Illness Representations and Psychological Adjustment of Greek Cancer Couples (PDF)

J Behav Med (2016) 39:85–93 DOI 10.1007/s10865-015-9664-z Illness representations and psychological adjustment of Greek couples dealing with a recently-diagnosed cancer: dyadic, interaction and perception-dissimilarity effects Zoe Giannousi1 Evangelos C. Karademas2 Georgia Dimitraki2 Received:...

J Behav Med (2016) 39:85–93 DOI 10.1007/s10865-015-9664-z Illness representations and psychological adjustment of Greek couples dealing with a recently-diagnosed cancer: dyadic, interaction and perception-dissimilarity effects Zoe Giannousi1 Evangelos C. Karademas2 Georgia Dimitraki2 Received: January 16, 2015 / Accepted: July 30, 2015 / Published online: August 11, 2015 Ó Springer Science+Business Media New York 2015 Abstract The aim was to examine the impact of the Introduction dyadic, interaction and dissimilarity effects of the illness representations on the psychological health of recently In the experience of illness trajectory, the significant role of diagnosed cancer patients and spouses in Greece. The family and partner cannot be ignored (Rosland et al., sample consisted of 298 individuals nested in 149 couples. 2012). The impact of the illness experience on partners’ Effects were examined with the Actor–Partner Interde- health and quality of life should not also be overlooked pendence Model. Both actor (i.e., within person) and (Applebaum & Breitbart, 2013; Given et al., 2001). There partner (i.e., between partners) effects were detected for is mounting evidence that patients’ and their partners’ ill- both patients’ and spouses’ psychological symptoms. The ness-related self-regulation mechanisms form a bidirec- negative association of patients’ psychological symptoms tional system that affects both parts (Berg & Upchurch, with their representations of illness coherence was weak at 2007; Badr et al., 2010; Hagedoorn et al., 2008). Never- the higher and medium levels, and stronger at the lower theless, the relation between the ways that partners and levels of spouse corresponding representations. Patient– patients understand and react to illness, as well as its partner discrepancy in perceived illness consequences was impact on the well-being of both members of the couple, associated with more psychological symptoms in patients. only recently have become the focus of research. Adaptation to cancer is a dyadic process within the context The contribution of self-regulation and its main aspects of which patient and partner psychological well-being is (e.g., illness representations, action plans) in adaptation to affected by each other’s understanding of illness. Thus, the illness has been shown in a number of studies and a variety parallel examination of the illness representations of both of different illnesses, such as heart disease, cancer, chronic partners is needed from the early phases of the illness fatigue syndrome, psoriasis, and diabetes (Hagger & trajectory. Orbell, 2003; Martin & Leventhal, 2004). The majority of these studies have employed the Common Sense Self- Keywords Cancer
Oncology
Illness representations
regulation Model (CS-SRM), which suggests that patients Dyadic effects
Perception dissimilarity effects construct parallel cognitive and emotional representations of illness, in order to make sense and effectively deal with the disease (Leventhal et al., 1980, 1992). Regarding can- cer, there is evidence that several illness representations are related to the coping behaviors patients adopt, their quality of life and psychological adjustment (Millar et al., 2005; & Zoe Giannousi Rozema et al., 2009; Llewellyn et al., 2007; Giannousi [email protected] et al., 2010). 1 Bank of Cyprus Oncology Centre, 32, Akropoleos Av., However, adaptation to illness is not an isolated process. 2006 Nicosia, Cyprus According to the CS-SRM, the entire process is affected and 2 Department of Psychology, University of Crete, Heraklion, moderated by several factors, including the social context such Greece as the family (Leventhal et al., 1980, 2005). Caregivers/partners 123 86 J Behav Med (2016) 39:85–93 form their own illness representations which may serve as an cancer. To our knowledge, only one of these studies has important source of information for the patients and affect their focused on recently diagnosed patients and their partners self-regulation process and outcomes (Weinman et al., 2003). (Karademas & Giannousi, 2013). That study included a Several studies have examined this issue. rather small number of couples and examined only illness Initially, research focused on the differences between representations of control. Provided that initial reactions to patients’ and partners’ illness representations, and on their cancer diagnosis are crucial for subsequent adjustment possible impact on health. Overall, illness representations (Carver & Antoni, 2004; Hack & Degner, 2004), more dissimilarity was associated with worse psychological research is needed in order to better understand the pro- adaptation for both (Benyamini et al., 2007; Figueiras & cesses underlying the couple’s adaptation at early phases of Weinman, 2003; Heijmans et al., 1999; Twiddy et al., illness. In this regard, the aim of this study was to examine 2012). As far as cancer is concerned, there is evidence that the dyadic effects of a broad range of illness representa- patient–partner illness representation dissimilarity is rela- tions on psychological adjustment of recently diagnosed ted to higher levels of psychological distress for both (Merz Greek cancer patients and their spouses. et al., 2011; Romero et al., 2008). Additionally, interaction as well as representation dis- Yet, patient–partner dyads interact in a manner that similarity effects were examined as the two types of effects includes more aspects than only (dis)similarities in illness may convey different type of information. Dissimilarity representations. As Bodenmann (2005) suggested, the ‘in- scores reflect the degree to which patients and their part- dividual processes’ in each member of the couple are ners hold (dis)similar representations, as well as whether mutually affected and inter-connected in complex ways. this impacts well-being and adaptation to illness. Yet, Existing research showed that the ways each partner dissimilarity scores do not allow for the examination of the understands illness is important for both partners’ illness potential moderation effects of illness representations adaptation, while there is evidence of significant interactions between partners (i.e., whether the illness representations between patients’ and partners’ illness representations. of the one member of the couple can affect the relationship Fagundes et al. (2012) found, in a sample of prostate cancer between the illness representations and adaptation to illness patients-partners dyads, that when a member of the couple of the other member). This may be achieved through the reported higher levels of avoidance, then the other member corresponding interaction effects. Thus, the parallel experienced more negative affect irrespectively of own level examination of both types of effects is probably important of avoidance. Wu et al. (2013) showed in a prospective study as it may permit a more thorough understanding of the that higher levels of spouse beliefs about treatment control mechanisms that connect patients and partners. led to beliefs that illness may be shorter, which in turn led to Provided that illness representations and illness adap- improved general quality of life in prostate cancer patients. tation are strongly connected (Leventhal et al., 1980, Also, Dempster et al. (2011) in a study of oesophageal 1992), as well as that patients’ and partners’ illness adap- cancer survivors and their carers, found that survivors felt tation mechanisms are interrelated (Bodenmann, 2005), our more anxious when both partners held a strong represen- first hypothesis was that each member’s illness represen- tation of illness consequences, and more depressed when tations would be related to personal as well as partner there was a discrepancy between own and carer’s illness psychological symptoms. The second hypothesis was that coherence. Finally, Karademas & Giannousi (2013) found illness representation dissimilarity is related to more psy- that both partners’ representations of control over illness chological symptoms for both, as found in previous studies were associated with patients’ symptoms of anxiety and (Merz et al., 2011; Romero et al., 2008). It is possible that depression. They also found that patients and spouses differences in illness representations reflect differences reported less psychological symptoms, irrespectively of also in the ways patients and their partners react to the their own levels of perceived control, when their partners’ illness experience, as well as difficulties in the couple’s reported higher levels of the corresponding representations. communication regarding illness. These differences may be Generally, it seems that patients’ and partners’ reactions to distressing for the patients and their partners and, thus, illness represent a crucial source of information for the affect their well-being in a negative way. other member of the couple that may affect their well- The final hypothesis was that patients’ and spouses’ being (Bodenmann, 2005; Dimitraki & Karademas, 2014). illness representations impact their partners’ self-regulation Yet, it is worth noting that those studies examining dyadic process and, thus, affect the relation of partner illness effects do not also examine the possible impact of patient– representations to own psychological symptoms. As Dim- partner representation dissimilarity, although both types of itraki and Karademas (2014) suggested, patients’ and effects seem to be important for illness adaptation. partners’ illness representations may function in a syner- The above mentioned studies underline the significance getic manner. It is possible for each member of the couple of both partners’ representations in their adaptation to to use the illness representations of the other member, not 123 J Behav Med (2016) 39:85–93 87 only as a source of information, but also as a way to val- Table 1 Participants’ characteristics idate own representations, or even as a way to enhance Patients Spouses well-being. For example, even though dissimilarity in ill- ness representations may sometimes be distressing (see Mean age 58.93 years (SD = 13.76, 58.17 years (SD = 11.75; also our first hypothesis), it is still possible for a person to min = 33, max = 85) min = 28, max = 83) use the more ‘‘positive’’ illness representation of their Sex Males: 90 (60.4 %) Males: 59 (39.58 %) partner so as to counterbalance own ‘‘negative’’ represen- Females: 59 (39.58 %) Females: 90 (60.4 %) tation and preserve a sense of control over the situation. In Education 9-Year mandatory: 48.32 % 9-Year mandatory: 55.03 % this respect, and as found in previous studies (Wu et al., High school: 34.23 % High school: 31.55 % 2013; Karademas & Giannousi, 2013), we expected that a Higher education: 17.45 % Higher education: 13.42 % more positive illness representation of one member could buffer the detrimental relation of less positive representa- tion of the other to own psychological symptoms. Table 2 Disease characteristics Cancer type Early-stage Metastatic Lung 50 (33.55 %) 9 (18 %) 41 (82 %) Methods Breast 37 (24.84 %) 9 (24.32 %) 28 (75.68 %) Participants and procedure Gastrointestinal 34 (22.82 %) 6 (17.65 %) 28 (82.35 %) Others types of cancer 28 (18.79 %) 10 (35.71 %) 18 (64.29 %) The study was conducted at the Department of Medical Total = 149 34 (22.82 %) 115 (77.18 %) Oncology of the University Hospital of Heraklion, Crete. Greek cancer patients who received the diagnosis of cancer less than a month ago and their treatment begun within the the measures of the study separately. The study was last few weeks, as well as their spouses were invited to approved by the hospital’s Ethics Committee. participate in the study. Inclusion criteria for the patients Two hundred and twenty-three patients were identified were age over 18, a first-time cancer diagnosis, ability to as eligible for participation. However, 38 refused partici- understand the study protocol and provide informed con- pation because they were not interested or felt unable, sent, being married for a year or more and consenting to while in 36 of the remaining couples only one member their spouse’s participation. A recent diagnosis of cancer returned all questionnaires completed. The final sample was chosen to control for possible differences in psycho- consisted of 298 individuals nested in 149 couples. logical adaptation or coping which would be affected by a Tables 1 and 2 present the socio-demographic and illness longer period of illness experience by participants. In diagnosis characteristics of the participants. addition, the understanding of the processes underlying the early reactions to cancer diagnosis is important as they are Measures crucial for subsequent adjustment (Carver & Antoni, 2004; Hack & Degner, 2004). Also, Crete is an island where Illness representations were assessed with the Revised Ill- society emphasizes the more ‘‘traditional’’ family bonds, ness Perception Questionnaire (IPQ-R; (Moss-Morris et al., and is characterized by a lack of a gay–lesbian community. 2002). A slightly re-worded version was used to assess Thus, we did not come across any homosexual couples. spouses’ own representations. We assessed seven dimen- Also, most couples traditionally decide to marry. sions of illness representations as included in the IPQ-R: Medical files were used to identify potentially eligible timeline acute/chronic (chronicity) (six items; e.g., My participants. Subsequently, a research assistant approached illness will last for a long time [spouses’ version: My patients who met the inclusion criteria while waiting for spouse’s illness will last for a long time]); timeline-cyclical their appointment with their doctor before starting their (four items; e.g., My illness is very unpredictable) [spou- treatment and invited them and their spouses to participate ses’ version: My spouse’s illness is very unpredictable]); in the study. The Cretan society, as a rural one, is rather consequences (six items; e.g., My illness has major con- collectivistic and puts emphasis on family support, espe- sequences on my life [spouses’ version: My spouse’s ill- cially under difficult conditions (Georgas et al., 2006; ness has major consequences on my life]); personal control Kafetsios & Nezlek, 2012). Thus, most patients usually (six items; e.g., The course of my illness depends on me visit hospital facilities accompanied by their partners or a [spouses’ version: The course of the illness depends on close family member. This renders easy access to a couple. me]); treatment control (five items; e.g., My treatment can Patients and spouses who agreed were asked to complete control my illness [spouses’ version: The treatment can 123 88 J Behav Med (2016) 39:85–93 control his/her illness]); illness coherence (five items; e.g., level of the scores across the items of a scale; spread, which I have a clear picture or understanding of my illness refers to the variability of the scores across the items; and [spouses’ version: I have a clear picture or understanding shape, which refers to the pattern of differences between the of my spouse’s illness]); and, emotional representations scores across the items (Kenny et al., 2006). The more (six items; e.g., My illness makes me feel afraid [spouses’ similar the level, the spread and the shape, the greater is the version: My spouse’s illness makes me feel afraid]). The dyadic similarity. Distance scores were calculated as the answers were formed on a five point Likert type scale square root of the sum of the squared differences between ranging from 1 (strongly disagree) to 5 (strongly agree), patients’ and partners’ scores in the items of each illness with higher scores indicating a stronger illness represen- representation scale (Kenny et al., 2006). These scores were tation. Reliabilities (Cronbach’s alpha) ranged from.69 to included in the APIM analyses as independent variables.88 for patients and from.70 to.85 for spouses (Table 3). together with the actor and the partner main effects in new Psychological symptoms The Hospital Anxiety and models predicting psychological symptoms. A post hoc Depression Scale (Zigmond & Snaith, 1983; for the Greek examination revealed a statistical power greater than.90 at adaptation see, Mystakidou et al., 2004) was used to assess an alpha level equal to 5 % and a medium effect size. psychological symptoms. Anxiety and depression symp- toms are measured with 7 items each (e.g., I feel tense or ‘wound up’; I still enjoy the things I used to enjoy;). The Results answers are on a four point Likert type scale with higher scores indicating more intense symptoms. All items’s Preliminary analyses responds were added up to create an overall score (Cron- bach’s alpha =.81 for the patients and.78 for the spou- To provide a broader picture of the relations between the ses). factors examined in the study, as well as to examine for variances that should be controlled for in the subsequent Analyses APIM analyses, a series of preliminary analyses were conducted. A MANOVA of patient illness representations Dyadic responses were examined with the Actor–Partner and psychological symptoms across patient gender, edu- Interdependence Model (APIM) (Kenny et al., 2006). cation level, the type of diagnosis and the early/metastatic According to APIM, the effects of a person’s own char- type of cancer was performed. No statistically significant acteristics on own outcomes is referred to as the actor differences were observed [Wilks ks \.97; F(8, effect, while the effects of the characteristics of one 108) \ 1.13, ps [.05, g2s \.08]. A similar analysis of member of the couple on the other member’s outcomes is spouse illness representations and psychological symptoms referred to as the partner effect. Following Kenny et al. across spouse gender, education level, the type of diagnosis (2006), we applied structural equation modeling (SEM) and the early/metastatic type of cancer also did not reveal using LISREL 8.54 (Joreskog & Sorbom, 1993) to run the any statistically significant differences [Wilks ks \.95; APIM analyses. Also, in order to examine the interaction F(8, 108) \ 1.10, ps [.05, g2s \.08]. Patients’ age was effects between patients’ and spouses’ representations on significantly related only to illness coherence (Pearson’s psychological symptoms, we followed the procedure pro- r =.25, p \.01), while spouses’ age was not related to posed by Cook and Kenny (2005): To reduce multi- own illness representations or mental health. Thus, all collinearity due to high correlations between interaction subsequent analyses were pooled over gender, education terms and the independent variables from which they are level and type of illness. Further, a series of paired t tests created, each of the variables was centered. Patient and showed that spouses reported higher levels of timeline- spouse scores were first summed up and the combined cyclical, consequences, treatment control and emotional mean was then subtracted from the individual patient and representations, as well as lower representation of personal spouse scores. The products of each partner’s illness rep- control over illness [ts(148) [ |2.18|, ps\.05; see Table 1]. resentations by the other partner’s corresponding repre- Spouses also reported more psychological symptoms than sentation were calculated. Finally, interaction terms were patients [t (148) = -2.04, p \.05], as also reported in included as independent variables together with the actor previous studies (Applebaum & Breitbart, 2013). and the partner main effects in new models predicting psychological symptoms. Actor and partner effects To assess dissimilarity scores, we used ‘distance’ scores which reflect the three factors that affect the size of the Table 1 presents the correlation coefficients (Pearson’s r) measure and, thus, the degree of (dis)similarity between of all variables. Patients’ psychological symptoms were partners’ scores: level, which refers to the mean/average positively related to their spouses’ representations of 123 J Behav Med (2016) 39:85–93 89 Table 3 Descriptive statistics and intercorrelations of patient and spouse illness representations and psychological symptoms (N = 149 couples) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 1. P. timeline 1.00 2. P. timeline.26* 1.00 cyclical 3. P. consequences.42**.24* 1.00 4. P. personal -.47** -.26** -.38** 1.00 control 5. P. treatment -.17*.06 -.20*.05 1.00 control 6. P. illness -.17* -.07.11.24** -.28** 1.00 coherence 7. P. emotional.27**.14.45** -.17* -.21**.14 1.00 represent. 8. S. timeline.19*.05.13 -.05 -.36**.02.01 1.00 9. S. timeline.21**.46**.18* -.15 -.23** -.01.06.37** 1.00 cyclical 10. S..22**.02.39** -.11 -.42**.13.25**.43**.38** 1.00 consequences 11. S. personal -.36** -.23** -.30**.45**.28**.30** -.17* -.27** -.22** -.28** 1.00 control 12. S. treatment -.36** -.15 -.33**.27**.61**.09 -.15 -.36** -.31** -.43**.59** 1.00 control 13. S. illness -.13 -.04 -.06.30** -.01.46**.10.03 -.05 -.01.39**.26** 1.00 coherence 14. S. emotional -.05.03.15.01 -.25**.26**.25**.16*.19*.36**.06 -.17*.27** 1.00 represent. 15. P. psychol..39**.21**.43** -.46** -.01 -.25**.29**.14.18*.32** -.42** -.32** -.35**.08 1.00 symptoms 16. S. psychol..16*.11.19* -.01 -.17*.01.09.24**.31**.39** -.32** -.38** -.11.42**.38** 1.00 symptoms Mean 16.72 10.46 16.85 21.23 16.23 16.10 19.64 17.26 11.32 18.68 20.31 18.00 16.79 22.15 14.92 16.79 SD 4.90 3.23 4.10 4.99 2.54 5.36 5.17 5.26 4.01 4.65 4.86 3.82 5.59 4.27 10.56 9.59 P. patient, S. spouse * p \.05; ** p \.01 timeline-cyclical and consequences, and negatively to ness coherence. Spouses’ symptoms were negatively spouses’ representations of personal, treatment control, and related to patients’ representations of personal and treat- illness coherence. Also, spouses’ psychological symptoms ment control. were positively associated with patients’ representations of illness timeline and consequences. Dissimilarity and interaction effects Figure 1 presents the results of the SEM analyses employed to test for the APIM effects (see first hypothesis). As far as patient–spouse illness representations dissimilarity Both actor (i.e., within person) and partner (i.e., between scores is concerned (the second hypothesis), and after con- partners) effects were detected for both patients’ and trolling for the actor and partner main effects, only one spouses’ psychological symptoms. Regarding actor effects, statistically significant effect was found. The patient–spouse patients’ higher levels of psychological symptoms were dissimilarity in representations of illness consequences was associated with lower levels of own illness representations positively related to patients’ psychological symptoms of personal and treatment control and higher levels of (b =.10, T = -2.05, p \.05). That is, the greater the emotional representations. Spouses’ symptoms were rela- discrepancy in perceived illness consequences, the more ted to lower levels of own representations of personal symptoms reported by the patients. No dissimilarity effects control and higher levels of emotional representations. were detected for spouses’ psychological symptoms. Regarding partner effects, patients’ symptoms were related Regarding the interaction effects (see third hypothesis), positively to spouses’ representations of illness conse- only one statistically significant interaction was found. quences, and negatively to spouses’ representations of ill- Spouse representations of illness coherence moderated the 123 90 J Behav Med (2016) 39:85–93 Fig. 1 Beta-coefficients from Personal control – patients the SEM analysis testing for actor and partner effects on psychological symptoms -.04* (N = 149 couples). Note: For Treatment control – patients -.10* clarity reasons, covariances between explanatory variables Patients’ symptoms.54.03* as well as non-significant paths are omitted from the figure. Emotional repres. – patients.04* Dotted lines represent partner -.04* effects represent partner effects. *p \.05; **p \.01 Consequences – spouses.16** -.05* -.09* Personal control – spouses -.06* Spouses’ symptoms. 58.10* Illness coherence – spouses Emotional repres. – spouses effects of patient corresponding representations on own symptoms, while treatment control was also related to psychological symptoms (b =.02, T = 2.29, p \.05). As patients’ symptoms. Such findings were rather expected, shown in Fig. 2, the negative association of patients’ psy- provided that according to the CS-SRM (Leventhal et al., chological symptoms with their representations of illness 1980; Rozema et al., 2009) the role of control and emo- coherence is weak at the higher and medium levels of tional reaction is central in illness adaptation. Previous spouse corresponding representations, and stronger at the studies (Millar et al., 2005; Rozema et al., 2009; Llewellyn lower levels of spouse representations. In other words, et al., 2007; Giannousi et al., 2010) have shown that a more when spouses reported higher levels of illness coherence, ‘positive’ representation of cancer as a controllable con- patients reported relatively low levels of symptoms irre- dition is related to less psychological symptoms, whereas a spectively of own illness coherence representations. No more intense emotional impact of the disease is related to statistically significant interaction was found with respect more symptoms. to spouses’ psychological symptoms. Furthermore, several partner (i.e., between partners) effects were found. Patients reported better psychological health when their spouses perceived cancer as a less Discussion burdensome (i.e., with less consequences) and under- standable condition, while spouses reported better psy- The present study aimed to examine at a dyadic level chological health when patients perceived their condition (a) the impact of illness representations of recently diag- as more controllable. In other words, a more ‘positive’ or nosed Greek cancer patients and their spouses on their optimistic appraisal of cancer diagnosis by one member of psychological well-being; (b) the possible impact of the couple appeared to help the other maintain higher patient–spouse illness representations discrepancy on psy- levels of psychological health, at least at the onset of chological symptoms of both; (c) the interactions between illness. This is an important finding, provided that psy- patients’ and partners’ illness representations. To our chological well-being has been associated with mortality knowledge, this is the first study to parallel examine dya- in cancer patients (Satin et al., 2009) as well as quality of dic, interaction and dissimilarity effects in a sample of life in patients and partners (Roland et al., 2013; Ellis, couples dealing with a recent diagnosis. Overall, the results 2012). provided partial support to all hypotheses. Overall, these findings emphasize the interdependence The findings revealed several actor and partner effects between patients and partners and indicate that, although for patients and spouses. Regarding actor (i.e., within personal illness representations are really important for person) effects, personal control and emotional represen- each member of the couple (Leventhal et al., 1980, tations were related to patients’ and spouses’ psychological Weinman et al., 2003), adaptation to illness is also affected, 123 J Behav Med (2016) 39:85–93 91 Fig. 2 Graphical representation 22,00 of the interaction effects Spouse illness coherence between patient and partner 21,00 high med low representations of illness Patients' psychological symptoms coherence on patients’ psychological symptoms 20,00 19,00 18,00 17,00 16,00 15,00 14,00 low med high Patient illness coherence even from the early phases of illness, by the ways the other allows patients to preserve a sense of control over illness, member represents illness. Moreover, representations and which in turn may facilitate their adaptation to it (Dimitraki behavior do not develop in isolation, but in constant & Karademas, 2014). However, not all studies suggest the interaction with family members and partners (Boden- same, as there is evidence that the minimization of illness mann, 2005). Probably, each person reckons his/her part- severity by spouses may have negative impact on patients’ ner’s understanding of illness as an important source of adaptation (Heijmans et al., 1999). It is possible that the information which is used to develop or adapt own repre- impact of this interaction depends on several factors, sentations of illness (Dimitraki & Karademas, 2014). including type and stage of illness, and personal needs Another interesting finding was that, even after con- (Heijmans et al., 1999; Karademas & Giannousi, 2013; trolling for actor and partner effects, one statistically sig- Dimitraki & Karademas, 2014). Nevertheless, the findings nificant interaction effect and one dissimilarity effect were of this study suggest that, at least at the onset of cancer, detected. Although just a few of those effects were found, spouses’ positive illness representations are beneficial for they indicate that, at least under certain conditions, the patients’ adaptation to illness. impact of the combination of patient and partner illness It is worthy of note that both the statistically significant representations on adaptation to illness is important in interaction and dissimilarity effects, referred to patients. ways beyond actor and partner effects. Specifically, the This may reflect that, in comparison to their spouses, discrepancy between patients’ and partners’ representations patients are more sensitive to any differences in the illness of illness consequences (with partners reporting more representations of the dyad. In any case, these effects consequences) was associated with more psychological underscore once more the strong connection between symptoms in patients. This is in accordance with previous patients’ and spouses’ adaptation to illness processes. studies in couples dealing with cancer (Merz et al., 2011; This study is faced with certain limitations. First of all, it Romero et al., 2008). It is possible that dissimilarity in was a cross-sectional study and, consequently, no causal ways of understanding illness may reflect different illness- effects between illness representations and psychological related behaviors (Benyamini et al., 2009), which may health could be tested. Second, patients were suffering cause patients additional psychological distress. from various types and stages of cancer. Third, only psy- Additionally, when spouses reported higher levels of chological health was examined. Other important indica- illness coherence, patients’ corresponding representations tors of adaptation to illness and well-being were not were not related to their psychological health. Thus, lower assessed taking into consideration the additional burden to levels of illness coherence were not associated with worse patients, especially those for whom their performance psychological health. This finding is in accordance with status allowed only one short meeting. In addition, the previous studies which suggest that when partners report quality of marital functioning was not examined, although more positive representations of illness, patients experience this may impact patient–spouse communication regarding higher levels of well-being (Figueiras & Weinman, 2003). illness and their reactions. Finally, our sample consisted of Probably, a more ‘positive’ spouse illness representation couples married for a year or more. Other types of romantic 123 92 J Behav Med (2016) 39:85–93 relationships or very recently married couples were not associations with received and provided social support and included. Also, the participants were living in a rather undermining. Psychology and Health, 22, 765–785. Berg, C. A., & Upchurch, R. (2007). A developmental-contextual collectivistic than individualistic society. Thus, the gener- model of couples coping with chronic illness across the adult life alizability of our findings may be limited by these limita- span. Psychological Bulletin, 133, 920–954. tions. Bodenmann, G. (2005). Dyadic coping and its significance for marital Despite these limitations, we believe that the findings of functioning. In T. A. Revenson, K. Kayser, & G. Bodenmann (Eds.), Couples coping with stress: Emerging perspectives on this study have significant implications. A cancer diagnosis dyadic coping (pp. 33–50). Washington, DC: American Psycho- and the subsequent medical treatment is a very taxing sit- logical Association. uation for both partners. Although the focus of most studies Carver, C. S., & Antoni, M. (2004). Finding benefit in breast cancer has been on individual’s processes (the patient), adaptation during the year after diagnosis predicts better adjustment 5 to 8 years after diagnosis. Health Psychology, 23, 595–598. to illness at the onset of diagnosis and thereafter refers to Cook, W. L., & Kenny, D. A. (2005). The Actor–Partner Interde- patients and their families and, especially, their partners. pendence Model: A model of bidirectional effects in develop- Within this context, and in order to more thoroughly mental studies. International Journal of Behavioral understand patients’ and partners’ illness-related behavior, Development, 29, 101–109. Dempster, M., McCorry, N. K., Brennan, E., Donnelly, M., Murray, adaptation to illness and well-being, the parallel examina- L. J., & Johnston, B. T. (2011). Illness perceptions among carer- tion of the illness representations of both partners emerges survivor dyads are related to psychological distress among as a necessity for early as well as later phases of illness oesophageal cancer survivors. Journal of Psychosomatic trajectory. Research, 70, 432–439. Dimitraki, G., & Karademas, E. C. (2014). The association of type 2 Present findings emphasize the importance of early diabetes patient and spouse illness representations with their engagement of partners in order to deal with cancer well-being: A dyadic approach. International Journal of Behav- effectively. They could also be used to tailor interventions ioral Medicine, 21, 230–290. focused on partners’ representations and re-forming them Ellis, J. (2012). The impact of lung cancer on patients and carers. Chronic Respiratory Disease, 9, 39–47. to more adaptive ones. This, in turn, may facilitate patients’ Fagundes, C. P., Berg, C. A., & Wiebe, D. J. (2012). Intrusion, adaptation to cancer as well. So far, there are several avoidance, and daily negative affect among couples coping with programs addressed solely to patients or both (Baum & prostate cancer: A dyadic investigation. Journal of Family Andersen, 2001; Manne & Ostroff, 2008). However, to our Psychology, 26, 246–253. Figueiras, M. J., & Weinman, J. (2003). Do similar patient and spouse knowledge, there are no programs addressed only to part- perceptions of myocardial infarction predict recovery? Psychol- ners. ogy and Health, 18, 201–216. Georgas, J., Berry, J. W., van de Vijver, F. J. R., Kagitcibasi, C., & Compliance with ethical standards Poortinga, Y. H. (2006). Families across cultures: A 30-nation psychological study. Cambridge: Cambridge University Press. Conflict of interest Zoe Giannousi, Evangelos C. Karademas and Giannousi, Z., Manaras, I., Georgoulias, V., & Samonis, G. (2010). Georgia Dimitraki declare that they have no conflict of interest. Illness perceptions of Greek patients with cancer: A validation of the Revised Illness Perception Questionnaire. Psycho-Oncology, Human and animal rights and Informed consent All procedures 19, 85–92. followed were in accordance with ethical standards of the responsible Given, B., Kozachik, S., Collins, C., Devoss, D., & Given, C. W. committee on human experimentation (institutional and national) and (2001). Caregiver role strain. In M. Maas, K. Buckwalter, M. with the Helsinki Declaration of 1975, as revised in 2000. Informed Hardy, T. Tripp-Reimer, & M. Titler (Eds.), Nursing care of consent was obtained from all patients for being included in the study. older adults diagnosis: Outcome and interventions (pp. 679–695). St Louis, MO: Mosby. Hack, T. F., & Degner, L. F. (2004). Coping responses following breast cancer diagnosis predict psychological adjustment three References years later. Psycho-Oncology, 13, 235–247. Hagedoorn, M., Sanderman, R., Bolks, H. N., Tuinstra, J., & Coyne, J. Applebaum, A., & Breitbart, W. (2013). Care for the cancer C. (2008). Distress in couples coping with cancer: A meta- caregiver: A systematic review. Palliative & Supportive Care, analysis and critical review of role and gender effects. Psycho- 11, 231–252. logical Bulletin, 134, 1–30. Badr, H., Carmack, C. L., Kashy, D. A., Cristofanilli, M., & Hagger, M. S., & Orbell, S. (2003). A meta-analytic review of the Revenson, T. A. (2010). Dyadic coping in metastatic breast Common-Sense Model of illness representations. Psychology cancer. Health Psychology, 29, 169–180. and Health, 18, 41–184. Baum, A., & Andersen, B. L. (2001). Psychosocial interventions for Heijmans, M., de Ridder, D., & Bensing, J. (1999). Dissimilarity in cancer. Washington, DC: American Psychological Association. patients’ and spouses’ representations of chronic illness: Explo- Benyamini, Y., Gozlan, M., & Kokia, E. (2009). Women’s and men’s rations of relations to patient adaptation. Psychology and Health, perceptions of infertility and their associations with psycholog- 14, 51–466. ical adjustment: A dyadic approach. British Journal of Health Joreskog, K. G., & Sorbom, D. (1993). LISREL 8 user’s guide. Psychology, 14, 1–16. Lincolnwood, IL: Scientific Software International. Benyamini, Y., Medalion, B., & Garfinkel, D. (2007). Patient and Kafetsios, K., & Nezlek, J. B. (2012). Emotion and support spouse perceptions of the patient’s heart disease and their perceptions in everyday social interaction: Testing the ‘‘less is 123 J Behav Med (2016) 39:85–93 93 more’’ hypothesis in two cultures. Journal of Social and Moss-Morris, R., Weinman, J., Petrie, K. J., Horne, R., & Cameron, Personal Relationships, 29, 165–184. L. D. (2002). The Revised Illness Perception Questionnaire Karademas, E. C., & Giannousi, Z. (2013). Representations of control (IPQ-R). Psychology and Health, 17, 1–16. and psychological symptoms in couples dealing with cancer: A Mystakidou, K., Tsilika, E., Parpa, E., Katsouda, E., Galanos, A., & dyadic-regulation approach. Psychology and Health, 28, 7–83. Vlahos, L. (2004). The Hospital Anxiety and Depression Scale in Kenny, D. A., Kashy, D. A., & Cook, W. L. (2006). Dyadic data Greek cancer patients: Psychometric analyses and applicability. analysis. New York: Guilford. Supportive Care in Cancer, 12, 821–825. Leventhal, H., Diefenbach, M., & Leventhal, E. A. (1992). Illness Roland, K. B., Rodriguez, J. L., Rees Paterson, J., & Trivers, K. F. cognition: Using common sense to understand treatment adher- (2013). A literature review of the social and psychological needs ence and affect cognition treatment. Cognitive Therapy and of ovarian cancer survivors. Psycho-Oncology, 22, 2408–2418. Research, 16, 223–230. Romero, C., Lindsay, J. E., Dalton, W. T., Nelson, D. V., & Friedman, Leventhal, H., Halm, E., Horowitz, C., Leventhal, E. A., & Ozakinci, L. C. (2008). Husbands’ perceptions’ of wives’ adjustment to G. (2005). Living with chronic illness: A contextualized, self- breast cancer: The impact on wives’ mood. Psycho-Oncology, regulation approach. In S. Sutton, A. Baum, & M. Johnston 17, 237–243. (Eds.), The SAGE handbook of health psychology (pp. 197–240). Rosland, A.-M., Heisler, M., & Piette, J. D. (2012). The impact of London: Sage. family behaviors and communication patterns on chronic illness Leventhal, H., Meyer, D., & Nerenz, D. (1980). The common-sense outcomes: A systematic review. Journal of Behavioral Medicine, representations of illness danger. In S. Rachman (Ed.), Medical 35, 221–239. psychology (Vol. 2, pp. 7–30). New York: Pergamon. Rozema, H., Völlink, T., & Lechner, L. (2009). The role of illness Llewellyn, C. D., McGurk, M., & Weinman, J. (2007). Illness and representations in coping and health of patients treated for breast treatment beliefs in head and neck cancer: Is Leventhal’s cancer. Psycho-Oncology, 18, 849–857. common sense model a useful framework for determining Satin, J., Linden, W., & Phillips, M. J. (2009). Depression as a changes in outcomes over time? Journal of Psychosomatic predictor of disease progression and mortality in cancer patients: Research, 63, 17–26. A meta-analysis. Cancer, 115, 5349–5361. Manne, S. L., & Ostroff, J. S. (2008). Coping with breast cancer: A Twiddy, M., House, A., & Jones, F. (2012). The association between couples-focused group intervention. Therapist guide. New York: discrepancy in illness representations on distress in stroke Oxford University Press. patients and carers. Journal of Psychosomatic Research, 72, Martin, R., & Leventhal, H. (2004). Symptom perception and health 220–225. care-seeking behavior. In T. J. Boll, J. M. Raczynski, & L. Weinman, J., Heijmans, M., & Figueiras, M. J. (2003). Carer C. Leviton (Eds.), Handbook of clinical health psychology (Vol. perceptions of chronic illness. In L. D. Cameron & H. Leventhal 2, pp. 299–328). Washington, DC: American Psychological (Eds.), The self-regulation of health and illness behaviour (pp. Association. 207–219). New York: Routledge. Merz, E. L., Malcarne, V. L., Ko, C. M., Sadler, M., Kwack, L., Wu, L. M., Mohamed, N. E., Winkel, G., & Diefenbach, M. A. Varni, J. W., & Sadler, G. R. (2011). Dyadic concordance among (2013). Patient and spouse illness beliefs and quality of life in prostate cancer patients and their partners and health-related prostate cancer patients. Psychology and Health, 28, 355–368. quality of life: Does it matter? Psychology and Health, 26, Zigmond, A. S., & Snaith, R. P. (1983). The Hospital Anxiety and 651–666. Depression Scale (validity and reliability). Acta Psychiatrica Millar, K., Purushotham, A. D., McLatchie, E., George, W. D., & Scandinavica, 67, 361–370. Murray, G. D. (2005). A 1-year prospective study of individual variation in distress, and illness perceptions, after treatment for breast cancer. Journal of Psychosomatic Research, 58, 335–342. 123