Open access Protocol Multicentre, randomised, single-­blind, parallel group trial to compare the effectiveness of a Holter for Parkinson’s symptoms against other clinical monitoring methods: study protocol Alejandro Rodríguez-­Molinero ‍ ‍,1 Jorge Hernández-­Vara,2 Antonio Miñarro,3 Carlos Pérez-­López,1 Àngels Bayes-­Rusiñol,4 Juan Carlos Martínez-­Castrillo,5 David A Pérez-­Martínez,6 Monitoring Parkinson’s patients Mobility for therapeutic purposes research group To cite: Rodríguez-­Molinero A, ABSTRACT Hernández-­Vara J, Miñarro A, Strengths and limitations of this study Introduction In recent years, multiple studies have et al. Multicentre, randomised, aimed to develop and validate portable technological single-­blind, parallel group trial ►► First clinical trial to assess efficacy of a Parkinson’s devices capable of monitoring the motor complications to compare the effectiveness Holter to improve patients’ motor symptoms. of Parkinson’s disease patients (Parkinson’s Holter). The of a Holter for Parkinson’s ►► Three-­ arm trial comparing the symptomatic con- symptoms against other effectiveness of these monitoring devices for improving trol of patients monitored with a Parkinson’s holter, clinical monitoring methods: clinical control is not known. monitored with a patient’s diary or not monitored. study protocol. BMJ Open Methods and analysis This is a single-­blind, cluster-­ ►► Patients are blind to the study arm. 2021;11:e045272. doi:10.1136/ randomised controlled clinical trial. Neurologists from ►► Neurologists are not blind to the study arm. bmjopen-2020-045272 Spanish health centres will be randomly assigned to one of ►► Observer bias could happen in some secondary out- three study arms (1:1:1): (a) therapeutic adjustment using ►► Prepublication history and comes, which are measured by the neurologists. additional supplemental material information from a Parkinson’s Holter that will be worn by for this paper are available their patients for 7 days, (b) therapeutic adjustment using online. To view these files, information from a diary of motor fluctuations that will be INTRODUCTION please visit the journal online. completed by their patients for 7 days and (c) therapeutic Parkinson’s disease (PD) is the most common To view these files, please visit adjustment using clinical information collected during the journal online (http://​dx.​doi.​ form of chronic and progressive hypoki- consultation. It is expected that 162 consecutive patients org/​10.​1136/​bmjopen-​2020-​ will be included over a period of 6 months. netic syndrome among the elderly popu- 045272). The primary outcome is the efficiency of the Parkinson’s lation and is the second most common Holter compared with traditional clinical practice in neurodegenerative disease after Alzheimer’s Received 28 September 2020 Accepted 27 June 2021 terms of Off time reduction with respect to the baseline disease.1 In early stages, PD responds well (recorded through a diary of motor fluctuations, which to dopaminergic therapy; however, as the will be completed by all patients). As secondary disease progresses, the duration of the effect outcomes, changes in variables related to other motor decreases and motor complications develop complications (dyskinesia and freezing of gait), quality due to ‘wearing off’ effects (end-­of-­dose dete- of life, autonomy in activities of daily living, adherence rioration) or due to a delayed or no response to the monitoring system and number of doctor–patient to medication, which requires frequent thera- contacts will be analysed. The noninferiority of the peutic adjustments to achieve good symptom Parkinson’s Holter against the diary of motor fluctuations in terms of Off time reduction will be studied as the control throughout the day.2 Despite all ther- © Author(s) (or their exploratory objective. apeutic adjustment efforts, 90% of patients employer(s)) 2021. Re-­use have motor complications or fluctuations Ethics and dissemination approval for this study has permitted under CC BY-­NC. No been obtained from the Hospital Universitari de Bellvitge after 10 years.3 These fluctuations consist of commercial re-­use. See rights and permissions. Published by Ethics Committee. The results of this study will inform the changes between periods called Off, in which BMJ. practical utility of the objective information provided by the medication has no effect and mobility For numbered affiliations see a Parkinson’s Holter and, therefore, the convenience of is difficult, and periods called On, in which end of article. adopting this technology in clinical practice and in future patients can move fluidly because the medi- clinical trials. We expect public dissemination of the results cation is having its best effect.4 In addition, in Correspondence to in 2022. Dr Alejandro Rodríguez-­ the transition between these two states (On Trial registration NCT04176302; https://​clinicaltrials.​gov/​ Molinero; and Off) or during the period of maximum show/​NCT04176302 ​rodriguez.​molinero@​gmail.​com medication effect, patients may present Rodríguez-­Molinero A, et al. BMJ Open 2021;11:e045272. doi:10.1136/bmjopen-2020-045272 1 Open access with dyskinesias, that is, involuntary movements of the As secondary objectives, besides security issues and user head, torso or extremities, which may interfere with the satisfaction with the Parkinson’s Holter, the following effi- patient’s activity.5 cacy results will be measured: number of medical contacts, Motor complications in patients with advanced disease adherence to monitoring system, severity of motor are not easy to control; they can have a variable char- complications, severity of freezing of gait, quality of live acter, fluctuating, as mentioned, throughout the day and performance in activities of daily living performance. and between different days. The chronology of symp- toms throughout the day and between different days is of great value for the precise adjustment of the medica- METHODS AND ANALYSIS tion dosage, adapting the scheduled doses to the most Study design prevalent symptoms in the postdose period. However, A single-­blind, cluster-­randomised controlled clinical trial neurologists do not currently have detailed information with three arms (1:1:1): group A (therapeutic adjustment on their patients’ symptom chronology; therefore, they using information from a Parkinson’s Holter); group B have serious difficulties in obtaining good results with (therapeutic adjustment using information from a diary medication adjustments. Currently, the information avail- of motor fluctuations) and group C (the therapeutic able to neurologists on the hourly course of symptoms adjustment is not supported by additional information, comes from the patient’s self-­report during consultation, other than the clinical information collected during or in the best case, from diaries kept by the patient at consultation). home noting their motor state (On or Off) periodically Study setting and duration (eg, every hour).6 Although the latter method continues The study will last a maximum of 9 months for each to be the reference standard in research and care, it patients (3 months from inclusion to basal visit at has serious limitations, as patients often forget to make maximum and 6 months of follow-­up period). The first notes (especially when they are Off), many do not recog- patient was included in November 2019; the estimated nise their motor state well, and few can adhere to such a last visit for the last patient is March 2022. Neurologists laborious system beyond a few days.7 Thus, a system for from at least 40 hospitals in Spain will participate in the measuring motor fluctuations, that is objective, does not study. require intervention on the part of the patient and can, therefore, be part of their day-­to-­day for the long term, Investigational device if necessary, can be of great utility in clinical practice to The Parkinson’s Holter is a commercial product (STAT-­ help optimise medication regimens and improve disease ON) manufactured by Sense4Care SL (​www.​sense4care.​ control.8 com). This medical device is intended to ambulatory During the last decade, our research group has devel- monitor motor manifestations and activity of Parkinson’s oped a system for monitoring patients with PD based patients. The Holter records motor fluctuations (On on accelerometry that can be comfortably worn at the and Off periods) during daily activities,14 in addition to waist during daily activities. This system is capable of dyskinesias, bradykinesia and freezing of gait episodes9–11 detecting various motor symptoms, including bradyki- (figure 1). Holter’s data are stored in its internal memory nesia, freezing of gait and dyskinesia,9–11 establishing the and can be downloaded by users (patients or neurologists) chronology of motor fluctuations (On and Off periods) to any mobile phone that has the application provided and detecting falls.12 13 This system, which, henceforth, by the manufacturer installed. This application produces will be generically referred to as Parkinson’s Holter, is reports in PDF, like the ones shown in figures 2 and 3. possibly the only such system that is easy to carry, is vali- The first report (figure 2) shows a summary of the dated under real conditions of use and provides suffi- data obtained from the patient during the time moni- cient information to improve the medication regimen. tored, including the number of freezing of gait episodes However, it remains a hypothesis that detailed knowl- detected and the percentage of time in On, in Off and in edge of the motor symptoms of patients leads to better status intermediate between the two. The graph shown in disease control, thanks to optimisation of the thera- figure 3 is the most important for clinicians, since it shows peutic regimen. To confirm or refute this hypothesis, we the time course of the different motor symptoms, over propose a clinical trial in which the clinical effectiveness of this device will be analysed in patients with moderate PD and motor fluctuations. The primary objective of this trial is to compare the clinical outcomes in patients with PD, measured as changes from baseline to last visit in daily Off time, in three different arms, according to different sources of information in regards of motor fluctuations: (1) Parkin- son’s Holter, (2) patient’s diary and (3) no information (the only information that the patient can provide at the visit). Figure 1 Parkinson’s Holter. 2 Rodríguez-­Molinero A, et al. BMJ Open 2021;11:e045272. doi:10.1136/bmjopen-2020-045272 Open access recommends using it for 7 days to capture the specific changes in motor manifestations and patient’s routines, which often occur on the weekend. The Parkinson’s Holter user manual is available as online supplemental material. Participants The target population is patients with PD and difficult-­to-­ control motor fluctuations. The neurologists participating in the study will select patients from those undergoing follow-­up in their outpa- tient clinic. In line with the clinical use envisaged for Parkinson’s Holter, neurologists are advised to offer the study to those patients who could benefit from daily monitoring of their motor symptoms, in order to better control them. It is planned to include 162 patients who Figure 2 Parkinson’s Holter summary data table. meet all the following inclusion criteria: (1) idiopathic PD according to the clinical criteria of the Brain Bank of a week of time. It needs to be taken into account that a the UK,15 (2) moderate to severe disease (Hoehn & Yahr proportion of the time monitored cannot be classified in ≥2, in the Off state),16 (3) motor fluctuations present, any of these three motor states (On, Off or intermediate with at least 2 hours/ day in the Off state. The time in states). As a result, the sum of the time in each motor off will be estimated by the neurologist in a first stage state does not reach 100%. Time without classification (according to the clinical information available) and will (represented in grey in figure 3) corresponds to the time be later confirmed by means of a patient’s diary, which all in which there is not enough data for the device algo- candidates will fill in at home before the baseline study rithms to reach a conclusion, which occurs frequently visit (see the Procedures section). To be included in the in prolonged periods of rest of the patient (in some study, previously informed patients will agree to partici- patients this happens in Off, but this information must be pate voluntarily and sign a written consent form. confirmed by the neurologist, through an interview with Patients who are unable to walk independently or with the patient). Hoehn & Yahr=5, patients participating in another clin- The Parkinson’s Holter must be used a minimum of 3 ical trial, patients with acute intercurrent disease, patients days, for calibration reasons, and has no upper temporary with psychiatric or cognitive disorders preventing collab- limit of use (it can be used indefinitely). The manufacturer oration (mini-­ mental status examination <24)17 and Figure 3 Parkinson’s Holter weekly record. Rodríguez-­Molinero A, et al. BMJ Open 2021;11:e045272. doi:10.1136/bmjopen-2020-045272 3 Open access patients with difficulty understanding the study proce- whose neurologist has been assigned to group B will also dures will be excluded. fill in the diary in the intermediate visits of the study. The neurologists will be professionals who care for The Parkinson’s Holter will be delivered to patients patients with PD and who recognise the potential of by courier along with the user manual and a quick start recruiting five patients with difficult-­ to-­ control motor guide. There will be a technical assistance telephone line fluctuations at the time of recruitment foreseen in the at their disposal to answer questions on how to handle the study. device. The device will have been previously configured, so that patients only have to turn it on the first time it is Interventions and randomisation taken out of the box by pressing the only button on the Prior to each visit with their neurologist, all patients partic- device. From that time on, the device will turn on and ipating in the study will be monitored using a Parkinson’s off autonomously depending on the movement detected Holter during 7 days at home. In addition, all patients by its sensors, so patients do not have to perform any of the study will keep a diary of motor fluctuations for 7 other operation. The device will have a charged battery days at home, prior to the first and last study visit to the and autonomy longer than 7 days, so no charger will be neurologist. The Holter and the diary will be delivered provided nor will patients have to worry about recharging and collected by courier. the batteries. After the last day of use, the device will be The neurologists participating in the study will be picked up by courier and transported to the centre that randomly assigned to one of the following three groups: manages the deliveries (which is a centre independent ►► Group A: for therapeutic adjustment, neurologists of the sponsoring entity) to download the collected data. will have access to the information from the Parkin- Simultaneously, patients will fill out a diary of motor son’s Holter (study device) and to the information fluctuations at home. The motor fluctuation diary was collected during consultation. designed by the researchers (figure 4), and the neurolo- ►► Group B: for therapeutic adjustment, neurologists gists participating in the study will explain to the patients will have access to the information from the diary of how to fill it out. To do this, the neurologists will follow a motor fluctuations (reference standard) and to the common procedure that involves showing instructional information collected during consultation. In this videos to patients that provide examples of the different specific group, patients will fill a motor fluctuations phases (On/Off) and motor complications. The diary diary, prior to every scheduled visit (not only in the of motor fluctuations will be collected by courier on first and last visits). the same day as the Holter device. All patients’ diaries ►► Group C: for therapeutic adjustment, neurologists will will be reviewed by a devoted team at baseline. Those only have access to the information collected during diaries with completeness problems, duplicates (simul- a typical consultation, without information from the taneous On and Off entries) or mayor inconsistencies Holter’s Parkinson or diary of motor symptoms (tradi- will be dismissed, and the investigator will be contacted tional clinical practice). to make a decision on the convenience of repeating The staff responsible for implementing the randomi- the diary, after retraining the patient or excluding the sation sequence will receive the patient’s clinical infor- patient. Patients who have less than 2 hours Off in the mation by courier: (1) Holter with data stored on the first study diary (before the baseline visit) will be consid- memory card and (2) patient’s diary of motor fluctua- ered screening failures and will not be able to continue tions. This staff will be responsible for sending this infor- the study. mation to the patient’ neurologist by encrypted email The results of the measurements taken at home (Holter and before the next appointment: information from the or diary of motor fluctuations) will be sent to the corre- Parkinson’s Holter, diary of motor fluctuations or no sponding neurologists by encrypted email before their additional information. The randomisation sequence will next consultation with the patient. All the neurologists have been performed by independent staff with the help will receive specific training in interpreting the Parkin- of a table of random numbers and following a balanced son’s Holter data and will have a manual and an explana- blocks model, whose size and composition will not be tory video available during the study time. revealed to the researchers or to the staff responsible for The home monitoring procedure will be repeated implementing the sequence.18 systematically before each appointment with the neurol- ogist. The study’s first follow-­up visit will take place in Procedures week 12 (±2 weeks) after the baseline visit. The study’s All study patients will wear the sensor 7 days before prior last evaluation will be carried out by week 26 (±2 weeks). consultation with the neurologist, although this informa- The neurologist is free to schedule intermediate appoint- tion will not be shown to the neurologist if they are not ments if necessary, before which the home monitoring expected to see it by randomisation arm (group A). Simi- process will also be repeated. The efficacy variables larly, all patients will keep a diary of motor fluctuations described in the next section will be recorded at each prior to the first and last consultation with the neurolo- study evaluation and at the last appointment, usability gist, although the information will not be shown to the and satisfaction questionnaires will also be administered neurologists, unless they belongs to group B. Patients to both the patients and neurologists (table 1). 4 Rodríguez-­Molinero A, et al. BMJ Open 2021;11:e045272. doi:10.1136/bmjopen-2020-045272 Open access Figure 4 Page 1 of the diary of motor fluctuations. At the end of the study, the neurologists will receive ►► Record of therapeutic changes. the complete information from the records of all their ►► Record of prescribed exercise programmes. patients (regardless of the study group to which they ►► Adherence to the motor fluctuations recording system belong) by email, including the diaries of motor fluctu- (On/Off diary and Parkinson’s Holter). ations filled out at home and the complete information ►► Motor complications (Unified Parkinson’s Disease from the Parkinson’s Holter. Rating Scale (UPDRS) part IV,21 administered by the In this study, there are no concomitant treatments neurologist). prohibited, although information systems or patient ►► Daily On time: through a diary of motor fluctuations monitoring systems, other than those tested, cannot be (patient’s diary).19 used. ►► Presence and severity of freezing of gait episodes: Outcome variables and measurement instruments Freezing of Gait Questionnaire (FOG-­ Q, adminis- The efficacy of clinical control will be measured using the tered to the patient by phone).22 following variables. ►► Quality of life: using the 39-­item Parkinson’s Disease Primary: Questionnaire (PDQ-39, self-­ administered by the ►► Daily Off time: through a diary of motor fluctuations patient).23 21 (On/Off).19 20 ►► Autonomy in activities of daily living: UPDRS part II Secondary: (administered by the neurologist). ►► Number of medical visits and telephone contacts for In addition, a record of adverse effects during the medication adjustment. study period will be kept and the usability of and user Rodríguez-­Molinero A, et al. BMJ Open 2021;11:e045272. doi:10.1136/bmjopen-2020-045272 5 Open access Table 1 Schedule of the study evaluations. Baseline Visit week Unscheduled Visit week Inclusion evaluation 12±2 visit 26±2 Inclusion criteria X   Informed consent X   Sociodemographic data X   Year of diagnosis X   Hoehn and Yahr Scale X   Baseline treatment X   Freezing of Gait Questionnaire X   X   X   X   Unified Parkinson’s Disease Rating Scale X   X   X   X   39-­item Parkinson’s Disease Questionnaire   X   X   X   X   Diary of motor fluctuations X   X   X   Parkinson’s Holter X   X   X   Record of health visits and contacts X   X   X   Record of therapeutic changes/exercise programmes X   X   X   Adherence X   X   X   Record of adverse effects X   X   X   Usability and satisfaction X   satisfaction with the Parkinson’s Holter will be evaluated Blinding could be broken in the event the patient’s using the System Usability Scale (SUS)24 and the Quebec physician deems it vital to access any of the study infor- User Evaluation of Satisfaction with Assistive Technolo- mation (especially the patient’s diary filled out at home) gies scale (QUEST),25 respectively. because the patient’s clinical situation requires it. This Other PD-­related data will be recorded as control vari- fact will be recorded for later exclusion from all analyses ables (year of PD diagnosis, stage according to the Hoehn potentially affected by the infringement of the protocol & Yahr scale in the Off state,15 patient sociodemographic data (age, sex, educational level) and neurologist data: Sample size age, sex, years of practice, type of activity (consultation, Assuming a mean reduction from baseline of 75 min of ward etc) and number of patients treated per year at each Off time daily (SD 130) between arm A and C, a sample care level. size of 49 patients per group would provide 80% power to show superiority at a significance level alpha of 5% (two Monitoring sided). All study data and procedures will be supervised by an Unassessable patients will be those that signed the independent monitor. The supervision will be carried informed consent form (inclusion visit) but are lost to out in accordance with Best Clinical Practices, ISO follow-­up before the baseline visit. The rest of the subjects 14155:2011. will be assessable even if they are not adherent to the motor fluctuation measurement systems. To cover loss to Blinding follow-­up and unassessable patients, the sample size will The participating patients are responsible for recording be increased by 10%, so that, in principle, 162 patients the main variable (Off time) in their diary of motor fluc- will be necessary (54/arm). A standard method to handle tuations. Patients will be blinded to the neurologist’s missing data (last observation carried forward) will be randomisation arm, who will not disclose what infor- used. mation is available to adjust the therapeutic regimen. The inclusion of 40 physicians is proposed, assuming Patients are also responsible for recording the On time that each physician will include four or five patients in (diary of motor fluctuations) and the variables related to the study FOG-­Q and quality of life (PDQ-39); therefore, there is blinding to these data. The neurologists are responsible Data analysis plan for collecting the UPDRS data and recording the thera- In the patient’s diary (main outcome variable), lost data peutic changes and adverse effects; therefore, there is no will be imputed, by interpolation between equal data, blinding to these secondary variables. The data analysts provided that the period without data does not exceed will also be blinded to the type of intervention in each the hour of duration. No other lost data of the study will group. be imputed. 6 Rodríguez-­Molinero A, et al. BMJ Open 2021;11:e045272. doi:10.1136/bmjopen-2020-045272 Open access A fixed effects analysis of variance (ANOVA) with the Universitari de Bellvitge (code AC012/19). Any protocol baseline Off time as a covariate will be used to test the change that may increase the risk or present new risks for superiority of group A versus group C in the overall anal- the patient, or that may affect the validity of the study, ysis and the noninferiority of group B in the per-­protocol must be approved by the sponsor in writing before being analysis. implemented. All study participants will sign the written A descriptive analysis of all the variables included in the consent form, after being properly informed by a study study will be performed. For the quantitative variables, local investigator. robust estimators of central tendency (mean, winsorized In all of the reports and communications related to the mean, trimmed mean, Huber estimator) and of sample study subjects, the subjects will be identified only by their variability (SD, standardised median absolute deviation, case numbers. Data will be handled strictly in accordance sample quasi-α-Winsorised-­standard deviation, weighted with the professional standards of confidentiality, under root mean variance and the adjusted percentage root the terms stipulated in Regulation (EU) 2016/679 of the mean variance) will be used. CIs will be calculated European Parliament and the Council of 27 April 2016 by applying bootstrap or resampling methods. The on Information Protection (General Data Protection maximum, minimum, skewness and kurtosis of the distri- Regulation). butions will be calculated. For comparison of two related The sponsor has a civil liability insurance policy that means, the Wilcoxon test or the robust generalisation of covers the potential damages for participants that could repeated measures ANOVA will be used. derive from the application of this protocol. For qualitative variables, the frequency of the distribu- The results will be disseminated to the scientific tions will be calculated with percentages. For compar- community in the form of a publication, preferably in an isons, Pearson’s χ2 or McNemar’s test will be used as open access journal, and to the general population, by appropriate. press release for the national media. Various Spanish and The total score on the usability and user satisfaction European patient associations will receive direct commu- scales (SUS and QUEST) will be calculated according nication of the results. to the instructions of each instrument, and a descriptive analysis of these results will be performed for the overall sample. The results for the usability of and the physician DISCUSSION satisfaction with the device will be analysed for the overall This study will evaluate the efficacy of a PD symptom sample. monitoring device for improving the clinical control of Finally, a descriptive analysis of the frequency and patients. This improvement will be measured in the form severity of the adverse effects and device-­related adverse of a reduction in the daily Off time and according to effects will be performed. other health outcomes as well as the neurologists’ and Patients lost to follow-­up will be included in the analysis patients’ satisfaction with the device. if at least one therapeutic adjustment was made before Although multiple studies have explored the validity of dropout. The baseline data of the patients lost before this various devices for monitoring PD symptoms, currently point, will be also analysed in order to study the potential there is no evidence of the therapeutic efficacy of moni- impact of these dropouts in the balance between groups, toring by such means.26 That the developed devices regarding the main confounding factors. correctly monitor motor symptoms, does not necessarily The analysts will be blinded to the type of diagnostic imply that this monitoring improves clinical control. intervention in each group. This is the first study to examine the efficacy, in terms of clinical control, of these new sensors. Additionally, the Patient involvement same data may be used to test the efficacy of motor fluc- Patients were not involved in the design, recruitment or tuation diaries, considered a reference standard, which choice of outcome measures of this research protocol. have been previously validated but for which there are However, patients played a central role in the devel- also no available clinical efficacy studies.19 The results of opment of the Parkinson’s Holter, carried out by the this study will provide information on the practical utility research team in previous research projects. Selected of the objective information that these devices provide groups of patients, who were involved from first stages, and, therefore, on the convenience of adopting this tech- contributed to identify needs and use cases, provided nology in clinical practice, in future clinical trials and in information on their symptoms and feedback on design various studies on PD. and usability, which have served to improve the product It is important to clarify that although the Parkinson’s in various iterations. Parkinson’s patient associations will Holter has a fall detection functionality, it has not been be involved in development of the dissemination plan of fully implemented in the study (the verification step by the results. the user was omitted), so the information related to falls will not be analysed. Ethics and dissemination This study has some limitations, such as the lack of This protocol and the informed consent form were blinding of the neurologists, which is inherent to the approved by the Ethics Committee of Hospital objective of the study: neurologists must necessarily know Rodríguez-­Molinero A, et al. BMJ Open 2021;11:e045272. doi:10.1136/bmjopen-2020-045272 7 Open access 2 the monitoring information that has been assigned to Department of Neurology, Hospital Universitari Vall d'Hebron and them by chance. This could lead to a greater effort to opti- Neurodegenerative Diseases Research Group, Barcelona, Spain 3 Department of Genetics, Microbiology and Statistics, Faculty of Biology, Universitat mise the medication regimen by neurologists with access de Barcelona, Barcelona, Spain to Holter data and by neurologists with access to the diary. 4 Parkinson's and Movement Disorders Unit, Hospital Quirón Teknon, Barcelona, While this phenomenon is not due to a Hawthorne effect Spain 5 (neurologists try harder because they know they are being Movement Disorders and Neurodegenerative Diseases Unit, Hospital Ramón y observed in the study), it is not necessarily a negative Cajal, Madrid, Spain 6 Neurology Service, Hospital Universitario 12 de Octubre, Madrid, Spain phenomenon, since it is possible that part of the improve- ment potentially produced by these means of monitoring Collaborators This research is being conducted by the 'Monitoring Parkinson’s is due to the neurologist’s increased attention to the case. patients Mobility for therapeutic purposes' (MoMoPa) research group, which That is, it is possible that the diary or Holter produce includes, in addition to the authors of this papers: Hospital de Sant Joan Despí better clinical results not only because of the information Moisès Broggi (Nuria Caballol Pons, Anna Planas-­Ballvé), Hospital Universitari Mútua Terrassa (Mariateresa Buongiorno, Pau Pastor, Ignacio Alvarez), Hospital they produce but also because they encourage neurolo- Universitario de Toledo (Núria López Ariztegui, Mª Isabel Morales Casado), Hospital gists to better adjust medication, which is one of the posi- Universitario Ramón y Cajal (Gema Sánchez), Hospital General de l'Hospitalet tive effects that should be included in the observation. (María Asunción Ávila Rivera), Terapia Integral Uparkinson (Anna Prats), Hospital In contrast, neurologists may in fact be subjected to General Universitario de Elche (María Álvarez Saúco), Hospital de la Santa Creu i Sant Pau (Alexandre Gironell Carreró), Hospital Universitario 12 de Octubre the aforementioned Hawthorne effect.27 Given that (Álvaro Sánchez-­Ferro, Antonio Méndez Guerrero), Hospital Sant Camil (Elisabet the protocol is identical in all arms of the study, if the Franquet Gomez), Hospital de Tortosa Verge de la Cinta (Sonia Escalante Arroyo), Hawthorne effect is symmetrical, that is, if it has the same Instituto de Biomedicina de Sevilla, Hospital Universitario Virgen del Rocío/CSIC/ consequences in all arms, it will not affect the relative Universidad de Sevilla (Laura Muñoz-­Delgado, Daniel Macías-­García, Silvia Jesús, Astrid Adarmes-­Gómez, Pablo Mir), Hospital Clínico de Valencia (José Mª Salom comparisons between arms. However, if the effect is more Juan, Antonio Salvador Aliaga), Hospital General de Alicante (Silvia Martí Martínez, marked in any of the arms (eg, in the case of neurologists Carlos Leiva Santana), Hospital del Mar (Victor M. Puente Pérez, Irene Navalpotro who do not have additional information but who particu- Gómez), Hospital Vall d'Hebron (Sara Lucas del Pozo), Hospital Universitario larly strive due to being observed in the study), then the Fundación Alcorcón (Lydia Vela Desojo), Hospital Álvaro Cunqueiro (Antonio Koukoulis Fernández, Mª Gema Alonso Losada), Hospital Universitario de Burgos differences observed in the study may vary with respect to (Mª Esther Cubo Delgado), Hospital Universitario Marqués de Valdecilla (Jon Infante the real ones in clinical practice. Ceberío, María Sierra Peña, Isabel González Aramburu, Mª Victoria Sánchez Peláez), In addition, observer bias may occur in this study Hospital Universitario Infanta Sofía (Marina Mata Álvarez-­Santullano, Carmen Borrúe because the neurologists, who know the information they Fernández, Mª Concepción Jimeno Montero), Clínico Virgen de la Victoria (Mª José Gómez Heredia, Francisco Pérez Errazquin, Lina Carazo Barrios), Hospital Royo have managed, are also responsible for applying some Villanova (Alfredo López López), Hospital de Llíria (Mª Pilar Solís Pérez), Hospital instruments to measure the secondary outcomes.28 That Univ Lucus Augusti (Rubén Alonso Redondo, Jessica González Ardura), Hospital is, knowledge of the study arm can lead to changes in the Universitario Donostia (Javier Ruiz Martínez, Ana Vinagre Aragón, Ioana Croitoru), way the UPDRS is applied or interpreted, for example. Hospital Universitario Puerta de Hierro Majadahonda (Pilar Sánchez Alonso, Elisa This bias has been reduced as much as possible by Gamo Gonzalez, Sabela Novo Ponte), Hospital Moraleja (Esteban Peña Llamas), Hospital Alcázar de San Juan (Esther Blanco Vicente, Rafael García Ruiz, Ana Rita removing the responsibility of applying the scales from Santos Pinto, Marta Recio-­Bermejo), Hospital Virgen de la Arrixaca de Murcia (José the participating neurologists: the scales will be self-­ López Sánchez, Judith Jiménez Veiga), Hospital Regional de Málaga (Teresa Muñoz applied or applied by telephone by a blinded evaluator, Ruiz, Lucía Flores García), Hospital Clínico San Carlos (Rocío García-­Ramos, Eva except for the UPDRS, which requires a physical exam- López Valdés), Hospital German Trias i Pujol (Lourdes Ispierto González, Ramiro Álvarez Ramo, Dolores Vilas Rolan), Hospital Comarcal de l'Alt Penedès (Esther ination by the neurologist. In any case, the results to Catena Ruiz), Hospital Universitari General de Catalunya (Ernest Balaguer, Antonio which the neurologists were not blinded will be analysed Hernández Vidal), Hospital Universitari de Girona Doctor Josep Trueta (Berta Solano with techniques that attempt to determine the presence Vila, Anna Cots Foraster, Daniel López Domínguez), Hospital la Princesa (Lydia of this bias: observer bias tends to more strongly affect less López‐Manzanares) severe patients; therefore, if the intervention is effective Contributors AR-­M conceived and designed the study, and drafted this paper. JH-­V, AB, JCM-­C and DADAP-­M contributed to the study design. CP-­L contributed to only in less severe patients, the possible presence of this study logistics preparation, including software for managing Parkinson’s Holter data bias will be reported.29 during the trial. AM contributed to the statistical analysis plan. All authors have read Finally, the duration of the clinical review has not been and approved the manuscript. considered as a variable, thus, there will not be possible to Funding This work is supported by AbbVie S.L.U, the Instituto de Salud Carlos III draw conclusions on the time consumed in patient atten- [DTS17/00195] and the European Fund for Regional Development, 'A way to make Europe'. tion in the different study arms. In conclusion, this clinical trial has been designed to Competing interests AR-­M and CP-­L are shareholders of Sense4Care, the company that will market the tested device in the short term. AR-­M participated determine whether automated symptom monitoring with other authors in obtaining funding for the study and in the protocol design. systems (Parkinson’s Holter) improve the clinical control Given his conflict of interest, he will manage the project as the sponsoring centre’s of patients with motor fluctuations. We expect the first coordinator but will not participate in the data collection, study monitoring, statistical analysis or interpretation of results. CP-­L only contributed to study results in 2022. logistics preparation, including software for managing Parkinson’s Holter data during the trial. Author affiliations 1 Àrea de Recerca, Consorci Sanitari de l'Alt Penedès i Garraf, Vilafranca del Pendès, Patient consent for publication Not required. Spain Provenance and peer review Not commissioned; externally peer reviewed. 8 Rodríguez-­Molinero A, et al. BMJ Open 2021;11:e045272. doi:10.1136/bmjopen-2020-045272 Open access Data availability statement Data sharing not applicable as no datasets generated disease: validation study under real conditions of use. JMIR Rehabil and/or analysed for this study. Assist Technol 2018;5:e8. 13 Rodríguez-­Molinero A, Samà A, Pérez-­Martínez DA, et al. Validation Supplemental material This content has been supplied by the author(s). It has of a Portable Device for Mapping Motor and Gait Disturbances in not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been Parkinson’s Disease. JMIR Mhealth Uhealth 2015;3:e9. peer-­reviewed. Any opinions or recommendations discussed are solely those 14 Rodríguez-­Molinero A, Pérez-­López C, Samà A, et al. A kinematic of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and sensor and algorithm to detect motor fluctuations in Parkinson responsibility arising from any reliance placed on the content. Where the content disease: validation study under real conditions of use. JMIR Rehabil includes any translated material, BMJ does not warrant the accuracy and reliability Assist Technol 2018;5:e8. 15 Hughes AJ, Daniel SE, Kilford L, et al. Accuracy of clinical diagnosis of the translations (including but not limited to local regulations, clinical guidelines, of idiopathic Parkinson’s disease: a clinico-­pathological study terminology, drug names and drug dosages), and is not responsible for any error of 100 cases. Journal of Neurology, Neurosurgery & Psychiatry and/or omissions arising from translation and adaptation or otherwise. 1992;55:181–4. Open access This is an open access article distributed in accordance with the 16 Hoehn MM, Yahr MD. Parkinsonism: onset, progression and Creative Commons Attribution Non Commercial (CC BY-­NC 4.0) license, which mortality. Neurology 1967;17:427–42 http://www.​ncbi.​nlm.​nih.​gov/​ pubmed/​6067254 permits others to distribute, remix, adapt, build upon this work non-­commercially, 17 Folstein MF, Folstein SE, McHugh PR. "Mini-­mental state". A and license their derivative works on different terms, provided the original work is practical method for grading the cognitive state of patients for the properly cited, appropriate credit is given, any changes made indicated, and the use clinician. J Psychiatr Res 1975;12:189–98 http://www.​ncbi.​nlm.​nih.​ is non-­commercial. See: http://​creativecommons.​org/​licenses/​by-​nc/​4.​0/. gov/​pubmed/​1202204 18 Altman DG, Bland JM. Statistics notes: how to randomise. BMJ ORCID iD 1999;319:703–4 https://www.​ncbi.​nlm.​nih.​gov/​pmc/​articles/​ Alejandro Rodríguez-­Molinero http://​orcid.​org/​0000-​0002-​9678-​2654 PMC1116549/ 19 Hauser RA, Deckers F, Lehert P. Parkinson's disease home diary: further validation and implications for clinical trials. Mov Disord 2004;19:1409–13. REFERENCES 20 Hauser RA, Friedlander J, Zesiewicz TA, et al. A home diary to assess 1 Hirtz D, Thurman DJ, Gwinn-­Hardy K, et al. How common are the functional status in patients with Parkinson's disease with motor "common" neurologic disorders? Neurology 2007;68:326–37. fluctuations and dyskinesia. Clin Neuropharmacol 2000;23:75–81 2 Kalia LV, Lang AE. Parkinson's disease. Lancet 2015;386:896–912. http://www.​ncbi.​nlm.​nih.​gov/​pubmed/​10803796 3 García-­Ruiz PJ, Del Val J, Fernández IM, et al. What factors influence 21 Fahn S. Elton R M of the UDC. In: Fahn S, Mardsen CD, Jenner P, motor complications in Parkinson disease?: a 10-­year prospective eds. Recent Developments in Parkinson’s Disease. New York, 1986. study. Clin Neuropharmacol 2012;35:1–5. 22 Giladi N, Tal J, Azulay T, et al. Validation of the freezing of gait 4 Fahn S, Oakes D, Shoulson I, et al. Levodopa and the progression of questionnaire in patients with Parkinson's disease. Mov Disord Parkinson's disease. N Engl J Med 2004;351:2498–508. 2009;24:655–61. 5 Ahlskog JE, Muenter MD. Frequency of levodopa-­related dyskinesias 23 Martínez-­Martín P, Frades Payo B. Quality of life in Parkinson's and motor fluctuations as estimated from the cumulative literature. disease: validation study of the PDQ-39 Spanish version. The Grupo Mov Disord 2001;16:448–58 http://www.​ncbi.​nlm.​nih.​gov/​pubmed/​ Centro for study of movement disorders. J Neurol 1998;245 Suppl 11391738 1:S34–8 http://www.​ncbi.​nlm.​nih.​gov/​pubmed/​9617722 6 Reimer J, Grabowski M, Lindvall O, et al. Use and interpretation of 24 Brooke J. Sus: a retrospective 2013;8:29–40 http://www.​usab​ilit​ypro​ on/off diaries in Parkinson's disease. J Neurol Neurosurg Psychiatry fess​ionals.​org 2004;75:396–400. 25 Demers L, Weiss-­Lambrou R, Ska B. Development of the Quebec 7 Papapetropoulos SS. Patient diaries as a clinical endpoint user evaluation of satisfaction with assistive technology (quest). in Parkinson's disease clinical trials. CNS Neurosci Ther Assist Technol 1996;8:3–13. 2012;18:380–7. 26 Odin P, Chaudhuri KR, Volkmann J. Viewpoint and practical 8 Maetzler W, Domingos J, Srulijes K, et al. Quantitative wearable recommendations from a movement disorder specialist panel on sensors for objective assessment of Parkinson's disease. Mov Disord objective measurement in the clinical management of Parkinson’s 2013;28:1628–37. disease. NPJ Park Dis 2018;4. 9 Sam A, Pérez-­López C, Rodríguez-­Martín D. Estimating bradykinesia 27 Parsons HM. What happened at Hawthorne?: new evidence severity in Parkinson’s disease by analysing gait through a waist-­ suggests the Hawthorne effect resulted from operant reinforcement worn sensor. Comput Biol Med 2017;84. contingencies. Science 1974;183:922–32. 10 Rodríguez-­Martín D, Samà A, Pérez-­López C, et al. Home detection 28 Hróbjartsson A, Thomsen ASS, Emanuelsson F, et al. Observer bias of freezing of gait using support vector machines through a single in randomized clinical trials with measurement scale outcomes: waist-­worn triaxial accelerometer. PLoS One 2017;12:e0171764. a systematic review of trials with both blinded and nonblinded 11 Pérez-­López C, Samà A, Rodríguez-­Martín D, et al. Dopaminergic-­ assessors. CMAJ 2013;185:E201–11. induced dyskinesia assessment based on a single belt-­worn 29 Szklo M, Nieto FJ. Epidemiología Intermedia : Conceptos Y accelerometer. Artif Intell Med 2016;67:47–56. Aplicaciones. Díaz de Santos, 2003. Available: https://www.​ 12 Rodríguez-­Molinero A, Pérez-­López C, Samà A, et al. A kinematic casadellibro.​com/​libro-​epidemiologia-​intermedia-​conceptos-​y-​ sensor and algorithm to detect motor fluctuations in Parkinson aplicaciones/​9788479785956/​927995 [Accessed 4 Dec 2017]. Rodríguez-­Molinero A, et al. BMJ Open 2021;11:e045272. doi:10.1136/bmjopen-2020-045272 9