Abstract

Objective. The aim of the present study was to investigate the relationship between individual coping (IC) and dyadic coping (DC), more specifically the respective predictive impact of DC and sociodemographic, partnership-related, and medical variables on IC in haematological cancer patients. Methods. In this multicenter longitudinal study, we examined DCI (Dyadic Coping Inventory) and IC (Freiburg Questionnaire on Coping with Illness) in 214 haematological cancer patients after their treatment and six months later. Associations between the coping constructs were calculated using Pearson correlations. Multiple linear regression analyses were conducted to examine the predictive impact of patient-specific variables on IC strategies. Results. Pearson correlation revealed numerous significant associations between DC and IC at the baseline and follow-up, but none of them yielded r ≥  (). Regression analyses showed a predictive impact of supportive DC on compliance () and of common DC on active IC (). Concerning patient-specific variables, age stood out as a positive predictor of trivialization () and compliance () and psychological distress as a predictor of depressive IC (), distraction (), and trivialization (). Conclusion. Our results indicate that patients’ personal characteristics, such as age and psychological distress, outweigh the influence of DC on IC in haematological cancer patients.

1. Introduction

Haematological cancer encompasses numerous subtypes, ranging from aggressive to chronic, and each subtype has unique characteristics and rates of progression. Compared with solid tumors, haematological cancer is known for its aggressive nature, its lengthy duration, and serious side effects [1, 2]. These often result in a variety of physical, psychological, emotional, and social consequences for the patient [3]. The highly aggressive treatments, such as high-dose chemotherapy, total body irradiation, and haematopoietic stem cell transplantation (HSCT), which requires mandatory isolation of patients, are considered to be particularly stressful for both patients and their partners [46]. In order to cope with these challenges, patients can resort to both individual and dyadic coping strategies. The latter describes the way couples respond to a shared stressor through a collaborative process [7]. The present longitudinal study aims to increase our understanding of the interplay between dyadic coping (DC) and individual coping (IC) in patients with haematological cancer.

It was not until the early 1990s that research in Europe and the United States began to conceptualize coping in partnerships as a dyadic rather than an exclusively individual phenomenon that results from the interactions of the two partners [8]. In recent years, this view has become increasingly influential and has been the central focus of various empirical studies [912]. DC describes the interaction of both the patients’ and the partners’ coping efforts and can be defined as “one partner’s attempt to reduce the external stress perceived by his or her partner, and a joint effort to cope with stress that arises within the relationship” [13]. The Dyadic Coping Inventory (DCI) is a widely used international measurement tool for assessing DC, which distinguishes between positive DC (supportive DC, delegated DC, common DC) and negative DC (use of meaningless flowery phrases, ambivalent and hostile behavior). The construct implies a sequence of events stemming from the stressed person’s expression of stress, to the partner’s perception of stress, and to the partner’s coping response [8]. In contrast, traditional models, such as the transactional stress theory by Lazarus and Folkman, focus on the individual-centered perspective of stress management, considering coping efforts as an individual process [14]. Based on this model, Muthny developed the Freiburg Questionnaire on Coping with Illness (FKV) [15], which examines cognitive, emotional, and behavioral coping mechanisms of individuals suffering from chronic disease (e.g., cardiac disease, cancer, and multiple sclerosis) [16].

There is extensive evidence for the importance of both DC and IC in adaptation to illness and cancer patients’ well-being. Literature reveals that IC behaviors, such as problem-solving, are associated with better physical and mental health outcomes. IC strategies, such as avoidance, denial, or giving-up, on the other hand, are associated with higher levels of psychological distress, fatigue, anxiety, depression, and lower physical well-being in a variety of solid tumors, including head and neck cancer [17, 18], breast cancer [1921], gynecologic cancer [22, 23], and prostate cancer [24]. Regarding DC, problem-solving, emotion-regulation-oriented, and common coping were found to be positively associated with physical and mental health in cancer patients, whereas hostile DC was associated with poorer health (e.g., [9, 10, 25, 26]). However, controversy exists regarding the interplay of these two coping constructs [27]. According to the latest model, the dyadic regulatory connectivity model (DR-CM), the use of IC and DC depends on the characteristics and the situation of the individuals involved [28]. This is consistent with Berg and Upchurch’s Developmental Contextual Model (DCM), which states that illness appraisals, coping, and, ultimately, outcome can be influenced by contextual factors (e.g., culture, age, gender, type or stage of illness, and quality of marriage) [29]. A number of studies support this assumption by demonstrating the influence of patients’ individual characteristics such as age [30, 31], socioeconomic status (SES) [32], level of psychological distress [33], or relationship satisfaction [34] on coping behavior in chronic disease such as breast cancer, prostate cancer, congenital heart disease, and malignant glioma. Our decision of identifying DC at the baseline as a predictor for IC at follow-up is also based on the DCM considering DC as “potentially the first line of coping for couples in dealing with stressful events” [29]. With this in mind, the purpose of this study is to examine whether and how DC and certain patient-specific characteristics at the baseline predict IC at follow-up in a sample of 214 haematological cancer patients.

In a study of 92 newly diagnosed female (predominantly breast) cancer patients, Paschali et al. found that more active IC behaviors (such as problem-solving) were positively associated with supportive and common DC, whereas IC behaviors such as avoidance and rumination were negatively associated with supportive and common DC. The study also examined correlations between potential influencing factors with DC and IC, finding no significant differences in both coping constructs between the level of education, presence of children, relationship duration, and patient age [27]. Two other studies found that adaptive IC was positively related to positive and common DC and negatively related to negative DC, whereas maladaptive IC was negatively associated with common DC [35, 36]. According to Paschali et al., positive IC and common coping were associated with higher levels of emotional well-being, and according to Brandão et al., the influence of IC on cancer-related well-being and quality of life is persistent, underlining the importance of both IC and DC for the well-being of cancer patients [27, 37]. However, the cross-sectional design of these studies does not allow for drawing causal conclusions. Moreover, two of them examined participants’ coping behaviors in relation to everyday stressors, whereas the interplay of DC and IC in relation to cancer remains largely unexplored.

To date, this is the first empirical study on the relationship between DC and IC in haemato-oncological patients. Knowledge about the interplay between these two coping constructs and possible predictors of IC may help clinicians improve psycho-oncological treatment and better understand the determinants of IC in general. We aimed to investigate(a)How DC and IC strategies change over time(b)Associations between baseline (t1) DC and IC and six-month follow-up (t2) DC and IC(c)The predictive impact of baseline DC as well as sociodemographic (gender, age, and SES), partnership-related (relationship duration and relationship satisfaction), and medical variables (haematopoietic stem cell transplantation and psychological distress) on follow-up IC

2. Methods

2.1. Participants and Procedures

This study is based on data on haemato-oncological patients from three university cancer centers in Germany (Leipzig, Ulm, and Regensburg). The data were collected through the research project “Dyadic coping in haematological patients over time” (2012–2015) funded by the Deutsche José Carreras Leukämie-Stiftung (grant: DJCLS R 12/36). The Medical Faculty of the University of Leipzig provided the ethics approval (no. 298-12-24092012), which complies with the Helsinki Declaration of 1975 as revised in 2000.

In order to be eligible to participate in the study, patients needed to have a confirmed diagnosis of a haematological neoplasm (ICD-10: C81–C96 and D46, newly diagnosed or recurrence), be aged between 18 and 75 years, and living in a partnership. Furthermore, sufficient knowledge of the German language and the partner’s willingness to participate were necessary to be included in the study. Exclusion criteria were severe cognitive and/or physical impairments and severe mental disorders.

Patients were informed about the study by hospital staff. After giving permission to be further contacted, study documents including study information, consent sheets (informed consent), and written questionnaires were sent to the patients.

568 patients met the inclusion criteria at t1 (after the treatment) and were informed by staff and contacted by research team members. In total, 330 of 568 patients (58.1%) participated at t1 and 238 (41.9%) did not for two main reasons: 56.2% of these nonparticipants were no longer available or provided no response, whereas 38.0% had no interest in the study.

Standardized questionnaires that assessed sociodemographic, couple-related, and disease-related characteristics, as well as coping and mental health, were sent to the patients, with stamped and addressed envelopes. Approximately half a year after t1, the follow-up measurement (t2) was conducted via mail and/or online survey. 217 patients took part in the follow-up (participation rate: 65.8%), of which 214 had complete data for DCI and FKV-15 at both t1 and t2. Figure 1 provides a flowchart of the sample and reasons for nonresponse.

2.2. Assessments
2.2.1. Sociodemographic and Medical Data

Sociodemographic data were assessed via self-report, and medical data were accessed through medical records. The SES results from the patients’ education/qualification which combines the highest school-leaving qualification and the highest vocational training [38]. As the purchasing power of nominal income depends upon the number of people in the household, household income was weighed by person per household. It was calculated by allocating the first person in the household a weighting of 1 and assigning each additional person a weighting of 0.5 [39].

2.2.2. Dyadic Coping Inventory

To assess DC, we used the Dyadic Coping Inventory (DCI) [8], a 37-item instrument that measures the degree to which couples support and actively help each other in stressful situations using five subscales: stress communication, supportive DC, negative DC, delegated DC, and common DC [40]. Stress communication is the ability to communicate emotion- and problem-oriented stress (four items, e.g., “I show my partner through my behavior when I am not doing well or when I have problems”). Supportive DC involves acts of support by one partner for the benefit of the other (five items, e.g., “I show empathy and understanding to my partner”). Negative DC includes ambivalent DC (supportive and/or joint coping actions are perceived as stressful or unnecessary by either partner), hostile DC (stress signals from one trigger a direct hostile response in the other), and avoidance of DC (superficial coping actions without actual emotional involvement of the performer, disregarding the other’s needs) (four items, e.g., “When my partner is stressed, I tend to withdraw”). In delegated DC, coping with the stressful situation is taken over by the supporting partner (two items e.g., “I take on things that my partner would normally do in order to help him/her out”). Common DC describes coping efforts in which the couple attempts to achieve common goals through mutual commitment (five items, e.g., “We engage in a serious discussion about the problem and think through what has to be done”). In contrast to supportive or delegated DC, both partners are mutually and directly involved [8, 9, 41]. Two additional single items assess satisfaction with and efficacy of DC. Items are rated on a five-point Likert scale that represents the frequency of a certain coping style used from 1 (very rarely) to 5 (very often). With the exception of the two single items and common DC, all subscales are assessed in two item parallel versions that measure respondents’ own behavior and their perception of their partner’s behavior [42]. The present study focuses on patients’ own coping behavior. Partners’ DC was not included because the FKV-15, whose interaction with the DCI will be examined, captures only the patients’ perspective. The focus is therefore exclusively on the patients’ own DC and IC behavior. For the current study, the internal consistencies (Cronbach’s α) of the DCI subscales were acceptable to good, with α = 0.78 to α = 0.84 (t1) and α = 0.73 to α = 0.87 (t2).

2.2.3. Freiburg Questionnaire on Coping with Illness (Short Form)

To assess IC, we used a short version of the German-language Freiburg Questionnaire on Coping with Illness (Freiburger Fragebogen zur Krankheitsverarbeitung), known as the FKV-15. The original version of the questionnaire (FKV-102) was developed by Muthny [15]; Hardt et al. introduced the shorter version with 15 items [43]. Widely employed in German speaking countries, the questionnaire is used to assess a broad range of coping strategies of individuals suffering from chronic disease (e.g., cardiac disease, cancer, and multiple sclerosis) [16]. The FKV-15 includes five subscales: depressive IC (to withdraw and react depressively, e.g., “Feeling sorry for oneself”), active IC (to approach one’s disease in an active, goal-oriented manner, e.g., “Determined to fight the disease”) distraction (to purposefully distract oneself from the disease, e.g., “Trying to gain distance”), trivialization (to downplay problems caused by the disease, e.g., “Downplaying the importance and scope.”), and compliance (to follow the doctor’s recommendations, e.g.: “Exactly following the medical advice”) [43]. Items are answered on a five-point Likert scale showing to what extent each one applies ranging from 1 (not at all) to 5 (very much) [44]. In contrast to the DCI, the FKV-15 measures only the patients’ coping behavior without the involvement of the partner. Unlike the DCI, which measures coping behavior in relation to any stressor, the FKV is aimed exclusively at individuals suffering from a medical condition. For the current study, the internal consistencies of the FKV-15 subscales ranged between 0.70 and 0.82 at both t1 and t2.

2.2.4. Partnership Questionnaire (Short Form)

Relationship satisfaction was assessed with the PFB-K, a short version of the PFB [45]. The PFB-K is composed of three subscales (quarrelling, tenderness, and togetherness/communication) with three items each on a four-point Likert scale. The response options range from 0 (never/very rarely) to 3 (very often). Another six-point item measures general relationship satisfaction from 0 (very unhappy) to 5 (very happy). The total value of the PFB-K is calculated by summing the subscale scores. In the present sample, the internal consistencies of the total value were 0.71 (t1) and 0.74 (t2).

2.2.5. Patient Health Questionnaire

Patients’ depression and anxiety were assessed using the PHQ-4, a very brief self-report screening instrument that consists of a 2-item depression scale (PHQ-2) and a 2-item anxiety scale (GAD-2) [46]. Its questions aim to assess how often these emotions have occurred over the previous two weeks from 0 (not at all) to 3 (nearly every day). In order to determine a total score for psychological distress, the four item scores are added together, reaching a total score between 0 (not at all distressed) and 12 (very distressed). A cut-off score of ≥6 distinguishes between those who are considered to be mentally stressed and those who are not [47]. The internal consistencies within our sample were good with α = 0.88 (t1) and 0.89 (t2).

2.3. Statistical Analysis

Differences in the frequency distribution of sociodemographic and clinical parameters were calculated by using chi-square tests; mean differences were determined using the t-test for independent samples (completer vs. noncompleter). Paired-sample t-tests were utilized to assess differences in means of DC and IC subscales over time. Δt2/t1 was calculated by comparing the mean values of the IC and DC subscales at t1 and t2. Effect sizes were determined according to the calculation for repeated measures within one group (prepost). A correction was made to correct the standard deviation used in the calculation of the effect size. The correction factor includes the correlation between premeasurement and postmeasurement. According to Morris, the standard deviation used was the value of the pretest since this value was not influenced by the intervention [48, 49]. Based on the number of tests, the significance level was Bonferroni-adjusted [50].

Correlation analyses between patients’ DC and IC at t1 and DC and IC at t2 were calculated according to Pearson. In order to classify correlation-based effect sizes, we followed Cohen’s recommendations (r ≥ 0.1 small effect, r ≥ 0.3 medium effect, and r ≥ 0.5 large effect) [51].

Multiple linear regression analyses were conducted to examine the predictive impact of sociodemographic, partnership-related, and medical variables, as well as DC subscales at t1 on IC at t2. Categorical variables were dummy-coded as follows: (a) gender (1 = female, 0 = male), (b) SES (1 = high, 0 = others), and (c) stem cell transplantation (1 = yes, 0 = others). Unstandardized (B) and standardized (β) correlation coefficients were reported.

Statistical analyses were conducted with IBM SPSS Statistics 28.

3. Results

3.1. Sample Characteristics and Drop-Out Analyses

As shown in Table 1, 61.7% of the 214 study participants were male. The average age of respondents was 57 years. 37.3% were employed, and 56.0% were retired. Relationship duration was >30 years on average. 47.2% had a medium SES. Household income weighted by household members was between 1,000 and 2,000€ for 64.3% of the patients. Chronic leukaemia was the most common diagnosis (23.8%). 57% of the patients underwent HSCT. Time since diagnosis was ≤two years for 47.7% of patients. There were no significant differences between completers and noncompleters.

3.2. Changes in DC and IC in Haematological Cancer Patients over Time

Figure 2 shows changes in DC and IC over time with bars representing mean values (0–5) of the DCI and FKV-15-subscales. Results illustrate nonsignificant changes in DC between t1 and t2. The corresponding effect sizes were small with d ≤ 0.15. In terms of IC, active IC significantly decreased over time (∆t2/t1 = −0.16). The corresponding effect sizes were also small with d ≤ 0.2.

3.3. Correlations between DC and IC

The DC subscales at t1 displayed numerous moderate to weak significant correlations with the IC subscales at t1 (see Table 2). The DC subscales at t1 were strongly positively associated with the corresponding DC subscales at t2 (see Table 3), with r values ranging from 0.53 to 0.68. All other t1 DC subscales were weakly to moderately correlated with those at t2, with the exception of a strong correlation between supportive DC (t1) and common DC (t2) (r = 0.60). The IC subscales at the baseline were moderately to strongly positively associated with the corresponding subscales at follow-up, with r values ranging from 0.45 to 0.67. We found weak to moderate correlations between the other IC (t1) and IC (t2) subscales.

There were several significant correlations between t1 DC and t2 IC, but none yielded  ≥ 0.34, as in the case of common DC (t1) and active IC (t2). Regarding the associations of t1 IC and t2 DC, the strongest correlation was between active IC (t1) and common DC (t2) (r = 0.34).

3.4. Regression Analysis
3.4.1. Sociodemographic Variables

Patients’ age at t1 turned out to be a significant positive predictor for trivialization IC (β = 0.37) and compliance IC (β = 0.29) at t2; the older the patients, the more they tended to use the respective coping strategies (see Table 4). Furthermore, high SES emerged as a significant negative predictor for trivialization IC (β = −0.16). No significant predictive influence of gender on patients’ IC behavior was found.

3.4.2. Partnership-Related Variables

Patients’ relationship duration was found to be a significant negative predictor for trivialization IC (β = −0.21). Moreover, high relationship satisfaction turned out to be associated with significantly less depressive IC (β = −0.17).

3.4.3. Medical Variables

Patients who had undergone stem cell transplantation were found to use significantly less trivialization IC than those who had not received stem cell transplantation (β = −0.23). Furthermore, patients’ psychological distress turned out to be a significant positive predictor of depressive IC (β = 0.4), distraction IC (β = 0.19), and trivialization IC (β = 0.17).

3.4.4. Dyadic Coping

Regarding the predictive impact of DC on IC, supportive DC (t1) was a significant positive predictor of compliance IC (t2) (β = 0.20). Moreover, we found a significant positive prediction between common DC (t1) and active IC (t2) (β = 0.25).

Overall, the explained variance of the models () is good with the highest for depressive IC at t2 (0.32).

4. Discussion

The purpose of this investigation was to examine the relationship between IC and DC as well as possible predictors of IC in patients dealing with haematological cancer. Longitudinal data from 214 patients were analyzed after their treatment (t1) and about six months later (t2), investigating (a) changes in coping mechanisms over time; (b) correlations between IC and DC at both t1 and t2; and (c) examining whether DC as well as sociodemographic, partnership-related, and medical variables at the baseline may predict follow-up IC.

Regarding our first research question (a), we found no significant changes in DC between the baseline and follow-up, which is consistent with Bodenmann’s analysis describing the DCI construct and its subscales as being stable over time [8]. Vaske et al. also reported no significant changes of patients’ DC in a sample of chronic obstructive pulmonary disease (COPD) [52]. Concerning IC, our analyses revealed that patients engaged in significantly less active coping at t2 than at t1. This is in line with the results of other follow-up studies on coping behavior among cancer patients [16, 53]. In the present study, t1 coincides with the end of treatment, just after the patients’ lives had been dominated by hospitalization and medical interventions. The patients’ constant interaction with medical specialists providing expert knowledge may have encouraged them to cope in an active manner and with fighting spirit. Six months later, the patients’ need to actively confront their diagnosis may have given way to their desire to “get back to normal” [54, 55]. Considering that the majority (70.1%) of respondents reported no changes in their disease, in terms of remission, relapse or another form, compared with t1, active coping efforts may have been replaced by a growing desire for normalcy in the face of a certain level of health stability.

Regarding correlations of coping constructs over time, the strongest (but still only moderate) associations were found between active IC (t1) and common DC (t2) (r = 0.34) and between common DC (t1) and active IC (t2) (r = 0.34). A high level of common DC suggests a high level of patient-partner interaction and a perception of the cancer diagnosis as a “we-disease” [56]. Such cooperation may enable patients to draw strength to actively deal with their disease even months after their treatment. At the same time, actively coping patients may increasingly involve their partners in coping efforts, leading to greater cohesion within the relationship and the more frequent use of common DC. These considerations are consistent with those of Bodenmann, who found that more active management of a health problem can facilitate open communication between the patient and the partner and contribute to the development of DC efforts [57]. In general, our analyses revealed a link between certain IC strategies considered positive or functional (active IC, compliance IC) with positive DC strategies (stress communication, supportive DC, common DC). This is in line with the observations by Paschali et al. who found that higher engagement in active coping was associated with supportive and common DC in a sample of 92 women recently diagnosed with cancer [27].

With regard to our third research question (c), regression analyses revealed that only two DC styles had a significant predictive impact on t2 IC. First, supportive DC at t1 predicted patients’ compliance (β = 0.20). A possible explanation for this finding is that patients who provide a high level of supportive DC also perceive a high level of responsibility towards their partner for fighting the disease, which leads them to follow therapeutic instructions. Second, t1 common DC significantly predicted active IC (β = 0.25). As assumed before, a high level of common DC may serve as a source of strength for patients, enabling them to deal more actively with the situation rather than resigning themselves to their fate. These observations suggest that compliance and active IC involve the partner more than comparative intrapsychic processes, such as depressive IC or trivializing, in which patients primarily engage with individually.

Regarding sociodemographic variables, older patients were found to trivialize more than younger ones (β = 0.37). The fact that the major part of their lives and accomplishments are already behind them could lead to a certain indifference or a tendency to downplay the disease, as severe diagnoses such as cancer are more likely among older people [30]. Furthermore, the patients in our sample were more compliant with increasing age. This could be related to the “obedience to authority” among older people who may feel morally obligated to follow the instructions of their physician. Patients with high SES were found to trivialize less than those with low or medium SES (β = −0.16). This is consistent with a study from Ouwehand et al. who found that people with high SES use more proactive coping mechanisms than those with low SES, hypothesizing that education enables people to learn to cope proactively by accumulating important resources [58]. People with low SES have been shown to be exposed to significantly more stressors in terms of finances, social relationships, employment, and health than people with higher SES [59, 60]. The living conditions of people with low SES may therefore not allow them to deal intensively with their disease in the face of existential needs, leading to trivializing coping.

Higher relationship satisfaction was found to be a negative predictor of depressive IC (β = −0.17). In another study, high marital quality was found to be negatively associated with depressive symptoms among both women and men [61], suggesting that this may also apply to depressive coping behavior.

Regarding medical variables, the highest predictive impact was found between psychological distress at the baseline and depressive IC at follow-up (β = 0.40). Given that the PHQ-4 used to assess psychological distress includes an ultrashort depression screener (PHQ-2), a high level of psychological distress indicates the presence of depressive symptoms, which in turn could lead to increased use of depressive coping strategies. Furthermore, our analyses indicated that the higher the patients’ psychological distress, the more they trivialize (β = 0.17) and distract themselves from their disease (β = 0.19). This is consistent with previous studies that found a strong relationship between increased emotional distress and the use of coping strategies characterized by avoidance and resignation [62, 63]. Regarding our second medical predictor variable, the present study found that patients who underwent HSCT trivialized significantly less than those who did not receive HSCT (β = −0.23). HSCT is often considered as the only possible cure for patients with haematological neoplasms and requires hospital isolation due to the high risk of life-threatening infections [64]. In addition, serious, sometimes, life-threatening side effects and complications, such as graft-versus-host disease, are associated with the treatment [65]. Close follow-up appointments, active medication management, and strict adherence to a healthy lifestyle are required in the following months, which can be a serious burden for those affected [66]. The constant confrontation with their own mortality and vulnerability is unlikely to allow patients to downplay their situation.

4.1. Strengths and Limitations

Unlike the majority of other studies dealing with coping in cancer patients, the present study is characterized by its longitudinal design permitting us to draw causal conclusion about the relation of DC and IC beyond the snapshot of a cross-sectional design. Furthermore, the sample size of 214 participants is large enough to provide adequate statistical power. As recommended in a previous study [67], our research precisely distinguished between specific coping mechanisms rather than dichotomizing only adaptive/maladaptive or positive/negative coping, thus providing concrete insights into the relationship between specific IC and DC strategies.

Nevertheless, the results of this study cannot be generalized to all types of haematological cancer. Due to the partly very small diagnostic groups in our sample, an evaluation of the results according to individual subtypes of haematological neoplasms was not possible. As haematological neoplasms differ in terms of severity, prognosis, and treatment, drawing further distinctions would be useful in subsequent studies. Regarding the characteristics of the nonresponders at t2, their rate was higher in the subsample with acute leukaemia (32.7%) than in the subsample with chronic leukaemia (10.9%; ). This shows that severe diagnoses tend to be underrepresented in our sample.

4.2. Clinical Implications

The results of this study have practical implications. As supportive and common DC promote the use of IC strategies considered positive (active IC, compliance IC), patients should be encouraged to engage in such positive coping behaviors. Moreover, psycho-oncologists should pay particular attention to the patients’ psychological distress since it emerges as the most common predictor of dysfunctional coping behavior, such as depressive IC or trivialization. Considering that a high level of psychological distress indicates symptoms of anxiety and depression, it may be useful to screen for the latter as part of psycho-oncological treatment. Furthermore, psycho-oncologists should consider patients’ age and SES, as older patients and patients with low SES are more likely to trivialize.

4.3. Conclusions

The present study shows that there are numerous intercorrelations between the two coping constructs and that DC partially predicts IC, namely, that supportive DC leads to increased compliance and common DC to increased active IC. Given these facts, the need for early partnership interventions in favor of improved dyadic and, consequently, improved individual coping becomes clear. Treatment-outcome studies that examine interventions to improve coping skills in couples facing cancer could help establish better psycho-oncological treatment programs. However, the results of the present study suggest that IC in haematological cancer patients is predicted by numerous sociodemographic, partnership-related, and medical factors, whose overall influence outweighs that of DC. In particular, advanced age stood out as a predictor for increased trivialization, psychological distress as a predictor for increased depressive IC, and having had a HSCT as a predictor for decreased trivialization. These findings suggest that the way a person copes with his or her disease is the result of a variety of factors that have not been sufficiently researched. Comparable studies with other cancer types are needed to identify potential further influencing factors on coping behavior and to uncover possible similarities and differences.

Data Availability

The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.

Conflicts of Interest

The authors declare that they have no conflicts of interest.

Acknowledgments

This study was funded by the “José Carreras Leukämie-Stiftung” from 2012 until 2015 (grant: DJCLS R 12/36). Open Access funding was enabled and organized by Projekt DEAL. The authors specially thank all the patients for participating in this investigation.