Neoplastic disease and cancer-related treatment may significantly affect the well-being and quality of patient life. At the beginning, cancer research focused predominantly on the negative psychological changes, i.e., depression or distress (Epping-Jordan et al., 1999), but later studies revealed that cancer patients may also experience positive changes known as ‘post-traumatic growth (PTG)’ (Danhauer et al., 2013, 2015; McDonough et al., 2014). Although positive and negative perceived psychological changes in cancer continue to be investigated, only a few authors attempted to analyze them simultaneously (Schroevers et al., 2011). Furthermore, different approaches are often adopted, rendering it impossible to compare the findings. Thus, the existence of specific predictors for the perceived direction of change remains to be elucidated. In this study, we attempted to address this matter by testing whether perceived positive (post-traumatic growth, PTG) and negative (post-traumatic depreciation, PTD) changes in women after mastectomy have unique or shared predictors in terms of coping strategies and effectiveness of social support attempts called ‘social support effectiveness’ (SSE). Moreover, by applying a person-centered approach, which assumes that a population is heterogeneous in relation to the effect of the predictors on the outcome (Laursen and Hoff, 2006), we tested whether it was possible to identify homogeneous subgroups with particular intensity of both, PTG and PTD symptoms, and whether these subgroups differ in terms of coping strategies and SSE.

Negative changes in breast cancer (BC) women often go beyond the sphere of physical health, as the disease may also cause psychological problems and increase social demands. Various authors have demonstrated BC, and especially mastectomy, to be associated with negative feelings, including depressive symptoms, anxiety, distress (Epping-Jordan et al., 1999; den Heijer et al., 2012), post-traumatic stress symptoms (Hahn et al., 2015), lower quality of life (King et al., 2000; Janz et al., 2005), and changes in personal and social beliefs (Andrzejczak et al., 2013).

However, an accumulating body of evidence indicated that positive psychological effects may also occur during or after BC treatment (Sears et al., 2003; Bozo et al., 2009; Chan et al., 2011; Danhauer et al., 2013; Ruini et al., 2013; McDonough et al., 2014; Soo and Sherman, 2015). These positive changes, which are considered to be the result of coping with a traumatic event, are labeled as ‘PTG’ (Tedeschi and Calhoun, 2004). Data on the relationship between PTG and other well-being measures remain inconsistent (Sumalla et al., 2009; Ruini et al., 2013; Soo and Sherman, 2015). A recent meta-analysis of cancer patients revealed weak negative associations between PTG and depression and distress, weak positive associations with optimism, and no correlation with anxiety or physical quality of life (Shand et al., 2015), which calls into question the status of PTG: it remains unclear whether PTG and psychopathology constitute two ends of the same continuum or whether it is unrelated to adjustment. PTG has evoked much controversy regarding its nature and scope (Sumalla et al., 2009), chief among them the fact that most measuring tools include questions only about the positive (subjective) changes. As such, they fail to reflect the full scope of experiences and feelings of the respondents (Park and Lechner, 2014). Positive and negative experiences in any given set of circumstances, even within the same domain (e.g., social), are not mutually exclusive (Park and Lechner, 2014). Negative changes, which are considered to be the result of coping with a traumatic event, have been defined as PTD (Baker et al., 2008). Baker et al. (2008), designed a PTD measure based on the existing PTG assessment (PTGI; Tedeschi and Calhoun, 1996). Identical factors are involved in the measurement of both, perceived positive and negative changes as the effects of a stressful encounter, addressing the same domains and reducing the likelihood of positive or negative response bias (Baker et al., 2008). To the best of our knowledge, PTD in women with BC has not been investigated. Studies in other groups of patients demonstrated both constructs to co-exist (Purc-Stephenson et al., 2015), independently (Cann et al., 2010; Forgeard, 2013), and PTD to have good construct validity (Kunz et al., 2016).

Since PTG and PTD measure distinct and independent dimensions, it seems valid to check whether the conditioning mechanisms also differ. Perhaps separate factors are associated with positive and negative psychological changes. Successful identification of these factors might broaden our knowledge of PTG and PTG, resulting in more adequate preventive strategies. According to the abovementioned definition, these changes should be the result of coping with a traumatic event, which makes their relationship with the coping strategies particularly interesting. Various studies have indicated that in women with BC or after mastectomy, PTG is positively associated with problem–focused coping (active coping, planning, and seeking instrumental support) and positive emotion–focused coping (positive reframing, humor, seeking emotional support, turning to religion, and positive rumination), while being negatively related to negative emotion–focused coping (brooding) (Chan et al., 2011; Danhauer et al., 2013; Soo and Sherman, 2015). Similar findings have been reported by studies on different cancers (Schroevers et al., 2011; Tallman, 2013). Regardless, analyses including all coping strategies demonstrate that positive emotion–focused group (Schroevers et al., 2011) and problem–focused coping strategies (Tallman, 2013) were the best predictors. A recent meta-analysis of PTG correlates in 70 cross-sectional studies in cancer patients revealed its moderate relationship with positive reframing and religious coping (Shand et al., 2015). Similar dependence can be found in short-term (Scrignaro et al., 2011) and long-term longitudinal studies (Sears et al., 2003; Danhauer et al., 2013, 2015).

The relationship between PTD and coping remains to be fully elucidated as data are scarce. So far, the main focus has been on the relationship between PTD and rumination, indicating a positive link between PTD and intrusive rumination (Cann et al., 2010; Forgeard, 2013). In another study, PTD was positively related to self-distraction but unrelated to positive reframing, acceptance, and active coping (Schroevers et al., 2011). Taken together, the data suggest that negative perceived changes are related to negative emotion–focused and avoidance–focused coping.

Social support constitutes another important factor as far as positive and negative psychological changes in cancer patients are concerned (Shand et al., 2015). Social support is considered be to an important resource in the process of coping with adversity (Lazarus and Folkman, 1984) and, together with coping strategies, could be a significant predictor of PTG or PTD. Positive perceived changes in women with BC or post-mastectomy are linked with more perceived (Schroevers et al., 2010; Danhauer et al., 2013; Soo and Sherman, 2015) and received social support (Bozo et al., 2009; Scrignaro et al., 2011; McDonough et al., 2014). These effects were observed in a number of cross-sectional and longitudinal studies, and are supported by review (Helgeson and Cohen, 1996) and meta-analytic studies (Shand et al., 2015). Regardless, contradictory evidence has also been reported. According to a longitudinal study on predictors of PTG patterns by Danhauer et al. (2015), BC women with the highest as well as the lowest levels of PTG reported the highest perceived social support. Perceived or reported received support is not equivalent to received effective support (Rini and Dunkel-Schetter, 2010). Ineffective support may diminish the ability to cope with the diagnosis and treatment, and hinder adaptation to disease. Cancer patients who report the received support to be satisfactory adapt to cancer better (Manne and Badr, 2010). On the other hand, there is evidence that dissatisfaction with the received support did not predict cancer-related positive changes 8 years later (Schroevers et al., 2010). Nevertheless, our knowledge about SSE remains insufficient to determine whether it is related solely to adaptation or also to growth through adversity.

Our cross-sectional study attempted to address the abovementioned issues. We analyzed both, PTG and PTD within the same domains. As positive and negative perceived changes can coexist in cancer patients, their interrelationships were investigated. We expected that both constructs to be independent and that the population of women with BC or post-mastectomy to be heterogeneous in terms of perceived positive and negative psychological changes, i.e., different individuals may experience different proportion of both, positive and negative changes. This is in line with the person-centered approach. In contrast to variable-centered approach, where a population is assumed to be homogeneous in relation to the effect of predictor on the outcome, the person-oriented approach assumes that a population is heterogeneous in this respect (Laursen and Hoff, 2006). We also analyzed the mechanisms of differentiation for PTG and PTD. Specifically, the aims of the study were to investigate whether (1) coping strategies and SSE were specific or shared predictors of PTG and PTD in post-mastectomy women; (2) sample was heterogeneous in terms of PTG and PTD levels, and whether it was possible to identify homogeneous subgroups with different levels of both, PTG and PTD symptoms, and (3) these subgroups would differ in terms of the coping strategies and SSE. We hypothesized that PTG would be positively associated with problem-focused coping, positive-emotion focused coping and SSE, and negatively associated with negative-emotion-focused coping or avoidance coping, in contrast to PTD. The analysis of predictor character (specific or shared) was of an exploratory nature. We also hypothesized that several subgroups with different levels of PTG and PTD would be found, and that coping strategies and SSE would be linked do different PTG × PTD classes. Specifically, we expected that high problem-focused coping, positive-emotion focused coping and SSE and low negative-emotion-focused coping or avoidance coping to predict inclusion into high PTG subgroups. Opposite associations were expected for high PTD subgroups.

Descriptive statistics and correlations between the variables are presented in Table 1. The level of PTG was significantly higher as compared to PTD (t₈₃ = 14.69, p < 0.001). PTG and PTD were not significantly correlated.

The aim of the study was to investigate the similarities and differences between PTG and PTD predictors in women after mastectomy. The heterogeneity in the perceived changes, i.e., identification of homogeneous classes with different PTG and PTD scores and class membership correlates, was also addressed. The following significant unique predictors were revealed: time since diagnosis and positive emotion–focused coping were predictors for PTG, while emotion– and avoidance–focused coping and low SSE were predictors for PTD. The study participants were subdivided into four classes, based on the interaction of high vs. low PTG and PTD. However, only emotion–focused coping differentiated class membership.

Post-traumatic growth and PTD were found to be uncorrelated, which is consistent with the findings of earlier studies (Cann et al., 2010; Forgeard, 2013). The results suggest these variables to be independent dimensions, with separate trajectories of development in the cognitive and behavioral sphere. They also are consistent with data on the independence of positive and negative affect (Larsen and McGraw, 2011), or positive and negative processes, thus supporting the theory of multipolar rather than bipolar adjustment properties. In consequence, coping with traumatic events might result in both, growth and depreciation, or either, or neither.

Another finding of our study is that fact that not all women experienced psychological changes following mastectomy in the same manner. Four different classes of change were identified in the post-mastectomy patients. A little over half of the women experienced only growth, with the remaining subjects reporting no change, mixed perceived changes, or only negative perceived changes. Interestingly, an analysis of the entire sample (based on mean scores according to a variable-centered perspective where the assumption is homogeneous of population (Laursen and Hoff, 2006) revealed that the affected women had higher PTG than PTD, which may indicate excellent adaptation to mastectomy. However, this does not account for approximately 14% of women with only PTD, 15.5% with both, PTG and PTD, and 18% with neither PTG nor PTD. In fact, psychological changes following mastectomy varied with each individual, indicating that the group lacked homogeneity in terms of variables. Taking that into account, it is possible to avoid the mistake of psychological change-related generalizations about the entire group. It also provides proof that PTD is not a simple opposite of PTG and that both these variables constitute largely independent dimensions.

The first group had high PTG and low PTD scores, signifying the highest potential for personal development in that group. Also, these women were probably the most motivated to introduce changes in their personal life, including compliance with the doctor’s orders and consistent care about their health and well-being. The second group had low PTG and low PTD scores. That group experienced the least significant consequences at the cognitive-emotional level. No perceived change might indicate ‘hedonic adaptation’ or ‘satisfaction treadmill’, i.e., returning to baseline, pre-trauma well-being (Lyubomirsky, 2011). Women from the third group had relatively high PTG and high PTD, and were most probably in the process of attaining mental balance. As such, they might alternately experience symptoms of PTG and PTD. Their scores indicate significant dynamics of the cognitive-emotional processes. An analysis of the mean results in the subgroups showed that the group with mixed perceived changes had the highest PTG and PTD scores. Coexisting PTG and PTD provide evidence for the high complexity of the coping processes, and the need for a comprehensive analysis of the potential changes, for example using the person-centered approach (Laursen and Hoff, 2006). The results might offer supporting arguments for the debate on the real vs. illusory facet of PTG (Maercker and Zoellner, 2004; Zoellner et al., 2008; Frazier et al., 2009; Pat-Horenczyk et al., 2015; Lahav et al., 2016). Women from the fourth group had low PTG and high PTD, and experienced the highest level of discomfort. The perception of high PTD, together with relatively low PTG symptoms might be considered a proof of the positive–negative asymmetry, i.e., negativity outweighing positivity (Baumeister et al., 2001).

Another finding to support the independence of PTG and PTD was that their correlates—in terms of coping strategies and SSE—were unique. Higher positive emotion–focused coping was indicative of PTG, while negative emotion–focused and avoidance–focused coping, as well as low SSE, predicted PTD, after controlling for demographic and medical-related factors, which is consistent with earlier studies (Cann et al., 2010; Schroevers et al., 2011; Forgeard, 2013; Tallman, 2013). However, these results do not support data on the relationship between PTG and problem–focused coping. Thus, it seems safe to conclude that both ‘results of coping’ had their own, specific determinants. As expected, PTG was positively related to behaviors such as positive reframing, acceptance, and religion, which was confirmed in a previous meta-analysis (Shand et al., 2015). Such behaviors, often referred to as ‘adaptive,’ allow for a different perception of traumatic events. The affected patients perceive them as meaningful. PTD, in turn, was positively correlated with maladaptive strategies, i.e., evoking or supporting negative emotions or diverting one’s attention from a problematic situation, for example by denial. Interestingly, only maladaptive strategies differentiated homogeneous group membership with different PTG and PTD scores. Such behavior pattern was characteristic for women with low PTG and high PTD, thus allowing to predict depreciation rather than growth. Our findings are consistent with earlier studies which confirmed the existence of the negativity bias (Baumeister et al., 2001), implying that negativity is more powerful and thus manifests itself more strongly.

Social support effectiveness effects may be treated in a similar manner. Low effectiveness of social support attempts positively correlated with the negative outcomes, i.e., PTD, which is consistent with earlier findings (Manne and Badr, 2010; Rini and Dunkel-Schetter, 2010). It suggests that SSE is more closely related to the return to baseline well-being (adaptation) or with worsening function (depreciation) than growth. However, a comparison of homogeneous groups did not support that hypothesis. The role of SSE in the coping process requires further analysis.

As for background variables, longer time since diagnosis was associated with PTG, which is consistent with other reports (Sears et al., 2003; Tedeschi and Calhoun, 2004; Danhauer et al., 2013). No relationship was found between demographic and medical-related variables and homogeneous subgroup membership. Our results might also be related to other variables which are relevant to post-traumatic changes, such as event centrality (Johnson and Boals, 2014) or self-efficacy (Hobfoll et al., 2007), but which were not accounted for in this study.

Our study is not without limitations. Due to the cross-sectional design of the study, we were not able to draw conclusions about the cause-and-effect relationships between the investigated variables. Furthermore, we lacked data on the pre-cancer levels of well-being of the study participants. There was also a relatively large variability in terms of time since diagnosis and time since mastectomy. Furthermore, the evaluation of the changing direction in growth and depreciation was subjective and, thus, distorted. The shortcomings of PTGI are well-known (Frazier et al., 2009), and the same applies to the measurement of PTD. Moreover, the sample size, especially in case of the MLR analyses, could have been insufficient, resulting in underpowered results and failure to identify significant effects. These limitations, along with the lack of analyses of specific domains of PTG and PTD, indicate the necessity of further studies.

Despite these limitations, we are of the opinion that our findings have cast new light on the understanding and practical implications of positive and negative perceived changes following mastectomy. The results may lead to the development of more tools to provide adequate psychological support for women after mastectomy, with special focus on the reduction of maladaptive coping strategies. Apart from developing strategies to prevent PTD or reduce its intensity, our study may raise awareness about the diversity of psychological reactions in patients, thus preventing stereotypical and routine approach to their problems. Particularly women from the mixed (high PTG/PTD) and the high PTG groups should receive psychological care and counseling. The former are most probably in the process of attaining mental balance and, thus, ought to be offered access to a psychologist who would facilitate constructive changes in their functioning. The latter, on the other hand, experience the highest level of mental discomfort and, as such, ought to receive psychological care and psychiatric consultation.

Substantial contributions to the conception or design of the work (AK); or the acquisition, analysis (AK), or interpretation of data for the work (AK, KB-M, and MK). Drafting the work or revising it critically for important intellectual content (AK, KB-M, and MK). Final approval of the version to be published (AK, KB-M, and MK). Agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved (AK, KB-M, and MK).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.