Evaluating Perceptions of Self-efficacy and Quality of Life in Patients Having Coronary Artery Bypass Grafting and Their Family Caregivers

Background Self-efficacy is a critical factor for quality of life in patients who undergo coronary artery bypass grafting, as well as for their family caregivers. However, there is lack of knowledge about whether patients' self-efficacy and caregivers' perceptions of patient self-efficacy are associated with quality of life in patient and caregiver dyads. Objectives The aims of this study were to compare self-efficacy and quality of life between patients and family caregivers and to examine whether patients' and caregivers' perceptions of patient self-efficacy were associated with their own and their partner's quality of life in patient and caregiver dyads who were waiting for patients' coronary artery bypass grafting. Methods In this cross-sectional study, 84 dyads (85% male patients and 87% female caregivers) completed the Cardiac Self-efficacy Scale, which consists of self-efficacy for controlling symptoms and self-efficacy for maintaining function subscales, and the Short-Form 12 Health Survey for quality of life. Data were analyzed using the Actor-Partner Interdependence Model. Results Caregivers rated patient self-efficacy for maintaining function higher than did patients themselves and caregivers' perceptions were positively correlated with patients' physical health. Patients' self-efficacy for maintaining function exhibited an actor effect on their own mental health. There were no other actor or partner effects of self-efficacy on quality of life. Conclusions Differences between patients' and caregivers' perceptions of patient self-efficacy for maintaining function should be addressed before surgery to reduce discordance. Patients' self-efficacy for maintaining function was associated with their own quality of life. There was no partner (relationship) effect of self-efficacy on quality of life. More research is needed in this area.


Introduction
Coronary artery bypass grafting (CABG) is a surgical treatment option for patients with advanced atherosclerotic coronary heart disease. Quality of life of patients awaiting CABG is poor and it has been affected by physical factors such as poor physical health 1 and severity of angina, 2 and mental health, 3 including low self-efficacy. 4 High levels of self-efficacy have been shown to promote health behaviour change, support self management and improve health status through reducing symptom burden and physical limitations in patients with coronary artery disease. [4][5][6][7][8] Socioeconomic deprivation is also a predictor of poor cardiovascular outcomes in patients undergoing CABG. 9,10 Self efficacy as a concept is derived from Bandura' social cognitive theory of behaviour; defined as an individual's confidence in his or her ability to perform a given task. 11,,12 The theory of self-efficacy proposes that an individual's perceptions of his or her ability to perform certain health behaviours influences their health outcomes. 11,12 Patient recovery and adjustment after CABG, although largely determined by their physical condition and treatment, may be influenced by perceived self-efficacy. Patients with similar levels of physical impairment may achieve different functional outcomes, depending on their 4 perceived self-efficacy. [13][14] Furthermore, a spouse's or partner's confidence in the patient's capabilities can influence health-related outcomes as well. 15,16 There may be differences between patients' and caregivers' perceptions of patient selfefficacy and this could influence the level of support provided to the patient, and also patient and caregiver outcomes 17,18 Poorer quality of relationship between caregiver and patient, greater patient symptoms and caregiver strain are associated with caregivers overestimating patient self-efficacy. 17 Whilst substantive research has examined the patient and caregiver relationship in heart failure, [19][20][21][22][23] and whether spouse confidence predicts patient survival following heart failure, 24 the effect of self-efficacy in patient-partner dyads in CABG has been rarely examined. 14 Previous self-efficacy research has mostly involved a single assessment of either patients or caregivers. [5][6][7]14,[25][26][27][28][29][30] Such an individualised approach ignores the interdependency of behaviours or beliefs within the patient and partner relationship. 31 Because patients and family caregivers are affected by the patients' health status, interactions in patient and caregiver dyads are inevitable. The relationship between patient and caregiver is nonindependent. The Actor-Partner Interdependence Model (APIM), based on Interdependence theory, allows investigators to examine the inter-relatedness of variables in dyads. 32 It provides insights into dyadic interactions by taking both the individual and family caregiver contribution into account in a single regression model. In the APIM, the association between a predictor (independent variable) and outcome (dependent variable) for members of a dyad is composed into two distinct parts: the actor effect is the impact of a person's own predictor variable on his or her outcome. The partner effect is the impact of a person's predictor variable on his or her dyadic partner's (family caregiver's) outcome. [32][33][34] No pre-operative studies of CABG were found that examined the relation between patients' and caregivers' perceptions of patient self-efficacy and quality of life at the dyadic level. This study aimed to compare patients' and caregivers' perceptions of patient self-efficacy and quality of life before CABG; to examine whether patients' and caregivers' perceptions of patient self-efficacy were related to their own, and their partner's quality of life before CABG.

Design, sample and setting
This was a secondary analysis of cross-sectional data from a study of patients and family caregivers recruited from a regional cardiology centre in Scotland. 13 The population consisted of patients due to have a first time elective CABG procedure, aged 40 -80 years of age, with stable angina pectoris -Canadian Cardiovascular Score (CCS) ii, iii, or iv) or grade ii -iv moderate to severe coronary artery disease, confirmed by coronary angiography as greater than 70% stenosis or 50%, if left main stem disease. Spouses, partners and close family members (hereafter referred to as family caregivers) were invited to participate in the study providing they lived in the same household as the patient and were identified by them as their primary carer. Patients were excluded if they were having emergency surgery, and patients and caregivers excluded if there were any major co-morbidities such as stroke 6 or cancer, or psychological or communication limitations likely to affect their ability to consent.

Procedure
After we received approval from the University and local Research and Ethics Committees, patients and their family caregivers were recruited prior to their first visit to the surgical out-patient clinic. Study information and consent forms were posted out to the participants with the patient's clinic appointment card. Following receipt of the signed consent forms, questionnaire packets were distributed to the participants at the clinic visit, or mailed to their home address. Patients and caregivers were asked to complete the questionnaires separately from each other and to refrain from discussing their answers. Completed questionnaires were returned to the investigator by mail or at the clinic. A reminder letter was sent after 2 weeks.

Self-efficacy
Patients' and caregivers' perceptions of patient self-efficacy were assessed using the 16 item Cardiac Self-Efficacy scale, 35 containing two sub-domains: self-efficacy for controlling symptoms (SE-CS) and self-efficacy for maintaining functioning (SE-MF). All items are rated on a five-point Likert scale ranging from 0 (not at all confident) to 4 (completely confident).
The scores for SE-CS range from 0 to 32 and the scores for SE-MF range from 0 to 20, with higher scores indicating greater self-efficacy. The scale measures patient's belief in their 7 ability to perform certain behaviour rather than the actual measure of a given behaviour. In this study, the introduction of the scale was modified to fit the context relevant to caregivers. The validity and reliability of the Cardiac Self-Efficacy scale has been established in research. [26][27][28][29]35 No studies were found that had used the scale with caregivers. In this study, the Cronbach alpha for SE-CS was 0.75 for patients and 0.74 for caregivers; SE-MF was 0.79 for patients and 0.76 for caregivers.

Quality of Life
Patients' and caregivers' own quality of life was assessed using the Medical Outcomes Short- converted to t-scores and standardised against UK population data. Totalled scores ranged from 0 to 100, with higher scores indicating better physical or mental health. The psychometric properties of the SF-12 have been well established in research. [37][38] In this study, the Cronbach alpha for the physical component score was 0.77 for patients and 0.72 for caregivers; the mental component score was 0.78 for patients and 0.78 for caregivers.

Sociodemographics and clinical characteristics
8 Sociodemographics and past medical history were collected in brief separate interviews with the participants, using a structured questionnaire. Occupation was identified in accordance with the Office of National Statistics. 39 Social deprivation was identified using an index which takes account of income, residential postcode etc. 40 Categories range from 1 (most affluent) to 7 (most deprived). Clinical characteristics were identified from the patient's clinical records.

Data analysis
Sociodemographics, self-efficacy and quality of life were compared using the paired sample t test, or chi-square statistics. Pearson's product moment correlations were used to identify associations among continuous variables. Multilevel dyadic modelling i.e. the actor-partner interdependence model (APIM) regression for distinguishable dyads was used, based on interdependence theory. [32][33][34] In this study, the actor effect measured the impact of patient self-efficacy on his or her own quality of life; and the impact of caregivers' perception of patient self-efficacy on his or her own quality of life. The partner effect examined the impact between each person's perceptions of patient self-efficacy on his or her partner's quality of life.
For the dyadic analysis, all data were restructured to a pairwise dyadic data set. Grandmean centred scores were created that were standardised using z scores to obtain unstandardised and standardised regression coefficients for the actor and partner effects.
The residual structure was treated as heterogeneous compound symmetry. 32 Four separate 9 APIM models were computed; physical health was regressed on SE-MF; mental health was regressed on SE-MF; physical health was regressed on SE-CS; and mental health was regressed on SE-CS. All analyses were performed using SPSS version 21.0 for Windows, with p < 0.05 indicating statistical significance. A power calculation was not performed as this was a secondary analysis of data. The data came from a study of 84 patients having CABG and their caregivers. 13 In this analysis, we used multilevel dyadic modelling i.e. the APIM to evaluate perceptions of self-efficacy on the quality of life of patients and family caregivers.
Previous research using the APIM has shown that 40 dyads was sufficient for conducting the dyadic analysis. 23 Given our actual sample of 84 patients and caregivers is larger we hope to achieve at least the same power.

Characteristics of the participants
A total of 84 patient-caregiver dyads participated in the study (Table 1). There were 79 patient-spouse or partner pairs and five patient-family pairs. Most patients were male (85%) aged 64.5 years (SD 9.22). Most caregivers were female (87%) aged 61.0 years (SD 10.80).
Additional information on the participants' characteristics is shown in Table 1.

Differences for perceptions of self-efficacy and quality of life
Patients' SE-CS was low and caregivers perceptions of patient's SE-CS was similarly low (p =0.164) ( Table 1). Patients' SE-MF and caregivers' perceptions of patient SE-MF were particularly low; there was a significant difference between them for perceptions of SE-MF (t = 2.51, p = 0.014), but not for SE-CS (t = 1.40, p = 0.164) ( Table 1).
In order to further examine differences between patients' and caregivers' perceptions of patient self-efficacy new variables were computed for each patient and caregiver dyad, by subtracting the caregiver score from the patient score. Based on qualitative observations of scores being the same, higher or lower, patient-caregiver dyad members with the same score (i.e. no difference in self-efficacy) were coded as 0; one person (i.e. the caregiver) in the dyad with a higher score in self-efficacy than the patient was coded as 1; and one person (i.e. the patient) in the dyad with a higher score in self-efficacy than the caregiver was coded as 2. Forty-three patients (51%) had higher scores for SE-CS than the caregivers; 33 caregivers (40%) had higher scores for SE-CS than the patients; and 8 patient-caregivers (9%) had the same score. Thirty-nine caregivers (46%) had higher scores for SE-MF than the patients; 25 patients (30%) had higher scores for SE-MF than the caregivers; and 20 patientcaregivers (24%) had the same score.
The patients' physical health was particularly poor pre-operatively, and poorer still compared to the caregivers (t = 7.48, p < 0.001) ( Table 1). The patients' and caregivers' scores for mental health were similarly low (t = 1.10, p = 0.275).

Correlations between ratings of self-efficacy and quality of life
Both patients' and caregivers' ratings for patient SE-MF were positively weakly correlated with the patients' physical health (r = 0.39, p <0.001 and r = 0.29, p = 0.007, respectively) ( Table 2). In addition, caregivers' ratings for patient SE-MF were weakly positively correlated with their own mental health (r = 0.23, p = 0.005). There were moderate to strong positive correlations for patients' and caregivers' perceptions of patient SE-CS and SE-MF. There were significant correlations between patients' physical health and mental health; and between patients' mental health and caregivers' physical and mental health; and between caregivers' physical and mental health ( Table 2).

Self-efficacy and quality of life in dyadic relationships
Patients' SE-MF exhibited an actor effect on their mental health (Table 3, Figure 1). Figure 1 shows the actor effect of the patient's SE-MF on his or her own mental health. Patients with higher SE-MF had better mental health. There was no partner effect of the patient's SE-MF on the caregiver's mental health. (Table 3) Thus, patients' SE-MF did not impact the caregiver's mental health. With respect to caregiver's perception of patient SE-MF, there was no actor effect on their own mental health, or partner effect on the patient's mental health (Table 3, Figure 1). Thus, caregiver' perception of patient SE-MF did not impact their own, or the patients' mental health. There were no actor effects or partner effects found for patients' and caregivers' SE-MF on their own, or their partner's physical health (Table 3), Also,, there were no actor effects or partner effects found for patients' and caregivers' SE-CS on their own, or their partner's physical or mental health (Table 3),

Discussion
This study was unique in that it compared patients' and caregivers' perceptions of patient self-efficacy and quality of life before CABG. It also examined interdependence between patients' and caregivers' in their perceptions of patient self-efficacy. Patients' SE-MF was particularly low which may be linked to their poorer physical health before CABG. 3,41 Previous research has shown that patients' low self-efficacy is related to increased symptom burden, impaired physical function and poorer quality of life, independent of disease severity and depression. 35 Evidence from the Heart and Soul Study showed that patients with stable coronary artery disease have low SE-MF. 7 Our patients awaiting CABG had lower scores for SE-MF compared to previous research. 7,42 In this study, our patients also reported low SE-CS which may be related to symptom burden and poor mental health. It is possible though that the patients' poorer mental health came first and contributed to their low self-efficacy. 8 Our results indicate there were some similarities and differences between the patients and caregivers in their perceptions of patient self-efficacy, based on our qualitative observation of scores being higher or lower. Only 9% of patient-caregiver dyads had the same scores for SE-CS, although more patient and caregiver dyads (24%) had the same scores for SE-MF.
Notably, 46% of caregivers' rated patient SE-MF higher than the patients themselves, indicating some over-optimism on the part of the caregiver which could have a detrimental 13 effect on the patient. 14 In contrast, 51% of patients scored higher for SE-CS than the caregivers, suggesting some underestimation of the patient's capacity to self-manage. Our findings are consistent with other studies that have found patient and caregiver incongruence. 21 Such incongruence may cause conflict and distress in relation to self-care and advance care planning. 21 Our findings reiterate the significance of considering both patients and caregivers perspectives ,which is especially important in the education and preparation of patients awaiting CABG.

Further, our results indicate that both patients' and caregivers' perceptions of patient SE-MF
were significantly positively correlated with the patients' physical health. Previous longitudinal research has shown that spousal confidence in the patient's ability to perform specified behaviours is related to patient outcomes. 15,24 The caregivers' ratings for patient SE-MF were correlated with their own mental health. No dyadic studies of patients awaiting CABG were found for comparison of our results. Previous studies of self-efficacy have mainly focused on its role in cardiac rehabilitation, [42][43] or after myocardial infarction, 8 or coronary revascularization. 35,44 In caregivers, studies of self-efficacy or caregivers' confidence in their partner ( i.e. the patient) have rarely been examined. 14-16 The importance of patient and caregiver dyads in heart failure has been given much more attention, 21,24,45 and there have been studies of heart failure dyads using the APIM, which have identified actor-partner effects of self care and depression and anxiety on quality of life. [19][20]23 Our study was novel in that it used the APIM as a way of examining the dyadic effect of selfefficacy on patient and family caregiver quality of life in CABG. The results revealed an actor effect of patients' SE-MF on his or her own mental health but not the caregivers' mental health. This indicates that self-efficacy was based more on the 'self' than on the dyad, which is consistent with Bandura's proposal that personal information has the most potential to impact self-efficacy beliefs. 12 Other studies have found that patients and caregivers influenced one another's mental and physical health, but not their self-efficacy. 18 Our finding of an actor effect of patients' SE-MF on their mental health is consistent with previous research that has identified patient self-efficacy is significantly related to their mental health. 14 It was an interesting finding that patients' SE-MF and their physical health were significantly correlated in simple correlation, but yet there were no actor or partner effects. Other studies have found positive correlations between self-efficacy and physical health albeit post-operatively, and the APIM was not used. 43 This may be explained by the fact that in this type of analysis the researcher is examining associations controlling for both partner and role, so it is possible for a non-significant simple correlation to be a significant regression coefficient. To our knowledge, this is the first study to examine cross-sectionally pre-operative cardiac self-efficacy and quality of life in patients and caregivers at the dyadic level. Further research using the APIM is needed which may lead to a better understanding of the interaction in dyad members. The aim would be to work with the dyad to build selfefficacy and optimise the patient's physical and mental health and functioning before surgery.

Limitations
There were limitations to this study. First, it was a secondary data analysis using crosssectional data which meant that the direction of causality of associations could not be determined. Second, the study sample was relatively small which limits the generalizability of the findings. This makes it difficult to know whether our null results i.e. no partner effects indicate unimportant dyadic relations or insufficient power. Further study is needed to support or refute our findings. Third, length of marriage or cohabitation and marital quality of the respondents was not known.

Conclusions
Patients' SE-MF was particularly low pre-operatively which may be related to perceptions of impaired physical function and poorer quality of life. Differences between patients' and caregivers' perceptions of patient SE-MF should be addressed before surgery to help promote patient functioning. Whilst the patients' SE-MF predicted with their own quality of life using the APIM, there was no dyadic effect. Further research is needed in this area.

What's New and Important:
 Patients' self-efficacy for maintaining function was particularly low before coronary artery bypass grafting (CABG), which may be linked to their impaired physical function and perceived quality of life. Use of a quality of life measure and Cardiac Self-Efficacy scale may be useful as part of pre-operative assessment.
 Differences between patient and caregiver dyads in their perceptions of patient self efficacy may lead to caregivers underestimating the patient's capacity to self manage. Addressing these differences is especially important in the education and preparation of patients awaiting CABG.
 Patients' self-efficacy for maintaining function impacted on their own mental health, but not the caregiver's mental health. There were no other actor effects or partner effects of self-efficacy on quality of life. More dyadic research is needed in this area.