Selection of social comparison standards in cardiac patients with and without experienced defibrillator shock

Patients with an implantable cardioverter-defibrillator (ICD) often report psychological distress. Literature suggests that patients with physical disease often compare their well-being and coping to fellow patients. However, we lack knowledge on social comparison among patients with ICD. In this study, we examined psychological distress and social comparison selection in patients with (ICD+) and without experienced ICD shocks (ICD−). We theorized that relative to ICD− patients, those with ICD+ display higher levels of psychological distress and thereby compare more frequently with fellow patients with more severe disease, but better disease coping and try to identify more strongly with these standards to improve their own coping. We recruited 92 patients with (ICD+, n = 38) and without an experienced ICD shock (ICD−, n = 54), who selected one of four comparison standards varying in disease severity and coping capacity. Relative to ICD−, ICD+ patients reported higher levels of device-related distress, but there were no significant differences in anxiety, depression, or quality of life. ICD+ patients selected more often comparison standards with poor coping and, irrespective of standard choice, displayed more negative mood following comparison. Our results show that ICD+ patients tend to perform unfavorable comparisons to fellow patients, which might explain higher psychological distress and worse coping. These findings warrant further research into social comparison as a relevant coping mechanism in ICD patients.


Comparison standard selection
The case vignettes presented to the patients were fictitious and in analogy to studies by Arigo et al. 25 and Corcoran et al. 26 .There were four different vignettes to select reflecting two different levels of disease severity (mild, severe) and two types of coping (good, bad).Participants were asked to select one of the four vignettes they would like to read, based on brief descriptions.In particular the vignettes represented respectively a patient with mild disease and good coping (i.e., representing a comparison standard that is upward in terms of illness and upward in terms of coping, abbreviated UIUC standard), a patient with mild disease and bad coping (i.e.upward illness/ downward coping, UIDC standard), a patient with severe disease and good coping (downward illness/upward coping, DIUC standard) and a patient with severe disease and bad coping (downward illness/downward coping, DIDC standard).The four brief descriptions were presented in a multiple-choice format on a single screen, requiring participants to select only one of them.The order of the options was randomized for each participant.

Comparison with the selected standard
Following standard selection, participants read the respective vignette and were asked to provide self-ratings relative to the standard (relief, anxiety, happiness about personal health, fear of deterioration, coping, focusing on similarity/differences) on a six-point Likert scale (1 = not at all; 6 = very much) using 7 items in total.The focus on similarities and the contrast between the standard and themselves was assessed by analogy with Arigo et al. 25 .

Current mood and subjective wellbeing
Current mood was assessed by a self-assessment manikin (SAM; Bradley and Lang 29 ; Sierra et al. 30 ) 9-point scale (1 = very bad; 9 = very good) at the beginning of the survey as well as after the case vignette with the selection of standard.Current mood SAM scores were lost for seven participants (3 in the ICD+ group) due to experimenter error.These cases were omitted from the analyses involving current mood but were retained in all other analyses.

Depression and anxiety levels
We assessed and quantified depressive symptoms with the Patient Health Questionnaire (PHQ-9) 31 and symptoms of anxiety with the Generalized Anxiety Disorder Scale (GAD-7) 32 .The items ask for the occurrence of the respective symptoms in the last two weeks.The PHQ-9 uses 9 items and the GAD-7 uses 7 items, all of which are scored on a 4-point Likert scale based on the frequency of symptoms.For both questionnaires good psychometric properties have been reported [31][32][33][34][35] .

Quality of life
The German version of the Short Form Health Survey (SF-12) questionnaire was used to assess the overall quality of life 36,37 and has been extensively and used in German speaking as well as international populations. 36,38The SF-12 makes use of 12 items, with 6 items using a 5-point Likert-scale, 2 items with a 3-point Likert scale and 4 dichotomous items.The SF-12 comprises two component scores, the Physical Component Summary (PCS) and the Mental Component Summary (MCS) with higher scores indicating a higher quality of life in the respective area.The scores range from 0 to 100 and are designed to have a mean score of 50 and a standard deviation of www.nature.com/scientificreports/ 10 for a representative US population.Scores > 50 therefore lie above the average health status while scores < 50 lie below 36,38 .

Florida patients acceptance survey (FPAS)
The modified FPAS 39 was used to assess patient acceptance of the ICD implanted.The instrument uses 15 items and comprises four underlying latent variables.These are Return to Function (RTF), Device-Related Distress (DRD), Positive Appraisal (PA), and Body Image Concerns (BIC).All the 15 items are answered on a Likertscale reaching from 1 (strongly disagree) to 5 (strongly agree).Subscales for each of the factors as well as a total sum score were calculated, with higher scores indicating higher levels of acceptance.An exception is the DRD subscale, where scores are separately calculated to indicate higher levels of distress.The scale has been evaluated extensively 40,41 .

Coping strategies and reaction to stress exposition
Coping behavior was assessed with the Coping Strategies Inventory Short-Form (CSI-SF) 42 .Sixteen items (5-point Likert scale) are used to assess 4 different factors of coping strategies, namely problem-focused engagement (PFE), problem-focused disengagement (PFD), emotion-focused engagement (EFE) and emotion-focused disengagement (EFD).Two overarching scales, total engagement (E) and total disengagement (D) are derived by summing the respective subscales (point range 4-20 for each subscale).The CSI-SF has been validated in various large cohorts in different languages, including German 43 .

Frequency of upwards and downward comparisons
The frequency of different types of appetitive and aversive comparisons in relation to well-being and the engendered affect were assessed with the Comparison Standards Scale-Well-being (CSS-W) 22 .The CSS-W uses 14 items to assess comparative thoughts in relation to the current well-being over the last three weeks by asking about the frequency of comparisons to different standards on a 6-point Likert scale (ranging from 0 to 5).The assessed types of comparison are social, temporal, counterfactual, and criteria-based.Seven items address appetitive comparisons and seven items address aversive comparisons.The two subscales are summarized separately using sum scores, with higher scores indicating more frequent comparison.Initial results suggest the CSS-W to be a valid assessment tool that relates significantly to mental health complaints 22 .

Procedure
Participants were asked for their participation in the study during a routine visit at our outpatient ICD clinic.
After receiving information about course and aim of the study and giving informed consent, participants completed all measures (for an overview, see Fig. 1).Participants were instructed to select one of four case vignettes they were most interested in reading.Afterwards they filled out the questionnaires in the order mentioned above and were thanked and debriefed.They did not receive compensation for completing the study.For overview of the order of measures see Fig. 1.In brief, participants were presented the SAM, CSS-W, PHQ-9 and GAD-7 before choosing the respective case vignette.Afterwards, they were presented with SAM, question regarding identification and contrast with the presented vignette, CSI-SF, FPAS and SF-12.All questionnaires were administered in pen-and paper format.The results of the questionnaires were complemented with medical data of the patients obtained from the patient data management system of our clinic.

Statistical analyses
Between-group effects, mainly focusing on mean or proportion differences between ICD+ and ICD− (cf.H1-H4), were analyzed using independent-samples t-tests and analyses of (co-)variance (ANOVA) to be complemented with post-hoc t-tests in case of significant interactions, and chi-squared tests.When the assumption of equality of variances was violated (i.e., significant Levene's test for equality of variances

Habitual comparison frequency
There was no difference in comparison frequency between the two patient groups.While ICD+ patients numerically selected aversive (i.e., mostly upward) comparisons more often (M = 11.9,SD = 6.9) than ICD− patients (M = 10.7,SD = 7.9), the difference was not significant, t (90) = − 0.780, p = 0.437, d = − 0.165.Concerning the choice of appetitive comparison both groups did also not differ significantly (ICD+: M = 8.9, SD = 5.5; ICD−: M = 8.3, SD = 5.9), t (90) = -0.529,p = 0.598, d = − 0.112).In terms of the absolute number of patients, 97.4% of ICD+ patients reported at least one type of aversive comparison, whereas 94.7% reported at least one type of appetitive comparison.Similarly, 88.9% of the ICD− patients reported at least one type of aversive comparison and the same percentage of patients reported at least one type of appetitive comparison.
With respect to social comparison frequency only, 76.3% of the ICD+ patients reported at least one aversive comparison and 81.6% of the ICD+ patients reported at least one appetitive comparison.On the other hand, 61.1% of the ICD− patients reported at least one aversive comparison and 66.7% of the ICD− patients reported at least one appetitive comparison.Overall, the mean of aversive (i.e., upward) social comparisons in ICD+ patients was somewhat higher (M = 3.1, SD = 2.7) than in ICD− patients (M = 2.6, SD = 2.7), yet, the difference was not significant, t (90) = − 0.790, p = 0.432, d = − 0.167.Likewise, the mean of appetitive social comparisons in ICD+ patients (M = 3.4,SD = 2.7) was not significantly higher than in ICD− patients (M = 3.2, SD = 2.9; t (90) = − 0.424, p = 0.673, d = − 0.090).

Standard selection
Chi-squared test on the proportion of ICD+ and ICD− patients selecting the four comparison standards (severity: high/low; coping: good/bad) revealed a significant group difference (χ 2 = 9.973, df = 3, p = 0.019).The respective choice frequencies are displayed in Table 2.

Mood following standard selection
To analyze the impact of comparison standard choice, we treated the chosen levels of severity and coping as between-subjects in the following analyses.Note that severity and coping were analyzed separately because cell sizes were too small to enter both factors into the same model (see Tables 2, 3, 4).

Discussion
The current study aimed to explore psychological well-being and coping behavior as well as antecedents and consequences of social comparison behavior among ICD patients with and without prior ICD shock experience.The main results are as follows.First, there were no significant differences between the two patient groups concerning anxiety, depression and quality of life.Similarly, there were no significant differences in coping behavior and habitual comparison frequencies.However, ICD+ patients had lower device acceptance due to higher device-related distress (cf.H1).Second, ICD+ patients did not differ from ICD− patients concerning the selected standard in terms of disease severity but only in terms of disease coping.That is, although ICD+ patients Table 2. Crosstable for the relation of prior experience of ICD shock and comparison standard selected.The difference between the selected standard in both ICD shock groups was significant (χ 2 = 9.973, df = 3, p = .019).ICD, implantable cardioverter-defibrillator; UI, upward illness; DI, downward illness; UC, upward coping; DC, downward coping.Table 4. Crosstable for the relation of prior experience of ICD shock and the coping performance presented in the standard selected.The difference between the selected standard in terms of coping performance in both ICD shock groups was significant (χ 2 = 7.922, df = 1, p = .005).ICD, implantable cardioverter-defibrillator.www.nature.com/scientificreports/chose a severe course more often and ICD− patients chose a mild course more often (cf.H2).Meanwhile, ICD+ patients selected standards with bad coping more often than ICD− patients (H3).Finally, despite comparable mood levels at the beginning of the study, ICD+ patients displayed significantly worsened mood after presentation of the case vignette than ICD− patients, irrespective of their comparison choice (H4).Arigo et al. reported that patients with a more severe course of a type 2 diabetes had worse quality of life and higher levels of depression 44 .In comparison, we only found trends towards worse psychological health in ICD+ patients who also have a physically more severe disease in comparison to ICD− patients (H1).This might be influenced by the fact that the modifiability of the disease course is different in diabetes compared to ICD patients.Patients with diabetes can directly impact measurable surrogate parameters of disease severity such as HbA1c by their own behavior and therapy adherence.ICD patients, however, have less influence on the occurrence of arrhythmias and ICD shocks and might therefore experience less self-efficacy, which is known to be closely associated with depression and anxiety [45][46][47] .Self-efficacy has been shown to be a relevant mechanism in self-management and quality of life in different patient groups 48 .Thus, the role of perceived self-efficacy in ICD patients in well-being and coping behavior warrants further examination.
We found subtle differences in the choices of comparison standards.However, these were not entirely as predicted: The results did not support the prediction that ICD+ patients would select a standard with a higher disease severity more often (H2), although a trend in this direction was present in the data.Our hypothesis was founded on results from results from Arigo et al. 25 who reported patients with type 2 diabetes commonly selected a better-off standard and justified it by the need for more information about successful coping.The results are in line with the Identification/Contrast Model of comparison 15 that proposes that a downward comparison with greater identification leads to worse quality of life and less motivation, whereas identification with an upward standard increases quality of life and motivation.In our study, we found that ICD+ patients identified more strongly with the worse-off standards in line with results from Arigo et al. 44 and selected standards with bad coping more often.
Against our expectation that ICD+ patients would select a standard with good coping more frequently than ICD− patients (H3), we found evidence for the opposite pattern.Poor psychological health and low self-efficacy might therefore play a role in selection standards with poor rather than good coping.The role of the ICD shock status as a relevant feature differentiating between upward and downward comparing patients is illustrated by the interesting finding that there was no relevant influence of the patient history leading to ICD implantation (primary vs. secondary prevention, aborted SCD), but only for the course of disease after ICD implantation on psychological distress.This should draw further attention to shock status as a distinctive feature in ICD patients in future studies.Practical implications should therefore contain a candid screening for common mental health complaints in this population.For example, the Patient Health Questionnaire represents a time efficient screening self-report for common mental health disorders and the average time required of the physician to review the questions is less than two minutes 31,32 .Furthermore, patients reporting mental health complaints should be encouraged to seek professional mental health.Importantly, the first period after implantation as well as first treatment of an arrhythmia by the ICD are probably the most vulnerable phases for patients and therefore demand more attention from the medical personnel.
Concerning the patients' mood and the influence of the case vignette presented, our study results align with prior studies such with Corcoran et al. 26 , who reported a strong mood effect of the presented case vignette in women with breast cancer regardless of identification or contrast effects and regardless of the disease course.In our study, the influence of social comparison on mood was more negative in ICD+ than in ICD− patients, and thus, depended on disease severity (H4).This might possibly be explained by the fact, that ICD+ patients have already experienced arrhythmias and ICD shocks first-hand, whereas ICD− patients have not.Accordingly, a case vignette of a fellow patient with experienced arrhythmias and ICD shocks can provoke concrete negative memories only in the ICD+.Additional evidence from Arigo et al. 44 illustrated that the selection of the worse-off standards worsened mood and stress levels and reduced self-care motivation.In our study, ICD+ patients had more negative mood following comparison selection than ICD− patients, irrespective of the specific selected standard.This might be due to the heterogeneity regarding age, disease severity and awareness of the ICD as a life-saving device but also a potential source of pain and distress (Supplementary Information).

Limitations
Our study has some limitations worth to be mentioned.Comparison selection in a laboratory setting might not reflect comparison selection in real life, which further increased by the fact that the number of patients in our study was rather small, yet comparable to similar studies 25,26 .Furthermore, case vignettes varied in course of disease and coping success but did not mention information about prognosis or further medical measures exceeding the ICD performance and ICD perception.To exclude possible manipulation of identification or contrasting with the standard presented we left out information such as age, underlying heart disease and medication.However, identification or contrasting might be stronger in a more detailed description of the case vignette (e.g., younger patients reading about a young patients) 13 , which might be addressed in future studies.Furthermore, the generalizability might be hampered by the rather homogenic study population, in which most of the patients were of Caucasian origin with a high socioecomonic status.As there were for example no African-American, Latin American and Asian participants included, cultural influences and different role perceptions are not represented in the results of the study.Finally, replication in studies with larger samples is needed.

with experienced ICD shock (n = 38) Patients without ICD shock (n = 54) Difference statistic
Vol:.(1234567890) Scientific Reports | (2024) 14:5551 | https://doi.org/10.1038/s41598-024-51366-3 close to ± 1), as these typically do not inflate Type 1 error rate in ANOVA.Nonetheless, we complemented the respective analyses with non-parametric Mann-Whitney U-tests and Quade's tests.Since none of these tests yielded different results (in terms of significance levels), they are not reported below.Alpha was set at 0.05 (twotailed) for all analyses.All statistical tests were performed using IBM® SPSS Version 28.

Table 3 .
Crosstable for the relation of prior ICD shock experience and disease severity of the selected standard.The difference between the selected standard in terms of disease severity in both ICD shock groups was not statistically significant (χ 2 = 3.236, df = 1, p = .072).ICD, implantable cardioverter-defibrillator.