The primary outcome will be the difference in change in the Hospital Anxiety & Depressional Score (HADS): depression score (HADS-D) and HADS anxiety score (HADS-A) between intervention and control groups at 3 months.25 Mood disorders are commonly experienced by PwMS, significantly affecting quality of life,25 amenable to identification and treatment through the chronic disease management model entrenched within a nurse practitioner’s approach to patient care.
Secondary outcomes include difference in change in: (a) HADS-D and HADS-A scores at 6 months; (b) European Quality of Life Score (EQ5D) at 3 and 6 months,33 a scale used by Alberta Health Services to measure impact of healthcare delivery models; (c) Modified Fatigue Impact Scale (MFIS) score at 3 and 6 months,35 a severely disabling symptom commonly experienced by those living with MS; (d) Caregiver Health Related Quality of Life (CAREQOL-MS), a caregiver quality of life questionnaire for those informal caregivers helping PwMS who can experience burden and lack of support, at 3 months and 6 months,36 participant diary of outpatient healthcare interactions during the study period and provincial database search for inpatient/emergency admissions, to see if the intervention of a NP decreases other costs of healthcare providers in the care of PwMS; in addition to (e) a Consultant Satisfaction Questionnaire (CSQ) at 6 months,37 to see if participants perceive NP-led care to be equivalent to that provided by their usual care (community neurologist, family physician and registered nurse.
A longer-term follow-up will be conducted at 12 months to evaluate the long-term differences of the six-month intervention in change of the patient-reported outcome measures (HADS, MFIS, EQ5D, and CAREQOL-MS). By then, all participants would have seen the NP, and the goal of the late follow-up would be to see if there was (1) long-term impact of the 6 month NP intervention (from those in the original intervention group), and (2) to check the larger patient number impact of the NP intervention upon those within the control group.
If participants score very low or severe on their scales around anxiety, depression, fatigue, quality of life and/or their caregivers score as very low on their questionnaire, their community neurologist and/or family physician will be contacted to see the participant on an urgent basis.
Sample size & power analysis
Honarmand and Feinstein (2009),39 validated the use of HADS in PwMS [Baseline scores and standard deviation (SD)]. Using the validation information in the study by Honarmand and Feinstein and the following assumptions of 80% power and a two-sided alpha of 0.05, a total sample size of 200 (100 in each group) will be required to detect 1.5 difference40 between the intervention and the control groups. The same size has been calculated for both HADS-A and HADS-D. This study will use the sample size calculated for HADS-A, as it required a larger sample size and to ensure there will be sufficient power for both HADS-A and HADS-D. This sample size will be inflated to 220 to account for possible dropouts, losses to follow-up, and withdrawals of consent.
Feasibility
We investigated the feasibility of this intervention by searching community neurologists’ electronic medical record databases to identify the number of PwMS typically followed by each of the seven neurologists. The number of PwMS in these practices do fluctuate over time (movement of PwMS in and out of the province, referral to another neurologist, or loss to follow-up etc.). There were greater than 2000 PwMS followed by community neurologists as of January 2017. There is also an established NP practice in the tertiary MS clinic at the University of Alberta, seeing more than 400 PwMS per year. Thus, we have determined that we will be able to recruit the required 200 PwMS for the study.
Preliminary Screening
Prior to conducting statistical analysis, preliminary screening will be conducted using SAS 9.4 software (SAS Institute Inc, Cary, NC, USA) to ensure that all the enrolled patients meet the eligibility inclusion and exclusion criteria and confirm the participants provide informed consent.
Statistical Analysis
Data analysis will be performed using the computer R 3.4.0 software (Vienna, Austria; https://www.R-project.org/) and SAS 9.4 software (SAS Institute Inc. Cary, NC, USA). Patient demographic and clinical characteristics will be analyzed using descriptive statistics.
The primary outcome of difference in change of HADS-D and HADS-A from baseline to 3 months between the intervention and control, will be tested using an Independent T-test. The last observation will be carried forward in the case of missing data. Secondary outcomes will be analysed using the same methods as described below. The CSQ will be treated as continuous variables with an overall satisfaction score as a sum of the sub-scales for each question in the CSQ.
Categorical variables will be reported using frequency and percentage and continuous variables will be reported using mean (SD) or median [Interquartile range (IQR)] as appropriate. Univariable analysis will be conducted to determine if there is a statistical significance between the outcomes (e.g., baseline to three months and six months, respectively). Chi-square and Independent T-tests will be used for the univariate analysis, or where appropriate the non-parametric tests (Fishers test and Wilcoxon rank test). ANOVA and generalized linear models will be used to test for overall differences between the intervention and control at the different timepoints and variances amongst the variables by group and timepoints. Post-hoc Tukey HSD (Honestly Significant Difference) Test will be performed afterwards with adjustments made using the Bonferroni method. All test assumptions will be checked during the data analysis process. Statistical significance will be set at p values less than 0.05.
Validity and Reliability
The questionnaires used in the study (EQ5D, HADS, MFIS, and CAREQOL-MS) have been tested for validity and reliability in studies involving PwMS. Three papers report total HADS scores in PwMS and were considered in the study design of this study.25,27,28 The EQ5D quality of life measure is commonly used in the MS population,33 and the MFIS is a standard measurement of fatigue in PwMS.35,41 The CAREQOL-MS, a caregiver quality of life survey for those caring for PwMS is also commonly used in this patient population.36,42 Satisfaction with healthcare provider care will be measured by the overall mean score of the CSQ completed at the 6 month follow-up visit. The CSQ is a self-administered tool with 18 questions, it measures 3 factors of the healthcare provider interaction: (1) professional aspects; (2) depth of patient relationship with provider; and (3) perceived length of consultation. The higher the score indicates a higher level of patient satisfaction.24,37
To reduce possible biases in the study, we will use objective measures to examine the effects of NP-led care in comparison to standard care. It is important that all participants are offered an opportunity to see a NP either within 4 weeks for the intervention arm vs at 6 months for the control arm. Thus, those participants randomized to the control arm will be provided with an appointment with the NP at 6 months. Then, 12-month follow-up of all participants will be conducted, to see if the timing of intervention makes a difference over the long-term.
Ethical considerations
The research protocol was reviewed and approved by the Health Research Ethics Board of the University of Alberta (approval number Pro00069595), initially on March 30, 2017, with protocol modification to version 2 approved on Jan 11, 2018. PwMS will be able to be referred to the NP by community neurologists and family physicians outside of participating in the study if they decline to participate in the study.
Funding
Funding was obtained through the University Hospital Foundation (Pro00069595), and partially matched by the Strategy for Patient-Oriented Research, Canadian Institutes of Health Research)
Data security and storage
Electronic data will be collected and stored using the RedCap secure website, housed behind the Faculty of Medicine and Dentistry firewall and secure server. Data will be stored under secure conditions for 10 years at EPICORE Centre, in accordance with EPICORE Centre's standard operating procedures. The questionnaires and consent forms will be completed electronically by the participants, participants will be offered paper copies if required. Electronic files will be kept on a secure server (Faculty of Medicine and Dentistry firewall) with password protection. Access will be limited to directly involved personnel. Paper copies will be kept in locked cabinets within the EPICORE Centre as according to their standard operating procedures.
A master list of participant name and personal health care number with associated research number, will be kept in a separate locked cabinet. Once secondary outcome blinded analysis has been completed, it will be destroyed. The research patient data will also be kept in a separate locked cabinet. Both of these cabinets are located in a locked room with limited access to research personnel only. Electronic files will be kept on a secure server (Faculty of Medicine and Dentistry firewall) with password protection. Access will be limited to those directly involved in this study (e.g., statistician, research assistant, and database manager). The electronic data information will be deleted after the 7-year post study completion requirement has been completed.
Data Monitoring
There will be an interim analysis completed after the 3-month follow up has been completed. The study involves PwMS completing validated surveys of their depression and anxiety scores which will be reviewed routinely by the NP and neurologist. If a PwMS depression and anxiety score falls drastically, the NP or neurologist will follow-up with the patient and provide any necessary care or referral. Quality assurance of the data will be performed by the data entry researcher and the project manager for the trial. Any and all amendments will be communicated to all investigators, research ethics office, and the participants. The study investigators will make the final decision to terminate the trial if needed. Since this a smaller, single centre study, we do not have plans for independent audit of trial conduct.
Electronic data will be collected and stored using the RedCap secure website which is housed behind the Faculty of Medicine and Dentistry firewall and secure server. Paper copies will be kept in locked cabinets within the EPICORE Centre as according to their standard operating procedures. Participants will be sent a reminder by the EPICORE team to complete the questionnaires at baseline, 3 months, 6 months, and at 12 months to encourage participants to complete follow-up visits. The first or last observation will be carried through to account for any missing data during the data analysis.
Confidentiality
Initially, identifying information will be collected to allow the research assistant to look for data required to assess for secondary outcome information (i.e., admission to hospital and/or emergency visits, MS relapse, infection or adverse events from medication). Participants will be offered to complete the questionnaires via secure internet website, however, some may prefer a paper copy in which case the participant’s full address and phone number may be required for participants that opt for paper questionnaires to be mailed to them for follow-ups. Therefore, depending on patient preference, we will collect the email or postal addresses for reminders to be sent to fill out the questionnaires.
Patients will complete diaries of phone calls and visits to GP, neurologist, emergencies, HealthLink, and other healthcare professionals (such as physiotherapists, occupational therapists, pharmacists, dieticians, homecare, social workers, etc.). This information will be deidentified once collected.
Identifying information will be removed once the above data are obtained. Charts will be reviewed weekly to ensure all data are complete and all identifying information will be removed by a research assistant and coded with only a research number. A master list coordinating research numbers and patient’s healthcare number will be kept separate from patient data in a separate locked cabinet which will be retained until the end of the study. At the conclusion of the study, this list will be destroyed.
Dissemination
This randomized control trial will be published once data analysis has been completed and all investigators will be invited to contribute to the manuscript. A copy of the publication and findings will be provided to the participants. At this time, there are no plans or arrangements for data sharing. We plan to disseminate the results at the European Committee for Treatment and Research in Multiple Sclerosis (ECTRIMS) conference.