Implementing a case management intervention for frequent users of the emergency department (I-CaM): an effectiveness-implementation hybrid trial study protocol

In line with FUEDs’ specific needs, the CM intervention aims at redirecting and reorienting FUEDs to a range of services within the hospital and community-based settings to improve the quality of care for these patients and reduce the ED overcrowding that negatively impacts all ED patients [12]. The CM intervention targets several mechanisms through which the CM intervention is expected to decrease FUEDs ED visits and related costs and to increase their quality of life. In addition to cares coordination, these mechanisms include health care empowerment (i.e., being engaged, informed, collaborative and committed to one’s health care) [17], perceived self-efficacy (i.e., the belief in one’s ability to perform a task) [18], and health literacy (i.e., skills to meet the complex demands related to health) [19].

Building on previous findings related to CM effectiveness [12], our research team developed and tested a CM intervention tailored to FUEDs in Lausanne University Hospital [3, 20]. Based on FUEDs needs and characteristics [21, 22] this intervention targets several specific components. First, the case managers provide counseling and education on health care utilization, substance abuse and the social determinants of health related to FUEDs, using skills such as motivational interviewing and cross-cultural competences [23, 24]. Additionally, the case managers enable access to concrete services, such as social services (i.e., providing support on income assistance, stable housing, improved health insurance coverage), and refer FUEDs to psychiatric, substance abuse treatment, and/or medical services (e.g., GP or medical specialist) on a case-by-case basis. Another key element of the CM intervention is to connect all health care or social service providers within the hospital and in the community, promoting continuity of care and improving the FUEDs’ ability to navigate the complex health care system. Along with this key dimension of the CM intervention, case managers send on a case-by-case basis letters summing up their intervention to FUEDs’ general practitioners or other involved health-care providers.

Previous research among FUEDs has established CM intervention effectiveness in improving housing and environmental quality of life and reducing homelessness [12, 13]. There is however, mixed evidence of CM intervention effectiveness in reducing ED use. On one hand, systematic reviews conducted in 2011 [12] and 2016 [14] on interventions including CM, care navigation, patient education and disease specific management programs, found that CM intervention only consistently reduced frequent ED use. On the other hand, results yielded in the few randomized clinical trials conducted to date are mixed [3, 15, 25, 26]. For instance, a recent randomized clinical trial conducted by our research team tested CM intervention efficacy [3]. FUEDs were randomized to the CM intervention or control groups. After 12 months, participants in the intervention group made 2.71 (+/− 0.23) ED visits versus 3.35 (+/− 0.32) visits in the control group (ratio 0.81, 95% CI 0.63–1.02), corresponding to a 19% absolute reduction in ED use. Of important note however, findings indicated that CM intervention effect on the number of ED visits was not significant. Therefore, although most findings support CM intervention effectiveness in reducing ED use and improving FUED’s quality of life, research further evaluating these questions is needed. Accordingly, we selected a hybrid design for this current study, examining both implementation and clinical outcomes [27].

Numerous conceptual frameworks exist to guide the design of research aiming to implement evidence-based practices in real-world settings. In this study, we used the Generic Implementation Framework [28] to guide the overall research protocol design and the RE-AIM (reach, efficacy, adoption, implementation and maintenance) [29] framework to guide the implementation outcomes evaluation.

To the best of the authors’ knowledge, no previous study has focused on the implementation of a CM intervention for FUEDs in Switzerland. French-speaking Switzerland includes 7 cantons and covers 23% of the extension of the country (with 2 million persons in 2015). ED overuse is of practical concern to public health advisors and clinicians. Research focusing on both implementation and clinical outcomes may provide key insights into the mechanisms and processes for disseminating and implementing a CM intervention for FUEDs into broader practice in the real-world settings. This study aims at 1) implementing a CM intervention for FUEDs in public hospitals with EDs in the French-speaking region of Switzerland, 2) studying the process of implementation of the intervention and, 3) studying CM intervention’s effectiveness among FUEDs. Regarding the intervention effectiveness assessment, we expect that FUEDs will evince improvements in clinical outcomes (e.g., higher scores in quality of life) and decreases in ED visits after receiving the CM intervention.

This research project features a Type 2-hybrid study design focusing on both implementation and clinical outcomes [27]. The implementation part of the study uses mixed methods a) to describe both qualitatively and quantitatively factors that may influence implementation process and b) to assess implementation outcomes. The clinical part of the study uses a within-subject design (pre-post intervention) to evaluate participants’ trajectories after receiving the CM intervention.

All community and academic public hospitals with an adult ED opened 24 h/ 7 days per week throughout the French-speaking region of Switzerland interested in implementing the CM intervention are eligible. There are no exclusion criteria.

Patients visiting the ED implementing the CM intervention are eligible if they are ≥18 years, able to communicate in a language that is spoken by the local team or a professional interpreter and have ≥5 ED visits in the past 12 months recorded. Exclusion criteria include presenting less than 2 vulnerability dimensions (i.e., social, somatic, mental, risk behaviors) besides frequent ED use, being unable to give informed consent, planning to stay in Switzerland less than 18 months, not expected to survive at least 18 months, awaiting for incarceration or being currently incarcerated, having a family member already enrolled in the study.

Given that this research study aims to disseminate and implement a CM intervention, we are neither aware of the number of hospitals that will eventually participate, nor of the number of participants who will be enrolled in each hospital.

The CM intervention, as developed by our research team, typically includes the following steps. First, health-care providers and/or administrative staff detect FUEDs in the ED through an electronic alert in patients’ files. They then notify the case manager that a FUED is in the ED. Next, if the FUED visiting the ED agrees to do so, the case manager makes a first evaluation of the FUED’s situation in the ED, aiming to identify his main problems, the involved health-care network and the priorities that need to be addressed. After this first contact and if the FUED agrees ambulatory care, the case manager writes an intervention report and sets an appointment with the patient. Notably, case managers send on a case-by-case basis letters summing up their intervention to FUEDs’ general practitioner or other involved physicians. Ambulatory care consists of setting a number of appointments with the case manager on a case-by-case basis (in the ED, in the community or health-care network, or at the patient’s home depending on situation and needs). Concretely, the case manager, together with the patient, identifies objectives based on the patient’s specific needs, which will define the total number of appointments needed. Ambulatory care aims at reaching these objectives through appointments with the involved health-care network and referral to specialized services fitting the patient’s needs. Staff involved in this step include the case manager, specialized workers from the health-care network, and sometimes, the patient’s relatives. Whenever possible, the patient is asked to be active in the decisions taken to address his social and medical needs (patient empowerment). The intervention concludes when the patient is integrated into a coordinated and functional health-care network.

The whole project should be completed in 3.5 years over five phases (see Fig. 1). We present below the study procedure over the five phases. Table 1 presents a summary of the implementation strategies by phase. We will use hereafter the term I-CaM research team to refer to staff working on the research project in the home institution, including 2 research assistants (bachelor-level students in Medicine or Social Sciences), a research nurse (nurse, MPH), a scientific collaborator (Ph.D), and the research project leader (Associate Prof, MD, MSc).

This project addresses a critical area of ED health services overuse using a practical solution, while taking an innovative research approach that will contribute to the growing literature on this topic. In terms of research feasibility and innovation, implementation science is an emerging field and serves as an ideal theoretical foundation to complement the study of the key clinical outcomes in diverse settings. The hybrid study design, and rooting our project in the established GIF and Re-AIM frameworks, provides a novel and feasible research approach. Findings of the current research project will provide useful information related to the mechanisms and processes for implementing a CM intervention for FUEDs into broader practice in the real-world settings.

This research project is also likely to have social and economic implications. ED overuse and overcrowding is a critical problem, and this project promotes a potential solution. At the Lausanne University Hospital ED, FUEDs accounted for 4.4% of ED patients yet 12.1% of all ED visits (i.e., in 2008–2009; n = 5813) [4]. There is a strong link between chronic disease and frequent ED use, driven largely by the medical and social complexity of these patients. Addressing ED overuse through CM redirects FUEDs to more appropriate forms of care, including primary care or substance abuse treatment, while remaining cost-effective. By developing a CM intervention program and studying its implementation at multiple sites throughout the French-speaking region of Switzerland, this project will provide insights and lessons for the broader use of CM interventions for health care services overuse throughout Switzerland.

Finally, the CM intervention targets a defined population— FUEDs —that is frequently highly vulnerable in terms of medical complexity, suffering from multiple chronic medical conditions, mental health and substance abuse treatment needs, as well as modifiable social determinants of health, including homelessness or low household income. Therefore, this project also addresses a medically and socially complex patient population with chronic health care and social service needs, as well as emphasizing improved communication and care coordination within interdisciplinary care teams.

There are several expected limitations in this current research project that were carefully considered while designing the study. Regarding clinical outcomes, the design of this study allows patients’ evolution to be evaluated after receiving the CM intervention (i.e., pre-post intervention; within-subjects design) in the sites implementing the intervention. A well-known limitation related to this design is regression to the mean. In fact, we previously conducted research using gold-standard randomized controlled trial to test the CM intervention efficacy [3] and the proposed within-subjects’ design is consistent with Stages I and II treatment development [51]. That said, interpretation of pre-post intervention changes on health outcomes will be made with caution. Next, self-reported data can be subject to reporting bias [52, 53]. It can however, be reliable when measures are developed for and piloted on the target population, timeframes are manageable, confidentiality is ensured and the target behavior is not stigmatized [54, 55]. The above conditions were carefully considered when designing the study, thereby minimizing the risk of self-report bias. Furthermore, recruitment and retention can be a challenge with the target population (i.e., FUEDs), resulting in missing data. However, previous work with a similar population, has demonstrated the possibility to attain acceptable 1-year retention rates [3]. Procedures for maximizing study retention will therefore be inspired from the latter study. For instance, we will use incentives to help retention over time (gift cards of increased value for each follow-up assessment completed: 10, 20 and 20 CHF for 3, 6 and 12-month follow-up assessment). Additionally, specific steps will be taken to address any bias that may occur because of missing data as described previously. Finally, although not known at this stage, the patient sample size will be larger than the hospital sample size; therefore, the implementation outcomes analyses will not have the same level of analysis and will mostly remain descriptive. To address this limitation, we will use mixed methods (i.e., qualitative and quantitative methods) to assess the implementation process and outcomes.