 The aim of this subanalysis of the PRIAMO study was to evaluate, in the subgroup of patients with PD in early disease stage: (1) the prevalence of psychotic symptoms at baseline and the new incident cases after 1 and 2 years of follow-up; and (2) the clinical variables associated with psychotic symptom development.
Materials and methods
Patients
In the PRIAMO study, patients with different forms of parkinsonism (diagnosed previously or at baseline) were enrolled and followed-up for 2 years from July 2006 to July 2008. Participating centres (see supplementary data available online only) were selected on the basis of previous participation in clinical studies and proven ability to administer study tools. Diagnosis of idiopathic PD was based on Gelb et al criteria.15 Study methods and patient baseline features have been extensively described elsewhere.13 14 Each patient underwent a baseline and two follow-up visits, at 12 months (±4) after the baseline visit and 9–16 months after the first follow-up visit, respectively (for further details, see supplementary material available online only). PD associated psychosis was specifically evaluated by means of a clinician rated questionnaire and was defined as the presence of at least one of the following symptoms, selected on the basis of the existing literature: illusions (misinterpretations of existing stimuli), hallucinations2 3 (defined as hallucinatory symptoms), jealousy ideas and persecutory ideas2 9 (defined as delusional symptoms). Illusions which were clinically judged to be due to a primary visual disorder were not considered as part of the PD associated psychosis spectrum. To be considered as PD associated psychosis, symptoms had to arise after the onset of PD and to have been recurrent or continuous for at least 1 month. Both the patient and caregiver contributed to the assessment of such symptoms. All clinicians who had to administer the questionnaire participated in a training session before starting enrolment for the PRIAMO study.
For this subanalysis, only patients with early stage PD (namely, Hoehn and Yahr (HY) stage ≤2) and with no missing answers to questions related to psychotic symptoms in the clinician rated questionnaire were considered. The study was approved by the ethics committees of the participating centres and all patients provided written informed consent.
Statistical analysis
Subjects without psychotic type symptoms over a follow-up period of 24 months were compared with incident cases with psychosis in terms of the following clinical variables: age, gender, disease duration and age at disease onset, Unified Parkinson's Disease Rating Scale part III (UPDRS-III), modified HY stage, Mini-Mental State Examination (MMSE), 17 item Hamilton Depression Rating Scale, sleep disturbances, including insomnia, REM sleep behavioural disorder (RBD) (defined according to the American Academy of Sleep Medicine)16 and other sleep disturbances (including excessive daytime sleepiness and restless legs). Moreover, they were compared for the presence of the following NMS based on a semistructured interview (see supplementary material available online only): apathy, attention/memory impairment, psychiatric symptoms in the anxiety/depressive domain (including anhedonia, anxiety, panic attacks, aggressive behaviour, suicidal ideas, nervousness, fear without reason, sadness/depression) and five senses alterations (including one of the following: smell and taste dysfunction and dyplopia). Continuous normally distributed variables are expressed as mean±SD and comparisons between groups were performed using the parametric Student's t test (or in the case of deviation from normality, a non-parametric Mann–Whitney test on median). Differences between categorical variables were tested by means of χ2 or Fisher test. A p value of 0.05 was considered significant. To test the association between PD associated psychosis and the considered clinical variables, logistic regression models were used applying the algorithm proposed by Bursac.17 Clinical variables having a relevant weight on the Bursac's interactive model were considered as predictors in the final logistic regression models. In particular, we set the confounding level to 15% and the non-candidate inclusion level to 0.15. Analyses were performed with SAS V.9.1 (SAS Institute Inc).
Results
Of 1142 PD patients enrolled in the PRIAMO study, 495 early stage PD patients with HY stage ≤2, assessed at all time intervals and with no missing answers in the questionnaire evaluating psychotic symptoms, were eligible for this subanalysis. Sample selection in showed in figure 1. Clinical features of these patients at baseline were: mean disease duration 3.4 years (IQR 1.5–6.1); mean age at disease onset 61.4 (SD 10.0) years, mean UPDRS-III score 18.1 (SD=9.1). HY stage was distributed as follows: in 26.1%, HY=1; in 30.9%, HY=1.5; in 43%, HY=2. Follow-up visits at 12 months and 24 months were conducted after a mean of 1.03 (SD 0.08, range 0.82–1.34) and 2.04 (SD 0.10, range 1.74–2.42) years.
Flowchart illustrating how Parkinson's disease patients with Hoehn and Yahr (HY) stage ≤2 were selected from the original PRIAMO study sample. PD, Parkinson's disease.
Prevalence and incidence of PD associated psychosis
Prevalence at baseline was 3.0%, whereas 12 and 24 month incidences were 5.2% and 7.7%, respectively. The 24 month prevalence of PD associated psychosis was 10.5%. Of 480 PD patients who were psychosis free at baseline, 443 patients never had psychotic type symptoms during the 24 month follow-up period (hereafter known as ‘non-PSY-PD’) whereas 37 developed psychotic type symptoms (hereafter known as ‘PSY-PD’) over the follow-up period. The frequency of hallucinatory and delusional symptoms is shown in table 1. The number of patients with PD associated psychosis only at 12 month was 12/37 (32.4%); 13 patients had psychotic type symptoms at either 12 or 24 months and 12 patients only at 24 months.
Frequency of psychotic symptoms in patients with psychosis at baseline and incident psychosis at 12 and 24 months according to their phenomenology
PSY-PD versus non-PSY-PD analysis
At baseline, PSY-PD had a modest but significantly longer disease duration, higher prevalence of sleep disturbances (different from insomnia and RBD) and were more frequently prescribed dopamine agonists at baseline. The groups did not statistically differ for the other variables, and specifically for HY and MMSE score (table 2).
Univariate statistics at baseline comparing parkinsonian patients with psychosis and parkinsonian patients without psychosis
At the 12 month follow-up visit, PSY-PD presented with worse disease severity (by HY score), lower MMSE score and higher HRDS score than non-PSY-PD (table 3). Slight but significant differences were also present for RBD and other sleep disturbances. Furthermore, psychiatric symptoms in the anxiety/depressive domain were more frequent in PSY-PD.
Univariate statistics at 12 and 24 months comparing parkinsonian patients with psychosis and parkinsonian patients without psychosis
At the 24 month follow-up, worse disease severity (according to HY stage) and lower MMSE score (table 3) were confirmed as features of PSY-PD. Moreover, significant differences were also observed in UPDRS-III score, RBD, insomnia, other sleep disturbances and five senses disturbances. Again, PSY-PD complained more frequently of non-specific psychiatric symptoms. Finally, PSY-PD were more likely to have l-dopa prescribed than non-PSY-PD (table 3).
Factors associated with development of PD associated psychosis
Clinical features associated with the development of new PSY-PD cases during the 24 month period were explored by means of a multivariate logistic regression model. Only subjects with valid data for all of the variables were considered in the model; we included 450/480 subjects (93.8%), 37 with PSY-PD. After application of the Bursac algorithm, the final main effect model included sleep disturbances different from insomnia and RBD, disease duration, age at disease onset and dopamine agonists (table 4).
Clinical factors associated with the development of psychosis over 24 months: final logistic regression model
In order to test the association between clinical variables and PSY-PD at the 24 month follow-up visit, we took into account seven variables in the logistic model: HY stage, UPRDS-III score, MMSE score, apathy, psychiatric symptoms in the depressive domain as previously described and treatment prescribed at the 24 month follow-up visit (ie, l-dopa (yes/no), dopamine agonists (yes/no) and other parkinsonian drugs (yes/no)). HY stage (p=0.002), MMSE score (p=0.006) and presence of psychiatric symptoms in the depressive domain (p=0.033) were all significantly associated with PSY-PD. Moreover, less frequent prescription of other antiparkinson drugs was significantly related to PSY-PD (p=0.033).
Subanalysis in PD without any dopaminergic pharmacological therapy at study entry
As PSY-PD were more frequently prescribed dopamine agonists at baseline, we conducted a supplementary analysis in order to rule out the confounding effect of treatment regimen. Patients were stratified at baseline on the basis of absence of dopaminergic pharmacological therapy in the 12 months before the start of the study (non-PSY-PD: 49/441=11.11%; PSY-PD: 2/37=5.41%). Fisher's exact test was not significant (p=0.41), indicating that the two groups were not significantly different. Finally, in the logistic regression analysis, elimination of the ‘prescribed dopamine agonists’ variable confirmed ‘disease duration’ as the only significant variable (p=0.003) when we applied this subanalysis. On the basis of these data, we can assume that the treatment regimen was not a determinant confounding factor.
Discussion
Our 2 year study, conducted on a large population of PD patients with an early disease stage (HY≤2) and relatively short disease duration, showed a prevalence of psychotic type symptoms as low as 3% at baseline and an increasing incidence of new PSY-PD cases, reaching 7.7% at the 2 year follow-up. Longer disease duration and prescription of dopamine agonists at baseline were the clinical factors associated with the development of PSY-PD over the 24 month period. At such follow-up time, worse disease severity, decline in cognitive performances, presence of depressive symptoms and anxiety were associated with its development.
Psychotic symptoms, mainly visual hallucinations, have frequently been reported in PD and their prevalence (ranging from 16%18 to 40%) has been related to disease severity and duration.2 10 Although psychosis tends to occur later in the course of the disease,10 a few studies have demonstrated that a small proportion of PD patients might develop hallucinations at earlier stages6 11 12 18: less than 4% of PD patients developed visual hallucinations within the first 5 years of the disease in a retrospective autoptic study6; in a community based study, hallucinations and vivid dreams were reported, respectively, in 5/235 (2.1%) and 27/235 (7.2%) PD patients with HY stage <2; finally, 14.6% of patients treated with pramipexole and 8.0% of patients treated with l-dopa developed hallucinations within 4 years of therapy in an early PD sample.12 Predictors of hallucinations in such population were having more than five comorbid illnesses, worse cognitive function at baseline and older age.11 Our data confirm that PD associated psychosis is not frequent in earlier disease stages, with increasing prevalence over time.
Similarly to advanced PD stages,3–5 19 a significantly longer disease duration at baseline and a more severe disease progression during the 2 years of follow-up were associated with the development of psychotic type symptoms in early stage PD. Specifically, a faster disease progression was revealed by higher HY stages and UPDRS-III scores at the 2 year follow-up in the PSY-PD group. Along with disease severity and duration, dopaminergic treatment is generally considered as causally related to the development of PD associated psychosis.3 Thus it is not surprising that significantly more frequent prescription of dopamine agonists at baseline was found in PSY-PD over the subsequent 24 months whereas l-dopa was more frequently prescribed to PSY-PD at the 24 month follow-up. On the other hand, at the 24 month follow-up period, other antiparkinson drugs (including anticholinergics, MAO-B inhibitors, COMT inhibitors and amantadine) were less frequently prescribed to PSY-PD, reflecting the clinical practice to first discontinue such drugs for their hallucinogenic potential and mild effect in improving motor features.6 20
Although several studies have shown that dopaminergic drugs have a facilitating effect on PD associated psychosis, other factors such as cognitive impairment,3 18 21 depression,18 22 23 sleep disorders8 24 and visual disturbances25 26 has been linked to it in the advanced disease stages.
The novel data of our study are that, beyond the facilitatory effects of dopaminergic drugs, cognitive decline and depressive/anxiety symptoms were disease related factors strongly associated with the development of psychotic type symptoms from the earliest disease stages. In our sample, cognitive impairment, assessed by MMSE, did not differ at baseline between PSY-PD and non-PSY-PD, but significantly worsened over time at the 12 and 24 month follow-up periods only in patients who developed PSY-PD. The reciprocal association between decline in cognitive function and psychosis has been established in many clinical studies in advanced PD: patients with PD and dementia have a higher prevalence of hallucinations3 21; moreover, non-demented PD patients with hallucinations show reduced performance on tasks exploring executive functioning27 and have a more rapid progression towards cognitive decline than patients without hallucinations28; finally, reduced phonological fluency was the only independent predictor of onset of hallucinations in a 2 year longitudinal neuropsychological study whereas hallucinations independently predicted development of diffuse cognitive impairment.29 In our study, Frontal Assessment Battery (FAB) score, which explores frontal lobe dysfunction, did not differ significantly between PSY and non-PSY-PD patients. The discordance between FAB score and MMSE decline in the PSY-PD group might be due to the deterioration of other cognitive domains, rather than executive functions; as an alternative explanation, it should be considered that, among the frontal domain tested in the FAB, only the go/no-go subtest has been previously correlated with development of psychotic type symptoms.29
Similar to cognitive impairment, psychiatric symptoms in the depressive domain (anhedonia, anxiety, panic attacks, aggressive behaviour, suicidal ideas, nervousness, fear without reason and sadness) were significantly associated with new incident PSY-PD cases in our study. Comorbid depression has been frequently recorded in PD with psychosis18 30 and depressive symptoms have been suggested to predispose to its development.23 Interestingly, Marsh et al demonstrated9 that PSY-PD have a high rate of comorbid psychiatric diseases, mainly in the affective domain, these symptoms being associated with greater cognitive impairment. Along with cognitive impairment and depression, sleep disorders, mainly RBD, have also been identified as risk factors for hallucinations,3 8 19 even though some authors have pointed out that they may not be sufficient to predict psychotic symptoms but probably only increase the probability that such symptoms will occur.31 In our population, sleep disturbances were more frequent in PSY-PD at baseline and, specifically, RBD progressively increased in frequency over time in patients who developed psychotic symptoms. Similarly, alterations of the five senses (reported as taste and smell abnormalities and dyplopia) were more frequently reported by psychotic patients at the 2 year follow-up. Ocular problems and visual disturbances have been frequently linked to hallucinations in PD: hallucinators tend to have ocular complaints more frequently,19 more selective dyplopia of isolated objects or persons32 and reduced visual acuity26 than non-hallucinators.
We acknowledge that our study presents some limitations which need to be addressed carefully. Importantly, we were not able to establish a possible association between onset of psychotic type symptoms and administered medications, as the case record form, designed for the original PRIAMO study purposes, did not record start and end of the different pharmacological treatments, nor the exact date of onset of psychosis. Moreover, diagnosis of PSY-PD was based on a structured questionnaire, as previously done in other clinical longitudinal studies,3 10 19 as there were no standardised criteria for its diagnosis at the time our study was designed. However, the potential of having included false positive cases, given the wide definition of PSY-PD and the inclusions of so called ‘minor hallucinations’ as illusions and false sense of presence, should have been minimised. Indeed, our structured questionnaire mirrored the recently diagnostic criteria for PD associated psychosis proposed by a working group from the National Institutes of Neurological Disorders and Stroke (NINDS) and the National Institute of Mental Health (NIMH) (at least one of the following symptoms: illusions, false sense of presence, hallucinations and delusions).33 Specifically, as in the NINDS/NIMH criteria,33 we chose to include illusions as part of the spectrum of psychotic type symptoms, as supported by recent literature suggesting that these symptoms are phenomenologically similar to hallucinations.3 7 33 34 Related to the small PSY-PD sample selected, we have to recognise that PSY-PD patients were grouped together regardless of whether they had minor type symptoms or more genuine psychotic phenomena; this meant that potential associations between specific symptoms (visual hallucinations rather than persecutory beliefs) and specific predictors could not be examined, due to power concerns.
We also acknowledge that a small proportion of the baseline sample might be contaminated by Lewy body dementia cases as hallucinations early in the course of the disease should alert the clinician towards this diagnosis.35 However, this bias should have been minimised as the diagnosis of PD always preceded the onset of psychosis in our sample, and at baseline none of the included patients had developed dementia. Finally, the lack of a control group from the general population represents another limitation as positive psychotic experiences might be experienced in the general population, although most of the subjects do not display persistence of this symptom.36
In conclusion, our results from a large PD population show that psychotic type symptoms may also occur in the early stages of the disease, although less frequently than in later stages. We observed that, beyond dopaminergic treatment, there are disease related factors, such as disease severity and the occurrence of cognitive and depressive symptoms, which may underlie the onset of psychosis from the earliest stages. Cognitive and affective function should be carefully monitored from the earliest phases of PD, in order to identify those subjects at risk of psychosis who are not usually treated when hallucinations are in their ‘minor’ form.34 As it has been shown that psychotic symptoms tend to progress in severity with PD duration and progression, they should be recognised and managed from their early appearance.37
Acknowledgments
The authors are grateful to Medidata for data collection and analysis and to Sara Rizzoli for help in the writing of the manuscript.
Appendix 1 The Priamo Study Group
Steering Committee:
Angelo Antonini Parkinson Institute, Milano,
Paolo Barone Università Federico II, Napoli,
Carlo Colosimo Università La Sapienza, Roma,
Roberto Marconi Ospedale della Misericordia, Grosseto,
Letterio Morgante Dipartimento di Neuroscienze, Scienze Psichiatriche ed Anestesiologiche, Università di Messina, Italy.
Participating centres:
Caravona N — Dipartimento Scienze Neurologiche, Sapienza Università di Roma, Roma,
Braga M — Neurologia, Ospedale Civile, Vimercate, Milano,
Pellicano C — Ospedale Sant'Andrea, II Facolta' di Medicina e Chirurgia, Sapienza Universita' di Roma, Roma,
Petretta V — Neurologia e Stroke, AORN San Giuseppe Moscati, Avellino,
Di Brigida G — Centro Parkinson — Dipartimento Neuroscienze, Università degli Studi di Genova, Genova
Zappulla S — Neurologia, Ospedale Umberto I, Enna,
Meoni S — Clinica Neurologica, Azienda Ospedaliera Universitaria, Firenze,
Pepe F — Fondazione Salvatore Maugeri IRCCS, Castel Goffredo, Mantova,
Di Giovanni M — UO di Neurologia, Ospedale Regionale, Aosta,
Modica C — UO Neurologia, Ospedale San Biagio di Marsala, Marsala, Trapani
Mucchiut M — Policlinico Universitario, Udine,
Bartalini S — Policlinico Le Scotte, Siena,
Cossu G — Azienda Ospedaliera G Brotzu, Cagliari,
Pennisi F — Ospedale di Castelvetrano, Castelvetrano, Trapani,
Bartolomei L, L'erario R — Ospedale Civile San Bortolo, Vicenza,
Tiple D — Dipartimento Scienze Neurologiche, Sapienza Università di Roma, Roma,
Petrone A — Neurologia, Presidio Ospedaliero Annunziata, Cosenza,
Avarello TP — Neurologia, Ospedale Villa Sofia CTO, Palermo,
Bentivoglio AR — Clinica Neurologica, Universita` Cattolica del Sacro Cuore, Roma,
Scala R — Neurologia, Ospedale S. Maria Loreto Nuovo, Napoli,
Gaglio RM — UO di Neurologia, Azienda Ospedaliera S.Giovanni di Dio, Agrigento,
Giglia L — Neurologia, Azienda Ospedaliera S. Elia, Caltanissetta,
Ceravolo R — Neurologia, Ospedale Santa Chiara, Pisa,
Nicoletti A — Clinica Neurologica I, Policlinico Universitario, Catania,
Trianni G — Neurologia, POF Ferrari, Casarano, Lecce,
Martinelli P — Università di Bologna, Bologna,
Rocco M — Universita` Federico II and IDC-Hermitage-Capodimonte, Napoli, Italy
Moschella V — Policlinico Tor Vergata, Roma,
Volpe G — Neurofisiopatologia Dipartimento Neuro Orto Traumatologia, Presidio Ospedaliero S. Luca, Vallo della Lucania, Salerno
Perini M — Ospedale S Antonio Abate, Gallarate,
Capus L — Ospedale di Gattinara, Trieste,
Simone P — Ospedale Casa Sollievo della Sofferenza, San Giovanni Rotondo,
Iemolo F — UO di Neurologia, Ospedale Guzzardi, Vittoria, Ragusa,
Grasso L — Dipartimento di Neuroscienze Ospedale della Misericordia, Grosseto
Savica R — Dipartimento di Neuroscienze, Scienze Psichiatriche ed Anestesiologiche Universita` di Messina,
Sensi M — Neurologia Dipartimento Neuroscienze applicate alla clinica, Ospedale Sant'Anna, Ferrara,
Baratti M — Ospedale Ramazzini, Carpi,
Pezzoli G — Parkinson Institute, Istituti Clinici di Perfezionamento, Milano,
Ceravolo MG — Clinica di Neuroriabilitazione, Ospedale Umberto I, Ancona
Del Dotto P — UO di Neurologia, Ospedale Versilia, Camaiore,
Scaglioni A — Ospedale di Vaio, Fidenza,
Soliveri P — Istituto Nazionale Neurologico C Besta, Milano,
Mauro A — Divisione di Neurologia e Neuroriabilitazione, Istituto Auxologico Italiano, IRCCS Piancavallo (Verbania),
Troianiello B — Istituto Clinico Città di Brescia,
Consoli D — Ospedale G Iazzolino, Vibo Valentia,
Cannas A — Clinica Neurologica, Policlinico Universitario, Monserrato, Cagliari,
Marano R — Azienda Ospedaliera Papardo, Messina.
Sponsorship:
Arina Dumitriu, Boehringer Ingelheim, Milano, Italy.
Project Management, Statistical Analyses and Data Management:
Simona Sgarbi, project leader MediData Studi e Ricerche, Modena
Andrea Rapisarda, clinical project manager MediData Studi e Ricerche, Modena
Sara Rizzoli, Lucia Simoni, statisticians MediData Studi e Ricerche, Modena
Luca Zanoli, clinical data manager MediData Studi e Ricerche, Modena
Alessandra Manfredi, clinical operation specialist MediData Studi e Ricerche, Modena, Italy
References
Footnotes
↵* For PRIAMO Study Group see appendix 1 at the end of the article.
Funding This study was supported by an educational grant issued by Boehringer Ingelheim, Italy.
Competing interests LM has received honoraria for lectures and educational activities from UCB Pharma, Glaxo Smithkline and serves on scientific advisory boards for Boehringer Ingelheim. CC received honoraria from serving on the scientific advisory board of Boehringer Ingelheim Pharmaceuticals, Ipsen Pharmaceuticals, Novartis Pharmaceuticals and UCB/Schwarz, and received royalties from CIC Edizioni Internazionali Publishers. RM received honoraria from serving on the scientific advisory board of Boehringer Ingelheim, serves as an editorial board member of the Journal of Neurodegenerative Disorders and has received honoraria for speaking engagements from Boehringer Ingelheim, Glaxo Smithkline, Lundbeck and UCB Pharmaceuticals. AA has received honoraria for consulting services and symposia from Abbott, Boehringer Ingelheim, GSK, Lundbeck, UCB, Novartis and Merck Serono. GM received institutional support from Ministero dell'Istruzione, dell'Università e della Ricerca (MIUR). MP received honoraria for speaking engagements from Boehringer Ingelheim. FEP received honoraria for speaking engagements from Boehringer Ingelheim, Novartis, GSK and received institutional support from Ministero dell'Istruzione, dell'Università e della Ricerca (MIUR). GA received honoraria from serving on the scientific advisory board of Boehringer Ingelheim and Novartis, serves as a member of scientific committee of the journal Equilibri, received honoraria for speaking engagements from Novartis, Lundbeck, Glaxo Smith Kline, UCB and Boehringer Ingelheim, and received institutional support from Ministero dell'Istruzione, dell'Università e della Ricerca. EB received funding for a trip from Novartis and Lilly and received honoraria for speaking engagements from Lilly, GSK, Boehringer Ingelheim and Sanofi Aventis. AB received honoraria for speaking engagements from Boehringer Ingelheim and received institutional support from MIUR. FM receives research support from Neureca Onlus, Milan. PB received honoraria for serving on the scientific advisory board of Boehringer Ingelheim, Merck Serono, Novartis Italy, UCB Italy and Lundbeck Italy, serves as an editorial board member of Journal of Neurodegenerative Disorders and Parkinson's Europe, received honoraria for speaking engagements from Boehringer Ingelheim, Merck Serono, Novartis Italy, UCB Italy and Lundbeck Italy, and received research support from Boehringer Ingelheim, Merck Serono, Novartis Italy, UCB Italy and Lundbeck Italy.
Ethics approval The ethic committees of each participating site approved the study.
Provenance and peer review Not commissioned; externally peer reviewed.