Nurse-led care versus physician-led care in the management of rheumatoid arthritis and psoriatic arthritis (StaerkeR): study protocol for a multi-center randomized controlled trial

The main hypotheses are that nurse-led care is (1) non-inferior to standard care by rheumatologists with regard to a reduction of disease activity after 1 year and (2) superior with regard to changes in health-related quality of life after 1 year of the intervention (hierarchical test).

The target group for this study are patients with rheumatoid arthritis or psoriatic arthritis. Their general health (including comorbidities) must be stable and disease activity should be low.

Inclusion criteria: stable rheumatoid arthritis or polyarticular psoriatic arthritis with the 28-joint Disease Activity Score (DAS28) < 3.2 (low disease activity) and no current adverse drug reactions.

Exclusion criteria: limited mobility, insufficient knowledge of German, highly active disease, or life-threatening disease.

The sample size was calculated by assuming that changes in the DAS28 scores in the nurse-led care group would not be inferior to those in the standard care group after 1 year of intervention, with a standard deviation of 1.7 and a non-inferiority margin of 0.4 [5]. Using a one-sided t-test of non-inferiority (α = 0.025) and setting the power to 90% result in a sample size of 380 patients for each study arm. Assuming a dropout rate of 5% finally results in a sample size of 400 patients per group. According to a British study from 2014, this sample size would be able to demonstrate the superiority of health-related quality of life in the nurse-led care group compared to the standard care group (α = 0.05). Thus, 800 patients will be recruited into the trial.

As a legal framework, this study is being conducted in cooperation with the health insurance company BARMER using selective contracts for the special nurse-led patient care according to §140a of the fifth Social Security Code (SBG V). To be eligible for the study, patients, thus, must have health insurance from BARMER.

Eight out of the 20 participating study centers were already participating in a selective contract with BARMER regarding the integrated care of patients with inflammatory rheumatic diseases. The involvement of a rheumatology nurse was not part of the contract at that time.

In total, 1140 patients signed the selective contract with the insurer. Of these, approximately 80% had rheumatoid arthritis and 10% had psoriatic arthritis. Analyses showed that most patients could be treated so that they were in a stable phase of their disease, thus making them eligible for the StaerkeR study.

Assuming a participation rate of 50% of those patients who have already signed the selective contract with BARMER and by increasing the number of participating study centers to about 20, recruiting the required sample size of 800 patients in 6 months is feasible.

Physicians will pre-screen all suitable BARMER patients with rheumatoid arthritis or psoriatic arthritis for eligibility. They will invite eligible patients for a visit to the practice to check final eligibility and to provide information about the study. Patients who are eligible and willing to participate in the trial must sign a written informed consent form.

The duration of the intervention is 1 year. All patients will be treated in their rheumatology practice or outpatient clinic, irrespective of group allocation. Adherence to the interventions will be monitored by participation in routine control checks.

There are two primary outcomes, which will be tested hierarchically: disease activity and health-related quality of life.

Disease activity is probably the most important patient-related outcome in chronic inflammatory arthritis. Thus, the mean change in disease activity in the course of 1 year was chosen as the first of the two primary outcomes. Disease activity is assessed using the well-established DAS28 [7]. DAS28 is a compound score comprising the number of tender joints (out of 28), the number of swollen joints (out of 28), the erythrocyte sedimentation rate, and the patient’s assessment of disease activity. The range of the score is 0 (no activity) to 10 (maximum disease activity).

We expect the intervention group to be superior with regard to changes in health-related quality of life without a relevant difference in disease activity after 1 year. Health-related quality of life will be assessed by the EQ-5D-5L [8], which has five dimensions: (1) mobility, (2) self-care, (3) usual activities, (4) pain and discomfort, and (5) anxiety and depression. Each dimension has five levels: (1) no problems, (2) slight problems, (3) moderate problems, (4) severe problems, and (5) extreme problems. This five-level scoring system has increased sensitivity and a reduced ceiling effect, amongst other advantages, compared to the three-level scoring system. For details on scoring and evaluation, see [8].

Secondary outcomes include other established assessments in rheumatology, as well as evaluations of organizational issues, patient satisfaction with medical treatment, and resource consumption.

One of the most important prognostic markers in rheumatoid arthritis is physical functioning, which will be measured by the validated Hannover Physical Functioning Questionnaire (FFbH) [9]. This assessment is comparable to the internationally applied Health Assessment Questionnaire (HAQ) [10]. Patients are asked, if they are able to perform 18 activities of daily living. They answer on a three-point scale (“yes,” “yes, but with difficulty,” and “no or only with assistance”). The patient’s physical activity will be assessed via the PRISCUS Physical Activity Questionnaire (PAQ) [11], which has ten items that assess the time spent in domestic activities (e.g., housework), sporting activities (e.g., riding a bicycle), and inactivity (e.g., sedentary activity) during the prior week. The tendency to depressive moods is assessed by the two-item Patient Health Questionnaire (PHQ-2) [12]. Further patient-related outcomes are the patient’s estimate of (1) disease activity (numeric rating scale 0–10 [NRS]), (2) pain (NRS), (3) fatigue (NRS), (4) sleep disturbances (NRS), and (5) duration of morning stiffness. The C-reactive protein (CRP) is a standard feature of laboratory analyses, and it will be assessed in the course of the intervention. Smoking status may influence the course of chronic inflammatory diseases and will be subject of patient education by the nurse. Therefore, it will be assessed as a secondary outcome.

In addition, patients will be asked about how long they need to wait for an appointment in their practice, the availability of the rheumatologist or the nurse for inquiries, their satisfaction with the information provided, their satisfaction with their relationship with the rheumatologist or the nurse, and their satisfaction with the coordination and communication done by the different professional groups.

These outcomes are compared to direct and indirect costs and other resource consumption (e.g. duration and costs of nurse-led care vs. standard care). A EULAR recommendation [4] is “to perform cost-effectiveness studies across different European countries, on the role of the nurse in basic and advanced practice.” There are already British [5] and Dutch [13] but no German studies on the costs and efficiency of nurse-led care in rheumatology. Within this study, an economic evaluation will be done using the Questionnaire for Health-Related Resource Use in an Elderly Population (FIMA) [14]. The components of FIMA are as follows: (1) outpatient medical care, (2) remedies, (3) nursing and domestic care, (4) semi-residential care (daycare) and short-term care, (5) benefits of statutory nursing-care insurance, (6) medication, (7) rehabilitation measures, (8) (partly) inpatient hospitalization, (9) medical aids, (10) relocation, (11) type of housing, and (12) periods of incapacity for work.

Baseline data for patient characterization comprise sociodemographic data (age, gender, academic qualifications, professional qualifications, occupation, and retirement status) and data on comorbidities. Newly diagnosed diseases will be continuously assessed within the course of the intervention.

CDAI [15] will be assessed at each visit. The score consists of the number of swollen joints, the number of tender joints, the patient’s global assessment of disease activity, and the evaluator’s global assessment (this will be either the rheumatologist or the nurse depending on group allocation in this study). CDAI will be used for participants with either rheumatoid arthritis or psoriatic arthritis (polyarticular manifestation). The score will be used to adjust patient treatment if necessary. Finally, the body surface area [16] is an established way to assess the percentage of skin affected by psoriatic arthritis.

Patients who are correctly enrolled via screening fax and have completed a telephone interview will be randomized by the data management center to either standard care or nurse-led care. Randomization will be performed using blocks with a variable block length stratified by center. The allocation sequence is accessible only to members of the data management team. Study centers will be informed by fax of the group allocation of each patient before their baseline visit.

In this study, trial participants and care providers cannot be blinded to group allocation. The outcome assessors will be blinded. Patients are asked not to tell the assessor or the call center agents to which group they are assigned. Data analysts are also blinded.

Study data will be collected at baseline after randomization, and after 13, 26, 39, and 52 weeks. Figure 1 is the schedule of enrolment, interventions, and assessments.

At the beginning and at the end of the intervention for a participant, the first primary outcome, DAS28, will be assessed by a blinded external assessor in the practice or outpatient clinic. The assessors are qualified assistant doctors from the Herne Rheumatology Center. They are trained in performing standardized assessments using DAS28 and have received a written copy of the standard operating procedures.

The other primary outcome, health-related quality of life, and secondary outcomes, including organizational procedures, patient satisfaction, and resource consumption, will be assessed via a computer-assisted standardized telephone interview at baseline as well as after 26 and 52 weeks.

The patient will be asked to complete questionnaires at baseline, after 6 months, and at the end of the intervention. Each questionnaire asks about secondary outcomes, such as smoking status, physical functioning, physical activity, depression, estimation of disease activity, intensity of pain, fatigue, tiredness, sleep disturbances, and duration of morning stiffness. The patient additionally provides sociodemographic information at baseline.

The rheumatologist or the nurse will document the participant’s age, gender, height, and weight at baseline. Furthermore, they will assess the CRP as a secondary outcome, (new) comorbidities, the CDAI (and body surface area as necessary), or death of the patient at each routine control examination every 3 months.

The Department of Medical Informatics, Biometry and Epidemiology (AMIB) at Ruhr-University Bochum will manage the data and undertake the statistical analysis. After each study visit, the study centers will send the relevant part of the patient’s case record via fax to the data management center. Practices may be asked in writing for corrections or additional information, as necessary.

All study data will be collected in strictly pseudonymous form. Patients may participate in the trial only if they sign the privacy statement and consent to their data being transferred for analysis. Only authorized personnel will make data entries in and corrections to case record forms. In addition to participating practitioners and rheumatology nurses, specifically trained medical staff may also assist in recording data.

Data processing will start after the last participant has had their final telephone interview and after data clearing. Double data entry, data verification, data coding, and consistency checks are standard data management processes that will be used to maximize data quality.

AMIB will store all study documents for at least 10 years. Thus, study results can be reproduced after the end of the funding period. Case record forms and written informed consent forms will also be stored for at least 10 years in the relevant rheumatology practice or clinic.

If nurse-led care is found to be non-inferior, its superiority over standard care will be tested with regard to health-related quality of life (EQ-5D-5L). This analysis will be performed with all randomized patients based on the intention-to-treat principle using multiple imputation.

The analysis of the secondary outcomes will use all available data for the per protocol population. Continuous data will be compared for the two groups using an analysis of covariance adjusting for baseline data. Analyses of categorical variables or proportions will be performed using a chi-squared test. An additional repeated-measures analysis will be done for all quantitative endpoints taken at different time points. Details of the primary, secondary, and all additional analyses are documented in the statistical analysis plan.

The delegation of a physician’s tasks to a specialized rheumatology nurse will be evaluated as follows:

The Coordinating Center for Clinical Trials at the Philipps University of Marburg is responsible for monitoring each participating study center based on its standard operating procedures for monitoring. It is independent of the trial sponsor and there are no competing interests. Audit visits will ensure that sites adhere to the study protocol. Data collection forms will be checked for completeness and plausibility. The monitoring staff will have direct access to source data (original medical records) during their visits to verify the existence of participants and to check the inclusion and exclusion criteria as well as written informed consent forms. The findings from each audit will be summarized in a report. Three audits for each participating study center are planned.

Due to its experience, the Coordinating Center for Clinical Trials will additionally support the coordinating center and the data management team at the University of Bochum in an advisory capacity regarding data monitoring and safety aspects, and will provide general trial oversight.

The project started in September 2017, and the nurses were trained in June and August 2018. Patient recruitment started in September 2018 and ended in August 2019. The data analysis will start one year later once the final participant has completed the intervention. The expected end date of the project is in September 2020.