Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2014 Dec 1.
Published in final edited form as: Health Psychol. 2012 Oct 22;32(12):10.1037/a0030189. doi: 10.1037/a0030189

Psychosocial Moderators of Presurgical Stress Management for Men Undergoing Radical Prostatectomy

Chelsea D Gilts 1, Patricia A Parker 2, Curtis A Pettaway 3, Lorenzo Cohen 4
PMCID: PMC3833972  NIHMSID: NIHMS495921  PMID: 23088178

Abstract

Objective

It is important to identify factors that predict who will benefit the most from psychosocial interventions in cancer populations.

Methods

This study examines the moderating effect of baseline social support (Social Support, SS; Dyadic Adjustment DA), distress (Brief Symptom Inventory, BSI; Impact of Event Scale, IES), and coping style (Brief COPE) on quality of life outcomes (SF-36 Physical Component Summary scores(PCS)) 1 year post surgery derived from a presurgical cognitive behavioral stress management program (SM; n = 23), supportive attention (SA; n = 37), or standard care (SC; n = 29).

Results

Moderation analyses indicated that men who reported low baseline social support (SS) and were in SM had increased PCS one year after surgery compared to men with low SS in the SC group (β = −0.39, p < .01), with SA having a non-significant intermediate effect. Men who reported high distress (BSI) at baseline and were in the SA group had increased PCS one year after surgery compared to those in the SC group (β = 24.80, p = .01), with SM having a non-significant intermediate effect. Mediation analyses suggested that neither SM nor SA improved QOL simply by increasing social support or decreasing general distress.

Conclusions

Distressed individuals may benefit more from unstructured discussion of distress whereas those low in social support may benefit more from a structured approach to learning coping skills.

Keywords: stress management, prostate, moderation, QOL

INTRODUCTION

The efficacy of psychosocial interventions for cancer populations has received considerable attention in the literature over the past several decades (see Dale, 2010 and Speigel, 2002 for reviews). Though the majority of studies investigating the benefits of managing stress occur following surgery, the period just prior to surgery can be a time of high stress for patients and may be critical to recovery (Manyande, 1992; Schneider et al., 2010). Presurgical stress management training has been shown to have positive physical and psychological benefits for prostate cancer patients (Parker et al., 2009; Cohen et al., 2011), open-heart surgery patients (Kshettry et al., 2006), abdominal surgery patients (Manyande, 1995) and breast cancer patients (Larson et al., 2000). Thus, well-timed, brief psychosocial interventions incorporated into the medical setting may be a feasible, cost effective approach to improving quality of life (QOL).

Targeting Interventions: Considering Moderation

Though the majority of studies have found some benefit to psychosocial interventions for adult cancer populations, results are far from conclusive and continue to be a source of controversy among researchers (Stefanek et al., 2006). Many researchers point to sample heterogeneity as a primary reason for outcome ambiguity (Sheard & Maguire, 1999; Stanton, 2005). Though demographic or initial medical factors do not appear to play a major role in predicting treatment efficacy (Penedo, 2007; Spiegel, 2007), pre-treatment psychosocial factors may affect long term QOL following intervention (Helgeson et al., 2006; Schneider et al., 2010).

While the direct association between psychosocial resources and QOL for persons diagnosed with cancer has been repeatedly demonstrated (Carver et al., 2006; Konski et al., 2006), there needs to be further research on the association between psychosocial resources and intervention effectiveness (Jacobson, 2007). There is evidence to suggest that psychosocially vulnerable patients, including those with high distress (Andersen et al., 2004; Carmack Taylor et al., 2007; Schneider et al., 2010) or low social support (Antoni et al., 2001; Scheier et al., 2007), may experience greater benefit from psychosocial interventions. A few studies, however, support the opposite conclusion that better psychosocially adjusted patients benefit more from interventions (Doorenbos et al., 2006; Hosaka et al., 2000) or are at least provided some prophylactic benefit (Kissane et al., 2007). It is also possible that there is a curvilinear relationship between psychosocial vulnerability and QOL following treatment, with very vulnerable and very well-adjusted patients being relatively unaffected by treatment (Lechner et al., 2006).

Due to the growing population of men diagnosed with cancer (CDC Cancer Statistics Working Group, 2010), the increasing popularity of psychosocial interventions for cancer patients, and the ever present need to optimize cost effectiveness of treatment, it has become crucial to examine potential moderators of treatment efficacy.

Mechanisms of Change: Considering Mediation

Exploring whom interventions most benefit begs the question of the intervention's mechanism of action. Although rarely the focus of clinical trials, it is likely that stress management (SM) interventions have a long term impact on psychosocial resources including marital adjustment, social support (Andersen et al., 2004), mood (Stanton et al., 2002), and coping skills (Manyande et al.,1995) which in turn impact QOL. Additionally, presurgical interventions are associated with more adaptive cortisol and adrenaline functioning (Bohnen et al., 1991; Steptoe, 1993) and improved immune functioning (Cohen et al., 2011; Witek-Janusek et al., 2008), which in turn can improve long term QOL (Carlson et al., 2003). There is a need for further clarification on the process by which SM interventions impact long term QOL, as this information can aid in improving intervention design as well as furthering the understanding of patient characteristics associated with benefits from SM.

The Present Study

Based on previous literature, the present study examined the association between perceived social support, relationship adjustment, general distress, intrusive thoughts and avoidance behaviors, and coping style with QOL following surgery and with response to the psychosocial interventions (SM or Supportive Attention (SA)). First, it was hypothesized that baseline psychosocial factors would correlate with long-term QOL, regardless of intervention type. Additionally, it was hypothesized that each of the five baseline psychosocial factors would moderate the effect of the intervention on QOL 1 year after surgery. The association between low baseline psychosocial resources and worse QOL that was expected be seen in the Standard Care (SC) and SA control groups was expected to be significantly decreased for individuals in the SM group. Finally, it was hypothesized that moderators of the intervention would also mediate intervention effect, indicating the intervention leads to improvements in the mediators, which in turn leads to improved QOL 1 year after surgery. Previously published findings from this dataset demonstrated that men in the SM group reported significantly reduced negative mood state (Profile of Mood States (POMS)) prior to surgery compared to men in SC and SA groups. Men in the SM group also had significantly higher QOL compared to men in SC group 1 year following surgery (Parker et al., 2009). The effect of SM on immune function has also been demonstrated in this dataset (Cohen et al., 2011). Though short term mood and immune outcomes were significantly impacted by the intervention, only QOL was found to be altered at 1 year follow up. Mediation and moderation of the acute intervention effects on mood and immune outcomes will be explored in a future paper. Research indicates that it can take up to a year following radical prostatectomy to have full recovery, and that further recovery beyond this point in minimal (Keller, Janetschek, & Abukora, 2005). Thus, data collected 12 months post-surgery was the focus of the present study exploring the possible moderating and mediating role of psychosocial resources on long term QOL.

METHOD

Participants

Participants were men with early-stage prostate cancer who were undergoing radical prostatectomy at 1 of 3 hospitals within the Texas Medical Center. Patients were included if they were 18 years of age or older, undergoing radical prostatectomy, English speaking, and able to come to the medical center four times prior to their surgery or live within 100 miles of the Texas Medical Center. Patients were excluded if they had any other surgeries in the preceding year, a major illness likely to limit survival to less than two years, medical conditions likely to affect outcome measures such as an autoimmune disease, endocrine abnormalities, chronic pain problems, or a current or past drug and alcohol dependence, had any major psychiatric diagnoses, were undergoing psychiatric or psychological counseling, or had severe cognitive dysfunction, detected by a score of < 6 on the Mini-Mental State Exam (Folstein et al., 1975).

There were a total of 89 participants (SM = 23, SA = 37, SC =29) who participated in their randomly assigned intervention and completed all measures at baseline and 1 year post surgery.

Procedures

Details of the study procedures have been reported elsewhere (Parker et al., 2009; Cohen et al., 2011). Briefly, after obtaining informed consented, self-report data was collected from patients at baseline before randomization, after intervention delivery but before surgery, the morning of surgery, 48 hours post-surgery, and 6 weeks and 6 and 12 months post-surgery. Blood samples for immune function analyses were provided at baseline and 48 hours after surgery.

A clinical psychologist provided patients in the SM and the SA groups their respective interventions on the same schedule. The presurgical SM intervention consisted of two 90-minute individual sessions and participants were given a corresponding Stress Management Guide. The sessions were cognitive-behavioral in nature, with a major focus on relaxation skills. The rationale for stress management was presented, focusing on decreasing stress, worry, and pain. After briefly discussing the patient's typical coping strategies and assessment of any prior use of relaxation techniques, the patient was introduced to relaxation training. Patients in the SM group had brief booster sessions with the clinical psychologist on the morning of surgery (before the assessment) and 48 hours after surgery to reinforce the use of relaxation strategies and the problem-focused coping strategies.

Patients in the SA group met with the same clinical psychologists for two 90-minute sessions and participated in a semi-structured interview focused on their psychosocial and medical history. These sessions were designed to provide empathy, encouragement, and extra attention from the medical community, but not to provide any active skills. Patients in the SA group also had brief booster sessions with the clinical psychologist on the morning of surgery (before the assessment) and 48 hours after surgery, during which empathy, but no specific relaxation strategies, was provided.

This study was approved by the Institutional Review Board of all participating hospitals.

Measures

The Interpersonal Support Evaluation List (SS; Cohen et al., 1985) was used to measure perceived level of social support. Patients rated the perceived availability of emotional/informational support, tangible support, positive interactions, and affectional support. Items were summed into a total score. Internal consistency for this measure is high, ranging from 0.91 to 0.96, and it has demonstrated good predictive validity of the associations between social support and health (Sherbourne & Stewart, 1991). In this study sample, the internal reliability was high (Cronbach's alpha = 0.94).

Patients' relationship with their partner was assessed using the 32-item Dyadic Adjustment Scale (DA; Spanier, 1976). Though some studies have indicated four subscales of the DA (consensus, satisfaction, cohesion, and affectional expression), more recent analyses yield a single factor of marital satisfaction. Good internal consistency has been demonstrated for this measure (0.96) (Kazak, Jarmas, & Snitzer, 1988), and was found in this study sample (Cronbach's alpha = 0.95).

Psychological distress was assessed using the 52-item Brief Symptom Inventory (BSI) Global Severity Index (BSIGSI) (Derogatis, 1975). The BSIGSI is well-established with internal consistency estimates of 0.81 to 0.85, and satisfactory estimates of construct validity (Derogatis & Melisaratos, 1983). In this study sample, the internal reliability was high (Cronbach's alpha = 0.93).

The Impact of Event Scale (IES) is a 15-item, self-report scale that assesses intrusive thoughts (intrusively experienced ideas or feelings) and avoidance behaviors (avoidance of certain feelings or situations) (Horowitz, Wilner, & Alvarez, 1979). Patients rated the items in relation to their cancer, and the scales were combined into a total score with higher scores indicating more cancer-related intrusive thoughts and avoidance behaviors. It has adequate internal reliability (0.82–0.86) and validity (Sundin & Horowitz, 2002), which was also demonstrated in this study sample (Cronbach's alpha = 0.90).

Coping style was assessed using the 20 item Brief Coping Operations Preference Enquiry (Brief-COPE) (Carver, 1997). The Brief-COPE measures a set of conceptually distinct coping subscales that include active coping, use of social support, turning to religion, positive reframing of the situation, and avoidant coping. Perczek and colleagues conducted a factor analysis on Brief-COPE responses of 172 men being tested for prostate cancer and determined that the subscales active coping, planning, acceptance, and positive reframing loaded onto one factor labeled “engagement coping,” and the subscales denial and behavioral disengagement loaded onto another distinct factor labeled “avoidant coping.” Due to the high demographic similarities between the sample of Perczek et al and the present sample, this two-factor structure was used in the present study. To confirm the fit of this two-factor structure for the current sample, we conducted a confirmatory factor analysis using the PROC CALIS procedure in SAS (Covariance Analysis of Linear Structural Equations), which tests the appropriateness of structural equation models using covariance structural analysis. Three fit statistics indicated acceptable fit, with values equal or greater than 0.90 (Bentler Comparative Fit Index = 0.970, Bentler-Bonett Non-Normed Index = 0.944, and Bentler-Bonett NFI = 0.900) confirming a good fit for the two-factor B-COPE (Hu & Bentler, 1999). Internal reliability for engagement type coping (Cronbach's alpha = 0.81) and avoidant coping (Cronbach's alpha = 0.67) were acceptable and comparable to that found by Perczek and colleagues (2002).

The Medical Outcomes Study 36-item Short Form Health Survey (SF-36; Ware & Sherbourne, 1992), was used to assess general health-related QOL. It assesses QOL in several domains including physical functioning, bodily pain, general health perceptions, vitality, social functioning, role-emotional, and mental health. In this study sample, the internal reliability for each of these subscales was acceptable (0.79–.90). There are two component scores derived from the subscales that measure mental health (MCS) and physical functioning (PCS). The focus of this paper is on the PCS as it was the only outcome for which there were group differences 1 year after surgery.

Analyses

Descriptive statistics were computed. We examined whether there were any differences in demographic (age, ethnicity, marital status, education) and clinical (Prostate Specific Antigen (PSA) at baseline, stage of disease) characteristics in the three groups using analysis of variance or chi-square tests and whether there were differences in the psychosocial measures at baseline. To determine the association between baseline psychosocial factors and QOL across groups, Pearson partial correlations were conducted between each baseline psychosocial adjustment measure and QOL at 1 year controlling for covariates (age, ethnicity, baseline PSA, and stage of disease). Moderation analyses were performed by regressing the QOL score at 1 year on each measure of baseline psychosocial adjustment, group assignment, the group by baseline psychosocial adjustment score, the respective baseline QOL measure, and covariates (age, ethnicity, baseline PSA, and stage of disease) using general linear model regression analyses. General linear model analyses include all patients in the analyses who completed the QOL measure at baseline and 1 year. We used the PROC GLM procedure in SAS V9.2 to run these analyses. Quadratic terms were added to PROC GLM to examine any curvilinear associations between psychosocial variables and QOL. None of the quadratic terms were significant indicating that the associations followed a linear trend. The group effect was treated as a classification variable using class statement in the PROC GLM procedure and the standard care group was the reference group. As these were a priori analyses, we used an alpha of .05 to determine significance. Where the interaction between group assignment and baseline psychosocial adjustment score was significant, t tests were used for post hoc group comparisons, and the Bonferroni correction was used to determine significance (alpha = .0167; Maxwell & Delaney, 2004). Mediation of intervention effect on QOL at 1 year by each psychosocial adjustment score at each time point (5 days prior to and 6 and 12 months following surgery) was explored using the PROC GLM procedure following the criteria outlined by Baron and Kenny (1986). Using G*Power (Faul et al., 2007), it was determined that given 3 groups and a total of 89 participants, we will be able to declare as statistically significant a medium effect size (Cohen's f2 > 0.15) assuming 80% power and a two-sided significance level of 0.05. Cohen's f2 is a common measure of effect size in tests of moderation (Aiken & West, 2001), and Cohen (1988) suggested that f2 effect sizes of 0.02, 0.15, and 0.35 are termed small, medium, and large, respectively.

RESULTS

Sample Characteristics

Measurements were obtained for 56% of the sample at 1 year (n = 89). There was no difference in follow-up rates between the intervention groups (p > .20) and no significant differences in demographic, medical, psychosocial, or PCS scores at any time point between those who dropped out of the study by 1 year follow up and those who did not (all p's > .29).

The demographic and medical characteristics are summarized in Table 1. The three groups were similar on all medical characteristics (including baseline medical comorbidities) and demographic variables. There were no statistically significant differences among the groups on any of the psychosocial variables at baseline (all p's < .20). Two individuals were determined to be outliers based on Grubb's test for detecting outliers, one due to high BSIGSI and SS scores and one due to a high avoidance copings score, and were dropped from analyses (Barnett and Lewis, 1994).

Table 1.

Demographic Information

Characteristic Standard Care Supportive Attention Stress Management
No. % No. % No. %
Mean Age 60.8 (6.4) 60.4 (8.0) 60.7 (8.0)
Ethnicity
 White 26 90 27 73 19 83
 African American 2 7 7 19 3 13
Married 24 83 33 89 20 91
Divorced 5 17 3 8 1 5
Education
 High School 8 27 4 11 2 10
 Some College 2 7 9 24 7 32
 College Grad 11 38 15 40 11 50
 Graduate Degree 6 21 9 24 1 5
Mean Prostate 7.9 (9.75) 6.4(3.9) 6.8(3.7)
Specific Antigen
Stage of Disease
 I 7 23 5 14 5 9
 II 19 65 27 75 17 74
 III 3 10 4 10 3 13
Type of Surgery
 Non-nerve sparing 8 28 10 29 4 17
 Nerve Sparing 19 68 21 62 16 70
 Nerve Graft 1 3 3 9 3 13
Open in a new tab

Baseline Predictors of QOL

Descriptive information on all potential predictors can be found in Table 2. Across intervention groups, a Pearson partial correlation procedure was used to determine the association between baseline psychosocial factors and QOL as measured by the PCS at 1 year controlling for age, ethnicity, stage of disease, baseline PSA and baseline PCS (see Table 3). Baseline perceived social support (SS; r = 0.23, p = .04) and dyadic adjustment (DA; r = 0.23, p = .04) were both moderately positively correlated with PCS scores 1 year after surgery. Baseline overall distress (BSIGSI; r = −0.20, p = .07) and intrusive thoughts and avoidance behaviors related to the upcoming surgery (IES; r = −0.23, p = .04) were negatively associated with PCS scores 1 year after surgery. Neither engagement coping (ENG; r = 0.11, p = .33) nor avoidance coping (AVD; r = −0.08, p = .47) were significantly associated with PCS scores 1 year after surgery.

Table 2.

Psychosocial Predictor Descriptive Statistics at Baseline and 1 Year Post-Surgery

Predictor Standard Care (n = 29) Supportive Attention (n = 37) Stress Management (n = 23)
M (SD) M (SD) M (SD)
SS
 Baseline 91.43 (17.25) 92.95 (13.76) 92.98 (15.82)
 1y 94.27 (16.63) 96.89 (13.88) 95.82 (16.68)
DA
 Baseline 113.60 (18.80) 109.11 (19.59) 113.48 (20.33)
 1y 113.27 (17.52) 106.43 (19.98) 113.33 (16.30)
IES
 Baseline 15.10 (13.56) 12.79 (11.76) 11.87 (10.07)
 1y 7.57 (9.94) 4.84 (8.45) 6.65 (8.49)
BSIGSI
 Baseline 0.22(0.23) 0.17 (0.18) 0.20 (0.19)
 1y 0.20 (0.27) 0.16 (0.19) 0.19 (0.23)
ENG
 Baseline 14.21 (5.13) 15.23 (5.50) 16.22 (4.88)
 1y 9.76 (5.11) 10.71 (4.69) 10.87 (5.91)
AVD
 Baseline 1.00 (1.51) 0.38(0.72) 0.22 (0.60)
 1y 0.43 (1.12) 0.46 (1.19) 0.22 (0.60)
Open in a new tab

There were no statistically significant differences among the groups on any of the psychosocial variables at baseline.

Table 3.

Baseline Psychosocial Predictors Correlated with PCS at 1-year

Predictor Pearson's Correlation p-value
SS 0.23 .04
DA 0.23 .04
BSIGSI −0.20 .07
IES −0.23 .04
ENG 0.11 .33
AVD −0.08 .47
Open in a new tab

Correlations controlled for age, ethnicity, stage of disease, baseline prostate-specific antigen (PSA), and baseline moderator.

Psychosocial Factors as Moderators of Intervention Effect

A general liner model in SAS was used to test a model of group and each baseline psychosocial measure predicting PCS at 1 year by entering the interaction between group and the psychosocial measure and controlling for age, ethnicity, stage of disease, baseline PSA, baseline PCS, group, and the respective psychosocial measure. Results of moderation analyses can be seen in Table 4.

Table 4.

ANOVA and Pairwise Comparisons of Baseline Psychosocial Factors Moderating Treatment Effect on PCS at 1 Year

Predictor Interaction SM v. SC SA v. SC SM v. SA
F(2,86) P β p β P β p
SS 4.00 .02 −0.39 .01 −0.15 .23 −0.24 .09
DA 1.05 .36 −0.15 .17 −0.03 .75 −0.12 .27
BSIGSI 3.29 .04 14.76 .18 24.80 .01 −10.04 .37
IES 2.34 .10 0.27 .19 0.32 .04 −0.05 .81
ENG 0.23 .79 0.30 .53 0.21 .60 0.09 .83
AVD 1.80 .17 −3.16 .29 3.60 .17 −6.76 .07
Open in a new tab

All analyses controlled for age, ethnicity, stage of disease, baseline prostate-specific antigen (PSA), and baseline moderator.

There was a significant interaction between baseline SS and group in predicting PCS (F(2,86) = 4.00, p = .02), representing a small-to-medium effect size (f2 = .102). Men who reported low baseline SS and were in SM had increased PCS 1 year after surgery compared to men with low SS in the SC group (β = −0.39, p < .01), with SA having a non-significant intermediate effect. Figure 1 shows that the negative effect of low SS is buffered for those in the SM group. In other words, the lower the perceived SS at baseline, the greater an individual benefited from SM compared to men in the SC group.

Fig. 1.

Fig. 1

Open in a new tab

Baseline perceived social support moderates the effect of intervention on PCS at 1 year

The slope of SM significantly differs from SC (p < .01). There is no significant difference between SA and SM or SC.

There was a significant interaction between baseline BSIGSI scores and group in predicting PCS (F(2,86) = 3.29, p = .04), representing a small-to-medium effect size. Men who reported high BSIGSI scores at baseline and were in the SA group had increased PCS scores1 year after surgery compared to those in the SC group (β = 24.80, p = .01). In other words, the higher the distress as measured by the BSIGSI at baseline, the greater an individual benefited from SA compared to men in the SC group, with the SM group having an intermediate non-significant effect. Figure 2 shows that the negative effect of high baseline distress as measured by the BSIGSI is buffered for those in the SA group. IES scores followed a similar but non-significant trend (p = .10; graphical data not shown).

Fig. 2.

Fig. 2

Open in a new tab

Baseline overall distress moderates the effect of intervention on PCS at 1 year

The slope of SA significantly differs from SC (p = .01). There is no significant difference between SM and SA or SC.

Contrary to hypothesis, there was no interaction between group and baseline dyadic adjustment scores (F(2,86) = 1.05, p = .36), or engagement (F(2,86) = 0.23, p = .79) or avoidant (F(2,86) = 1.80, p = .17) coping style in predicting PCS levels 1 year after surgery.

Psychosocial Factors as Mediators of Intervention Effect

Baron and Kenny's (1986) method was used to determine whether post-intervention psychosocial improvements mediated the interventions' effects on QOL. First, an association between intervention group and PCS at 1 year was established (F(2,81) = 3.67, p = .03). Specifically, those in SM reported higher PCS scores 1 year after surgery compared to those in SC (β = 6.23, p < .01). Patients in SA demonstrated an intermediate but nonsigificant effect on PCS scores. Next, an association between group and all possible mediators (SS, DA, BSIGSI, IES, ENG, and AVD) at 5 days prior to surgery as well as 6 and 12 months following surgery was tested, controlling for age, ethnicity, stage of disease, baseline PSA, and baseline measures of each mediator. There were no statistically significant associations between group and SS, DA, BSI, or coping measures at any time point. There was a significant association between intervention group and IES after the intervention 5 days before surgery, with those in SA (M = 11.22, SD = 11.79) reporting lower IES scores than those in SC (M =17.13, SD = 13.13) or SM (M = 15.74, SD = 12.38). However, change in presurgical IES did not significantly predict PCS (F(2, 87) = 2.62, p = .11), so the last step of Barron and Kenny's (1986) mediation testing method was not met.

DISCUSSION

Results are consistent with previous research indicating an association between presurgical psychosocial factors and post-surgical QOL in cancer patients, regardless of intervention. Men who reported high levels of social support and low levels of distress as measured by the BSIGSI at baseline tended to have high levels of QOL one year after surgery. It appears that, regardless of intervention, presurgical support and distress are strongly associated with QOL after surgery, even after taking presurgical QOL into account. Interestingly, coping styles were not associated with QOL after surgery. A meta-analysis by Penley, Tomaka, and Wiebe (2002) indicates a consistent and relatively large effect size for maladaptive coping strategies predicting poorer QOL, but a small to nonexistent effect size for adaptive coping strategies predicting improvement in QOL. Baumeister and colleagues outline similar findings on the relationship between coping and QOL in their presentation of the “bad is stronger than good” phenomenon (2001). Thus, while the lack of association between engagement coping styles and QOL is in line with previous research, the lack of association between avoidance coping styles and post-surgical QOL may be due to the limited range of avoidant coping styles reported in this sample. The low avoidant coping endorsement may be in part explained by the high educational attainment in the sample. Several studies have demonstrated a negative association between avoidant coping styles and level of education (Green, Wells, & Laakso, 2011; Zhang et al., 2008). There is no association between educational level and coping style in the current sample, but this could be due to the restricted range of education levels.

The primary aim of this study was to identify psychosocial factors that moderate patients' responses to brief presurgical intervention, with the idea that future intervention programs may “triage” cancer patients into the most appropriate intervention based on their current psychosocial state to maximize limited resources and to improve QOL benefits for patients. The results are consistent with previous research demonstrating psychosocial interventions most benefited the psychosocially vulnerable. Specifically, the SM intervention was significantly more beneficial for those low in perceived social support than was SC, demonstrating that SM buffered the negative effects of low social support on QOL. This suggests that men who are low in perceived social support and face an upcoming surgery may be benefited most by a structured, skills-based intervention. In particular, relaxation training, imaginal exposure of the day of surgery, and a discussion of skills for eliciting social support may be important intervention components for those low in perceived social support. Future component analysis studies can help identify the essential elements of CBT-based stress management interventions for cancer patients low in perceived social support. Importantly, participants high in perceived SS at baseline did not demonstrate significant QOL benefit in any intervention. It appears that there is a QOL “ceiling” that those initially high in SS experience following surgery. Thus, involving patients low in SS in a presurgical CBT-based intervention may be the most effective use of limited resources.

It appears that men reporting high baseline overall distress according to the Brief Symptom Inventory's Global Severity Index do not respond to interventions in the same manner as those low in perceived social support. Whereas those low in social support benefited more from a structured approach to learning coping skills, highly distressed individuals benefited more from unstructured attention. The SA program was most beneficial for those high in pre-surgical distress, as measured by the BSIGSI, with SA buffering the negative association between pre-surgical distress and QOL 1 year after surgery that is seen for the men undergoing standard care. Scores on the Impact of Events Scale, which assessed intrusive thoughts and avoidance behaviors related to their cancer and upcoming surgery, followed a similar, though non-significant trend.

A discussion with the clinical psychologists who administered the SA sessions indicated that though intervention time was used to discuss fears and concerns related to the upcoming surgery, many men also used the time to share information about their lives, families, and current projects. It may be that men who were particularly distressed benefited from the opportunity to tell their story, integrating their cancer experience with the rest of their lives. A large body of research on expressive writing supports the notion that individuals experiencing a major stressor benefit from the opportunity to form a narrative that works to integrate stressful events into their life story. For example, de Moor et al (2002) found that patients undergoing treatment for renal cell carcinoma who wrote in an unstructured manner about their diagnosis experienced better sleep quality and less daytime dysfunction than patients who wrote about neutral topics. This rational is of course speculative, as SA in this study was delivered in two brief sessions (total of 180 minutes) that did not directly address narrative formulation, whereas most studies on expressive writing involve four sessions lasting 20 minutes each (total of 80 minutes) explicitly focused on narrative formulation. Time to work on a “narrative” involving cancer treatment may be particularly important for individuals with high presurgical anxiety. Future research comparing professional attention (i.e., educational sessions) to life narrative creation may help to identify the essential elements of this intervention for cancer patients high in general distress. Additionally, it is possible that the reduced structure of SA allowed for greater expression of “non-specific factors” (e.g., empathy, warmth, and therapist-patient relationship) demonstrated to have clear impact on treatment outcomes in the psychotherapy literature (Lambert & Barley, 2001).

Contrary to hypothesis, relationship adjustment, coping style, and intrusive thoughts and avoidance behaviors related to their cancer were not significant predictors of intervention response. The null finding for relationship adjustment as an intervention moderator may be due to the relatively high dyadic adjustment score reported by this sample (M = 111.84, SD = 19.33) compared to Spanier's (1976) normative sample (M=101.5, SD = 28.3). Only 17 individuals (20% of the sample) reported DA scores below the mean of Spanier's distressed sample (M = 97, SD = 17.8). Similarly, limited endorsement of avoidance coping styles may explain why it was not a significant predictor of intervention response. As mentioned above, previous research suggests avoidance coping exerts a stronger and more consistent effect on QOL compared to engagement coping (Baumeister et al., 2001; Penley et al., 2002). The present sample reported low avoidance coping, which may explain why coping did not moderate intervention effect. Lastly, these men reported relatively low intrusive thoughts and avoidance behaviors related to their cancer, with a range of scores from 0 to 53, and a mean of 13.31 (SD = 11.92). In sum, the positive mental health reported by the participants and well as the relatively small sample size may explain why these hypothesized psychosocial factors were not detected as significant moderators of intervention effect.

Contrary to hypothesis, the factors that moderated the interventions (social support and distress) did not mediate the intervention's effect on QOL. It does not appear that stress management improved QOL simply by increasing social support or that supportive attention improved QOL simply by decreasing general distress. Rather, both interventions had a unique effect on QOL that is not accounted for by changes in perceived social support, marital adjustment, general distress, intrusive thoughts or avoidant behaviors. The interventions appear to affect QOL through different or more complex means than were tested in the present study.

It should be noted that the reason this paper focused on SF-36 PCS and not MCS was because PCS scores were the only outcome for which there were group differences 1 year after surgery. Reasons for this may be related to the high levels of mental health participants reported at study entry compared with normative data. There were no statistically significant changes in MCS scores over time, which may suggest a “ceiling effect.”

There are several limitations to recognize in this study. First, the majority of participants were white, non-Hispanic, married and highly educated. Thus, future research is needed to test the generalizability of these findings to more diverse populations. Additionally, there is a relatively limited range of distress and avoidant coping mechanisms reported by these men, which may be related to sampling limitation. Thirty-two out of the 42 men who refused to participate in the study, from whom we did not collect any data, indicated they were too busy to participate. It is possible that these men were experiencing more distress prior to the surgery due to a busy or stressful schedule, meaning the most distressed men may not have been involved in the study. The presence of this research problem is difficult to assess, but future studies could emphasize the brevity of the intervention as well as the potential to help manage the stress that surgery may add to an already stressful life. It should also be noted that a measure to assess the amount of “buy-in” to the treatment was not included in the present study. This may be an important factor in determining for whom a treatment works best as well as helping guide the development of future interventions to increase patient compliance.

Due to the relatively small sample size, replication is required to confirm the findings of this study. More research is needed to determine other intervention response predictors as well as to investigate what intervention components are most effective. These results suggest that providing interventions tailored for specific psychosocial vulnerabilities may be appropriate in cancer populations. Future research can assess the feasibility and effectiveness of targeted interventions as well as the means by which interventions exert their beneficial effects.

Acknowledgments

This project was supported by a research grant from the National Institutes of Health/National Cancer Institute (RO1 MH59432), the National Institutes of Health (CA016672), and a cancer prevention fellowship for Chelsea Gilts from the National Cancer Institute (R25T CA57730, Shine Chang, PhD, Principle Investigator).

We are indebted to Dr. Andy Baum who helped to start it all. We thank Drs. Richard Babaian, Louis Pisters, and Brian Miles for opening up this trial to their patients, Dr. Daniele Devine for study oversight, and Adoneca Fortier for helping with data collection.

REFERENCES

  1. Aiken LS, West SG. Multiple regression: Testing and interpreting interactions. Sage; Newbury Park, CA: 1991. [Google Scholar]
  2. Andersen BL, Farrar WB, Golden-Kreutz DM, Glaser R, Emery CF, Crespin TR, Shapiro CL, Carson WE., 3rd. Psychological, behavioral, and immune changes after a psychological intervention: A clinical trial. Journal of Clinical Oncology. 2004;22:3570–80. doi: 10.1200/JCO.2004.06.030. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Antoni MH, Lehman JM, Klibourn KM, Boyers AE, Culver JL, Alferi SM, et al. Cognitive-behavioral stress management intervention decreases the prevalence of depression and enhances benefit finding among women under treatment for early-stage breast cancer. Health Psychology. 2001;20(1):20–32. doi: 10.1037//0278-6133.20.1.20. [DOI] [PubMed] [Google Scholar]
  4. Barnett V, Lewis T. Outliers in Statistical Data: Wiley Series in Probability and Statistics. John Wiley & Sons Ltd.; Chichester, England: 1994. [Google Scholar]
  5. Baumeister RF, Bratslavsky E, Finkenauer C, Vohs KD. Bad is stronger than good. Review of General Psychology. 2001;5(4):323–370. [Google Scholar]
  6. Bohnen N, Nicolson N, Sulon J, et al. Coping style, trait anxiety, and cortisol reactivity during mental stress. Journal of Psychosomatic Research. 1991;35:141–147. doi: 10.1016/0022-3999(91)90068-y. [DOI] [PubMed] [Google Scholar]
  7. Carmack Taylor CL, de Moor C, Basen-Engquist K, et al. Moderator analyses of participants in the active for life after cancer trial: implications for physical activity group intervention studies. Annals of Behavioral Medicine. 2007;33:99–104. doi: 10.1207/s15324796abm3301_11. [DOI] [PubMed] [Google Scholar]
  8. Carver CS. You want to measure coping but your protocol's too long: Consider the Brief COPE. International Journal of Behavioral Medicine. 1997;4:92–100. doi: 10.1207/s15327558ijbm0401_6. [DOI] [PubMed] [Google Scholar]
  9. Carver CS, Smith RG, Petronis VM, Antoni MH. QOL among long-term survivors of breast cancer: different types of antecedents predict different classes of outcomes. Psycho-Oncology. 2006;15(9):749–758. doi: 10.1002/pon.1006. [DOI] [PubMed] [Google Scholar]
  10. Cohen L, Parker PA, Vence L, Savary C, Kentor D, et al. Presurgical Stress Management Improves Post-operative Immune Function in Men with Prostate Cancer Undergoing Radical Prostatectomy. Psychosomatic Medicine. 2011;73(3):218–225. doi: 10.1097/PSY.0b013e31820a1c26. [DOI] [PubMed] [Google Scholar]
  11. Cohen S, Mermelstein R, Kamarck T, Hoberman H. Measuring the functional components of social support. In: Sarason I, Sarason B, editors. Social support: Theory, research, and applications. Martinus Nijhoff; The Hague, The Netherlands: 1985. pp. 73–94. [Google Scholar]
  12. Coyne JC, Kagee A. More may not be better in psychosocial interventions for cancer patients. Health Psychology. 2001;20(6) doi: 10.1037//0278-6133.20.6.458. [DOI] [PubMed] [Google Scholar]
  13. Dale HL, Adair PM, Humphris GM. Systematic review of post-treatment psychosocial and behaviour change interventions for men with cancer. Psycho-Oncology. 2010;19(3):227–237. doi: 10.1002/pon.1598. [DOI] [PubMed] [Google Scholar]
  14. de Moor C, Sterner J, Hall M, Warneke C, Gilani Z, Amato R, Cohen L. A pilot study of the effects of expressive writing on psychological and behavioral adjustment in patients enrolled in a Phase II trial of vaccine therapy for metastatic renal cell carcinoma. Health Psychology. 2002;21:615–619. doi: 10.1037//0278-6133.21.6.615. [DOI] [PubMed] [Google Scholar]
  15. Derogatis LR. Brief Symptom Inventory. Clinical Psychometric Research; Baltimore: 1975. [Google Scholar]
  16. Derogatis LR, Melisaratos N. Brief Symptom Inventory: an introductory report. Psychological Medicine. 1983;13:595–605. [PubMed] [Google Scholar]
  17. Doorenbos A, Given B, Given C, Verbitsky N. Physical functioning: Effect of behavioral intervention for symptoms among individuals with cancer. Nursing Research. 2006;55:161–171. doi: 10.1097/00006199-200605000-00002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Faul F, Erdfelder E, Lang AG, Buchner A. G*Power 3: A flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behavior Research Methods. 2007;39:175–191. doi: 10.3758/bf03193146. [DOI] [PubMed] [Google Scholar]
  19. Folstein MF, Folstein SE, McHugh PR. Mini-mental state: A practical method for grading the cognitive state of patients for the clinician. Journal of Psychiatry Research. 1975;12(3):189–98. doi: 10.1016/0022-3956(75)90026-6. [DOI] [PubMed] [Google Scholar]
  20. Green HJ, Wells DN, Laakso LL. Coping in men with prostate cancer and their partners: A quantitative and qualitative study. European Journal Of Cancer Care. 2011;20(2):237–247. doi: 10.1111/j.1365-2354.2010.01225.x. [DOI] [PubMed] [Google Scholar]
  21. Helgeson VS, Lepore SJ, Eton DT. Moderators of the benefits of psychoeducational interventions for men with prostate cancer. Health Psychology. 2006;25(3):348–354. doi: 10.1037/0278-6133.25.3.348. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Horowitz M, Wilner N, Alvarez W. Impact of Event Scale: A measure of subjective stress. Psychosomatic Medicine. 1979;41:209–218. doi: 10.1097/00006842-197905000-00004. [DOI] [PubMed] [Google Scholar]
  23. Hosaka T, Sugiyama Y, Tokuda Y, Okuyama T. Persistent effects of a structured psychiatric intervention on breast cancer patients' emotions. Psychiatry and Clinical Neurosciences. 2000;54:559–563. doi: 10.1046/j.1440-1819.2000.00753.x. [DOI] [PubMed] [Google Scholar]
  24. Hu L, Bentler PM. Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling. 1999;6(1):1–55. [Google Scholar]
  25. Jacobsen PB. Screening for psychological distress in cancer patients: Challenges and opportunities. Journal of Clinical Oncology. 2007;25:4526–4527. doi: 10.1200/JCO.2007.13.1367. [DOI] [PubMed] [Google Scholar]
  26. Ketterer MW, Denollet J, Chapp J, Thayer B, Keteyian S, et al. Men deny and women cry, but who dies? Do the wages of denial include early ischemic coronary disease? Journal of Psychosomatic Research. 2004;56(1):119–123. doi: 10.1016/S0022-3999(03)00501-4. [DOI] [PubMed] [Google Scholar]
  27. Kazak AE, Jarmas E, Snitzer L. The assessment of marital satisfaction: An recalculation of the Dyadic Adjustment Scale. Journal of Family Psychology. 1988;2:82–91. [Google Scholar]
  28. Keller H, Janetschek G, Abukora F, et al. Technique of radical prostatectomy—a head to head comparison of retropubic, peineal and laparoscopic access—data on perioperative morbidity. European Urology Supplement. 2005;4(3):247. [Google Scholar]
  29. Kissane DW, Grabsch B, Clarke DM, Smith GC, Love AW, Bloch S, et al. Supportive-expressive group therapy for women with metastatic breast cancer: Survival and psychosocial outcome from a randomized controlled trial. Psychooncology. 2007;16(4):277–286. doi: 10.1002/pon.1185. [DOI] [PubMed] [Google Scholar]
  30. Konski AA, Pajak TF, Movsas B, et al. Disadvantage of men living alone participating in radiation therapy oncology group head and neck trials. Journal of Clinical Oncology. 2006;24:4177–4183. doi: 10.1200/JCO.2006.06.2901. [DOI] [PubMed] [Google Scholar]
  31. Kshettry VR, Carole LF, Henly SJ, Sendelbach S, Kummer B. Complementary alternative medical therapies for heart surgery patients: Feasibility, safety, and impact. The Annals of Thoracic Surgery. 2006;81:201–205. doi: 10.1016/j.athoracsur.2005.06.016. [DOI] [PubMed] [Google Scholar]
  32. Lambert MJ, Barley DE. Research summary on the therapeutic relationship and psychotherapy outcome. Psychotherapy: Theory, Research, Practice, Training. 2001;38(4):357–361. [Google Scholar]
  33. Larson MR, Duberstein PR, Talbot NL, et al. A presurgical psychosocial intervention for breast cancer patients: Psychological distress and the immune response. Journal of Psychosomatic Research. 2000;48:187–194. doi: 10.1016/s0022-3999(99)00110-5. [DOI] [PubMed] [Google Scholar]
  34. Lechner SC, Carver CS, Antoni MH, Weaver KE, Phillips KM. Curvilinear associations between benefit finding and psychosocial adjustment to breast cancer. Journal of Consulting and Clinical Psychology. 2006;74(5):828–840. doi: 10.1037/0022-006X.74.5.828. [DOI] [PubMed] [Google Scholar]
  35. Manyande A, Berg S, Gettins D, Stanford SC. Preoperative rehearsal of active coping imagery influences subjective and hormonal responses to abdominal surgery. Psychosomatic Medicine. 1995;57(2):177–182. doi: 10.1097/00006842-199503000-00010. [DOI] [PubMed] [Google Scholar]
  36. Manyande A, Chayen S, Priyakumar P, Smith CC. Anxiety and endocrine responses to surgery: Paradoxical effects of preoperative relaxation training. Psychosomatic Medicine. 1992;54(3):275–287. doi: 10.1097/00006842-199205000-00004. [DOI] [PubMed] [Google Scholar]
  37. Maxwell SE, Delaney HD. Designing experiments and analyzing data: A model comparison perspective. 2nd ed. Lawrence Erlbaum Associates; Mahwah, NJ: 2004. [Google Scholar]
  38. McNair DM, Lorr M, Droppleman LF. Profile of Mood States. Educational and Industrial Testing Service; San Diego, CA: 1981. [Google Scholar]
  39. Parker PA, Pettaway CA, Babaian RJ, Pisters LL, Miles B, Fortier A, Wei Q, Carr DD, Cohen L. The effects of a presurgical stress management intervention for men with prostate cancer undergoing radical prostatectomy. Journal of Clinical Oncology. 2009;27(19):3169–3176. doi: 10.1200/JCO.2007.16.0036. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Penedo FJ, Traeger L, Dahn J, Molton I, Gonzalez JS, Schneiderman N, et al. Cognitive behavioral stress management intervention improves QOL in Spanish monolingual Hispanic men treated for localized prostate cancer: Results of a randomized controlled trial. International Journal of Behavioral Medicine. 2007;14(3):164–172. doi: 10.1007/BF03000188. [DOI] [PubMed] [Google Scholar]
  41. Perczek RE, Burke MA, Carver CS, Krongrad A, Terris MK. Facing a prostate cancer diagnosis: who is at risk for increased distress? Cancer. 2002;94(11):2923–9. doi: 10.1002/cncr.10564. [DOI] [PubMed] [Google Scholar]
  42. Penley JA, Tomaka J, Wiebe JS. The association of coping to physical and psychological health outcomes: A meta-analytic review. Journal Of Behavioral Medicine. 2002;25(6):551–603. doi: 10.1023/a:1020641400589. [DOI] [PubMed] [Google Scholar]
  43. Scheier MF, Carver CS. Optimism, coping, and health: Assessment and implications of generalized outcome expectancies. Health Psychology. 1985;4(3):219–247. doi: 10.1037//0278-6133.4.3.219. [DOI] [PubMed] [Google Scholar]
  44. Schneider S, Moyer A, Knapp-Oliver S, Sohl S, Cannella D, Targhetta V. Pre-intervention distress moderates the efficacy of psychosocial treatment for cancer patients: A meta-analysis. Journal of Behavioral Medicine. 2010;33(1):1–14. doi: 10.1007/s10865-009-9227-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Sheard T, Maguire P. The effect of psychological interventions on anxiety and depression in cancer patients: Results of two metaanalyses. British Journal of Cancer. 1999;80:1770–1780. doi: 10.1038/sj.bjc.6690596. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Sherbourne CD, Stewart AL. The Medical Outcome Study Social Support Survey. Social Science and Medicine. 1991;32:705–714. doi: 10.1016/0277-9536(91)90150-b. [DOI] [PubMed] [Google Scholar]
  47. Spanier GB. Measuring dyadic adjustment: New scales for assessing the quality of marriage and similar dyads. Journal of Marriage and the Family. 1976 Feb;:15–28. [Google Scholar]
  48. Spiegel D. Effects of psychotherapy on cancer survival. Nature Review Cancer. 2002;2(5):383–389. doi: 10.1038/nrc800. [DOI] [PubMed] [Google Scholar]
  49. Spiegel D, Butler LD, Giese-Davis J, Koopman C, Miller E, DiMiceli S, et al. Effects of supportive-expressive group therapy on survival of patients with metastatic breast cancer: A randomized prospective trial. Cancer. 2007;110(5):1130–1138. doi: 10.1002/cncr.22890. [DOI] [PubMed] [Google Scholar]
  50. Stanton AL, Danoff-Burg S, Sworowski LA, Collins CA, Branstetter AD, Rodriguez-Hanley A, et al. Randomized, controlled trial of written emotional expression and benefit finding in breast cancer patients. Journal of Clinical Oncology. 2002;20:4160–4168. doi: 10.1200/JCO.2002.08.521. [DOI] [PubMed] [Google Scholar]
  51. Stanton AL. How and for whom? Asking questions about the utility of psychosocial interventions for individuals diagnosed with cancer. Journal of Clinical Oncology. 2005;23:4818–4820. doi: 10.1200/JCO.2005.01.913. [DOI] [PubMed] [Google Scholar]
  52. Stefanek ME, Jacobsen PB, Christensen AJ. The society of behavioral medicine's “great debate”: An introduction. Annals of Behavioral Medicine. 2006;32:83–84. [Google Scholar]
  53. Steptoe A. Stress, helplessness, and control: The implications of laboratory studies. Journal of Psychosomatic Research. 1983;27:361–367. doi: 10.1016/0022-3999(83)90067-3. [DOI] [PubMed] [Google Scholar]
  54. Sundin EC, Horowitz MJ. Impact of event scale: psychometric properties. British Journal of Psychiatry. 2002;180:205–209. doi: 10.1192/bjp.180.3.205. [DOI] [PubMed] [Google Scholar]
  55. U.S. Cancer Statistics Working Group . United States Cancer Statistics: 1999–2007 Incidence and Mortality Web-based Report. Department of Health and Human Services, Centers for Disease Control and Prevention, and National Cancer Institute; Atlanta, GA: 2010. [Google Scholar]
  56. Ware J, Sherbourne C. The MOS 36-item shortform health survey. Medical Care. 1992;30:473–483. [PubMed] [Google Scholar]
  57. Ware JE, Snow KK, Kosinski M, Gandek B. SF-36® Health Survey Manual and Interpretation Guide. New England Medical Center, The Health Institute; Boston, MA: 1993. 1993. [Google Scholar]
  58. Witek-Janusek L, Albuquerque K, Chroniak KR, Chroniak C, Durazo-Arvizu R, Mathews HL. Effect of mindfulness based stress reduction on immune function, QOL and coping in women newly diagnosed with early stage breast cancer. Brain, Behavior, and Immunity. 2008;22:969–81. doi: 10.1016/j.bbi.2008.01.012. [DOI] [PMC free article] [PubMed] [Google Scholar]
  59. Zhang X, Chen S, Li Z, Yao S. The relationship between college students' life intelligence and coping styles. Psychological Science (China) 2008;31(3):725–728. [Google Scholar]

RESOURCES