Social Comparisons and Quality of Life Following a Prostate Cancer Diagnosis

Abstract

Purpose:

The objective was to explore the relationships among cognitive appraisals of prostate cancer (challenge, threat, and harm/loss), social comparisons, and quality of life in men previously diagnosed.

Design, Sample, & Methods:

Men who had participated in prostate cancer support groups completed a cross-sectional questionnaire (N=189). Multivariable linear regression was used to evaluate social comparisons as mediators of quality of life while controlling for uncertainty and optimism.

Findings:

Positive and negative social comparisons were parallel mediators of the relationships between challenge or threat appraisals and quality of life, while only negative social comparisons mediated the relationship between harm/loss appraisals and quality of life.

Conclusions:

These findings demonstrate the importance of social comparisons in accounting for the effect of cognitive appraisals of prostate cancer on quality of life among men in support groups.

Implications for Psychosocial Providers:

Interventions to improve quality of life could address reduction of maladaptive comparisons, a strategy that could be tailored based on the patient’s appraisal of prostate cancer.

Background

Prostate cancer affects about one in seven men during his lifetime, and is the most common type of cancer among men in the United States (Siegel, Miller, & Jemal, 2016). Due to variability in the stage at which tumors are detected, the speed at which tumors progress, and the efficacy of treatments across individuals, there is often disagreement among clinicians as to an individual’s medical management of prostate cancer (Walsh & Worthington, 2007). Whether the patient and his physician opt for surgery to eradicate the cancer or an active surveillance approach, it is important to understand how these individuals experience life post diagnosis.

It has been demonstrated through population-based surveys that cancer survivors often experience an array of continued health problems including, for example, lesser physical and psychological well-being (Elliot et al., 2011). Many men diagnosed with prostate cancer seek the psychosocial support of other prostate cancer patients through in-person or online support groups or education networks. In addition, men may turn to these groups to fulfill informational needs regarding treatment options or to learn how to cope with the side effects of prostate cancer and treatment (Weber, Roberts, & McDougall, 2000). It has been shown that coping strategies play a key role in long-term adaptation to life with prostate cancer, as does the manner in which one judges and interprets the implications of prostate cancer, i.e., cognitive appraisals (Wooten et al., 2007).

Lazarus and Folkman (1984) developed the Transactional Model of Stress and Coping to explain how individuals react to stressful events in their lives, such as receiving a cancer diagnosis. This theory postulates that a stressful situation can be appraised as a harm/loss, a threat (a potential harm/loss that has not yet come to fruition), or a challenge that offers the possibility of growth or benefit. While some cancer-related stressors can be quite certain (e.g., objective measures of physical state), uncertainty pervades many aspects of life post diagnosis. One theory that speaks to the role of uncertainty in the context of coping is Social Comparison Theory (Festinger, 1954), which posits that an individual will compare himself with others in the absence of objective information. Social comparisons (SCs) may be used by individuals to reduce the sense of uncertainty produced by a threatening diagnosis such as cancer (Bellizzi, Blank, & Oakes, 2006), as they provide the individual with a standard by which he can evaluate his own status in terms of, for example, his psychological well-being and medical management.

The psychological outcome of a SC depends on two factors: the direction and the interpretation of the comparison. The direction of an individual’s comparison can be upward—when the target of comparison is doing better than the individual on the relevant dimension, or downward—when the target is doing worse. Interpretation refers to whether an individual identifies with the target of comparison—concentrating on the ways in which he is or may become similar to the target, or contrasts himself with the target—focusing on the differences between himself and the target (Van der Zee, Buunk, Sanderman, Botke, & Van den Bergh, 2000). Positive affective response (self-enhancement) most often results from upward identification and downward contrast; thus, these two forms of SC may be collectively referred to as positive SCs. Inversely, downward identification and upward contrast more often result in negative affective responses (self-defeat) and may be collectively referred to as negative SCs.

Importantly, it has been demonstrated that SCs can have a lasting impact on subjective quality of life (QoL; Buunk et al., 2012), which has been defined as “a person’s sense of well-being that stems from satisfaction or dissatisfaction with areas of life that are important to him/her” (Ferrans & Powers, 1992). Previous research has highlighted the influence of individual differences (e.g., current and future health expectations) on the interaction between SCs and QoL (Brakel, Dijkstra, & Buunk, 2014). One potential confounder of these relationships is dispositional optimism. Specifically, it has been suggested that prostate cancer patients with higher levels of optimism may find it easier to make treatment decisions and may be more satisfied with their choice (Orom et al., 2009).

The purpose of the current study was to better understand psychosocial predictors of QoL among men who participated in prostate cancer support groups, especially in regard to the role played by SCs. Informed by the aforementioned theories and recent research on their application, we specified the following hypotheses: (1) higher levels of uncertainty would be associated with a greater tendency to engage in all forms of SC; (2) higher levels of dispositional optimism would be associated with more positive SCs and less negative SCs; (3) the primary hypothesis, that SCs would be parallel mediators of the relationships between cognitive appraisals and QoL.

Method

Sample and Design

Eligible participants were men who had been diagnosed prior to age 70 with localized prostate cancer (stage T1b, T1c, or T2), and had never been diagnosed with another primary cancer other than non-melanoma skin cancer. Participants were recruited through support group meetings of the Us TOO Prostate Cancer Education and Support Network, emails sent to support group members from local facilitators and from the national Us TOO organization, and announcements in Us TOO newsletters.

This study was approved by the National Human Genome Research Institute (NHGRI) Institutional Review Board (Protocol #10-HG-N172).

Measures

Demographic Information was collected including age, race (American Indian or Alaska Native; Asian; Black or African American; Native Hawiian or other Pacific Islander; White; Other), ethnicity (Hispanic or Latino; Not Hispanic or Latino), marital status, income, and highest level of education. The time since initial diagnosis, stage of prostate cancer at initial diagnosis, mode of treatment for initial diagnosis, recurrence of prostate cancer (Y/N), mode of treatment for recurrence (if applicable), and family history of prostate cancer were recorded. This information was used to evaluate the variables’ potential as confounders in regression analyses and to assess external validity with published literature.

Dispositional Optimism was assessed using the Life Orientation Test-Revised (LOT-R; Scheier, Carver, & Bridges, 1994). Developed as a tool to evaluate individual differences in optimism, the LOT was revised slightly to orient each of the items to the future. This revised version was utilized, and consists of 10 items (four of which are fillers) formatted on a five-point Likert-type response scale ranging from “I agree a lot” (1) to “I disagree a lot” (5). Summed scores range from 6–30, with higher scores indicative of greater dispositional optimism. Cronbach’s alpha has been reported at .80 among prostate cancer patients assessed shortly after diagnosis (Steginga & Occhipinti, 2006), and was .87 in this study.

Perceived Uncertainty was assessed using the Mishel Uncertainty in Illness Scale-Community Form (MUIS-C; Mishel, 1997). This scale was developed for use among individuals living in their communities with a chronic condition, or with an acute but treatable disease that presents a possibility of recurrence. It is a 23-item Likert scale allowing for responses ranging from “strongly agree” (1) to “strongly disagree” (5). Summed scores range from 23–115, with higher scores indicative of greater uncertainty. While validity has only been demonstrated for the original MUIS, reliability of the MUIS-C has been demonstrated among prostate cancer patients with a Cronbach’s alpha of .91 (Bailey et al., 2011); it was .92 in this study.

Cognitive Appraisals were assessed using the modified Cognitive Appraisal of Health Scale (Kessler, 1998). Based on the Transactional Model of Stress and Coping, the original 28-item scale was developed to measure appraisals of a health-related event. Each of the five-point Likert items allows for responses ranging from “strongly disagree” (1) to “strongly agree” (5). Through investigation of the psychometric properties of the scale among prostate cancer patients, Ahmad (2005) demonstrated improved validity and reliability of a 13-item version of the measure consisting of three subscales: challenge (three items, α=.70), threat (five items, α=.74), and harm/loss (five items, α=.79). This abbreviated version was used in the present study, and the Cronbach’s alphas were .68 for challenge, .82 for threat, and .85 for harm/loss. Higher scores on each subscale are indicative of greater agreement with the appraisal.

Social Comparisons were assessed using the Identification/Contrast scale (Van der Zee et al., 2000), which measures four types of SC: upward identification, upward contrast, downward identification, and downward contrast. It consists of 12 Likert-type items formatted on a five-point response scale ranging from “not at all” (1) to “strongly” (5) how you feel or what you think. In addition to the four-factor model, validity of the scale has been demonstrated for a second-order, two-factor model in which upward identification and downward contrast are reduced to one factor (positive) and upward contrast and downward identification are reduced to a second factor (negative; Van der Zee et al., 2000). An exploratory factor analysis (EFA) was conducted on data from this study to assess whether the four types of SC could be adequately summarized by a lesser number of composite factors. Scores for each composite factor range from 6–30, with higher scores indicating a greater tendency to compare oneself to others. The Cronbach’s alphas for this study were .82 for positive SCs and .87 for negative SCs.

Quality of Life was assessed using the 66-item Quality of Life Index-Cancer Version III (QLI-CA; Ferrans, 1989). Participants rate their satisfaction with certain aspects of their lives on 33 Likert-type items formatted on a six-point response scale ranging from “very dissatisfied” (1) to “very satisfied” (6). They then indicate the importance they ascribe to each aspect on 33 Likert-type items formatted on a six-point response scale ranging from “very unimportant” (1) to “very important” (6). An overall QoL score across the four domains (health and functioning, psychological/spiritual, social/economic, family) was computed by weighting each satisfaction score by the corresponding importance score. Total scores range from 0–30, with higher scores indicative of better QoL. This version is similar to the original QLI, which has demonstrated acceptable construct and content validity as well as internal reliability with Cronbach’s alphas for the overall score ranging from .73–.99 across 48 studies (Ferrans and Powers, 2016). The Cronbach’s alpha was .94 for this study.

Statistical Analysis

The enrollment target was set at 155 participants, which power analysis indicated would allow detection of an 8% or greater contribution of the SC variables to the total variance in QoL with 80% power. Data were analyzed using SPSS 18.0. A maximum of 20% missing data was tolerated for each scale. All variables were treated as continuous; all p-values presented are two-tailed. Underlying second-order factors among the four types of SCs were evaluated using EFA with Varimax rotation if Barlett’s test of sphericity was significant (p<.05) and the Kaiser-Meyer-Olkin (KMO) measure of sampling adequacy >.5. Bivariate associations among all demographics, covariates, predictors, and outcome variables were tested using a Pearson’s correlation, t-test, or analysis of variance (ANOVA). Any variables significantly associated with outcome variables (p≤.2) were entered into a stepwise model selection procedure using backward elimination. The resulting multivariable linear regression models were used to evaluate predictors of SCs and QoL.

PROCESS for SPSS (version 2.15; Hayes, 2013) was used to evaluate three dual mediator models: each involved a single cognitive appraisal (challenge, threat, or harm/loss) as a predictor, significant demographic predictors, uncertainty, optimism, positive SCs and negative SCs as parallel mediators, and QoL as the outcome variable. Finally, bias-corrected 95% confidence intervals derived from 1000 bootstrap resamples were used to quantify the indirect effects in each model.

Results

During the recruitment period (August to December 2010), 162 individuals began the questionnaire online, and 27 individuals returned paper questionnaires by mail. Of these 189 individuals, 164 responded sufficiently to all components of the questionnaire.

Sample Characteristics

The average age of participants was 65±8 years and ranged from 44–87 years (n=162). The vast majority of participants who reported race (n=171) identified as only white (96.5%), five identified as only African American (2.9%), and one identified as white and African American (0.6%). All participants who reported ethnicity identified as non-Hispanic (n=171). Other demographic characteristics are shown in Table 1 (the marital status variable was dichotomized for analysis given the small proportion of respondents who were not married or in a marriage-like partnership). Notably, this sample was largely comprised of individuals who were highly educated and diagnosed with early stage disease.

Table 1.

Participant Demographics

Characteristic	Response Option	n	%
Income (n=161)	Less than $30,000	7	4.3
	$30,000–$50,000	22	13.7
	$50,001–$70,000	26	16.1
	More than $70,000	106	65.8
Marital Status (n=169)	Married or in a Marriage-like Partnership	142	84.0
	Not Married or in a Marriage-like Partnership	27	16.0
Highest Level of Education (n=169)	High School / GED	6	3.6
	Some College	26	15.4
	Bachelor’s Degree	55	32.5
	Post-Graduate Degree	80	47.3
	Technical School	2	1.2
Stage of Cancer at Diagnosis (n=147)	T1b	4	2.7
	T1c	105	71.4
	T2a	12	8.2
	T2b	7	4.8
	T2c	19	12.9
Time Since Diagnosis (n=167)	0–2 years	45	26.9
	2–5 years	50	29.9
	More than 5 years	72	43.1
Recurrence (n=168)	Recurrence of prostate cancer	28	16.7
	No recurrence of prostate cancer	140	83.3

Exploratory Factor Analysis

The results from the EFA of the four types of SC were consistent with previous reports (Van der Zee et al., 2000): the upward contrast and downward identification variables loaded highly on a factor denoted as negative SCs (loadings of .87 and .86, respectively), while the upward identification and downward contrast variables loaded highly on a factor denoted as positive SCs (loadings of .86 and .78, respectively). None of the SC variables loaded onto both factors. The overall KMO=.51, Bartlett’s test of sphericity was significant (p<.001), and the two composite factors captured 76% of the variance in the four types of SC. Thus, for the rest of the analyses, the four types of SC were collapsed into two factors: positive SCs and negative SCs.

Bivariate Analyses

The only significant difference in key variables across demographic characteristics was higher QoL among those who were married or in a marriage-like partnership (t=2.194, p=.036). Associations among key variables are presented in Table 2. As shown, uncertainty was inversely associated with positive SCs and directly associated with negative SCs, while optimism was directly associated with positive SCs and inversely associated with negative SCs. In turn, positive SCs were associated with higher QoL, whereas negative SCs were associated with lower QoL.

Table 2.

Correlations Between Key Variables

Variable	possible score	mean (SD)	Positive Social Comparisons	Negative Social Comparisons	Quality of Life
Challenge (n=170)	3–15	10.8 (2.6)	.299 (p<.001)	−.330 (p<.001)	.272 (p<.001)
Threat (n=167)	5–25	14.6 (4.6)	−.292 (p<.001)	.549 (p<.001)	−.461 (p<.001)
Harm/Loss (n=164)	5–25	14.1 (5.0)	−.289 (p<.001)	.476 (p<.001)	−.513 (p<.001)
Uncertainty (n=170)	23–115	56.0 (10.6)	−.194 (p=.011)	.446 (p<.001)	−.474 (p<.001)
Optimism (n=169)	6–30	22.9 (5.3)	.341 (p<.001)	−.413 (p<.001)	.477 (p<.001)
Positive Social Comparisons (n=186)	6–30	10.9 (2.4)	1	−.137 (p=.064)	.351 (p<.001)
Negative Social Comparisons (n=185)	6–30	6.2 (2.5)	—	1	−.510 (p<.001)
Quality of Life (n=186)	0–30	19.1 (2.0)	—	—	1

Multivariate Analyses

Three regression models (shown in Table 3) include as predictors uncertainty, optimism, and marital status. These predictors collectively accounted for 13.5% of variance in positive SCs, 29.2% of variance in negative SCs, and 38.0% of variance in QoL. These regression models were then used to evaluate positive SCs and negative SCs as parallel mediators (Figure 1): challenge, threat, and harm/loss were analyzed as separate predictors while controlling for aforementioned confounders by adding them individually to each of these three regression models.

Table 3.

Multivariable Linear Regression Models

Predictor	Outcome
	Positive Social Comparisons		Negative Social Comparisons		Quality of Life
	B (SE)	β	B (SE)	β	B (SE)	β
Uncertainty	−0.025 (0.016)	−0.114	0.086 (0.016)	0.363‡	−0.068 (0.012)	−0.372‡
Optimism	0.141 (0.033)	0.324‡	−0.153 (0.033)	−0.322‡	0.133 (0.024)	0.362‡
Marital Status	−0.287 (0.460)	−0.046	0.455 (0.455)	0.067	0.806 (0.327)	0.153*
[Constant]	9.401 (1.375)		4.543 (1.363)		19.264 (0.980)
	R2=.135 F(3,162)=8.45 p<.001		R2=.292 F(3,161)=22.12 p<.001		R2=.380 F(3,162)=33.15 p<.001

Note:

Significance at p < .05 is denoted by *,

at p < .01 by ^†,

and at p < .001 by ^‡.

Figure 1.

This model depicts positive and negative social comparisons as dual parallel mediators of the relationships between cognitive appraisals and quality of life. The indirect effect(s) of cognitive appraisals on quality of life are represented by the product(s) of a and b paths, while the direct effect is represented by the c’ path. The total effect of cognitive appraisals on quality of life is represented by the c path.

While challenge and threat were significant predictors of both positive SCs and negative SCs (paths a₁ and a₂, respectively), harm/loss was only a significant predictor of negative SCs. Additionally, both positive SCs and negative SCs were significant predictors of QoL in all three models (paths b₁ and b₂, respectively). Unstandardized path coefficients are presented on the model in Figure 1. Note that unstandardized indirect effects can be calculated by subtracting c’ from c. Standardized indirect effects (i.e., the standardized products of corresponding paths a and b) are presented in Table 4.

Table 4.

Standardized Indirect Effects of Social Comparisons

Predictor	Indirect effect [95% Confidence Interval]
	Total	Positive Social Comparisons	Negative Social Comparisons
Challenge	.102 [.038, .195]	.052 [.013, .119]	.050 [.010, .111]
Threat	−.117 [−.224, −.044]	−.036 [−.094, −.005]	−.080 [−.170, −.022]
Harm/Loss	−.102 [−.190, −.035]	−.030 [−.085, .008]	−.072 [−.145, −.019]

Discussion

Social Comparisons

Higher uncertainty was hypothesized to be associated with increased use of all types of SCs, as reducing uncertainty is arguably the primary motive underlying SC. Contrary to this hypothesis, men participating in prostate cancer support groups made more negative SCs and less positive SCs in conditions of greater uncertainty. As most participants had been diagnosed at least two years prior, those who reported high levels of uncertainty might have been unique in their prognosis or objective health in ways that would lead them to have a more negative perception of their status relative to others. It has been shown that in some contexts distressed individuals are less likely to engage in SCs such as emphasizing relative benefits of a stressful situation (Buunk & Ybema, 1995). Additionally, van der Zee and colleagues, among others, have demonstrated that cancer patients with high levels of neuroticism derive less benefit and more consequence from SCs (Van der Zee, Oldersma, Buunk, & Bos, 1998). These findings highlight the importance of evaluating additional predictors and outcomes of SCs, as well as how these relationships may be moderated across contexts.

The data supported the hypothesis that higher levels of dispositional optimism would be associated with both more positive SCs and less negative SCs. Additionally, EFA suggested that among men in this study, the four types of SC could be reduced to two underlying factors. Few other studies have quantitatively measured SCs, and among those that have, there has been little consistency in the SC measure. Although not one of its primary aims, the present study contributes to the SC literature by providing further empirical evidence for this two-factor structure of the Identification/Contrast scale. These findings further substantiate the incorporation of these distinct constructs into existing theory. Future studies should consider other variables including, for example, motives for employing SCs such as learning coping strategies through observation and modeling (Bennenbroek, Buunk, Van der Zee, & Grol, 2002). Associations among motives and other related variables may guide clinicians in tailoring their recommendations for effective coping and improving QoL.

Quality of Life

Challenge appraisals and positive SCs significantly predicted higher QoL. Conversely, threat and harm/loss appraisals significantly predicted lower QoL, as did negative SCs. The present study is the first to provide quantitative evidence of these relationships among men with prostate cancer, using a broad measure of QoL to encompass many domains of life. The observed relationship between positive SCs and QoL is also consistent with Downward Comparison Theory, which postulates that downward (positive) comparisons may help to restore one’s sense of well-being after a health threat (Wills, 1981).

Mediation

Positive SCs and negative SCs were hypothesized to be parallel mediators of the relationships between three cognitive appraisals and QoL. While indirect effects in the challenge model indicate that they were significant dual mediators of the relationship between appraisal of prostate cancer and QoL, direct and indirect effects with opposite signs suggest inconsistent mediation (MacKinnon, Fairchild, & Fritz, 2007). Essentially, while appraising one’s prostate cancer as a challenge might have been associated with a slightly negative direct effect on QoL, the total effect on QoL might have been positive due to robust indirect effects through more positive SCs and less negative SCs. Given this highly optimistic sample of men with predominantly early stage disease and successful treatment, it is conceivable that the total effect of a challenge appraisal on QoL could have been positive despite the stress of prostate cancer. In the threat model, on the other hand, the direct effect was not significantly different from zero, which suggests that SCs completely mediated the relationship between a threat appraisal and QoL. Finally, only negative SCs were found to be a unique mediator in the harm/loss model–i.e., the effect on QoL of appraising one’s prostate cancer as a harm/loss occurs in part through more negative SCs, and is not accounted for by positive SCs. This can be understood in light of the fact that negative illness representations have been shown to be associated with psychological distress, anxiety, and depression (Dempster et al., 2011).

Study Limitations

As the majority of participants were white, educated, and of a high socioeconomic status, further investigation in a more diverse sample is warranted. The incidence of prostate cancer among African Americans is significantly higher than among whites, and the mortality rate for African Americans is more than double that for whites (Howlader et al., 2015). Accordingly, as African American men with prostate cancer may view the disease as a greater threat to their health and their lives, these relationships may be quite different among other ethnicities. Additionally, the period immediately following diagnosis presents unique challenges that may not have been salient to many of the participants in our study: most had been diagnosed more than two years prior, and many more than five years prior. This study was further limited to men with localized prostate cancer. Thus, the results are not necessarily representative of men with a recent prostate cancer diagnosis or metastatic cancer.

Recruitment was conducted exclusively through support groups. This limits the generalizability of the findings, as few men who have been diagnosed with prostate cancer choose to participate in prostate cancer support groups (Weber, Roberts, & McDougall, 2000). In particular, selection bias may partially explain why positive SCs were more prevalent than negative SCs in our sample. In addition, among men who reported making negative SCs, there was a large amount of variability in reported QoL, and responses from a minority of participants seem to have negatively skewed the QoL results. Thus, the observed relationships among positive SCs, negative SCs, and QoL may not be entirely representative of men who attend prostate cancer support groups.

Clinical Implications

The present study provides evidence that, among men who participate in support groups following a diagnosis of localized prostate cancer, the SCs that a man makes may be an important predictor of his QoL. Both upward and downward comparisons may enhance one’s self-concept when interpreted positively. These findings are relevant to clinicians who help men adapt to the psychosocial implications of a prostate cancer diagnosis.

In clinical practice, patients often ask clinicians how their situation compares to others’ and what other patients have done in similar circumstances. The results of this study suggest that patients who are connected to a support network may benefit from clinicians’ efforts to modify SCs toward more positive interpretations. This may be accomplished by helping the patient avoid maladaptive comparisons, choosing a new dimension of comparison, and/or making a different comparison on the same dimension (Wood, 2000).

Clinicians should assess how their patient appraises his diagnosis, as well as how he is comparing himself to others and what he seeks with these comparisons. SCs may provide information about how an individual is doing relative to others, make one feel better about his or her situation, and help one adapt to a stressor (Buunk & Gibbons, 2007). When a patient perceives his diagnosis as a challenge, he may seek information that will help him to deal with his disease (Lazarus & Folkman, 1984). However, when the diagnosis is seen as a threat, the patient may be better served by evidence that he is coping well (Taylor, Wood, & Lichtman, 1983). By assessing such appraisals, clinicians will be better equipped to tailor mental health services to the individual.

A patient’s interpretation of a SC as positive or negative may be the most important determinant of its effect on his sense of well-being. However, Van der Zee and colleagues also recommend assessing the specific types of comparisons that a patient is making—i.e., upward identification or contrast, downward identification or contrast—to better understand why the patient feels better or worse as a result (Van der Zee et al., 2000). Next, the clinician should draw attention to SCs that do not appear to be meeting the patient’s needs. The clinician can assist the patient in making positive interpretations of SCs by putting a different focus on the comparisons that a patient is already making, and emphasizing additional areas in which the patient is doing well.

Conclusion

This cross-sectional study of SCs among men in prostate cancer support groups sheds light on the relationships among participants’ appraisals of their cancer, SCs to other men with prostate cancer, and subjective QoL. This study was novel in its use of a quantitative measure of SCs in this population, and the results add to the body of knowledge by providing further evidence that positive SCs and negative SCs are distinct cognitive processes, rather than opposite ends of a continuum. Multivariable linear regression models were used to demonstrate that while controlling for confounders, cognitive appraisals (challenge, threat, harm/loss) were significant predictors of SCs, and SCs were significant predictors of QoL. Finally, positive SCs and negative SCs were parallel mediators of the relationships between challenge or threat appraisals and QoL, while only negative SCs mediated the relationship between harm/loss appraisals and QoL. Interventions to improve QoL could address reduction of maladaptive SCs, a strategy which could ultimately be employed by clinicians who help men manage a prostate cancer diagnosis.