1. Introduction

Theory is widely regarded as integral to scientific progress. Using theories in research is thought to increase the quality and effectiveness of health interventions ^[1][2], making a theory-based intervention more likely to be effective than a purely empirical or pragmatic approach^[3]. Yet, even a cursory review of the published literature will demonstrate that investigators struggle with the practicalities of how to use theories to design interventions. Thus, it is unsurprising that some have entirely given up and have argued that theories are unnecessary^[4]. Yet, changing human behavior is particularly difficult. There are numerous theories of behavior change available to support the development of interventions; however, it is unclear how commonly those theories are used in behavior change interventions. Thus, the aim of this scoping review is to clarify how theory is being used in research evaluating interventions promoting self-care behaviors.

3. Discussion

The results of this scoping review illustrate that theories are rarely used to guide self-care research. Only one of every three self-care studies used a theory to underpin the rationale, the intervention, outcomes, and/or discuss the results. This rate is even lower than that recently reported in a review of nine systematic reviews on health behavior studies, where is was found that a theory was used by 47% of the study authors^[1][7].

Another major finding of this study was that a theory was most often used to a limited extent or even superficially; that is, a theory was used to inspire the researchers in the choice of certain concepts (e.g., self-efficacy) but not to guide their thinking about the rationale, development, and evaluation of the self-care intervention. Only eight studies fully used a theory in the rationale, intervention development, choice of outcomes, and discussion. One potential reason for the apparent lack of use of a theory is that it was used but space constraints limited the ability of the authors to describe their use of a theory. However, in case a reference was made to a published protocol or in-depth description related to the study, we did check these papers and use it in the analysis.

Another potential reason for the limited theory use could be that the researcher might not have known how to select an appropriate theory ^[12]. But even if some studies used a theory, the reasoning was not always clear, since some others just copied previous studies in the area, including the use of the same theory^[2]. In our study, we found that some researchers only loosely referred to a theory without actually using it. Others have found that some researchers believe that referencing a theory makes it seem more sophisticated, solid, or advanced^[1][2]. Some research fields have a stronger tradition of theory use in guiding interventional studies; therefore, researchers might feel forced to mention theory even if it was used in an unreflective or superficial way.

We also found that individual theories were often used in only a single study, often without an explicit rationale for choosing that particular theory. When use of a theory is not replicated in more than one study, this limits the field’s ability to reinforce, revise, or refine the theory and build a cumulative body of knowledge over time. Assuredly, some theories will turn out to be more useful than others. It is only with accumulating evidence that we will come to realize that certain theories are not effective in facilitating self-care behavior change. A smaller handful of well-developed, empirically tested and relatively useful theories might contribute more to development of knowledge than the current practice. For example, previously it was suggested that the transtheoretical model of behavior change by Prochaska and Diclemente ^[13] should be replaced by better models of behavior change^[14]. Yet, surprisingly, 10 relatively recent studies in our review used the outdated theory to underpin their study.

Some studies reported using more than one theory for their work, also without a clear rationale of why this was needed. Using multiple theories can result in combining contradictory constructs of behavior change, which may hamper our ability to capture the potential impact of a coherent theory^[15]. Further, using more than one theory has not been found to be more effective than using a single, well-selected theory ^[4].

Interestingly, in those situations in which investigators did use a theory, they mostly approached self-care deficiencies as a problem related to lack of knowledge. That is, there was an underlying assumption that if people are not performing self-care, they must need to learn how to do it. Yet the wider literature illustrates that lack of motivation to change one’s behavior is a much more common barrier to self-care than lack of knowledge^[16][17]. When a theory was used to underpin the intervention, specific evidence-based behavior change techniques such as goal setting, problem solving, action planning, and review of behavioral goals were used more often. These techniques reflect the concepts used most often in theories such as social cognitive theory and the transtheoretical model of behavior change.

Self-care research can be improved by more efficiently building on existing knowledge by learning from other researchers. Explicit use of appropriate theory in future self-care studies might improve the design of a self-care intervention and subsequently help to build cumulative knowledge^[1][7]. Choosing a relevant theory is challenging given the numerous theories and overlapping constructs^[12][16]. Guidance is limited in choosing a theory suited to the target population, behavior, and context, which provokes researchers to choose a commonly used theory that is traditionally used in their field^[1][2]. To advance the potential benefit of using theory, there is a need for an accessible source of potentially useful theories and a method of how to select them^[1]. A practical and systematic guide for designing and evaluating behavior change interventions is based on the theory-based behavior change heel ^[17][18]. Also, a web-based tool is available for researchers in which various theories of behavior change are linked to various behavior change techniques (https://theoryandtechniquetool.humanbehaviourchange.org/too).

We advocate that researchers consider using a theory as a guide to develop, evaluate, and optimize self-care interventions, allowing for appropriate use of a theory in future self-care research and accumulation of evidence over time^[19]. Doing so would involve a process of choosing an existing theory on the basis of the goal to be accomplished (e.g., understand behavior, change and maintain behavior, influence the response to illness)^[1][9][10].

Some limitations need to be considered. We were only able to extract the use of a theory as described in the primary publication or study protocol, which might have underestimated the actual use of a theory. We used a self-developed coding scheme that allowed us to explore the use of theory in studies testing self-care interventions. Using a taxonomy that guides the use of techniques to theoretical constructs would increase an in-depth understanding of the use of a theory.