Effects of care of multimorbid patients in general practices by advanced practice nurses (FAMOUS): study protocol for a nonrandomized controlled trial

Multimorbidity is a common phenomenon among patients in general practices [1, 2]. The term multimorbidity is not well defined. However, most definitions characterize multimorbidity as the cooccurrence of two to three chronic diseases. Because of the somewhat unclear definition, the prevalence rate of multimorbidity in primary care patients varies across studies [2]. One systematic review [2] showed a prevalence of multimorbidity ranging from 12.9% in participants aged 18 years and older to 95.1% in a population aged 65 years and older. Worldwide, more than half of people over the age of 65 years suffer from multiple chronic conditions, most of which are treated in primary care [3]. Multimorbid patients experience many symptoms at the same time. Often, these symptoms cannot be attributed to a causal disease. Therefore, some key problems arise, including functional difficulties, polypharmacy, treatment burden, mental health difficulties, fragmentation of care, reduced quality of life and increased health care utilization [1]. Addressing these challenges requires optimizing functional capacity, rigorous medication management, a greater focus on patient-centred care, and collaboration across health care services [1]. The management of polypharmacy is complex [4], and the implementation of patient-centred care faces several barriers [5]. Thus, these goals can hardly be achieved in the short consultation time a general practitioner (GP) has, which is less than 10 minutes in Germany [6]. In addition, there is an increasing shortage of GPs, especially in rural areas. In Germany, there were already 3,300 unfilled primary care physician positions in 2019, and there will be approximately 11,000 in 2035 [7]. To address this issue, the involvement of other health care professionals is recommended [1]. In many countries, advanced practice nurses (APNs) are successfully integrated into primary health care for multimorbid patients [8, 9]. A systematic review of patient safety, clinical outcomes, cost, and patient satisfaction shows that care offered by nurse practitioners in primary health care settings provides equally good or better outcomes compared with GPs [10]. Outcomes regarding the use of resources, such as the number of emergency visits, hospital admissions, and outpatient contacts, are varied [11]. In addition, findings on the cost effectiveness of integrating APNs in primary care are not very clear. Some studies indicate no increase in costs [11], while other sources have found evidence for cost savings [12].

In Germany, outpatient medical care is predominantly provided by general practitioners. Unlike in many other countries, APNs are not deployed in general practices in Germany. One reason for this is that for a long time, there were no APN education opportunities at the master’s level in Germany. However, academically qualified nurses are now available thanks to the degree programs that have been established in Germany for some time. The objective of this study is to evaluate the care for multimorbid patients by APNs deployed in general practices in rural areas in Germany.

To evaluate whether the integration of APNs in the primary care setting improves the care of multimorbid patients, a control group from German health care insurance was selected. Persons in the control group are considered to be receiving the current standard care. Therefore, a comparison between the new intervention and standard care is possible. To minimize the risk of confounding variables, participants in the intervention and control groups will be matched for the following variables: (1) age, (2) gender, (3) comorbidity and (4) living in a nursing home.

Our objective is to study whether the integration of APNs in the primary care of multimorbid patients leads to optimized care of the target group and to a reduction in the workload of the GPs. Improved care is expected to be reflected by a stabilization of the patients' health and home care situation. This will be expressed as the reduced use of emergency contacts (emergency services, emergency room, acute inpatient admission) and GP visits, thus reducing overall treatment costs. Finally, we expect to observe high satisfaction from all parties involved (physicians, patients, relatives, ANPs and medical assistants) with this type of care.

The FAMOUS trial is a prospective multicentre mixed methods study. Patients with at least three chronic conditions across nine participating general practices are eligible to be treated as an intervention group by a trained APN. All patients will be treated by APNs for twelve months. Data on the period of intervention are available as data from health insurance companies or the Association of Statutory Health Insurance Physicians (ASHIP) or data collected by the APN. The intervention group will be compared with a control group of patients from German health insurance who are matched according to age, sex, comorbidity and living in a nursing home. The matching will be performed at a ratio of one FAMOUS patient to two control patients.

Qualitative interviews will take place with GPs, APNs and patients to assess factors that are difficult to capture using quantitative data. This evaluation procedure will be carried out by an independent institution (IMBEI). In addition, the intervention will be assessed in more detail in an internal evaluation from the trial sponsor using a longitudinal design.

The study is being conducted in nine general practices in rural areas of Rhineland-Palatinate, Germany. GPs interested in participating in the study were invited to apply and were selected by the project team of Catholic University of Applied Sciences (CUAS). The main criteria for selection were the number of suitable multimorbid patients and the willingness of the practice to implement the intervention. The number of treatment cases in the participating general practices ranges between 2000 and 6000 patients per quarter. Approximately 30% of them are multimorbid. With the size of the practice, we have ensured that the number of potentially eligible patients is sufficient to achieve the number of cases targeted in the study.

Patients (or a legal representative) must provide written informed consent before any study procedures occur. The inclusion criteria are as follows: at least 18 years of age; medical diagnosis of at least three chronic diseases that have led to GP treatment in at least three out of four quarters in the last twelve months; and insured by statutory health insurance. The exclusion criterion is a life expectancy less than 12 months as assessed by the treating physician.

The initial data of the participating patients includes age, sex, medical diagnoses, and number and duration of illnesses. These data are collected during the enrolment phase. Each patient enrolled in the project takes part in APN care for twelve months. The initial assessment lasts on average 1.5 hours. Follow-up contacts for the implementation and continuation of the care plan usually take place every four weeks. Depending on the need, however, the intervals can also be significantly shorter or significantly longer. APN contact is held at least quarterly. Routine contacts last on average 30 minutes. Depending on patient concerns and needs, the time frame may vary.

Routine data are used to count emergency contacts as the primary outcome measure, which means that there is no time stress for the patients. Secondary data are supplied by the ASHIP and by one health insurance company. However, there are patients for whom not all data regarding hospital and emergency contacts are available from routine data. For them, data are collected by the APNs. These data are also used for the analysis of secondary endpoints and health economic analyses. In addition, 30 patients will be interviewed about their experiences after completion of the 12-month intervention. We do not foresee the need for follow-up data collection.

The assessment tools mentioned above are used to measure the health status of the patients in the internal assessment. Data will be collected at the beginning (t0) and at the end (t1) of the 12-month intervention. Patient satisfaction with APN care is measured using a project-specific questionnaire.

The participants’ timeline in relation to the individual steps that occur during enrolment, intervention and evaluation is shown in Table 2:

The number of cases was calculated on the basis of the number of emergency contacts after twelve months (primary outcome). Based on literature estimates, it was assumed that the expected number of 2.7 emergency contacts per patient per year could be reduced by approximately 15% to 2.295 emergency contacts. The alpha was assumed to be 0.05. Taking into account the cluster effect and a dropout of 30% and a target power of 80%, this results in 817 patients in the intervention group and 1634 patients in the control group.

Recruitment of patients will be carried out by the participating general practitioners through a direct approach. Patients trust their GPs, so a recommendation to participate made by primary care physicians is likely to result in a high participation rate. The recruitment process is supported by the offer of personal contact between APNs and potential participants in the enrolment phase. In addition, to support recruitment, the patients are provided with comprehensive information and educational material about the project in easy-to-understand language, informing them about the project, its objectives and the data collection and processing that will be carried out to achieve the project goals. Posters and flyers are placed in the waiting rooms of the participating GP offices and are customized with the individual data of the general practice and the APN.

For the independent evaluation, data from ASHIP and the health insurance fund will be used for primary and secondary endpoints. To fill in missing data, such as hospital visits for patients with other health insurance, APNs collect health-related data and document patients' emergency and routine visits to any health-related facility (doctor's office, hospital, emergency room). The health insurer provides a data pool for control group matching. All data are transferred to the IMBEI in pseudonymized form. Patients, GPs, and APNs are informed by mailed invitations about the possibility of interview participation and asked to complete a feedback form. Potential interviewees are called via the feedback form, and appointments are made. The qualitative interviews are conducted by telephone using a previously prepared standardized interview guide. There is a separate guide for each interview group, which is finalized after piloting. Interviews are audio recorded and transcribed verbatim.

As part of the internal evaluation, the health status of each participating patient was measured before (t0) and at the end of the patient-individual intervention phase (t1) using the assessment instruments mentioned in the intervention section above as we use the data from these instruments for care and research purposes. Their psychometric data are as follows: DemTect: interrater reliability r = 0.993, sensitivity 85.1%, specificity 97% [14]; MNA: internal consistency α = 0.701, test–retest reliability r = 0.697 [23]; NRS: with a cut-off point NRS > 4, the sensitivity of ‘unbearable pain’ is 83%, while the specificity is 96.7%, with a cut-off point NRS > 5, the sensitivity is 94%, while the specificity is 85% [24]; PDI: internal consistency α = 0.86, [17]; HAQ: internal consistency α = 0.78–0.98 [25, 26]. The psychometric results of the German translation are as follows: test–retest reliability 0.94, internal consistency α = 0.92, criterion validity 0.76 [27]; MILVA: internal consistency range between α = 0.61–0.94, sensitivity 72%, specificity 59%; [17]. PAINAD-G: internal consistency α = 0.85, interrater stability r = 0.80, retest reliability r = 0.90 [21]; DEMMI: internal consistency α = 0.83, interrater reliability ICC 0.94 [28].

For the internal evaluation of patients’ satisfaction with APN care, a standardized questionnaire is used before and at the end of the patient-individual intervention phase. We take the APN-BQ (Advanced Practice Nurse- counselling quality) questionnaire and modify it for project-specific concerns. The APN-BQ aims to measure the quality of advanced practice nurse counselling in home care settings. The APN-BQ has shown itself to be a reliable instrument with good content and construct validity. The value is α = 0.86 [29]. The questionnaire for relatives was adapted to the patient questionnaire, which made sense from a thematic point of view. The timing of the data collection runs parallel to the data collection from the patients. All GPs in the participating practices are questioned via standardized project-specific questionnaires at the beginning (t0) and the end (t1) of the project intervention. In addition, qualitative interviews are conducted with medical assistants as well as with the APNs. Two medical assistants from each participating general practice are interviewed before (t0) and at the end of the 24-month period of project intervention (t1), and all nine APNs are interviewed before (t0), after 12 (t1) and after 24 months (t2) of intervention. The interviews of both groups are guideline based.

Data from patients in the intervention group will be entered by the APNs. For this purpose, a special secure software that stores all data securely and provides secure data transfer to the trust centre has been developed. The IMBEI receives the pseudonymised data from the trust centre as well as from the health insurance funds and the ASHIP by means of Qiata File Transfer Appliances. All data, which have already been checked and cleaned by the ASHIP and the health insurance fund, as well as the data of the APN entries, are again subjected to a quality and plausibility check by the IMBEI.

Both quantitative and qualitative data received by the IMBEI are stored in a secure network at the University Medical Center of the Johannes Gutenberg University Mainz. The appropriate technical and organizational measures according to the General Data Protection Regulation (GDPR) are listed in a separate document. Upon completion of the project, all data will be archived and secured in accordance with applicable laws.

Additional analyses will take place for the individual endpoints, which are combined into the aggregate endpoint of emergence contacts. Subgroup analyses as well as sensitivity analyses will be performed.

The data for the analyses come from different sources. The data of the control group and the data of the insured persons of the participating health insurers are assumed to be complete. Data from the ASHIP are available for all patients. Therefore, it can be assumed that the outpatient care data of all patients are complete. Inpatient data of patients who are not insured by the participating health insurance company will be requested from the APN and entered into a database. Subgroup analyses were performed to assess the data quality of these data. Data can be supplemented by means of imputation procedures.

Due to the low risk, no data monitoring committee (DMC) will be established. Interim analyses will not be conducted.

Adverse events are not to be expected. APNs will be in regular contact with patients, so they should be able to detect any deterioration in their health status. If the APNs notice any unclear change in the health status of the patients, they will discuss this with the GPs. In addition, the GPs have regular contact with the patients. If any adverse event should occur, the APN will report to the mentor accompanying her or him.

The routine data provided by the health insurance company and ASHIP are controlled and quality-checked through well-established procedures. These high control standards are necessary beyond the research requirements because routine data are usually the basis for billing modalities.

The documentation form, which is used to document the patients' health status and the activities of the APNs, is continuously monitored by the experienced mentors who accompany the APNs. In the monthly structured evaluation meetings or - if necessary - in between, the mentors provide the APNs with feedback and discuss any need for improvement.

Mentors are part of the project team, which holds weekly meetings to discuss the progress of the project and to make steering decisions as necessary.

The protocol, informed consent forms, participant education and recruitment materials, and other requested documents were approved by the Ethics Committee of the German Society of Nursing Science. The ethical reference issued (No. 21–012). All methods were carried out in accordance with relevant guidelines and regulations. Informed consent was obtained from all subjects or their legal guardian.

Any change to the protocol that may affect the conduct of the study, potential benefit to patients, or patient safety, including changes in study objectives, study design, patient population, sample size, study procedures, or significant administrative aspects, will require a formal amendment to the protocol. Such a change will be submitted to the Ethics Committee of the German Society for Nursing Science and will only be implemented if approved.

Administrative changes to the protocol are minor corrections and/or clarifications that do not affect the conduct of the study. The Ethics Committee of the German Society for Nursing Science will be informed about these administrative changes.

Patients who meet the inclusion criteria are informed verbally and in writing by their GP. If desired, patients will have the opportunity to speak with their potential future APN. She or he will explain the APN care program again. General practitioners obtained informed consent from patients if they decided to participate in the study. Participants document their informed consent by signing a treatment contract. If necessary, the information is passed on to the legal representatives, and their signature on the treatment contract is obtained.

In accordance with the GDPR and the project-specific data privacy concept, all study-related information is kept securely at the study site. All participant information is kept in locked file cabinets. All reports, data collection, process, and administrative forms will be marked with a coded ID (identification number) only to maintain participant confidentiality. All records containing names or other personal identifiers, such as consent forms, are kept separate from study records, which are marked with a code number. All local databases are secured by a password protected access system. The health record consists of two parts: part one contains the data identifying the person, and part two contains the health-related data. The person-identifying data and the health-related data are hosted on different servers. Therefore, only pseudonymized data are made available to the external evaluator.

All authors have completed the International Committee of Medical Journal Editors (ICMJE) uniform disclosure form at http://​www.​icmje.​org/​disclosure-of-interest/​ and declare that they have no support from any organization for the submitted work (with the exception of the funding by the Innovation Fund); no financial relationships with any organizations that might have an interest in the submitted work in the previous three years; and no other relationships or activities that could appear to have influenced the submitted work.

The Principal Investigator will have access to the cleaned datasets by the independent evaluator. All other partners involved in the project may have access to the data upon request. There are no contractual agreements restricting this access.

During the intervention phase, care is provided by the APN within the framework of delegation. This means that the physicians are informed at all times about the patient's state of health and the treatment steps. After the end of the intervention, the patients returned to the normal care of general practice.

The study group aims to publish the results in scientific papers, present them at scientific congresses and use all appropriate channels from a scientific point of view to inform the scientific community. In addition, the study group will inform policy-makers and health care providers involved in this area about the outcomes. The various partners of the study group will be informed in advance of planned publications as contractually agreed upon.

The main publication will be prepared by the study director. All active project partners are coauthors, according to the guidelines of the ICMJE [30].

All partners involved in the project are allowed to publish the data obtained. To do so, they must first prepare an exposé that is submitted to the Publication Steering Committee, which is composed of the study director and the head of the independent evaluation. If the exposé is approved, an application for data export can then be submitted to the IMBEI. All publications must be submitted to and approved by the FAMOUS Publication Steering Committee prior to submission.