A growing recognition of the potential of rehabilitation has led to higher prioritization and acknowledgment of the need for high quality rehabilitation research (1, 2). Rehabilitation is, in essence, composed of a number of complex interventions (3)changing along the disease and recovery stages and involving shifting professionals. Thus, it is pivotal that rehabilitation research mirrors this complexity.

The benefits of a biopsychosocial approach in rehabilitation to reduce negative consequences of health issues has attracted growing attention and recognition (4). An increasing number of people need rehabilitation due to better medical treatment and demographic changes (5, 6). Worldwide, more than 2.4 billion people would benefit from rehabilitation (7). It has been established that rehabilitation may enhance self-efficacy and autonomy among people with disabilities (8) leading to decreasing health care needs. Consequently, rehabilitation may increase quality of life (QOL) and reduce the burden on health care systems and societies (5, 8).

According to WHO, rehabilitation addresses different aspects of the lived life and is “a set of interventions designed to optimize functioning and reduce disability in individuals with health conditions in interaction with their environment” (7). Thus, rehabilitation ranges from simple training that supports the individuals (here referred to as service users) to resume previous levels of functioning, to complex interventions for service users with multimorbidity and permanent loss of functioning that negatively affect several aspects of their everyday lives. The simultaneous parallel and linear interventions may take place over time and be considered as a continuous process involving physical, psychological and social aspects, especially for service users with limited resources, complex problems, and evolving needs. These aspects are reflected in the biopsychosocial understanding of the International Classification of Functioning, Disability and Health (ICF model) (9) and in established definitions of rehabilitation (7, 10). If rehabilitation is intended to improve functioning and facilitate recovery toward better QOL, autonomy and a meaningful life, then targeted coherent efforts are required, based on the service users' perspectives and life situations. Rehabilitation should thus be seen as a continuous collaborative process between the actively involved service user and an interdisciplinary group of social, vocational and healthcare professionals (11, 12). This complexity of independent variables and individualized approaches happening at different times makes 'rehabilitation research' complex and may explain why rehabilitation efforts are challenged by a limited knowledge base (2, 8). For example, in patients with low back pain, lack of rehabilitation research that covers broad aspects of functioning, QOL, autonomy and a meaningful life has been claimed (13). Similar problems have been described in stroke rehabilitation research (14). The lacking knowledge has implications for how rehabilitation research must be planned, executed and evaluated, if rehabilitation as evidence-based practice shall improve functioning, personal independence and a meaningful life. Thus, rehabilitation research must mirror the full rehabilitation process. Additionally, societal needs to optimize health-related costs imply that rehabilitation research should examine and test the effect of rehabilitation efforts as well as the individual service users' experiences of how the efforts contribute to increase and maintain functioning, QOL and autonomy. Complex rehabilitation interventions call for application of multiple research methods and designs from different scientific traditions to cover wider perspectives than traditionally explored by the randomized controlled trial (RCT) (2, 15). Moreover, long-term studies of multifaceted interventions are needed to examine all elements of a specific rehabilitation trajectory (2). Therefore, it becomes the responsibility of rehabilitation researchers to consider, optimally in collaboration with the service users, how rehabilitation research best meets the needs of the end users including service users, professionals, politicians, and administrators. The aim of the present paper is to offer recommendations for the broad field of rehabilitation research emphasizing the special characteristics of complex rehabilitation interventions that researchers should use to supplement established generic health-related research practices (16).

The Nominal Group Technique (NGT) was used to identify recommendations for rehabilitation research practice. This method is designed to explore opinions, generate ideas, and determine priorities in e.g., the health sector (17–20). Our group process followed the five steps process: (1): Introduction and explanation; (2): Silent generation of ideas building on individuals' knowledge, perspectives, and experiences; (3): Sharing individual contributions (round robin); (4): Group discussion for clarification; and (5): Prioritization of ideas (individual ranking).

The take-off of the nominal group process was a collaboration initiative between research, education and clinical practice in Central Denmark Region with focus on rehabilitation development in the region (21). Discussions were initiated at two meetings attended by a group of two representatives for service-users and 33 experts in rehabilitation (the expert group) from four independent sectors: hospitals; health and social care services; university, and university college. The expert group represented different backgrounds in relation to gender, education, role as clinician or researcher or patients' representative, organizational affiliation, and research tradition. From this expert group, six researchers with health and social care professional backgrounds representing the direct users were appointed to a working group to formulate targeted recommendations for rehabilitation research. The working group met four times from May to October 2019 to form the recommendations and at additional meetings to describe the final outcomes. The entire working group took part in all the five steps described below.

The NGT process resulted in six recommendations for rehabilitation research as presented in Table 1 and explained in detail below.

In this paper, we have raised the need for targeted recommendations in rehabilitation research and have attempted to establish such recommendations to potentially clarify and increase the quality of the research and to ensure a transparent process from idea to implementation of results.

Rehabilitation is recognized as an important part of health care services. However, there is a need for more high-quality research, that encompasses a broad field of disciplines and methodologies covering the full spectrum from basic to applied science, and involves many different specialists with different research traditions. Ideally, a scientific process uses the most appropriate design to answer or illuminate the research question, and no single research tradition or method can be recommended or stand alone in rehabilitation research. The randomized controlled trial (RCT), often the default choice for intervention studies, has been claimed unfeasible for some clinical questions (e.g., if a particular presentation or condition is heterogeneous or rare) (38). The WHO guideline from 2017 stressed the limitations of randomized controlled trials and suggested that results from case-, observational or longitudinal studies can capture how environmental factors impact interventions at health system-level) (39). Likewise other types of evidence, as qualitative studies, are needed, too. Other methodological questions to address in relation to effect studies have been pointed out in relation to Cochrane studies. They relate to heterogeneous patient populations, complex rehabilitation interventions that are difficult to standardize, and to often vaguely described control conditions (38). The challenge of applying appropriate designs implies a need for future scholarly work on design development for rehabilitation research.

Using our recommendations can be a way to establish common ground for future discussion and development of holistic rehabilitation research, e.g., by using the biopsychosocial perspective of the ICF and the new ICF-matrix and the perspectives of the people receiving rehabilitation services to define the specific area of interest in each research project. It is important to emphasize that these recommendations are intended to guide researchers, decision-makers and funders within rehabilitation and rehabilitation research, all of whom do not necessarily have long experience with the discipline.

Although the recommendations are targeting rehabilitation researchers, there should be no doubt that we find it absolutely crucial that end users, including service users and stake holders, participate in the rehabilitation research process. End users are to be involved at distinct stages of the process: (1) priority setting and formulating research questions (2) study design, data collection and analysis (3) dissemination of findings and knowledge translation. This will improve the relevance and quality of rehabilitation research. Thus, the ultimate aim of the recommendations is to increase the knowledge base for rehabilitation by improving the actuality, relevance and implementability of rehabilitation research. Therefore, an important part of the recommendations stresses the involvement of all end users in the field of rehabilitation, in line with Solvang et al.'s emphasis on identifying and involving all agents in the field of rehabilitation at micro, meso, and macro level (40).

The original contributions presented in the study are included in the article/supplementary material, further inquiries can be directed to the corresponding author.

All authors listed have made a substantial, direct, and intellectual contribution to the work and approved it for publication.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.