Guilt, shame and their associations with chronic diseases in Czech adults

ABSTRACT The interest in the association between feelings of guilt and shame and psychological health is growing. However, less attention so far has been paid to the associations with chronic disease. Therefore, the aim of this study was to explore whether feelings of guilt and shame are related to physical health in chronic disease and in selected ones. A nationally representative sample of Czech adults (n = 1000, age = 46.0 ± 17.3 years, 48.6% men) participated in the survey. Feelings of guilt and shame (items from The Positive and Negative Affect Schedule) and health problems in selected chronic diseases were measured. Respondents with higher feelings of guilt – but not of shame – were more likely to suffer from a chronic disease; we found this in arthritis, back pain, cardiovascular disease, asthma, cancer and depression or anxiety. The association was strongest in the case of cancer with odds ratios (OR) 5.83 (95% confidence interval (CI) 2.27–16.69). There were no significant associations in case of diabetes and stroke. Feelings of shame were not related to chronic diseases. Our findings suggest that feelings of guilt are associated with worse physical health. Further research is needed in this area.


Introduction
Both shame and guilt are connected with the experience of failure. Although they are often confused in common language, they cannot be combined into one homogeneous emotion. Quantitative (Cacciatore et al., 2013) as well as qualitative (Gibson, 2016;Tangney et al., 1996) studies show that guilt and shame are two separate emotions. They differ in their dynamics, context, and accompanying emotions. These two emotions also vary with interindividually different evoked behaviors in the situations and the character of the experience itself. Tangney et al. (1996) showed that interindividual differences can be explained more by the judgments and feelings connected with the experience and the subsequent behavior that provokes the situation than by the number of witnesses present, which contrasted the earlier convictions that shame is a more public emotion, whereas guilt is a more private affair (Ausubel, 1955).
Shame is associated with the desire to undo aspects of the self (Niedenthal et al., 1994). According to Velotti et al. (2017), frequent shame experiences can develop into the form of a tendency to feel shame in general. However, the concept of guilt as an urge to change some aspects of behavior, was related to a superego response to unacceptable impulses (Eisenberg, 2000). In this approach, guilt became a weaker contributor to psychopathology. The image of guilt is also considered as an adaptive emotion implying prosocial behavior (Tangney, 1995) and strengthening the sense of control in the world (Lindsayhartz, 1984). There are many studies focusing on relationships of guilt with poorer mental health, and some studies focusing on its relationships with physical health.
Many studies report an association of guilt and shame with a worse mental health especially in some chronic diseases (Crosskey et al., 2015;Dearing et al., 2005;Exline et al., 2000;Lee et al., 2001), e.g., in depression (Andrews et al., 2002;Bennett et al., 2010;Oconnor et al., 1997). Other studies have documented the relationship between shame with many other phenomena: low self-esteem, aggressiveness and hostility (Baumeister et al., 1994;Bennett et al., 2010;Garofalo et al., 2016;Orth et al., 2010). Personality shameand guilt-proneness had positive correlations with grief and psychological adaptation after loss among bereaved mothers (Barr, 2012;Barr & Cacciatore, 2007). Shame and guilt may exacerbate and perpetuate symptoms of posttraumatic stress disorder (Lee et al., 2001;Leskela et al., 2002). Shame-proneness was positively correlated with substance use problems, while guilt-proneness was inversely related (Dearing et al., 2005). Lower tendency to experience shame was significantly associated with more satisfaction and with less emotional exhaustion that could lead to burnout (Barnard & Curry, 2012;Crosskey et al., 2015).
Less attention has been paid to the potential impact of these individual emotions on physical health. However, there are some exceptions in this respect, e.g., studies describing the association of feelings of guilt related to punishing reactions from significant others with pain complaints among patients with chronic spinal cord injury pain (Conant, 1998). Furthermore, guilt was positively associated with the uncertainty about the diagnosis and disability among patients with low back pain (Serbic & Pincus, 2014;Serbic et al., 2016). Guilt and shame were also positively related to how the patients with hypercholesterolemia manage their condition (Frich et al., 2007) and to the belief that one caused one's own cancer (Abrams & Finesinger, 1953;LoConte et al., 2008). Nevertheless, some studies have not found any associations, e.g., Ten Klooster et al. (2014) found no significant relationship between guilt, shame and the severity of rheumatoid arthritis.
The inconsistency of these findings shows the need of further research in this area, especially with regards to physical health. Therefore, the aim of this study is to assess the associations of guilt and shame with self-rated health in chronic disease and in selected ones.

Participants and procedure
We obtained a national sample of the Czech population aged fifteen years and older using a two-step procedure. In the first step, the questionnaire and all further procedures were piloted among 109 participants. This led to the final version of the survey. A sample for the main analysis was derived using a Power analysis, which indicated that with 80% power, a sample size of 1000 respondents would be large enough to detect small effect sizes (OR = 1.43; OR = 0.70). In the second step, another 1215 participants were randomly chosen with the help of quota sampling; and asked to participate in a study on the problematics of health, life experiences and attitudes and lifestyle. Of these respondents, 215 (17.7%) refused to participate in the survey. Among the main reasons for refusal, participants reported a lack of time (45.6%), a lack of interest or reluctance to fill the survey (22.1%), the length and difficulty of the survey (19.1%), 8.8% declined to participate for considering the questions unnecessary or did not participate in research at all. Finally, 3.4% of respondents expressed concerns about misuse of the provided answers or stated other reasons. Data was collected by professionally trained administrators November and December 2014 by a standardized interview with the respondents (face-to-face). Participation in the survey was anonymous and voluntary. The final sample consisting of 1000 respondents; a representative sample of the Czech population aged fifteen years and over with regards to age (mean age 46, SD = 17.3), sex (48.6% men), education (primary 8.4%, secondary 76.1% and university 15.5%) and regional affiliation.
According to the Czech law by the time of the study a protocol needed approval of an Ethics Committee only if medicines were used; we carefully followed the latest version of the Declaration of Helsinki.

Measures
The Positive and Negative Affect Schedule (PANAS) is a 20-items scale comprising two mood scales, the Positive and the Negative Affect. Ten descriptors are used for each scale. The whole scale consists of 20-items. Respondents were asked to report the extent of feeling different emotions during the past week using a 5-point scale ranging from 'very slightly' or 'not at all' (1) to 'extremely' (5). For the purpose of this study, we chose the items 'guilty' and 'ashamed'. For each item, participants who chose the option 'not at all' or 'very slightly' were considered as not experiencing shame and guilt or experiencing them only to a very limited extent, while the rest as experiencing these feelings often. The items were translated from English to Czech independently by two native speakers who agreed on the translations used.
Responses to the item that evaluates self-rated health were also dichotomized; responses 'excellent', 'very good' and 'good' were considered as perceiving health as subjectively good, the options 'quite good' and 'bad' as poor.
Presence of a chronic disease was measured by the question: "Do you have a longstanding illness or disability? Please mark everything that applies to you", with following possibilities: arthritis, back pain, heart problems, asthma, cancer, depression or anxiety, diabetes, cerebrovascular accident/stroke, or none of those.
The background characteristics gender, age and other basic socio-demographic characteristics were obtained by the questionnaire.

Statistical analyses
First, we described the background characteristics of the sample and compared different sociodemographical groups using the Kruskal-Wallis test. Then, we assessed the associations of guilt, shame and any chronic disease; next between guilt, shame and eight kinds of chronic diseases using a multinomial logistic regression model adjusted for gender, age and education. Each of the independent variables was assessed in a separate model. All analyses were performed using the statistical software package IBM SPSS version 21.

Results
The description of the study population is shown in Table 1. Of the whole sample, 57.9% respondents reported the presence of some of the diseases (see Table 2). Furthermore, 17.7% reported feelings of guilt and 16.8% feelings of shame.
A non-parametric comparison between different sociodemographic groups showed no statistically significant differences (results not shown). Table 2 shows the prevalence of all chronic diseases and the selected ones in the sample and the prevalence of feelings of guilt and shame within the groups of respondents with chronic diseases.
Results of binary logistic regression adjusted for gender, age and education showed that respondents with higher feelings of guilt were two times more likely to report some chronic disease with odds ratio (OR) 2.0; 95%-confidence interval (CI) 1.37-2.92, p < 0.001). Furthermore, results of the multinomial regression adjusted for gender, age and education, as presented in Table 3, showed that respondents with higher feelings of guilt were also more likely to suffer from arthritis, asthma, back pain, cancer, cardiovascular disease, depression. The association was the strongest in the case of cancer with OR 5.83; 95%CI (2.17-15.69, p < 0.001). On the contrary, there were no significant associations of the feelings of shame with individual health problems.

Discussion
The aim of this study was to assess the associations of feelings of guilt, shame and physical health in chronic disease and in selected ones. We found no significant differences between the sociodemographic groups regarding the prevalence of feelings of guilt Note ᵃ Only the respondents with the occurrence of feelings of guilt or shame are presented. b Percent of the total number of people reporting increased feelings of shame and guilt c Participants may have more than one chronic disease Table 3. Associations of physical health with guilt and shame in any chronic disease and in selected chronic diseases: results of multinomial logistic regressions with the healthy respondents being the reference category, adjusted for age, gender and education, leading to odds ratios (OR) with 95% confidence intervals (95%CI). Notes: *p < 0.05, **p < 0.01, ***p < 0.001 and shame. However, we found strong associations of feelings of guilt and of self-rated health with having a chronic disease and with several selected ones (arthritis, asthma, back pain, cancer, cardiovascular disease, depression). Feelings of shame were not related to chronic diseases.
We found that higher feelings of guilt were associated with chronic disease and with six of the eight observed chronic diseases. The relationship that we found between guilt and several chronic diseases could be interpreted as an outcome of the influence of selfconscious emotions on the system supposed to regulate them. According to Dickerson et al. (2004) self-related emotions can cause changes in biological processes underlying psychological ones. A higher level of guilt was found to be associated with a significant increase of cortisol level among respondents exhibited to a social-evaluative situations  and (together with anger and pride) with cardiovascular reactivity and peripheral resistance (Herrald & Tomaka, 2002). The guilt-prone individuals might be more likely to experience higher productivity of the part of endocrine glands system, which may result in the form of increased susceptibility to inflammatory processes (Dickerson et al., 2004). They are involved in the development of asthma (Rosenkranz et al., 2005;Runeson et al., 2011) or atherosclerosis (Ridker et al., 1997;Ross, 1999).
On the other hand, our results need to be interpreted with caution as the crosssectional nature of the study does not allow us to conclude on causality. Serbic and Pincus (2014) reported a higher level of guilt among patients with chronic low-back pain due to pain intensity and disability, hence guilt could be seen as a consequence of suffering of an illness. Similarly, asthma, arthritis, cardiovascular diseases, depression and especially cancer (with its treatment and aftermath) may be limiting in full experiencing of life, so the patients may experience guilt, anger and frustration (e.g., (Conant, 1998;Else-Quest et al., 2009). The feelings of guilt might be also associated with the awareness of one's lifestyle and not following medical advice and/or violating the principles of healthy lifestyle (LoConte et al., 2008). The relationship of different lifestyles to various types of cancer is well documented (Baan et al., 2007;Center et al., 2009;Garcia-Alvarez et al., 2007), as well as to cardiovascular diseases (Gorostegi-Anduaga et al., 2017), back pain (Hoy et al., 2010) and other chronic diseases (Ross, 1999;Ten Klooster et al., 2014).
We also observed a higher prevalence of guilt in respondents with depression, which could be considered a specific case, because feelings of excessive or inappropriate guilt is one of the specific symptoms of major depressive disorder according to DMS-IV (Diagnostic and Statistical Manual of Mental Disorders). The relationship between depression and guilt is demonstrated also by O'Connor et al. (O'Connor et al., 2002). At the same time, shame-and guilt-proneness are by some authors (Young et al., 2016) considered as contributors to the development of depressive symptoms.
However, though we have found the association of guilt with physical health in a number of chronic diseases, we have not found any no association of shame with physical health in chronic diseases. This finding contradicts the frequent claim of the greater destructive influence of shame than guilt (Boudewyns et al., 2013;Tangney et al., 1996). The difference of our findings may indicate that our respondents did not manage to empathize to the state based on the word 'ashamed', or that they misunderstood the concept of shame, which is not in the Czech language so commonly used as the term 'guilt'.

Strenghts and limitations
This study has several important strengths, the most important being the large and representative sample with no missing values. A limitation is the cross-sectional design of the study, which does not allow to deduce the direction of causality. Some categories of diseases also had very few respondents (e.g., cancer), which could influence the statistical analyses, although the same outcome was found in chronic disease. Another limitation is the wording of the questions using the term 'shame', which could had a negative impact on the understanding of the statement. Our method of data collection (a face-to-face interview) could also possibly influence answers of the respondents in a socially desirable way. On the other hand, respondents were informed about maintaining anonymity and the trained administrators were neutral, unknown persons.

Implications
Our findings show that the feelings of guilt are associated with physical health in chronic disease while feelings of shame are not. Guilt can greatly influence the quality of life of people with a chronic disease, this is the direction most studied in the literature. However, it can also play a direct role in the development of the chronic disease and should therefore be considered an important topic in the clinical area. Addressing these negative emotions could represent an important aspect of a psychotherapeutic and also clinician's care when assessing and treating patients with chronic diseases. Thus, keeping in mind the importance of a possible role of mental phenomena in physical health is essential, regardless of the direction of a causality, and a cooperation between mental and physical health professionals should be developed and deepened. Clinicians and other healthcare workers should be adequately trained to recognize signs of psychological distress and should closely cooperate with psychologists and psychotherapists in order to address exaggerated feelings of guilt and shame. Next, healthcare providers should be encouraged to create a safe environment for their patients. The strength of the associations implies the importance of the topic and the need of a more detailed research in this area. Further research should assess these phenomena with the use of a more detailed questionnaire and focus on unravelling the causality.

Conclusions
The present study is one of the few that explores the relationship between guilt/shame and physical health in chronic disease and in selected ones in a representative sample. We found a positive association between guilt and in chronic disease and in six of the eight selected ones, but not between shame and chronic disease and the selected ones.

Disclosure statement
No potential conflict of interest was reported by the authors.

Funding
This work was supported by the Grant Agency of the Czech Republic [19-19526S].