|dc.description.abstract||Facilitating individual engagement in the rehabilitation process is vital if our investment in interventions is to achieve the desired outcomes. Studies of engagement in the cardiac rehabilitation (CR) context are limited, in that they lack a comprehensive and detailed basis for understanding and monitoring the whole process of engagement. The Model of Therapeutic Engagement (MTE) (Lequerica & Kortte, 2010) is the most comprehensive theoretical framework yet proposed to explain CR engagement, however, in the ten years since its first proposal, no research has investigated this multi-layered model empirically to determine its utility in the context of CR. The MTE defines the process of engagement by theorising a series of sub-models that focus on: 1) individual intention to engage in CR programs; 2) initiation of CR (i.e. actual attendance); and 3) maintenance of participation in CR programs over time (i.e. completion). Although the MTE is likely to be useful in understanding engagement in CR, it has been derived from a psychological orientation. Consequently, it gives little consideration to the role of socio-environmental factors which are considered vital in the fields of rehabilitation and health promotion. The overall aim of this empirical study was to evaluate the way in which the components of the MTE contribute to engagement outcomes and interact with each other, and also to examine the role of socio-environmental barriers in the MTE. This empirical study aims to evaluate the theory underlying the MTE by implementing it empirically, and thus develop a better understanding of the process of CR engagement. Through this approach, useful predictors that may act as management ‘levers’ can be identified, which if manipulated, can lead to better engagement. In turn, this could inform planning and design of future programs.
The research began with a synthesis of the existing evidence for each of the proposed relationships among variables at each of the three stages of the MTE within a CR setting (see Chapter Three). A model-centric systematic review was used to explicitly structure the evidence, according to each stage of the MTE. This review identified eight studies which focused on aspects of stage one of the MTE, four additional studies which were relevant to stage two of the MTE, and six studies which considered aspects of stage three of the MTE. The results showed that the propositions of the first stage of the MTE have been well supported in the literature. However, there has been limited research investigating the proposed relationships among the variables that define the second and third stages of the MTE. Importantly, the literature review revealed that research to date has failed to provide a holistic approach to the understanding of individual engagement in CR programs. Hence, this literature review provided a strong basis for designing a substantial empirical study that aimed, for the first time, to comprehensively consider all stages of the MTE. However, due to its complexity, the analysis of the whole MTE was broken into several components, as explained below.
Implementing a complex multi-component model such as the MTE, in an empirical study, raises significant analytical challenges. The MTE reflects the process that individuals follow during CR as this process unfolds over time. As would be expected, some individuals drop out. Changes in the sample during investigation of each stage of the MTE reflect the construct of interest (i.e., engagement) and, therefore, cannot be ignored by treating this as missing data or attrition. For example, not all those who are referred to CR will wish to attend; not all those who intend to engage in the CR program will initiate contact; attendance will fluctuate during CR; not all those who remain in the program will sustain their engagement.
These attrition issues make it difficult for empirical modelling to make use of a single model to reflect the whole process of CR. For that reason, a modular approach to analysis was chosen, where each module focuses on a different key stage of the MTE process: 1) individual intention to engage in CR programs; 2) initiation of CR (i.e. actual attendance); and 3) maintenance of participation in CR programs over time. Due to the emphasis on modelling relationship ‘pathways’ amongst variables, and the limited amount of data available, structural equation modelling (SEM) was determined to be suitable to provide an explanation of engagement, and therefore was applied to each of the three stages of the MTE.
In the initial empirical study, the first stage of the MTE was evaluated in a sample of 217 participants at one hospital in metropolitan Australia who were referred to CR following a cardiac event that required hospitalisation. The results (Chapter Seven) revealed that perceived self-efficacy and perceived need for rehabilitation positively impacted on intention to engage in CR, with moderate to large effect sizes. Perceived need and outcome expectancies were also strongly and significantly associated with each other. Contrary to the MTE, there were no significant relationships detected between outcome expectancies and intention to engage in the CR program, and between perceived self-efficacy and outcome expectancies. However, more detailed inspection (by supplementing SEM with bootstrap resampling) revealed that willingness to consider treatment acted as a mediator of the relationship between perceived self-efficacy and intention to engage in the CR program, through a small indirect effect that was significant and negative. The inclusion of this mediator doubled the variance explained by the relationship pathways.
A prospective study was then conducted to test the entire MTE in the subset of 101 participants who enrolled in, and commenced, the CR program. The subsequent findings (Chapter Eight) mirrored the effects found in the initial cross-sectional study (Chapter Seven). The effects remained the same in both direction and size in most cases. The main exception was the small significant negative relationship between perceived self-efficacy and outcome expectancies, which was much smaller and non-significant in the total sample. Another exception was the effect of perceived need on intention to engage which was found to be positive and significant in both studies, but larger in the cross-sectional analysis (Chapter Seven) compared to the prospective analysis (Chapter Eight). In this prospective study, the findings were consistent with all relationships proposed within the second and third stages of the MTE.
Finally, in Chapter Nine, the SEM was expanded to a multi-group analysis in order to examine the role of socio-environmental barriers. Despite being omitted from the original MTE, these barriers were found to be important in this first exploratory empirical analysis. The findings indicated that experiencing high-level socio-environmental barriers appeared to have a substantial moderating impact on the majority of the relationships proposed at the second and third stages of the MTE, with barriers tending to reduce engagement as would be expected. Significant differences were found between participants with high-level barriers and those with low-level barriers when comparing the effects, as shown by the beta estimate of the statistical model. Several relationships in the model were significantly weakened for participants with high levels of barriers compared to participants with low levels of barriers. Specifically, the strength of the relationship between intention to engage and actual involvement in preparation for the program was five times lower as was the relationship between their analysis of the CR experience and engagement. The strength of the relationship between CR initiation and engagement in the CR program was ten times lower and the relationship between CR maintenance and engagement was three times lower. These findings suggest that people with high levels of barriers are less likely to translate their intention to engage into actual engagement or involvement in preparing for CR. Even for those who initiated CR, or maintained CR over time, this was less likely to translate into deep levels of engagement. Importantly, their engagement over time was less likely to be influenced by their experience of rehabilitation, presumably reflecting the greater influence of barriers on their engagement. Thus, overall, a high level of socio-environmental barriers substantially diluted the relationships between variables in subsequent phases of CR, with the relationships being three to ten times stronger for individuals who had a low level of barriers.
In summary, this study showed that perceived self-efficacy and perceived need for rehabilitation positively impacted on intention to engage in CR in the total population of eligible participants. Willingness to consider treatment was a strong mediator that doubled the likelihood of self-efficacy influencing intention to engage in CR. Perceived need was associated with outcome expectancies, however, these expectancies did not directly impact on intention to engage in CR program. Perceived need reduced in importance for people who did engage in CR, as might be expected. For those who did engage in CR, sustained engagement was associated with individual analysis of experience. Barriers in the social and physical environment reduced the strength of the MTE relationships, so must be included in future articulations of the model. Addressing barriers would significantly enhance engagement in CR.
This study has indicated that the MTE can improve our understanding of the process of engagement in CR programs. Further, a deeper acknowledgment and understanding of socio-environmental barriers can improve the utility of the MTE. Based on this more holistic view of CR, as provided by the extended MTE, it is now possible to identify potential management levers that could be manipulated in order to enhance engagement. In particular, these results highlight that more attention is needed for raising the perceived need for CR and supporting the involvement of participants in preparation for their rehabilitation. These changes could significantly increase intention to engage in CR and actual initiation. Most importantly, the current study has confirmed the need to carefully review socio-environmental barriers that affect sustained engagement in CR. Prospective studies with larger samples than used in this first exploratory study, are required to determine whether these findings can be replicated, and are generally applicable to different populations of CR patients. Such a quantitative analysis could perhaps be supplemented by a qualitative study, to allow a deeper examination of the mechanisms underlying the apparent relationships among these variables. In addition, future studies must focus directly on exploring the role of other elements of engagement in order to provide a more comprehensive understanding of the engagement process. Specifically, greater consideration needs to be provided to CR barriers and facilitators, such as health professional-related factors (e.g. clinicians’ communicative and relational skills); factors associated with family caregivers of individuals; and mental distress factors to expand the MTE.||en_US