The NICE-PD study protocol will here be reported according to the SPIRIT 2013 Statement . Additional file 1 details the IPPCollapse-II SPIRIT checklist. The study is an 18-month, single-blind randomized controlled clinical trial that will be performed in eight community hospitals in the Netherlands. The participating centers can be found in Table 1. A total of 240 people with PD will be included (120 in each group), equally distributed over the participating hospitals. We have selected hospitals where, due to lack of sufficient PDNS staff, only a proportion of PD patients currently have access to PDNS care. This provides us with a unique opportunity to identify patients who at present have no access to PDNS care, and to randomize them within hospitals (at the patient level) between PDNS care and no nursing intervention. We have summarized the study design in Figure 1. The enrolment and assessments during the study period are shown in Figure 2.
Eligible patients will be allocated randomly to either PDNS care or usual care in a 1:1 ratio, using a computer-generated list of random numbers. An independent researcher (who will not perform study assessments) will perform the randomization in an online data management system. Subsequently, this researcher will contact the PDNS to inform them about which participants are randomized to the intervention group. The other participants will receive a letter or an e-mail stating that they have been assigned to the control group. To ascertain an equal representation of patients, we will stratify for gender and disease duration (according to pre-defined subgroups, i.e. disease duration <5 years, 5-10 years and >10 years). The PDNS intervention will follow the Dutch ‘Guideline Nursing care in PD’  (see the section on ‘intervention’). A blinded researcher will perform the clinical assessments at baseline (t0), after 12 months (t1) and after 18 months (t2). Patients and caregivers will also be asked to complete a set of questionnaires at t0, t1 and t2. Finally, every three months patients and their caregivers will complete a questionnaire about healthcare utilization, costs and productivity loss. Care providers (e.g. neurologists) will not be blinded for the assigned interventions. We do not foresee any reason why unblinding of participants would be necessary.
Inclusion and exclusion criteria
The inclusion and exclusion criteria for patients are kept purposefully broad, in order to represent the full diversity of the PD spectrum and thus generate results that apply to real clinical practice. All patients with PD regardless of disease severity or disease duration, male and female, aged 18 years or older at the time of PD diagnosis are eligible. We will only exclude patients that:
- do not have sufficient knowledge of the Dutch language to complete questionnaires;
- received care from a PDNS in the past two years;
- have a score of <18 on the Mini-Mental State Examination (MMSE18) and <12 on the
Frontal Assessment Battery (FAB19).
- have a type of atypical parkinsonism caused by medication (e.g. neuroleptics), a metabolic disorder (e.g. Wilson's disease), encephalitis or a neurodegenerative disorder (e.g.
multiple system atrophy, progressive supranuclear palsy, corticobasal syndrome).
- are residing in a nursing home or another type of residential care facility (because the PDNS is not operational there).
- have any other medical or psychiatric disorder that, in the opinion of the researcher, may compromise participation in the study.
Patients will be approached within each hospital using one of three scenarios.
- The involved neurologists in the hospitals identify eligible patients from their electronic patient file and inform these patients about the study in their clinic (when the patient is coming in for a consultation). lf the patient agrees to be approached by a researcher, they will be provided with the patient information letter.
- The neurologists identify eligible patients using their electronic patient file and subsequently approach them by directly sending out a letter including a short description of the study and a form in which patients can indicate if they want to receive any further information about the study or not. Only if patients actively indicate that they wish to be approached, the researcher will contact them by telephone and send the information letter to the patient.
- The research team organizes an information meeting for patients in the participating center (where also the PDNS and neurologist are present). Here, the patient information letter will be handed out directly. Importantly, patients will be given sufficient time to consider their participation. If they are interested, they will be contacted by the research team at least two weeks after the information session.
Training and coaching of Parkinson’s disease nurse specialists
Before the start of the study, we will organize a single training session with all participating PDNSs (one from each center). The goal of this meeting is to acquire commitment for the study and uniformity in workflow by reviewing the ‘Guideline Nursing care in PD’, to explain the study specifics and to discuss practical issues related to the study intervention. In addition, PDNSs will be closely coached in order to optimize the intervention and adherence to the guideline. Every month an experienced PD nurse from the Radboudumc will have an individual intervision session with each PD nurse, mainly to discuss difficult cases and to optimize the intervention and its uniformity. Finally, we will organize a video meeting every three months with all PDNSs to maintain commitment, support each other, discuss difficulties related to the study and to give each other advice [20,21].
Importantly, for the purpose of this study, we will implement an increase in nursing staff capacity for participating nurses. This will allow us to study the real impact of current usual care, which would not be achieved by adding a new set of specifically trained research nurses to the existing PDNS staff. The PDNSs are all graduated nurses (education level according to the European Qualifications Framework 6 or 7) with a certificate in Parkinson’s Nursing. Furthermore, they have achieved a standard of competences as described in the Guideline on PDNS care .
The PDNS intervention
The PDNS intervention will be performed according to the Dutch ‘Guideline Nursing care in PD’ published in 2015 . The intervention is not standardized, but tailored to the patients’ and caregivers’ needs. This includes the following:
- Assessment of individual care needs of people with PD and their caregivers. At the start of the study, the PDNS performs a specific nursing assessment related to the medical, physical, psychological and social domains.
- Development of a patient-centered treatment plan that supports the patient and caregiver in self-management. The PDNS composes a multidisciplinary plan, based on the results of the individual assessment, and as prioritized by the patient and caregiver (shared decision making). The treatment plan is developed according to the national self-management framework .
- Specific nursing interventions. The intervention varies across disease stages and is tailored to the specific problems and needs of individual patients and their caregivers. The Guideline on PDNS care describes general- and specific nursing interventions. General interventions consist of providing information and education, disease management (e.g. considering advanced treatment options such as DBS) and monitoring (e.g. of caregiver burden). Specific nursing interventions are described for the following areas: mental functions, fatigue, sleep, urogenital functions, sexuality, medication adherence, orthostatic hypotension, caregiver burden, coping, mobility, self-management and dietary issues (table 2 provides examples of such interventions).
- Collaboration with other healthcare professionals. The PDNS stimulates and supports multidisciplinary collaboration between healthcare professionals based on the individual patient-centered treatment plan. The PDNS also plays a pivotal role in the timely referral to other healthcare professionals.
The PDNS will maintain a pre-defined electronic study report according to a structured format for each PD patient, documenting the individual care needs, present symptoms, performed interventions and (changes in) the individual care plan. This report will be started at the initial assessment and updated at every follow-up contact with the patient, e.g. at the outpatient clinic, during a telephone consultation or at a home visit. This data will be purposefully collected for a possible process analysis at the end of the study.
Patients will have regular contact with their PDNS about the progress and realization of the personal goals, both during face-to-face contacts and by telephone, and sometimes during additional home visits. The frequency and type of contact will be optimized for each patient depending on disease stage and individual patient needs. The Guideline on PDNS care advises that each patient has a minimum of one contact with the PDNS each year . Currently in the Netherlands, patients are seen on average twice a year by their PDNS, with an additional two interim telephone consultations per year.
The control group will receive ongoing usual care which is medically comparable to the intervention group, but without a nursing intervention. This involves regular consultations with a neurologist in their own community hospital (typically 2-4 times per year, depending on patient preferences and health status). In addition, control patients will have no restrictions considering any other medical treatments (e.g. by a psychologist or social worker). Importantly, many key elements of care (including in particular the treating neurologist) remain comparable between the two arms because of the randomization at the patient level within hospitals.
Clinical assessment and outcome measures
At baseline, t1 and t2 all patients will visit their own hospital for the study assessments which are performed by a blinded researcher (PDQ-39, MDS-UPDRS and TUG). The patients and their caregivers will also complete home questionnaires. In addition, every three months, patients will receive a questionnaire at home regarding healthcare utilization, costs and productivity loss over the past three months. Caregivers will complete a cost questionnaire including healthcare utilization, costs and productivity loss specifically related to caregiver burden. To improve adherence, patients and caregivers can choose whether they prefer to fill out digital or paper-based questionnaires. Participants will be contacted by telephone when they do not complete the questionnaires within four weeks. All the outcomes, including secondary outcome measures, can be found in Table 3.
Similar to previous large randomized controlled trials in the field of PD [23,24], we will use two co-primary outcomes: quality of life and motor symptoms . For measuring quality of life, we will use the PDQ-39, which is the most widely used quality of life scale in PD and frequently used as outcome measure, e.g. in trials on DBS  and multidisciplinary care . Our second co-primary outcome measure is the severity of motor symptoms measured by the MDS-UPDRS part III. The MDS-UPDRS is a frequently used clinical rating scale and has been shown to be sensitive to change in clinical status . Both scales have been previously validated and are reliable and valid methods to either measure quality of life  or motor symptoms  in people with PD.
Data collection and management
Patients will be given a unique personal identification code not containing any information that refers back to the individual. The key-file, connecting personal identification codes to the individual patient, will be stored on a secure Radboudumc data server. Only the research team has access to this key. The key-file will be stored on a different server from the acquired study data for five years, allowing the research team to contact patients after they have finished the study. After five years, the key-file will be destroyed.
Data from all paper-based Case Report Forms (CRFs) completed by the researcher (PDQ-39, MDS-UPDRS and TUG) will be entered manually into an online certified data management system (Castor EDC). Online CRFs (the remaining questionnaires) will automatically be recorded in Castor. When patients or caregivers are not able to complete questionnaires online, they also have the opportunity to do this on paper. We will send out the questionnaires by post and patients can return the completed questionnaires using a self-addressed envelope. These questionnaires will be entered manually into Castor. Both online and paper-based CRFs only contain the personal identification code.
Clinical notes taken by the PDNS in the online study report will also not contain any information that refers back to the individual. PDNSs are instructed to make notes according to a pre-defined structured format, without mentioning personal information that traces back to an individual patient. The study report will be completed in Castor.
All serious adverse events (SAEs) will be collected and followed up by the investigators and documented in the electronic CRFs. Each SAE will be reported by the respective PDNS to the study team (DR) and the SAE will be notified to the local ethics committee as soon as the researcher has knowledge of the SAE, but no later than 24 hours after the researcher has become aware of the event. Other adverse events will not actively be inquired for during the study, because of the low risk associated with the trial. When a participant spontaneously reports an adverse event, it will be registered in the electronic CRF.
Sample size analysis
We performed a sample size calculation based on the PDQ-39 score. Based on observations in one of our previous studies in a similar population of PD patients where we evaluated multidisciplinary care , we found a mean improvement in PDQ-39 score in the intervention group of -2.5 (SD: 5.8) points and a mean deterioration in PDQ-39 score in the control group of +1.4 (SD: 8.6). We calculated the sample size based on a mean difference between groups of 3.9, with a standard deviation of 8.6 (the highest SD reported). Using a significance level of alpha=0.025 (instead of 0.05 because of two primary endpoints; PDQ-39 and MDS-UPDRS part III) and a power of 80%, a sample of 93 patients in each group would be needed. Considering an attrition rate of 20%, 117 patients are needed per group. We have rounded this up to 120 patients per group, which means a total of 240 patients. We expect this to be feasible, because following a baseline inventory, all centers indicated to be able to include at least 30 patients.
The economic evaluation investigates, alongside the clinical trial, the value for money of full implementation of the PDNS into PD care from a societal and healthcare perspective. We will take all relevant costs into account. The cost-effectiveness timeframe adheres to the clinical study protocol and evaluates cost-effectiveness up to 18 months after randomization. Cost will be measured using a healthcare utilization questionnaire (e.g. including medical consultations, hospital admissions, medication, travel costs, etc.) and a questionnaire measuring productivity loss while working of both patients and caregivers. Per item of healthcare consumption, standard cost-prices will be determined using the guideline for performing economic evaluations . If standardized prices are not available, full cost prices will be determined using activity-based costing. Costs will be analyzed using a mixed model approach or a general linear model approach with a gamma distribution using a log link to account for possible skewness of the cost data.
We will use a PD-specific quality of life measure (PDQ-39) and a generic health-related quality of life scale (EQ5D) to evaluate the quality of the health status of patients. The potential difference in Quality-Adjusted Life Years (QALYs) measured with the EQ5D will be analyzed with a regression approach. We will use a linear mixed model with repeated measurements to test for differences in quality of life (measured with the PDQ-39) between both groups. The same analysis will be used to measure differences between groups in the secondary outcome measures. We will include study center as a random effect and fixed effects for group, time and the interaction between group and time. Each of the outcomes will be included as dependent variable. Statistical analyses will be performed based on the intention-to-treat principle.
As mentioned previously, we hypothesize that both interventions (PDNS care versus no PDNS care) will yield equal costs, while PDNS care is more effective. If this hypothesis is confirmed, then the effect analysis is sufficient to show the efficiency of PDNS care. The design of the economic evaluation follows the principles of a cost-effectiveness analysis and adheres to the Dutch guideline for performing economic evaluations in healthcare .
Besides the overall cost-effectiveness evaluation, we will perform a pre-planned subgroup analysis based on disease duration (diagnosis made <5 years, 5-10 years or >10 years ago) to obtain more insight into the nursing interventions used in each disease stage and the effects of PDNS care in these different groups of patients. This subgroup analysis will be performed because, for example, for the more severely affected patients the nursing intervention is expected to become more intensive and possibly more effective, but also more expensive. When different patterns of this kind are found, this should be investigated further in future trials that are powered adequately to address such group differences.
The chief investigator has the overall responsibility for the conduct of the study. The study group has responsibility for the day-to-day management of the trial and consists of the following people. DLMR, HHL, RHH, MM, NMdV and BRB designed the study. DLMR and NMdV are responsible for day-to-day management of the trial, including the inclusion of participants and communication with participating centers, participants and the ethics committee. TvA, CCSD and HV have a more consultative role and provide substantial feedback to the trial procedures. There will be no independent data monitoring committee, due to the low risk associated with the trial. The results of the study will be sent for publication to a peer-reviewed medical journal. No professional writers will be involved. In addition, the results will be shared with trial participants via the Dutch Parkinson Association and via ParkinsonNet. We report no restrictions for publication.