Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2022 Mar 3.
Published in final edited form as: Psychol Trauma. 2020 Sep 3;13(6):703–712. doi: 10.1037/tra0000944

Measuring Positive Psychosocial Sequelae in Patients with Advanced Cancer

Allison J Applebaum 1, Allison Marziliano 2, Elizabeth Schofield 1, William Breitbart 1, Barry Rosenfeld 3
PMCID: PMC7925699  NIHMSID: NIHMS1634629  PMID: 32881572

Abstract

Objective:

Posttraumatic growth and benefit finding describe the potential for positive changes resulting from traumatic experiences, including cancer. In oncology, these constructs are increasingly examined concurrently using the Post-Traumatic Growth Inventory (PTGI) and the Benefit Finding Scale (BFS). However, distinctions between these constructs, and their corresponding scales, are not altogether clear, and the burden of administering two lengthy questionnaires is evident, particularly for patients at end-of-life.

Method:

Baseline data from 209 participants enrolled in a randomized controlled trial evaluating the efficacy of a psychosocial intervention were analyzed. We assessed the structure and covariance of all PTGI and BFS items using item response theory to determine the extent to which these measures overlap and the potential value of their concurrent administration in patients with advanced cancer.

Results:

Despite conceptual differences in posttraumatic growth and benefit finding, results indicated that these measures address the same underlying construct. We subsequently analyzed three abbreviated scales (7-,11-, and 16-items) that combine items from both scales to identify an optimal briefer combined scale. Results supported all three versions, with the 7- and 16-item measures appearing to have the best balance of content and concurrent validity, and the 11-item version optimizing information gained with brevity.

Conclusions:

These findings indicate that concurrent administration of the PTGI and BFS may be unnecessary given the high degree of overlap between these two measures, and that a brief subset of items may adequately evaluate positive change among patients with advanced cancer while reducing participant burden.

Keywords: posttraumatic growth, benefit finding, posttraumatic growth inventory, benefit finding scale, advanced cancer

A growing body of literature documents the potential for positive change as a result of traumatic experiences, such as natural disasters (Park, 2016), interpersonal violence (Elderton, Berry, & Chan, 2015), and war (Kılıç, Magruder, & Koryürek, 2016). Posttraumatic growth and benefit finding are two terms used to describe such positive changes. In theory, posttraumatic growth refers to positive changes in life perspective, relationships and self-perception that result from a traumatic event (Tedeschi & Calhoun, 1995). The term benefit finding has been described as the process by which individuals re-assign positive value to challenging, though not necessarily traumatic, events based on benefits he or she identifies, and has been used to describe a broader, less specific set of positive changes (Antoni et al., 2001; Carver & Antoni, 2004; Helgeson, Reynolds, & Tomich, 2006; Tomich & Helgeson, 2004). Although these terms are grounded in distinct theoretical frameworks, there is clearly conceptual overlap between them.

Not surprisingly, given the growing emphasis on positive psychology, multiple measures have been used to assess these constructs (Bernstein & Rubin, 2006; Connor & Davidson, 2003; Walsh et al., 2018). The most widely used measures of posttraumatic growth and benefit finding are the Posttraumatic Growth Inventory (PTGI; Tedeschi & Calhoun, 1996) and the Benefit Finding Scale (BFS; Tomich & Helgeson, 2004), respectively. The PTGI is a 21-item self-report instrument that rates perception of positive change and has been used in multiple studies of patients with cancer (Brunet et al., 2010; Casella-Grau, Ochoa, & Ruini, 2017). The BFS is a 17-item self-report measure of perceived benefits adapted for patients with breast cancer (Antoni et al., 2001) from Behr’s Positive Contributions Scale (Behr, Murphy, & Summers, 1991). The BFS assesses potential benefits that could result from the cancer experience, including shifting personal priorities, world views, relationships and sense of purpose in life.

Both posttraumatic growth and benefit finding have been increasingly used to describe the potential positive experiences of individuals with cancer, among whom posttraumatic stress symptoms are common (Cordova et al., 2017). For example, a systematic review of 72 studies that evaluated clinical correlates of posttraumatic growth in patients with a wide range of sites and stages of cancer (Casellas-Grau et al., 2017) found that posttraumatic growth was inversely associated with anxiety and depressive symptoms, and positively associated with hope, optimism, spirituality and meaning. Additionally, a number of studies in oncology have examined these two constructs concurrently (e.g., Baník & Gajdošová, 2014; Li et al., 2015; Jansen, Hoffmeister, Chang-Claude, Brenner, & Arndt, 2011; Mols, Vingerhoets, Coebergh, & Van de Poll-Franse, 2009). For example, one study in which both the PTGI and BFS were administered to 483 colorectal cancer patients five years after diagnosis (Jansen et al., 2011) found that moderate to high levels of posttraumatic growth and benefit finding were experienced by a significant proportion of participants and both were inversely associated with education and depression. Interestingly, the PTGI was found to be more strongly associated (positively) with cancer stage and illness burden than was the BFS. Importantly, the majority of these studies have been conducted primarily with patients with stage I and II cancers.

Despite occasional differences in findings from the PTGI and BFS (e.g., Andrykowski et al., 2017) the added benefit of administering both instruments is not clear and there is significant overlap in the underlying construct(s) (Harding, Sanipour, & Moss, 2014; Mols et al., 2009). For example, using the Chinese version of the Benefit Finding Scale, Li et al. (2017) highlighted three dimensions (personal growth, improved relationships and acceptance), two of which are strikingly similar to dimensions measured by the PTGI (i.e., relating to others and overall personal growth). Additionally, strong correlations between the two measures have been demonstrated when the two scales are administered concurrently. For example, Jansen et al., (2011) found a correlation of .71 between the total scores on these measures in a sample of colorectal cancer survivors (those who had been diagnosed with colorectal cancer but at the time of evaluation had no evidence of disease), further highlighting the conceptual overlap between the two constructs. Importantly, such correlations between the PTGI and BFS do not obviate a range of substantive differences in the theoretical frameworks underlying these constructs.

One factor that may differentiate posttraumatic growth and benefit finding is their temporal relationship to the identified stressor. Posttraumatic growth refers to changes in one’s capacity to cope with adverse events that can occur long after the experience of the initial stressor, whereas benefit finding typically refers to changes that occur immediately after the adverse event (Calhoun & Tedeschi, 1998; Cassidy, 2013). For example, a mixed methods study of posttraumatic growth in caregivers of patients who died of HIV disease (Cadell & Sullivan, 2006) suggested a positive association (both via quantitative measures and qualitative interviews) between reports of positive outcomes and time since the patients’ death. In their meta-analysis of 87 studies that assessed benefit finding, Helgeson, Reynolds and Tomich (2006) found that amount of time that had elapsed since a traumatic event (i.e., war, sexual assault, natural disasters) served as a moderator for the relationship between benefit finding and various psychosocial outcomes. Specifically, benefit finding was more strongly related to decreased depression and greater positive affect when more than two years had elapsed since the traumatic event, whereas it was more strongly related to global distress and anxiety when less than two years had elapsed. Importantly, most studies of posttraumatic growth and benefit finding in oncology settings have examined post-treatment cancer survivors, and survivors of early stage breast and colorectal cancers specifically, enabling researchers to examine posttraumatic growth and benefit finding over time. Few studies, however, have examined potential benefits among patients with advanced cancers.

Despite a shorter life expectancy and poorer quality of life, a growing body of literature indicates that patients with advanced and life-limiting cancers may indeed experience posttraumatic growth and benefit finding (Mosher et al., 2017; Posluszny, Baum, Edwards, & Dew, 2011; Tang et al., 2015). It is not clear from these studies, however, whether it is the initial cancer diagnosis or facing thoughts about impending death that would constitute the traumatic stressor and therefore serve as the catalyst for change (Tang et al., 2015). Moreover, additional variables such as perceived social support or symptom management may contribute to positive changes. Nonetheless, it is evident that patients with advanced cancers have the potential to derive meaning from their experience of illness, which subsequently can significantly and positively impact their quality of life, despite however limited it might be. In this vulnerable population, the concurrent administration of the PTGI and the BFS increases participant burden, perhaps without significant added value. Indeed, tension between balancing the desire for comprehensive data collection and the potential negative impact of that collection on patient participants is a key concern in palliative care research (Aktas & Walsh, 2011; Dean & McClement, 2002; Jordhoy et el., 1999; Maloney et al., 2013). The purpose of the present study was to explore the extent to which measures of posttraumatic growth and benefit finding differ from one another or tap a single overarching construct among patients with advanced cancers. This information could then be used to determine whether a more streamlined assessment could be devised that optimally assesses growth and benefit finding in patients with advanced cancer.

Method

Participants

The current analyses are based on baseline (pre-intervention) data collected from a randomized controlled trial evaluating the impact of a structured psychotherapeutic intervention in 253 patients with advanced and life limiting cancers (Breitbart et al., 2015). This study was conducted in a tertiary care cancer hospital in a major urban center and for which Institutional Review Board approval was granted (Protocol number 07–094). Participants were 209 patients with an advanced (Stage III or IV solid tumor) cancer diagnosis who were at least 18 years old, ambulatory, were cognitively intact and fluent in English. Only those participants who completed all or most items on the PTGI and the BFS at baseline were included. The authors of the PTGI and BFS gave permission to the study team to conduct the analyses presented here (V. Helgeson, personal communication, April 5, 2012; R. Tedeschi, personal communication, April 12, 2019). Participants included in the current study did not differ significantly on baseline demographics (i.e., sex, age, race, ethnicity, years of education) compared to participants in the parent study without PTGI and BFS collected at baseline.

Posttraumatic Growth Inventory.

The PTGI is a 21-item instrument that rates perception of positive change in response to a severe stressor (Brunet et al., 2010; Cordova et al., 2007; Sears, Stanton & Danoff-Burg, 2003). Although originally conceptualized as tapping three dimensions of growth (perceptions of self, relationship with others, and philosophy of life), Tedeschi and Calhoun (1996) identified five distinct factors: relating to others (7 items), new possibilities (5 items), personal strength (4 items), spiritual change (2 items) and appreciation of life (3 items). While initial psychometric evaluation of the PTGI was conducted among undergraduate students with and without experience of severely traumatic events, support for this 5-factor model has been found in studies of individuals experiencing a variety of traumatic events (Taku, Calhoun, Cann, & Tedeschi, 2008), among adults with a variety of chronic illnesses (Purc-Stephenson, 2014), adults with a diverse trauma history (Horswill et al., 2016), patients with breast cancer (Ramos et al., 2016) and caregivers of patients with cancer (Teixeira & Pereira, 2013). However, other research has supported a single factor model and a higher order factor encompassing the five correlated lower-order factors (Taku, Cann, Calhoun & Tedeschi, 2008). Responses to the PTGI range from 0, indicating “I did not experience this change as a result of my crisis” to 5, indicating “I experienced this change to a very great degree as a result of my crisis.” The total score is the sum of the 21 items (possible range = 0 to 105), and has demonstrated good internal consistency (Cronbach’s coefficient alpha = .90), but somewhat more variable internal consistency for the five factors: new possibilities (alpha = .84), relating to others (alpha = .85), personal strength (alpha = .72), spiritual change (alpha = .85), and appreciation of life (alpha = .67; Tedeschi & Calhoun, 1996). Subscores for the factors are reported as a mean of the associated items. In addition, test-re-test reliability over a two month interval was acceptable for the total score (r = .71), but more variable for the five factors, with correlations ranging from .37 to .74 (Tedeschi & Calhoun, 1996).

Benefit Finding Scale (BFS).

The BFS is a 17-item measure of perceived benefits that could result from the cancer experience, including shifting personal priorities, acceptance, daily activities, family, world views, relationships and purpose in life. The BFS was originally developed specifically for women who had survived breast cancer, and original psychometric testing was conducted among breast cancer survivors. However, its use (and wording) has been expanded to apply to patients with all sites and stages of cancer. The stem “having cancer” is followed by a list of positively-worded statements such as “has led me to be more accepting of things” and “has taught me to accept things I cannot change.” Responses range from 1 (“not at all”) to 5 (“very much”) and are summed to generate a total score ranging from 17 to 85. There is support for the reliability of the BFS in patients with a wide variety of cancer diagnoses, as evidenced by internal consistency that is consistently high across studies, ranging from .91 to .95 for the total score (Antoni et al., 2001; Tomich & Helgeson, 2004; Urcuyo, Boyers, Carver, & Antoni, 2005). Support for the factor structure of the BFS has been more mixed, as some research has characterized the BFS as a unidimensional construct (Pascoe & Edvardsson, 2015) whereas others have recommended a 2-factor model comprised of personal growth and growth in family relationships (e.g., Pakenham, 2005).

Statistical Analysis

We examined pairwise correlations between all BFS and PTGI items, irrespective of original instrument, as well as Cronbach’s coefficient alpha both within and across instruments. We then assessed the structure and covariance of all PTGI and BFS items using multi-dimensional Item Response Theory (IRT) graded response models via marginal maximum likelihood (using the SAS IRT procedure). IRT models are preferable to confirmatory factor analysis in this study due to the limited sample size (Forero & Maydeu-Olivares, 2009). We first compared confirmatory IRT models for three scenarios: 1) PTGI and BFS items all loading to a single factor, 2) PTGI and BFS items loading to separate factors, and 3) “self” and “relationship” items loading onto separate factors, regardless of instrument (i.e., drawing from the combined pool of PTGI and BFS items). These three models were selected based on our hypothesis that the two scales may form a single overarching construct (a single factor model), the “null” hypothesis that the two scales reflect distinct constructs (the 2-factor model), and past research suggesting an overarching distinction between self- and relationship-oriented growth. To compare these three confirmatory models, we use likelihood ratio tests. Once a structure was confirmed (e.g., unidimensional), items were grouped according to their levels of difficulty and then ordered within those groups according to trait discrimination, in order to aid in identifying a reduced set of items. Grouping of item difficulty was conducted by comparing item information distributions over the latent trait with Mood’s median non-parametric test (Siegal, 1956) and item slopes were used as the measure of item discrimination (Hays, Morales, & Reise, 2000). Specifically, each pairwise combination of items was tested for significant differences using the aforementioned tests and a group was defined as the collection of all items for whom pairwise combinations were not significantly different. Abbreviated item subsets were then based on extracting those items with the highest slope (i.e., maximum discrimination), using three different thresholds: 2.5 or greater (7 items), 2.25 or greater (11 items) and 2.0 or greater (16 items). Abbreviated items subsets and single instrument scores were compared for incremental validity with the full 38-item test information. Finally, we compared the original scales, the combined 38-item scale, and potential reduced scales on how they correlate to several external, conceptually-linked measures. These scales included the Hospital Anxiety and Depression Scale (Zigmond & Snaith, 1983) Anxiety and Depression subscales (HADS-A and HADS-D, respectively), the Life Orientation Test (a measure of dispositional optimism; Scheier et al., 1994), the Duke-UNC Social Support Questionnaire (Broadhead et al., 1988), the Hopelessness Assessment in Illness (Rosenfeld et al., 2011), the Beck Hopelessness Scale (Beck, Weissman, Lester, & Trexler, 1974), the FACIT Spiritual Well-being Scale (Brady, Peterman, Fitchett, Mo, & Cella, 1999), the Schedule of Attitudes Towards Hastened Death (Rosenfeld et al., 1999, 2000) and the Intrinsic/Extrinsic Religiosity Scale (Maltby, 1999). The Schedule of Attitudes Towards Hastened Death and Intrinsic/Extrinsic Religiosity Scale were added to the assessment after commencement of the study, thus were not available for a subset of participants who were enrolled prior to the protocol amendment date. All statistical analysis was performed using SAS version 9.4.

Results

Demographic characteristics of the study sample are shown in Table 1. Participants were predominantly female (68%), Caucasian (80%), non-Hispanic (88%), and partnered (59%). A variety of cancer diagnoses were represented including breast (26%), colon/rectum (16%), pancreas (17%), and lung (16%). The sample ranged in age from 27 to 86 years old, with a mean age of 57.6 years (SD=11.3 years). Participants were generally well-educated, with an average of 15.9 years of education (SD=2.4 years) and 70% of participants reported at least 16 years of education (i.e., a college degree). The subsample providing data on Intrinsic/Extrinsic Religiosity and Attitudes Towards Hastened Death (n = 140) was comparable on age, race, ethnicity, and years of education, but this subset contained disproportionately fewer men (26%) compared to the earlier enrolled participants who did not complete these instruments (26% vs. 45%, p < 0.01).

Table 1.

Participant Characteristics (N = 208a)

Variable n (%)
Sex
 Male 67 (32)
 Female 141 (68)
Race
 White 166 (80)
 African American 26 (13)
 Other or Unknown 16 (8)
Ethnicity
 Hispanic/ Latino 24 (12)
 Non-Hispanic/ Latino 183 (88)
 Missing 1 (<1)
Relationship Status
 Partnered 123 (59)
 Non-partnered 85 (41)
Cancer Diagnosis
 Breast 54 (26)
 Pancreas 36 (17)
 Colon/ Rectum 33 (16)
 Lung/ Bronchi 33 (16)
 Stomach 11 (5)
 Other 40 (19)
Mean (SD)
Age 57.6 (11.3)
Years of Education 15.9 (2.4)
BFS 60.9 (16.6)
PTGI Total 64.7 (23.8)
 Relating to Others 3.4 (1.1)
 New Possibilities 2.5 (1.4)
 Personal Strength 3.1 (1.3)
 Spiritual Change 2.8 (1.9)
 Appreciation of Life 3.5 (1.3)
Open in a new tab
a

Note. All demographic information is missing for one participant.

BFS total scores ranged from 19 to 85, with mean and median of 60.9 (SD=16.6) and 62, respectively. PTGI total scores ranged from 7 to 104 with mean and median of 64.7 (SD=23.8) and 67, respectively. Average BFS, PTGI total, and PTGI subscale scores were all just above the midpoint for each scale. Cronbach’s alpha for the 21 PTGI items was .94 and no item resulted in an appreciable increase of alpha when removed. Similarly, for BFS, Cronbach’s alpha was .94, with no increase in internal consistency if any item was removed. When all items from both instruments were combined, the internal consistency for the combined item pool was even higher, with a Cronbach’s alpha of .97. There was no increase in internal consistency if any items were removed. Median pairwise correlation of items from the PTGI was .44 and .49 for items from the BFS. Cross-instrument correlations were comparable to intra-instrument correlations, as depicted in Figure 1; the median pairwise correlation for cross-instrument items was .42.

Figure 1.

Figure 1.

Open in a new tab

Pearson correlation value for items within and between PTGI and BFS instruments.

Using exploratory IRT, the first eigenvalue (20.44) was over eight times larger than the second eigenvalue (2.41), providing support for the hypothesis that the pool of items is unidimensional. Further, comparing fit of the three a priori candidate models (i.e., single factor, factors distinguished by instrument, and factors distinguished by relation), the unidimensional model optimized both AIC and BIC criteria. Based on a likelihood ratio test, the fit of the unidimensional model was significantly better than the instrument-specific two factor model (p < .0001) but was not significantly better than the model distinguishing items based on improved relationships with others versus changes in one’s sense of self or self-perception (p = .78). Model fit details are given in Table 2. Interestingly, the majority of items (24) from the PTGI and BFS clustered under changes in one’s sense of self or self-perception (versus 11 items addressing relationships with others).

Table 2.

Log-likelihood comparisons for confirmatory models (N = 209 and 38 items, all models)

Log likelihood AIC BIC LRT
p-value
BFS vs. PTGI −10235.14 20892.28 21597.52 <.0001
Self vs. Relationship −10192.38 20806.77 21512.00 0.7840
Single Dimension −10176.90 20775.79 21481.02 REF
Open in a new tab

Note. LRT comparison is against the unidimensional model.

Development of an Abbreviated Measure

Given the high degree of overlap among the items and subscales that comprise the PTGI and BFS, coupled with the unidimensional structure supported by IRT analyses, we utilized several strategies to create an abbreviated instrument that maximizes test information while reducing participant burden. The first approach involved using pairwise comparisons of the 38 item information curves to cluster items by difficulty. This process resulted in eight groups of items, encompassing 24 of the 38 items in the two scales (i.e., 13 items were left ungrouped). Each group represented a set of items which had a roughly similar level of difficulty with respect to the single construct measured by the full battery. Within these groups, items were ordered by their model slope, a measure of information the item provides about the construct. The results are shown in Table 3.

Table 3.

IRT model information characteristics and slopes, by item information groupings (N = 209)

Item Description Information Mean (Median) Slope
Group 1
BFS 17 has helped me become a stronger person, more able to cope effectively with future life challenges. −0.56 (−0.6) 3.07
PTGI 10 I know better that I can handle difficulties. −0.52 (−0.5) 2.16
PTGI 16 I put more effort into my relationships. −0.51 (−0.5) 1.87
PTGI 21 I better accept needing others. −0.51 (−0.5) 1.60
PTGI 1 I changed my priorities about what is important in life. −0.53 (−0.6) 1.03
Group 2
PTGI 2 I have a greater appreciation for the value of my own life. −0.75 (−0.8) 2.06
PTGI 15 I have more compassion for others. −0.78 (−0.8) 1.52
PTGI 20 I leaned a great deal about how wonderful people are. −0.79 (−0.8) 1.51
BFS 8 has made me realize the importance of planning for my family’s future. −0.86 (−0.9) 1.41
BFS 4 has brought my family closer together. −0.83 (−0.8) 1.35
Group 3
BFS 12 has led me to meet people who have become some of my best friends. 0.45 (0.4) 1.61
PTGI 3 I developed new interests. 0.42 (0.4) 1.57
Group 4
PTGI 12 I am better able to accept the way things work out. −0.11 (−0.1) 2.28
PTGI 5 I have a better understanding of spiritual matters. −0.13 (−0.1) 2.27
PTGI 18 I have a stronger religious faith. −0.13 (−0.1) 1.86
Group 5
BFS 1 has led me to be more accepting of things −0.41 (−0.4) 2.75
BFS 9 has made me more aware and concerned for the future of all human beings. −0.44 (−0.4) 1.96
PTGI 8 I have a greater sense of closeness with others. −0.42 (−0.4) 1.65
Group 6
BFS 2 has taught me how to adjust to things I cannot change.  −0.47 (−0.5) 2.47
BFS 7 has shown me that all people need to be loved. −0.48 (−0.5) 2.41
PTGI 19 I discovered that I’m stronger than I thought I was. −0.48 (−0.5) 2.13
Group 7
BFS 3 has helped me take things as they come. −0.66 (−0.7) 2.19
BFS 5 has made me more sensitive to family issues. −0.67 (−0.7) 1.68
Group 8
BFS 6 has taught me that everyone has a purpose in life. −0.21 (−0.2) 2.82
BFS 11 has led me to deal better with stress and problems. −0.18 (−0.2) 2.57
Ungrouped Items
BFS 13 has contributed to my overall emotional and spiritual growth. −0.32 (−0.3) 3.11
BFS 16 has helped me become more focused on priorities, with a deeper sense of purpose in life. −0.69 (−0.7) 2.74
PTGI 11 I am able to do better things with my life. 0.13 (0.1) 2.53
PTGI 13 I can better appreciate each day. −0.62 (−0.6) 2.18
PTGI 4 I have a greater feeling of self-reliance. 0.05 (0.0) 1.86
BFS 10 has taught me to be patient. −0.23 (−0.2) 1.85
PTGI 9 I am more willing to express my emotions. −0.26 (−0.3) 1.76
PTGI 7 I established a new path for my life. 0.15 (0.1) 1.70
PTGI 17 I am more likely to try to change things which need changing. −0.37 (−0.4) 1.63
BFS 15 has helped me realize who my real friends are. −0.95 (−1.0) 1.63
BFS 14 has helped me become more aware of the love and support available from other people. −1.31 (−1.4) 1.51
PTGI 14 New opportunities are available which wouldn’t have been otherwise. 0.32 (0.3) 1.47
PTGI 6 I more clearly see that I can count on people in times of trouble. −0.99 (−1.1) 1.00
Open in a new tab

Note. Information mean and median are values for the Item Information Curve. Grouping is based on a series of Mood’s median pairwise comparisons of all items.

The resulting test information curves for a 16-item abbreviated item subset, compared to the full 38-item set and single-instrument information are displayed in Figure 2. The most parsimonious abbreviated subset included only seven items, six of which were drawn from the BFS (items 1, 6, 11, 13, 16, and 17) and one from the PTGI (item 11); all 7 items reflected changes in the patient’s sense of self or self-perception. These items represented 3 of the 8 item subsets, along with 3 of the 13 ungrouped items, and ranged in item difficulty from −0.69 to 0.13 (compared to a range of −1.31 to 0.45 for the 38-item battery). The decrease in total scale information compared to that for the full 38-item information was substantial, effectively reducing scale information by 62% (from 48.4 to 18.5), although scale length was reduced by more than 80%.

Figure 2.

Figure 2.

Open in a new tab

Abbreviated test information curves.

The addition of four more items, to an 11-item scale, included two more PTGI items (5 and 12), along with two more BFS items (2 and 7); three of these four items also tapped the “self” domain. The addition of these four items increased the total scale information substantially (to 26.2), reflecting only a 46% decrease from the original 38 items, despite eliminating 70% of the item pool. Although this subset of items included items from 3 more groups, these groups represented the mid-range of item difficulty, and therefore did not expand the range of item difficulty assessed. The final subset of items contained a total of 16 items, including four more items from the PTGI (2, 10, 13, and 19), along with BFS item 3, all of which tapped the “self” domain. These additional items also expanded the lower end of the item difficulty range to −0.75, enabling an assessment of a broader range of growth. The total scale information estimate for this subset of items was substantially higher (33.9), capturing more than 70% of the maximum scale information while still eliminating 57% of the items.

Correlational analyses and plots were also used to examine the loss of information from the three abbreviated versions of the combined item pool with external, conceptually-linked variables that have frequently been analyzed in prior psychosocial oncology research (i.e., social support, optimism, depression, anxiety, spiritual well-being, hopelessness, desire for hastened death, and religiosity). These analyses indicated virtually no loss of concurrent or discriminant validity for any of the three abbreviated scale versions (see Table 4). As evident from the pattern of correlations and plots, the three abbreviated scale versions generated virtually identical correlations to one another, and these correlations were comparable to those generated by the PTGI and BFS total scores, as well as a combined 38-item measure.

Table 4.

Correlation Analyses of PTG and BFS original, combined, and reduced instruments to other outcomes

Outcome n Mean (SD) Range PTGI Total BFS Total PTGI/BFS Combined 7-item Score 11-item Score 16-item Score
Social Support 203 3.25 (0.7) 0.9 – 4 0.17 (0.016) 0.13 (0.069) 0.16 (0.027) 0.16 (0.021) 0.14 (0.040) 0.14 (0.048)
Optimism 207 17.60 (3.6) 7 – 30 0.19 (0.007) 0.13 (0.068) 0.16 (0.018) 0.16 (0.020) 0.18 (0.010) 0.16 (0.019)
Anxiety 207 9.20 (4.3) 0 – 21 −0.01 (0.877) −0.05 (0.512) −0.03 (0.684) −0.04 (0.539) −0.04 (0.592) −0.04 (0.556)
Depression 207 7.17 (3.3) 0 – 17 −0.18 (0.009) −0.13 (0.059) −0.16 (0.018) −0.18 (0.010) −0.15 (0.034) −0.17 (0.013)
Spiritual Well-being 207 31.46 (10.0) 4 – 48 0.43 (<.001) 0.39 (<.001) 0.43 (<.001) 0.48 (<.001) 0.48 (<.001) 0.47 (<.001)
Hopelessness (Illness-specific) 209 7.52 (3.5) 0 – 15 −0.17 (0.015) −0.15 (0.030) −0.17 (0.016) −0.19 (0.006) −0.18 (0.008) −0.18 (0.010)
Hopelessness (general) 208 7.39 (3.6) 0 – 19 0.39 (<.001) 0.35 (<.001) 0.39 (<.001) 0.38 (<.001) 0.42 (<.001) 0.39 (<.001)
Religiosity IN 137 13.01 (4.2) 6 – 18 0.44 (<.001) 0.38 (<.001) 0.43 (<.001) 0.42 (<.001) 0.46 (<.001) 0.43 (<.001)
Religiosity EP 132 6.58 (2.2) 3 – 9 0.34 (<.001) 0.31 (<.001) 0.34 (<.001) 0.27 (0.002) 0.30 (0.001) 0.30 (0.001)
Religiosity ES 134 4.14 (1.7) 3 – 9 0.46 (<.001) 0.43 (<.001) 0.46 (<.001) 0.42 (<.001) 0.47 (<.001) 0.45 (<.001)
Religiosity EX 136 10.74 (3.2) 6 – 18 0.46 (<.001) 0.42 (<.001) 0.46 (<.001) 0.44 (<.001) 0.48 (<.001) 0.45 (<.001)
Religiosity TO 137 23.80 (6.8) 12 – 36 −0.31 (<.001) −0.24 (0.005) −0.29 (0.001) −0.30 (<.001) −0.29 (0.001) −0.30 (<.001)
Attitudes towards Hastened Death 136 2.04 (2.9) 0 – 18 0.39 (<.001) 0.35 (<.001) 0.39 (<.001) 0.38 (<.001) 0.42 (<.001) 0.39 (<.001)
Open in a new tab

Note. BFS items are scaled up to be on same range as PTGI Items for all combination PTGI/BFS summary measures (i.e., the last four columns in Table 4). “PTGI/BFS Combined” includes all 38 items from both instruments. The strongest correlation (or two) for each outcome is bolded. Subscales of the I/E scale are as follows: Religiosity EP – extrinsic personally oriented religiousness; Religiosity ES – extrinsic socially oriented religiousness; Religiosity EX – extrinsic socially oriented religiousness; Religiosity TO – I/E total score.

Discussion

This study explored the structure and covariance of the PTGI and BFS items to determine the overlap in these measures and whether a more streamlined assessment could optimally assess growth and benefit finding in patients with advanced cancer. Despite conceptual differences in PTG and BF, these measures appear to address the same underlying construct of positive psychological sequelae, as evidenced by the high correlations found here and in previous studies (e.g., Jansen et al., 2011), and by the results of our IRT analyses. Indeed, we observed a significant overlap between the items of the PTGI and BFS, with bivariate correlations of the BFS total score yielding strong, positive correlations with the PTGI total score and all five subscale scores.

Results of IRT revealed that the pool of PTGI and BFS items appear to be unidimensional and that the fit of the unidimensional model was significantly better than the instrument-specific two factor model. Future studies with larger samples will be needed to confirm if there indeed may be only minimal added value of concurrent administration of the PTGI and the BFS. There was, however, some support for a 2-factor model comprised of “self” and “relationship” domains that are measured by the combined pool of items. The 2-factor IRT analysis based on this distinction provided a comparable level of fit to the single factor model, suggesting that while the distinction between “type” of growth may have some theoretical relevance, it may not be central to the construct of growth after a cancer diagnosis.

Our efforts to identify an abbreviated item subset that taps into the overarching construct evaluated by the PTGI and BFS were generally successful. Three abbreviated scales, with 7, 11, and 16 items, were identified, each of which had nearly identical concurrent and discriminant validity with one another, as well as with each of the original scales and a summary score based on the combined pool of 38 items. The 7 and 11 item scales were drawn primarily from the BFS (6 of 7 items and 8 of 11 items). These findings suggest that BFS items, and in particular, those that target changes in the sense of self, may provide a better reflection of the overlapping construct measured by the two scales.

While there is no clear basis for determining which of the three streamlined versions is optimal, the 7-item version is likely adequate for use in studies where participant burden is at a premium, such as with patients with advanced cancer at end-of-life. On the other hand, the 11- item version draws more equally from both measures and provides nearly as much information as the original scales. Therefore, given the inevitable trade-off between scale length and scale information, and the importance of brevity in research involving patients with advanced cancer, the 11-item subsets may be an appropriate alternative, and minimizes the loss of information while still shortening patient burden.

Limitations and Future Directions

Several limitations of this investigation must be acknowledged. First, as this was a secondary data analysis with a modest sample size and restricted data available for analysis, all interpretations should be made with caution. Specifically, we did not have information regarding the accrual and refusal rate for participants in the parent trial, the type of treatment (e.g., curative versus palliative) participants were receiving, the time since patients’ diagnosis or documentation of an experience of trauma (or subsequent diagnosis of posttraumatic stress disorder) related to the cancer diagnosis. These latter two variables have been previously used to evaluate the constructs of PTG and BF, and historically have played central roles in how these constructs are operationalized. This information would have allowed us to more concretely establish the validity of the proposed scales. Second, our approach to item reduction relied wholly on statistical methods, versus qualitative approaches that would focus more comprehensively on the content of items. The inclusion of a mixed-methodological approach may have generated somewhat different findings, perhaps leading us to eliminate or retain additional items. Additionally, as this was a secondary data analysis of an already completed study, we were unable to examine multiple psychosocial (e.g., appraisals of threat, rumination, empathy, gratitude) or clinical (e.g., time since diagnosis of advanced cancer, type of treatment received (e.g., curative versus palliative)) variables that are conceptually related to PTG and BF. It will be important for such relationships to be evaluated in future studies.

Finally, the generalizability of our findings is limited due to the restricted demographic profile of participants (i.e., predominantly white, highly educated, identified as Catholic and Jewish) and that only cancer patients were included in this study; as such, the interpretations of the role of “type” of growth and other inferences about the covariance of PTG and BF items may not be generalizable to other contexts. Additionally, participants had a diagnosis of a Stage III or IV solid tumor cancer, a group within whom there is great variability among the type of care received. Our intention here was not to generalize to patients with all sites and stages of cancer, but to groups living with life limiting illnesses. Nevertheless, mean scores on the PTGI scores were on average higher than those reported among cancer survivors (Crawford et al., 2015; Cormio et al., 2017; Sharp et al., 2018), but similar to those of parents of children with cancer (Hullman et al., 2014). Scores on the BFS, on the other hand, were roughly similar to those reported in studies of colorectal cancer survivors (Jansen et al., 2011), but higher than those obtained by women with breast cancer (Wang et al., 2015). Thus, firm conclusions cannot be drawn about the extent to which sample demographic and clinical characteristics may have impacted study results.

As the field of psychosocial oncology continues to explore avenues to enhance the quality of life of patients with advanced and life-limiting cancers, it is necessary to refine assessment tools so that we can optimally capture the experience of benefit and growth without contributing unnecessarily to participant burden. Our results underscore the redundancy of concurrent administration of the PTGI and BFS and provide preliminary support for more streamlined assessment tools. Future studies are needed to further evaluate the performance of these abbreviated measures in a larger sample, where the collection of additional data (e.g., documentation of posttraumatic stress symptoms, time since diagnosis) and the inclusion of longitudinal assessments will further assist in establishing their construct validity.

Clinical Impact:

As the field of psychosocial oncology continues to explore avenues to enhance the quality of life of patients with advanced and life-limiting cancers, it is necessary to refine assessment tools to optimally capture the experience of benefit and growth without contributing unnecessarily to participant burden. Our results highlight the potential redundancy of concurrent administration of the Posttraumatic Growth Inventory (PTGI) and Benefit Finding Scale (BFS) and provide preliminary support for a more streamlined method of measuring the construct of positive change in response to adversity.

Funding disclosure:

This research was supported by funding from the National Institutes of Health (R01CA101940, PI: Barry Rosenfeld, Ph.D.; R01CA128187, PI: William Breitbart, MD; MD; P30CA008748, PI: Craig Thompson, MD).

References

  1. Antoni MH, Lehman JM, Kilbourn KM, Boyers AE, Culver JL, Alferi SM, … Harris SD. (2001). 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, 20(1), 20. [DOI] [PubMed] [Google Scholar]
  2. Aktas A, & Walsh D. (2011).Methodological challenges in supportive and palliative care cancer research. Seminars in Oncology, 38, 460–6. [DOI] [PubMed] [Google Scholar]
  3. Andrykowski MA, Steffens RF, Bush HM, & Tucker TC (2017). Posttraumatic growth and benefit-finidng in lunch cancer surrivors: The benefit of rural residence? Journal of Health Psychology, 22, 896–905. [DOI] [PubMed] [Google Scholar]
  4. Baník G, & Gajdošová B (2014). Positive changes following cancer: posttraumatic growth in the context of other factors in patients with cancer. Supportive Care in Cancer, 22(8), 2023–2029. [DOI] [PubMed] [Google Scholar]
  5. Bartl H, Hagl M, Kotoučová M, Pfoh G, & Rosner R (2018). Does prolonged grief treatment foster posttraumatic growth? Secondary results from a treatment study with long-term follow-up and mediation analysis. Psychology and Psychotherapy: Theory, Research and Practice, 91(1), 27–41. [DOI] [PubMed] [Google Scholar]
  6. Beck AT, Weissman A, Lester D, & Trexler L (1974). The measurment of pessimism: the Beck Hopelessness Scale. Journal of Consulting and Clinical Psychology, 42, 861–865. [DOI] [PubMed] [Google Scholar]
  7. Behr S, Murphy D, & Summers J (1991). Kansas inventory of parental perceptions. Lawrence: University of Kansas. [Google Scholar]
  8. Berntsen D, & Rubin DC (2006). The centrality of event scale: A measure of integrating a trauma into one’s identity and its relation to posttraumatic stress disorder symptoms. Behaviour Research and Therapy, 44(2), 219–231. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Brady MJ, Peterman AH, Fitchett G, Mo M, Cella D (1999). A case for including spirituality in quality of life measurement in oncology. Psychooncology, 8(5):417–28. [DOI] [PubMed] [Google Scholar]
  10. Breitbart W, Rosenfeld B, Pessin H,…. et al. (2015). Meaning-centered group psychoteherapy: an effective itnervention for improving psychological well-being in patients with advanced caner. Journal of Clinical Oncology, 33(7), 749–754. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Broadhead WE, Gehlbach SH, de Gruy FV, Kaplan BH (1988). The Duke-UNC Functional Social Support Questionnaire. Measurement of social support in family medicine patients. Med Care, 26(7):709–723. [DOI] [PubMed] [Google Scholar]
  12. Brunet J, McDonough MH, Hadd V, Crocker PRE, & Sabiston CM (2010). The Posttraumatic Growth Inventory: an examination of the factor structure and invariance among breast cancer survivors. Psycho-Oncology, 19, 830–838. [DOI] [PubMed] [Google Scholar]
  13. Cadell S, & Sullivan R (2006). Posttraumatic growth and HIV bereavement: Where does it start and when does it end? Traumatology, 12(1), 45. [Google Scholar]
  14. Calhoun LG, & Tedeschi RG (1998). Beyond recovery from trauma: Implications for clinical practice and research. Journal of Social Issues, 54(2), 357–371. [Google Scholar]
  15. Carver CS, & Antoni MH (2004). Finding benefit in breast cancer during the year after diagnosis predicts better adjustment 5 to 8 years after diagnosis. Health Psychology, 26, 595–598. [DOI] [PubMed] [Google Scholar]
  16. Casellas-Grau A, Ochoa C, & Ruini C (2017). Psychological and clinical correlates of posttraumatic growth in cancer. A systematic and critical review. Psycho-Oncology. [DOI] [PubMed] [Google Scholar]
  17. Cassidy T (2013). Benefit finding through caring: The cancer caregiver experience. Psychology & Health, 28(3), 250–266. [DOI] [PubMed] [Google Scholar]
  18. Connor KM, & Davidson JR (2003). Development of a new resilience scale: The Connor-Davidson resilience scale (CD-RISC). Depression and Anxiety, 18(2), 76–82. [DOI] [PubMed] [Google Scholar]
  19. Cordova MJ, Giese-Davis J, Golant M, et al. (2007). Breast cancer as trauma: posttraumatic stress and posttraumaticgrowth. Journal of Clinical Psychology in Medical Settings, 14, 308–319. [Google Scholar]
  20. Cordova MJ, Riba MB, & Spiegel D (2017). Posttraumatic stress disorder and cancer. Lacet Psychiatry, 4, 330–338. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Cormio C, Muzzatti B, Romito F, Mattioli V, & Annunziata MA (2017). Posttraumatic growth and cancer: a study 5 years after treatment end. Supportive Care in Cancer, 25(4), 1087–1096. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Crawford JJ, Vallance JK, Holt NL, & Courneya KS (2015). Associations between exercise and post,traumatic growth in gynecologic cancer survivors. Supportive Care in Cancer, 23, 705–14. doi: 10.1007/s00520-014-2410-1. Epub 2014 Aug 30. [DOI] [PubMed] [Google Scholar]
  23. Dean RA, & McClement SE (2002). Palliative care research: Methodological and ethical challenges. International Journal of Palliative Nursing, 8, 376–80. [DOI] [PubMed] [Google Scholar]
  24. Elderton A, Berry A, & Chan C (2015). A Systematic Review of Posttraumatic Growth in Survivors of Interpersonal Violence in Adulthood. Trauma Violence Abuse. doi: 10.1177/1524838015611672 [DOI] [PubMed] [Google Scholar]
  25. Forero CG, & Maydeu-Olivares A (2009). Estimation of IRT graded response models: Limited versus full information methods. Psychological methods, 14(3), 275. [DOI] [PubMed] [Google Scholar]
  26. Harding S, Sanipour F, & Moss T (2014). Existence of benefit finding and posttraumatic growth in people treated for head and neck cancer: a systematic review. Peer Journal. Feb 11;2:e256. doi: 10.7717/peerj.256. eCollection 2014. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Hays RD, Morales LS, & Reise SP (2000). Item response theory and health outcomes measurement in the 21st century. Medical care, 38(9 Suppl), II28. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Helgeson VS, Reynolds KA, & Tomich PL (2006). A meta-analytic review of benefit finding and growth. Journal of Consulting and Clinical Psychology, 74(5), 797. [DOI] [PubMed] [Google Scholar]
  29. Horswill SC, Desgagné G, Parkerson HA et al. (2014). Posttraumatic growth and hope in parents of children with cancer. Journal of Psychosocial Oncology, 32, 696–707. doi: 10.1080/07347332.2014.955241. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Jansen L, Hoffmeister M, Chang-Claude J, Brenner H, & Arndt V (2011). Benefit finding and posttraumatic growth in long-term colorectal cancer survivors: prevalence, determinants, and associations with quality of life. British Journal of Cancer, 105(8), 1158. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Jenkinson C, Fitzpatrick R, Garratt A, Peto V, & Stewart-Brown S (2001). Can item response theory reduce patient burden when measuring health status in neurological disorders? Results from Rasch analysis of the SF-36 physical functioning scale (PF-10). Journal of Neurology, Neurosurgery & Psychiatry, 71(2), 220–224. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Jordhoy MS, Kaasa S, Fayers PM, Ovreness T, Underland G, & Ahlner-Elmqvist M (1999). Challenges in palliative care research; recruitment, attrition and compliance: experience from a randomized controlled trial. Palliative Medicine, 13, 299–310. [DOI] [PubMed] [Google Scholar]
  33. Kılıç C, Magruder K, & Koryürek M (2016). Does trauma type relate to posttraumatic growth after war? A pilot study of young Iraqi war survivors living in Turkey. Transcultural psychiatry, 53(1), 110–123. [DOI] [PubMed] [Google Scholar]
  34. Li YC, Yeh PC, Chen HW, Chang YF, Pi SH, & Fang CK (2015). Posttraumatic growth and demoralization after cancer: The effects of patients’ meaning-making. Palliative and Supportive Care, 13(5), 1449–1458. [DOI] [PubMed] [Google Scholar]
  35. Maloney C, Lyons KD, Li Z, Hegel M, Ahles TA, & Bakitas M (2012). Patient perspectives on participation in the ENABLE II randomized controlled trial of a concurrent oncology palliative care intervention: benefits and burdens. Palliative Medicine, 27, 375–383. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Maltby J (1999). The internal structure of a derived, revised, and amended measure of the Religious Orientation Scale: the ‘Age-Universal’ I-E Scale-12. Social Behaivor and Personality, 27, 407–412. [Google Scholar]
  37. Mols F, Vingerhoets AJ, Coebergh JWW, & Van de Poll-Franse LV (2009). Well-being, posttraumatic growth and benefit finding in long-term breast cancer survivors. Psychology and Health, 24(5), 583–595. [DOI] [PubMed] [Google Scholar]
  38. Mosher CE, Adams RN, Helft PR, O’Neil BH, Shahda S, Rattray NA, & Champion VL (2017). Positive changes among patients with advanced colorectal cancer and their family caregivers: a qualitative analysis. Psychology & Health, 32(1), 94–109. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Pakenham KI (2005). Benefit finding in multiple sclerosis and associations with positive and negative outcomes. Health Psychology, 24, 123–132. 10.1037/0278-6133.24.2.123 [DOI] [PubMed] [Google Scholar]
  40. Park CL (2016). Meaning making in the context of disasters. Journal of Clinical Psychology, 72(12), 1234–1246. [DOI] [PubMed] [Google Scholar]
  41. Park CL, & Helgeson VS (2006). Introduction to the special section: growth following highly stressful life events--current status and future directions. Journal of Consulting and Clinical Psychology, 74(5), 791. [DOI] [PubMed] [Google Scholar]
  42. Pascoe L, & Edvardsson D (2014). Benefit finding in adult cancer populations: psychometric properties and performance of existing instruments. European Journal of Oncology Nursing, 18(5), 484–491. [DOI] [PubMed] [Google Scholar]
  43. Pascoe L, & Edvardsson D (2015). Psychometric properties and performance of the 17-item Benefit Finding Scale (BFS) in an outpatient population of men with prostate cancer. European Journal of Oncology Nursing, 19(2), 169–173. [DOI] [PubMed] [Google Scholar]
  44. Posluszny DM, Baum A, Edwards RP, & Dew MA (2011). Posttraumatic growth in women one year after diagnosis for gynecologic cancer or benign conditions. Journal of psychosocial oncology, 29(5), 561–572. [DOI] [PubMed] [Google Scholar]
  45. Purc-Stephenson RJ (2014). The Posttraumatic Growth Inventory: factor structure and invariance among persons with chronic diseases. Rehabilitation Psychology, 59, 10–18. doi: 10.1037/a0035353. Epub 2014 Jan 20. [DOI] [PubMed] [Google Scholar]
  46. Ramos C, Leal I, Marôco AL, & Tedeschi RG (2016). The Posttraumatic Growth Inventory: Factor Structure and Invariance in a Sample of Breast Cancer Patients and in a Non-Clinical Sample. Spanish Journal of Psychology, Oct 3;19:E64. [DOI] [PubMed] [Google Scholar]
  47. Rosenfeld B, Breitbart W, Stein K, …et al. (1996). Measuring desire for death among patients with HIV/AIDS: the schedule fo attitudes toward hastened death. American Journal of Psychiatry, 156, 94–100. [DOI] [PubMed] [Google Scholar]
  48. Rosenfeld B, Breitbart W, Galietta M, …et al. (2000). The schedule of attitudes toward hastened death: Measuring desire for death in terminally ill cancer patients. Cancer, 88, 2868–2875. [DOI] [PubMed] [Google Scholar]
  49. Rosenfeld B, Pessin H, Lewis C…et al. (2011). Assessing hopelessness in terminally ill cancer patients: development of the Hopelessness Assessment in Illness Questionnaire. Psychological assessment, 23(2), 325 – 336. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Rosenfeld B Personal Correspondence. April 5, 2019.
  51. Rosenfeld B Personal Correspondence. April 7, 2019.
  52. Sears SR, Stanton AL, Danoff-Burg S (2003). The yellow brick road and the emerald city: benefit finding, positive reappraisal coping, and posttraumatic growth in women with early-stage breast cancer. Health Psychology, 22, 487–497. [DOI] [PubMed] [Google Scholar]
  53. Scheier MF, Carver CS, Bridges MW (1994). Distinguishing optimism from neuroticism (and trait anxiety, self-mastery, and selfesteem)—a reevaluation of the life orientation test. Journal of Personality and Social Psychology, 67(6):1063–1078. [DOI] [PubMed] [Google Scholar]
  54. Sharp L, Redfearn D, Timmons A, Balfe M, & Patterson J (2018). Posttraumatic growth in head and neck cancer survivors: Is it possible and what are the correlates? Psychooncology, 27, 1517–1523. doi: 10.1002/pon.4682. Epub 2018 Apr 16. [DOI] [PubMed] [Google Scholar]
  55. Siegal S (1956). Nonparametric statistics for the behavioral sciences: McGraw-hill. [Google Scholar]
  56. Steffens RF, & Andrykowski MA (2016). Posttraumatic Growth Inventory: Overview. In Comprehensive Guide to Posttraumatic Stress Disorders (pp. 2203–2219): Springer International Publishing. [Google Scholar]
  57. Taku K, Calhoun LG, Cann A, & Tedeschi RG (2008). The role of rumination in the coexistence of distress and posttraumatic growth among bereaved Japanese university students. Death Studies, 32(5), 428–444. [DOI] [PubMed] [Google Scholar]
  58. Tang ST, Lin KC, Chen JS, Chang WC, Hsieh CH, & Chou WC (2015). Threatened with death but growing: changes in and determinants of posttraumatic growth over the dying process for Taiwanese terminally ill cancer patients. Psycho-Oncology, 24(2), 147–154. [DOI] [PubMed] [Google Scholar]
  59. Tedeschi RG, & Calhoun LG (1995). Trauma and transformation: Sage. [Google Scholar]
  60. Tedeschi RG, & Calhoun LG (1996). The Posttraumatic Growth Inventory: Measuring the positive legacy of trauma. Journal of traumatic stress, 9(3), 455–471. [DOI] [PubMed] [Google Scholar]
  61. Teixeira RJ & Pereira MG (2013). Growth and the cancer caregiving experience: psychometric properties of the Portuguese Posttraumatic Growth Inventory. Family Systems and Health, 31, 382–95. doi: 10.1037/a0032004. Epub 2013 May 20. [DOI] [PubMed] [Google Scholar]
  62. Tomich PL, & Helgeson VS (2004). Is finding something good in the bad always good? Benefit finding among women with breast cancer. Health Psychology, 23(1), 16. [DOI] [PubMed] [Google Scholar]
  63. Urcuyo KR, Boyers AE, Carver CS, & Antoni MH (2005). Finding benefit in breast cancer: Relations with personality, coping, and concurrent well-being. Psychology & Health, 20(2), 175–192. [Google Scholar]
  64. Walsh DM, Groarke AM, Morrison TG, Durkan G, Rogers E, & Sullivan FJ (2018). Measuring a new facet of post traumatic growth: Development of a scale of physical post traumatic growth in men with prostate cancer. PloS one, 13(4), e0195992. [DOI] [PMC free article] [PubMed] [Google Scholar]
  65. Wang Y, Zhu X, Yi J, Tang L, He J, Chen G, Li L, & Yang Y Benefit finding predicts depressive and anxious symptoms in women with breast cancer. Quality of Life Research, 24, 2681 – 2688.doi: 10.1007/s11136-015-1001-z. Epub 2015 May 24. [DOI] [PubMed] [Google Scholar]

RESOURCES