- Email: [email protected]
The Relationship Between Posttraumatic Stress Disorder, Mood States, Functional Status, and Quality of Life in Oncology Outpatients
520
Journal of Pain and Symptom Management
Vol. 44 No. 4 October 2012
Original Article
The Relationship Between Posttraumatic Stress Disorder, Mood States, Functional Status, and Quality of Life in Oncology Outpatients Jeffrey I. Gold, PhD, Marilyn ‘‘Marty’’ K. Douglas, DNSc, RN, FAAN, Mary Laudon Thomas, RN, MS, AOCN, Janette E. Elliott, RN-BC, MSN, AOCN, Stephen M. Rao, PhD, and Christine Miaskowski, RN, PhD, FAAN Keck School of Medicine (J.I.G.), University of Southern California, and Pain Management and Palliative Care Program, Department of Anesthesiology Critical Care Medicine, Children’s Hospital Los Angeles, Los Angeles; School of Nursing (M.K.D.), University of California, San Francisco, San Francisco; Veterans Affairs Palo Alto Health Care System (M.K.D., M.L.T., J.E.E.), Palo Alto; San Francisco Veterans Affairs Medical Center (S.M.R.), San Francisco; and Department of Physiological Nursing (C.M.), University of California, San Francisco, San Francisco, California, USA
Abstract Context. Oncology patients are at risk for developing posttraumatic stress disorder (PTSD) and other comorbid mood states, which are associated with decreases in functional status and quality of life (QOL). However, few studies have investigated the relationship between PTSD, other mood states, functional status, and QOL in oncology outpatients. Objectives. This study had four aims: 1) determine the percentages of patients with PTSD and partial PTSD; 2) evaluate for differences in demographic and clinical characteristics among patients with PTSD, partial PTSD, and no PTSD; 3) evaluate for differences in mood states, functional status, and QOL among the three PTSD groups; and 4) evaluate whether demographic and disease characteristics were predictors of PTSD. Methods. As part of a larger clinical trial that evaluated the effects of a cognitivebehavioral intervention on cancer pain management, 289 adult oncology patients (Mage ¼ 61.3, SD ¼ 11.6) completed self-report measures that assessed PTSD, other mood states, functional status, and QOL. Results. Forty-five percent of the sample met the diagnostic criteria for PTSD (n ¼ 78) and partial PTSD (n ¼ 53) and were younger than those with no PTSD. Patients with PTSD had a significantly lower Karnofsky Performance Status, shorter time since diagnosis, higher ratings of mood disturbance, lower ratings of functional status, and lower QOL than patients with no PTSD. A lower Karnofsky Performance Status, fewer months since diagnosis, and presence of bone metastases predicted a higher likelihood of being classified as having PTSD.
Address correspondence to: Jeffrey I. Gold, PhD, Keck School of Medicine, University of Southern California, Children’s Hospital Los Angeles, 4650 Sunset Ó 2012 U.S. Cancer Pain Relief Committee. Published by Elsevier Inc. All rights reserved.
Boulevard, MS #12, Los Angeles, CA 90027-6062, USA. E-mail: [email protected] Accepted for publication: October 27, 2011. 0885-3924/$ - see front matter doi:10.1016/j.jpainsymman.2011.10.014
Vol. 44 No. 4 October 2012 PTSD, Mood States, Functional Status, and QOL in Oncology Outpatients
521
Conclusion. Early identification and interventions may mediate the onset of psychological distress and improve health outcomes. J Pain Symptom Manage 2012;44:520e531. Ó 2012 U.S. Cancer Pain Relief Committee. Published by Elsevier Inc. All rights reserved. Key Words Cancer, posttraumatic stress disorder (PTSD), mood states, functional status, quality of life
Introduction Patients diagnosed and treated for cancer are at increased risk for psychological distress, such as depressive symptoms and posttraumatic stress disorder (PTSD). PTSD is an anxiety disorder that develops following exposure to an extreme traumatic stressor in which the person experienced, witnessed, or was confronted with an event or events that involved actual or threatened death or serious injury or a threat to the physical integrity of self or others (Criterion A1 from the Diagnostic and Statistical Manual of Mental Disorders, 4th ed [DSM-IV]).1 The person’s response to the event must include intense fear, helplessness, or horror (Criterion A2). Symptoms that can result from the traumatic event include persistent reexperiencing of the traumatic event (Criterion B), persistent avoidance of the stimuli associated with the trauma and numbing of general responsiveness (Criterion C), and persistent symptoms of hyperarousal (Criterion D). Because the DSM-IV criteria for PTSD were introduced as a nosological category,1 they have been seen in research as restrictive criteria for individuals who experience PTSD symptoms but do not completely fulfill the DSM-IV criteria.2 It is important to consider individuals who experience the distressing symptoms of PTSD, as an individual’s level of functioning may be impaired because of these symptoms. For this reason, partial PTSD is defined as meeting Criteria A1 and A2, as well as two of the three remaining PTSD criteria (i.e., reexperiencing, avoidance/numbing, or hyperarousal).3 Moreover, partial PTSD captures some of the debilitating symptoms that are experienced in PTSD. Findings from four reviews4e7 suggest that, in some patients, the diagnosis and treatment of cancer lead to the persistence of partial or fullblown PTSD symptoms. Although some cancers are not life threatening, the initial diagnosis of
cancer is interpreted as a life-threatening experience and may increase the risk of PTSD. In one study, 32% of Hodgkin’s disease survivors had either partial or full PTSD and reported significantly lower quality of life (QOL) scores than those without PTSD.8 Current levels of cancer symptoms and the PTSD subdimension of hyperarousal were identified as the strongest predictors of depression. Amir and Ramati9 found a significantly higher rate of PTSD and partial PTSD in 39 long-term survivors of breast cancer compared with 39 matched healthy women. Survivors reported significantly higher levels of emotional distress and lower physical and psychological QOL scores and exhibited poorer coping styles compared with the control group. Multiple regression analysis showed that chemotherapy and disease stage as well as the interaction between chemotherapy and disease stage were significant predictors of hyperarousal. A review on PTSD and cancer reported that individuals with poor premorbid physical health are at greater risk for having PTSD.6 Therefore, it is important to examine the impact of comorbidities on PTSD, given that PTSD in oncology patients is highly variable (i.e., full or partial PTSD), to provide appropriate interventions for potential impairments. In the cancer population, most studies of PTSD are focused on women with breast cancer. Within the veteran population, PTSD is studied within the context of combat-related PTSD and how it relates to substance abuse10,11 and other neuropsychiatric conditions12,13 (e.g., dementia, Korsakoff’s syndrome). However, no studies have evaluated the relationships between PTSD symptoms and other outcomes in a sample of oncology patients, the majority of whom received their care at Veterans Affairs (VA) health care settings. Although studies have noted a wide array of PTSD prevalence rates in patients with cancer
522
Gold et al.
and have reported on the impact of PTSD symptoms in cancer survivors,14 less is known about the occurrence and impact of PTSD symptoms on physical and psychological function. In one study, adults with lymphoma (n ¼ 44) who met criteria for full or partial PTSD reported lower physical activity/QOL.8 Given the paucity of research on PTSD in oncology patients, the purposes of this study among a mostly heterogeneous, male veteran, oncology outpatient population were to 1) determine the percentages of patients who met diagnostic criteria for PTSD, partial PTSD, and no PTSD (i.e., PTSD groups); 2) evaluate for differences in demographic (e.g., age, gender) and clinical (e.g., time since diagnosis, treatment) characteristics among the PTSD groups; 3) evaluate for differences in mood states, QOL, and functional status among the three PTSD groups; and 4) evaluate whether specific demographic and disease characteristics were predictors of PTSD group membership. It was hypothesized that patients with PTSD or partial PTSD would have significantly more mood disturbance, lower functional status, and QOL scores than patients with no PTSD. Additionally, it was hypothesized that demographic and disease characteristics would predict PTSD group membership.
Methods Sample and Settings This study was part of a larger randomized clinical trial that evaluated the effects of a cognitive-behavioral intervention for cancer pain management.15,16 A total of 289 patients participated in this study, of which 85% were recruited from VA medical centers. Patients were eligible if they were older than 18 years; had a diagnosis of cancer; had cancer pain; were able to read, write, and understand English; and had a minimum life expectancy of six months. Patients were excluded if they had a current cognitive or psychiatric condition that could jeopardize adherence to the protocol or had a comorbid pain syndrome that caused more severe pain than their cancer pain. Eligible patients were recruited from oncology and hematology clinics at four U.S. Department of Veterans Affairs medical centers, one public hospital, and one community group practice. Of the 406 patients approached, 84 declined to participate. A total
Vol. 44 No. 4 October 2012
of 322 patients responded and provided written informed consent. Four participants dropped out of the study before randomization. Twentynine participants dropped out of the study for a variety of reasons, including death. The institutional review boards at each of the study sites approved this study. The data presented in this article were obtained before the start of the intervention.
Study Procedures Oncologists or nursing staff identified potential participants to a research assistant who then approached the patient during an oncology clinic appointment and determined if they met the study’s inclusion criteria. These patients were invited to participate and if they agreed, written informed consent was obtained. At the time of enrollment, patients completed the baseline questionnaires. Patients’ medical records were reviewed for disease and treatment information.
Instruments Demographics. A demographic questionnaire obtained information on a variety of demographic characteristics: age, education, gender, ethnicity, marital status, and living arrangement. Disease and Treatment Characteristics. Disease and treatment characteristics were collected from medical records, including cancer diagnosis, stage of disease, treatment history, date of cancer diagnosis, and PTSD diagnosis. Patients were asked if they had been diagnosed with PTSD (but not entered into their medical record). Patients completed the Karnofsky Performance Status (KPS) scale, a measure widely used to subjectively estimate the functional status of patients with cancer.17 KPS scores are based on intervals of 10, ranging from normal functioning (100) to moribund (10), with a midpoint of requiring considerable assistance and frequent medical care (50). The KPS has acceptable testretest and interrater reliability as well as validity.17e19 PTSD ChecklistdCivilian Version. The PTSD ChecklistdCivilian Version (PCL-C) is a 17-item self-report inventory that corresponds to the DSM-IV symptoms of PTSD20 and is widely used for measuring PTSD symptoms or establishing the diagnosis of PTSD in adults.
Vol. 44 No. 4 October 2012 PTSD, Mood States, Functional Status, and QOL in Oncology Outpatients
It is a well-established, reliable, and valid measure of PTSD developed and validated by the National Center for PTSD.20,21 The PCL-C yields a total PTSD score and subscale scores for reexperiencing symptoms (Cluster B, five items), avoidance and numbing symptoms (Cluster C, seven items), and hyperarousal symptoms (Cluster D, five items). Although cancer does not qualify as a PTSD diagnosis in the International Classification of Diseases, Tenth Revision, the PCL-C was used to identify a subset of oncology patients who met PTSD criteria. The PCL-C was modified to specifically reflect symptoms related to the diagnosis and treatment of cancer. This modification was done to decrease attention on other potentially confounding traumatic events (e.g., combatrelated PTSD). Patients responded to each of the items in the following manner: ‘‘Please indicate how distressing each difficulty has been for you during the past 30 days with respect to being diagnosed and treated for cancer.’’ The patients responded to each item using a five-point Likert scale that ranged from 1 (not at all) to 5 (extremely). The PCL-C has been used with veterans20,22and chronically ill adults.23e25 Studies have established high correlations (i.e., greater than 0.85) between the PCL-C and an interviewestablished PTSD diagnosis. These studies have demonstrated strong test-retest reliability and strong internal consistency, with Cronbach’s alpha values that ranged from 0.92 to 0.97.20e22 In this study, the Cronbach’s alpha for the total score was 0.91. The DSM-IV diagnostic criteria also were used to create a diagnosis of PTSD or partial PTSD. To be placed into the PTSD group, the patient needed to endorse (i.e., have an item score of 3 or more on the PCL-C) in the following cluster symptom pattern: one item from Cluster B (reexperiencing), three items from Cluster C (avoidance/numbing), and two items from Cluster D (hyperarousal). Partial PTSD was met if the patient endorsed two of the three symptom clusters. These criteria were used to categorize patients into one of the three PTSD groups (i.e., PTSD, partial PTSD, or no PTSD) instead of cutoff scores because this approach reflects symptom endorsement from all necessary clusters.25,26 Profile of Mood StatesdShort Form. The Profile of Mood StatesdShort Form (POMS-SF) is
523
a 30-item instrument that evaluates six mood states (tension-anxiety, depression-dejection, anger-hostility, vigor-activity, fatigue-inertia, and confusion-bewilderment).27 Participants responded on a five-point Likert scale that ranged from 0 (not at all) to 4 (extremely). Subscale scores and a total mood disturbance (TMD) score are calculated with vigor weighted negatively. In a sample of 428 patients with cancer, Cronbach’s alpha values ranged from 0.78 to 0.91.27 Support for strong convergent and discriminate validity for the POMS-SF was found in correlations with the Center for Epidemiologic Studies Depression Scale, the KPS, the Medical Outcomes Study Short Form-20 Physical Functioning Scale, and the Bradburn Positive and Negative Affect Scales.27 In the present study, Cronbach’s alpha values ranged from 0.84 to 0.92 for the mood subscales and was 0.91 for the TMD score. Functional Assessment of Cancer Therapyd General. The Functional Assessment of Cancer TherapydGeneral (FACT-G) is a 27-item self-administered instrument that evaluates four domains of health-related QOL in patients receiving cancer treatment (physical, social, emotional, and functional well-being).28 Each item is rated on a five-point Likert scale that ranged from 0 (not al all) to 4 (very much). Items are summed to obtain subscale scores. The sum of all the subscales comprises the total score, which can range from 0 to 108, with a higher score indicating better healthrelated QOL. The validity and reliability of the FACT-G are well established in patients with cancer.28 In this study, the Cronbach’s alpha was 0.51 for the total score, 0.79 for the physical well-being subscale, 0.78 for the social/family well-being subscale, 0.63 for the emotional well-being subscale, and 0.80 for the functional well-being subscale. Functional Health StatusdShort Form-36 for Veterans. The Short Form-36 for Veterans (SF-36V) measures eight health status concepts (physical functioning, role limitations because of physical health problems, bodily pain, social functioning, role limitations because of emotional health problems, general mental health, vitality, and perception of general health).29 Scales are summed and linearly transformed on a 0e100 scale, with higher
524
Gold et al.
scores reflecting better health states. The validity and reliability of the SF-36V are well established.29 The Cronbach’s alpha values for the subscales ranged from 0.70 to 0.91.
Statistical Analyses Data were analyzed using the SPSS version 15.0 software for Windows (SPSS Inc., Chicago, IL). Descriptive statistics were generated for all demographic, clinical, and outcome variables. One-way analyses of variance and Chisquare analyses were performed to evaluate for differences in demographic, clinical, and outcome variables among the three PTSD groups (PTSD, partial PTSD, and no PTSD). All calculations used actual values. Adjustments were not made for missing data. One percent of the data on KPS scores was missing, 1.4% of the POMS-SF data was missing, and the subscales for SF-36V had missing data that ranged from zero to two subscales missing 1.4%. Data were missing for FACT-G (n ¼ 7) and the PCL-C (n ¼ 5). Therefore, the cohort for each analysis was dependent on the largest set of data across PTSD groups. If the overall analyses indicated differences among the three PTSD groups, Bonferroni correction post hoc analyses were conducted to determine specific group differences. All P-values were adjusted so that values of P < 0.05 were considered statistically significant. Binary logistical regression analyses were used to evaluate the unique contributions of demographic and clinical characteristics to PTSD group membership (i.e., no PTSD vs. partial/full PTSD). For these analyses, P < 0.05 was considered statistically significant.
Results Percentage of Patients With PTSD and Partial PTSD Based on the prespecified criteria, 27% (n ¼ 78) of the 289 patients met the criteria for PTSD and 18% (n ¼ 53) met the criteria for partial PTSD.
Demographic Characteristics As shown in Table 1, the sample was mostly Caucasian (60%), male (88%), not married (58%), living with someone (77%), not working (92%), and receiving care at a VA facility (85%). The average age was 61.3 years (SD ¼ 11.6), and
Vol. 44 No. 4 October 2012
patients had an average of 13 years of education (SD ¼ 2.8). No differences were found among the PTSD groups in any demographic characteristic except age. Patients in the PTSD and partial PTSD groups were significantly younger than those in the no PTSD group (F(3, 286) ¼ 4.2, P < 0.05).
Disease and Treatment Characteristics As shown in Table 1, 20% of the patients had received prior cancer treatment, and 32% were receiving current cancer treatment. Twentythree percent of the patients were diagnosed with lung cancer, 16% with prostate cancer, 12% with head and neck cancer, 8% with breast cancer (22 cases: 21 females and one male), 6% with myeloma, 5% with rectal cancer, and another 5% with colon cancer. The remaining sample (25%) had a variety of cancer diagnoses. Patients in the PTSD group had significantly lower KPS scores, and a significantly shorter amount of time had elapsed since their cancer diagnoses when compared with the patients with no PTSD (P < 0.01). Significant differences were noted between the groups based on the presence of bone metastases (P < 0.01), with a larger proportion of patients in the no PTSD group having evidence of bone metastases compared with the partial PTSD and PTSD groups (Table 1). Forty-one patients (14%) had a self-reported or medical record-documented diagnosis of non-cancer-related PTSD. No significant differences in total PTSD symptom score or other health-related outcomes were found between patients with and without non-cancer-related PTSD. Therefore, all patients (n ¼ 289) were included in the subsequent analyses.
Differences in Mood States As shown in Fig. 1, significant differences were found in all the POMS-SF subscale scores, as well as the TMD score, among the three PTSD groups (all P < 0.01). Post hoc contrasts determined that patients in the PTSD group scored significantly higher on all the POMSSF subscales, except vigor-activity, than patients in the partial PTSD group, who in turn scored higher than patients in the no PTSD group. Conversely, patients in the PTSD group scored significantly lower in the vigor-activity subscale than individuals in the partial PTSD group,
Vol. 44 No. 4 October 2012 PTSD, Mood States, Functional Status, and QOL in Oncology Outpatients
525
Table 1 Differences in Demographic, Disease, and Treatment Characteristics Among the PTSD Groups
Characteristic Age, years Education, years Karnofsky score Months since cancer diagnosis
Total Sample (N ¼ 289), Mean (SD) 61.3 13.2 75.0 33
(11.6) (2.8) (12.9) (47.3)
No PTSD (n ¼ 158), Mean (SD) 63.1 13.5 77.5 77.5
(11.3) (2.8) (12.7) (12.7)
Partial PTSD (n ¼ 53), Mean (SD) 59.4 12.6 74.2 74.2
(12.6) (2.7) (12.9) (12.9)
PTSD (n ¼ 78), Mean (SD) 59.5 13.0 70.8 70.8
(12.6) (2.7) (12.2) (12.2)
n (%)
n (%)
n (%)
n (%)
Type of care VA care No VA care
247 (85) 42 (15)
136 (86) 22 (14)
46 (87) 7 (13)
65 (83) 13 (17)
Gender Male Female
254 (88) 42 (15)
139 (88) 22 (14)
47 (89) 7 (13)
68 (87) 13 (17)
Ethnicity Caucasian Ethnic minority
173 (60) 116 (40)
96 (61) 62 (39)
33 (62) 20 (38)
44 (56) 34 (44)
Marital status Married Not married
119 (42) 168 (58)
65 (41) 93 (59)
21 (40) 32 (60)
33 (42) 45 (58)
Living arrangement Alone With somebody
65 (23) 224 (78)
35 (22) 123 (78)
9 (17) 44 (83)
21 (27) 57 (73)
Employment Working Not working
23 (8) 263 (92)
13 (8) 145 (92)
3 (6) 49 (93)
7 (9) 69 (89)
Cancer diagnosis Lungc Other
67 (23) 222 (77)
36 (23) 122 (77)
10 (19) 43 (81)
21 (27) 57 (73)
a
4.2 2.0 7.5b 7.5b
3<1 3<1 3<1 c2 (df) 1.01 (2)
0.07 (2)
0.81 (2)
0.05 (2)
1.81 (2)
0.51 (2)
1.81 (2)
12.56b (2)
Evidence of bone metastases Yes No
99 (34) 189 (66)
68 (43) 89 (56)
11 (21) 42 (79)
20 (26) 58 (74)
Prior cancer therapy Yes No
59 (20) 230 (80)
35 (22) 123 (78)
10 (19) 43 (81)
14 (18) 64 (82)
Current cancer therapy Yes No
Post Hoc
F
0.66 (2)
5.68 (2) 91 (32) 196 (68)
59 (37) 98 (62)
12 (23) 41 (77)
20 (26) 57 (73)
PTSD ¼ posttraumatic stress disorder; VA ¼ Veterans Affairs. a P < 0.05. b P < 0.01. c Most frequent type of cancer diagnosis in this sample.
who scored lower than patients in the no PTSD group.
Differences in FACT-G and SF-36V All measures of health-related QOL were significantly different among the three PTSD groups (Table 2), with higher subscale and total scores in the no PTSD group than in the PTSD group. All measures of function were significantly different among the three PTSD
groups, except for the SF-36V physical function score and its associated component score (Table 2). Again, patients in the no PTSD group reported better function than those in the PTSD group.
Predictors of PTSD Group Membership A binary logistic regression was performed, with age and gender entered first as control variables and with KPS score, months since
526
Gold et al.
Vol. 44 No. 4 October 2012
Fig. 1. Differences in Profile of Mood States subscale (a) and total (b) scores among the three PTSD groups. All values are plotted as means and SDs. For all subscales and total mood disturbance scores except vigor: full PTSD group > partial PTSD group > no PTSD group (all P < 0.05). For vigor, full PTSD group < partial PTSD group < no PTSD group (all P < 0.05). PTSD ¼ posttraumatic stress disorder.
diagnosis, cancer type (lung vs. other), and presence of bone metastases entered as predictors of PTSD diagnosis (no PTSD vs. partial/ full PTSD). Although no significant differences were found for gender and cancer type, they were controlled for because previous research demonstrated their association with PTSD.7,30 The final model with all predictors was significant and explained only 12.7% of the variance in PTSD diagnosis (Cox and Snell R-square). The nonsignificant Hosmer and Lemeshow Chi-square statistic demonstrated a good model fit. The overall classification rate for the final model with all predictors included was 65.5%, with 70.4% of the patients without PTSD and 59.7% of the patients with PTSD (partial or full PTSD) correctly classified (Table 3).
As shown in Table 3, independent of age and gender, KPS score, months since diagnosis, and presence of bone metastases were significantly related to PTSD diagnosis. For every 1% decrease in KPS score, patients were 4% more likely to be classified as having PTSD (partial or full). For every additional month since diagnosis, patients were 1% less likely to be classified as having PTSD (partial or full). Finally, patients with bone metastases were 2.2 times more likely to be classified as having PTSD (partial or full) than those without bone metastases.
Discussion This study is the first cross-sectional examination of PTSD, comorbid mood states,
Vol. 44 No. 4 October 2012 PTSD, Mood States, Functional Status, and QOL in Oncology Outpatients
527
Table 2 Differences in FACT-G and SF-36V Scores Among the PTSD Groups (n ¼ 289) Instruments
Total Sample, Mean (SD)
No PTSD (n ¼ 154), Mean (SD)
Partial PTSD (n ¼ 53), Mean (SD)
PTSD (n ¼ 78), Mean (SD)
F
Post Hoc
22.7 (5.4)
15.9 (5.9)
18.5 (4.9)
13.6 (5.8)
12.2 (5.3)
44.1b
2 and 3 < 1
19.1 (6.8)
20.2 (5.8)
21.5 (5.6)
20.2 (5.8)
17.8 (5.6)
10.8b
3<1
b
Normative Data,a Mean (SD)
FACT-G subscales Physical wellbeing Social family wellbeing Emotional wellbeing Functional wellbeing FACT total score SF-36V subscales Physical Functioning Role Physical Bodily Pain General Health Vitality Social Functioning Role Emotional Mental Health Physical component score Mental component score
19.9 (4.8)
16.6 (5.0)
19.0 (3.9)
15.4 (4.4)
12.7 (4.7)
61.2
3<2<1
18.5 (6.8)
13.1 (5.8)
15.4 (5.6)
11.8 (5.0)
9.4 (4.3)
37.5b
3<2<1
b
3<2<1
80.1 (18.1)
65.9 (16.3)
74.4 (13.2)
61.0 (14.0)
52.1 (11.9)
79.9
56.8 (28.9)
40.7 (26.5)
42.7 (27.1)
42.7 (26.8)
35.4 (24.5)
2.2
39.7 55.4 51.4 52.6 67.4
33.7 34.2 43.6 36.5 50.1
38.3 36.7 50.4 40.3 57.9
30.6 32.6 42.7 35.4 45.8
26.5 30.4 30.2 29.8 37.2
(38.4) (27.2) (23.2) (23.25) (28.2)
(25.8) (18.1) (22.1) (20.5) (27.2)
(28.1) (19.0) (22.1) (22.6) (27.9)
(23.3) (18.2) (21.7) (16.9) (27.7)
(20.4) (15.3) (15.8) (16.2) (18.6)
6.2b 3.4c 25.4b 7.2b 17.9b
1>3 1>3 1 and 2 > 3 1>3 1 and 2 > 3
59.6 (37.5) 69.0 (22.8) 37.1 (11.9)
56.6 (32.4) 64.5 (20.1) 31.3 (8.4)
68.5 (31.6) 73.6 (18.1) 31.3 (8.4)
47.0 (28.5) 58.5 (18.7) 32.5 (7.8)
39.3 (26.2) 50.3 (14.3) 30.6 (7.4)
28.6b 51.4b 0.9
1 and 2 > 3 1>2>3
47.8 (12.2)
42.5 (12.4)
48.2 (11.8)
38.3 (10.4)
33.9 (8.1)
51.4b
1 and 2 > 3
FACT-G ¼ Functional Assessment of Cancer TherapydGeneral; SF-36V ¼ Short Form-36V; PTSD ¼ posttraumatic stress disorder; FACT ¼ Functional Assessment of Cancer Therapy. a Normative data compared with cancer patients.43,44 b P < 0.01. c P < 0.05.
functional status, and QOL in a heterogeneous sample of patients with cancer. The first aim of the study was to determine the percentage of patients who met diagnostic criteria for PTSD. Approximately 45% of these patients had PTSD symptoms associated with their cancer experience. This percentage is higher than previous reports in oncology patients that ranged from 2% to 22%.8,9,14,23,24,30e36 A
number of reasons may account for this higher percentage of patients having PTSD or partial PTSD, including the fact that the majority were veterans. In general, veterans, with or without combat exposure, have a higher incidence of lifetime PTSD (30% in men and 27% in women) compared with the lifetime prevalence of PTSD among men (3.6%) and women (9.7%) in the general population.37 However,
Table 3 Results of Binary Logistic Regression Analysis Predicting PTSD Diagnosis (No PTSD vs. Partial/Full; n ¼ 281) Characteristics Age Men (Ref.) Karnofsky score Months since diagnosis Lung cancer (Ref.) Bone metastases (Ref.)
B
SE
OR
95% CI
Wald Statistic
0.20 0.16 0.40 0.01 0.25 0.80
0.01 0.42 0.01 0.01 0.31 0.29
0.98 0.85 0.96 0.99 1.28 2.24
0.96e1.00 0.38e1.92 0.94e0.98 0.99e1.00 0.70e2.36 1.27e3.94
2.61 0.15 14.67c 5.69a 0.63 7.76b
PTSD ¼ posttraumatic stress disorder; OR ¼ odds ratio; Ref. ¼ reference group. a P < 0.05. b P < 0.01. c P < 0.001.
528
Gold et al.
a validated measure focused specifically on symptoms associated with the cancer experience was used in an attempt to decrease symptoms associated with non-cancer-related events. Additionally, it should be noted that when the distributions of PTSD, partial PTSD, and no PTSD were examined between the patients from non-VA facilities and VA facilities, no significant differences were found. Given that patients with previous chart- or self-reported PTSD were not significantly different from patients without previously documented PTSD, these results suggest that PTSD or partial PTSD related to the cancer experience may account for differences in mood, functional outcomes, and QOL in patients with cancer. A cancer-specific PTSD measure and the inclusion and analysis of patients with partial PTSD symptoms may account for the increased occurrence of PTSD rates. An examination of partial PTSD provided critical information about the patients, who would have been classified as not having PTSD. An evaluation of the oncology patients’ military experience, regardless of care setting, may provide important information to clinicians. Further research is warranted in this area to evaluate the rates of PTSD and partial PTSD in VA and non-VA facilities to facilitate the identification and treatment of these high-risk patients. The second aim of this study was to evaluate for differences in demographic and clinical characteristics among the PTSD groups. No differences were found among the PTSD groups in any demographic characteristic except age. Specifically, patients in the PTSD and partial PTSD groups were significantly younger than those in the no PTSD group. All patients enrolled in this study had pain associated with their cancer experience. Therefore, the relationship between pain and functional status among the three PTSD groups was examined and significant differences emerged among the PTSD and the no PTSD groups. Patients with PTSD demonstrated poor levels of physical functioning in a group of patients with relatively low SF-36V scores. This study confirmed the relationship between increased pain and PTSD. This finding is consistent with recent theories that, in medical conditions such as cancer or other unintentional injuries, pain may trigger PTSD symptoms or pain may be a cooccurring condition.38,39 Furthermore, findings
Vol. 44 No. 4 October 2012
from studies in patients with chronic pain following motor vehicle injures,40 in major burn victims,41 and in methadone-maintained patients42 support an association between pain and PTSD. As hypothesized, patients with full PTSD or partial PTSD reported higher levels of mood disturbance than those without PTSD. As expected, the partial PTSD and PTSD groups had less vigor-activity compared with the no PTSD group. In addition, patients with PTSD or partial PTSD reported a higher number of comorbid conditions. All the between-group differences in POMS-SF subscale and total scores were both statistically significant and clinically meaningful as reflected by effect sizes that ranged from 0.14 to 1.44. Similarly, mood disturbance scores in patients with partial PTSD and PTSD, when compared with other published POMS scores in patients with cancer,27,32 were found to be consistently higher, with effect sizes ranging from 0.15 to 0.80. These results suggest that patients with partial and full PTSD not only have the disturbance of PTSD symptoms, but also experience higher levels of mood disturbance that may be related to their diagnosis of cancer. The hypothesis that patients with PTSD and partial PTSD would have decreases in functional status and QOL was confirmed. All the between-group differences in functional status and QOL outcomes represent both clinically and statistically meaningful differences (i.e., effect sizes ranged from 0.22 to 1.37). In addition, compared with normative values for the FACT-G and SF-36V,43,44 patients with PTSD or partial PTSD had significantly lower scores than the patients with no PTSD. However, patients with no PTSD were similar to the normative group on the SF-36V and the FAGT-G.43,44 Whereas the SF-36V is a generic measure of function, the FACT-G is a QOL measure that was specifically developed and normed for patients with cancer. Findings from this study suggest that patients with PTSD and partial PTSD have extremely poor functional status and QOL. Lastly, it was hypothesized that demographic and disease characteristics would predict PTSD group membership. Given that the present study found that time since diagnosis, KPS score, and the presence of bone metastases explained only a small portion of the variance in PTSD status
Vol. 44 No. 4 October 2012 PTSD, Mood States, Functional Status, and QOL in Oncology Outpatients
(i.e., 13%), additional research is warranted to determine what other demographic and clinical factors place patients with cancer at higher risk for PTSD. Additionally, a higher percentage of patients in the no PTSD group (43%) had evidence of bone metastases compared with the partial (21%) and full (26%) PTSD groups. This finding suggests that although having bone metastases was shown to be a risk factor, having bone metastases does not necessarily mean that the patient will develop PTSD. Therefore, future research should examine possible mechanisms that link the association between bone metastases and the development of PTSD. Previous research has demonstrated that factors such as social support, being non-Caucasian, living alone, and lower education are predictors of PTSD in patients living with breast cancer23,31,45 and in childhood cancer survivors.46 In addition to identifying factors predictive of functional decline in this population, the identification of protective and/or posttraumatic growth needs to be considered.47e49 Several study limitations need to be acknowledged. Findings on the prevalence of PTSD warrant replication in larger samples of patients with cancer who receive care within and outside the VA health care system. In addition, veteran status should be evaluated in all patients with cancer to determine its impact on PTSD occurrence rates. The findings are generalizable to oncology patients in the U.S. However, because the sample was primarily male, the findings may not be generalizable to women with cancer, particularly women with nonbreast cancer. Because the study was cross sectional, additional investigation is warranted on how PTSD symptoms change over time. Additionally, the reliability for the total score (0.51) and the subscale of emotional well-being (0.63) of the FACT-G was lower than expected. It is unclear why the internal consistency was lower, given that the FACT-G was shown to be reliable and valid in patients with cancer. Lastly, the identification of other relevant predictors of PTSD may require a larger sample. Despite these limitations, findings from this study suggest that PTSD symptoms are prevalent in patients with cancer and that these symptoms have a negative impact on patients’ mood, functional status, and QOL. These findings warrant replication in larger and more heterogeneous samples, including different care settings, so
529
that specific and tailored interventions can be designed and tested in future investigations.
Disclosures and Acknowledgments This research was supported by the Department of Veterans Affairs, Veterans Health Administration, Health Services Research and Development Service (project no. NRI-97026). The views expressed in this article are those of the authors and do not necessarily represent the views of the Department of Veterans Affairs. The authors declare no conflicts of interest.
References 1. American Psychiatric Association. Diagnostic and statistical manual of mental disorders, 4th ed. Washington, DC: American Psychiatry Association, 1994. 2. Mylle J, Maes M. Partial posttraumatic stress disorder revisited. J Affect Disord 2004;78:37e48. 3. Carlier IV, Gersons BP. Partial posttraumatic stress disorder (PTSD): the issue of psychological scars and the occurrence of PTSD symptoms. J Nerv Ment Dis 1995;183:107e109. 4. Gurevich M, Devins GM, Rodin GM. Stress response syndromes and cancer: conceptual and assessment issues. Psychosomatics 2002;43:259e281. 5. Neel ML. Posttraumatic stress symptomatology and cancer. Int J Emerg Ment Health 2000;2:85e94. 6. Smith MY, Redd WH, Peyser C, Vogl D. Posttraumatic stress disorder in cancer: a review. Psychooncology 1999;8:521e537. 7. Kangas M, Henry JL, Bryant RA. Posttraumatic stress disorder following cancer. A conceptual and empirical review. Clin Psychol Rev 2002;22:499e524. 8. Geffen DB, Blaustein A, Amir MC, Cohen Y. Post-traumatic stress disorder and quality of life in long-term survivors of Hodgkin’s disease and nonHodgkin’s lymphoma in Israel. Leuk Lymphoma 2003;44:1925e1929. 9. Amir M, Ramati A. Post-traumatic symptoms, emotional distress and quality of life in long-term survivors of breast cancer: a preliminary research. J Anxiety Disord 2002;16:195e206. 10. Bremner JD, Southwick M, Darnell A, Charney DS. Chronic PTSD in Vietnam combat veterans: course of illness and substance abuse. Am J Psychiatry 1996;153:369. 11. Breslau N, Davis GC, Schultz LR. Posttraumatic stress disorder and the incidence of nicotine, alcohol, and other drug disorders in persons who have experienced trauma. Arch Gen Psychiatry 2003;60: 289e294.
530
Gold et al.
12. Koenen KC, Driver KL, Oscar-Berman M, et al. Measures of prefrontal system dysfunction in posttraumatic stress disorder. Brain Cogn 2001;45: 64e78. 13. Yaffe K, Vittinghoff E, Lindquist K, et al. Posttraumatic stress disorder and risk of dementia among US veterans. Arch Gen Psychiatry 2010;67: 608e613. 14. Black EK, White CA. Fear of recurrence, sense of coherence and posttraumatic stress disorder in haematological cancer survivors. Psychooncology 2005;14:510e515. 15. Fahey KF, Rao SM, Douglas MK, et al. Nurse coaching to explore and modify patient attitudinal barriers interfering with effective cancer pain management. Oncol Nurs Forum 2008;35:233e240. 16. Thomas ML, Elliott JE, Rao S, et al. A randomized clinical trial of education or motivational-interviewingbased coaching compared to usual care to improve cancer pain management. Oncol Nurs Forum 2012; 39:39e49. 17. Conhill C, Verger E, Salamero M. Performance status assessment in cancer patients. Cancer 1990; 65:1864e1866. 18. Grieco A, Long CJ. Investigation of the Karnofsky performance status as a measure of quality of life. Health Psychol 1984;3:129e142. 19. Schag CC, Heinrich RL, Ganz PA. Karnofsky performance status revisited: reliability, validity, and guidelines. J Clin Oncol 1984;2:187e193. 20. Weathers FW, Litz BT, Huska JA, Keane TM. The PTSD Checklist-Civilian Version (PCL-C). Boston, MA: National Center for PTSD, Boston Veterans Affairs Medical Center, 1994. 21. Berlant J, van Kammen DP. Open-label topiramate as primary or adjunctive therapy in chronic civilian posttraumatic stress disorder: a preliminary report. J Clin Psychiatry 2002;63:15e20. 22. Dobie DJ, Kivlahan DR, Maynard C, et al. Posttraumatic stress disorder in female veterans: association with self-reported health problems and functional impairment. Arch Intern Med 2004;164: 394e400. 23. Andrykowski MA, Cordova MJ. Factors associated with PTSD symptoms following treatment for breast cancer: test of the Andersen model. J Trauma Stress 1998;11:189e203.
Vol. 44 No. 4 October 2012
26. Akechi T, Okuyama T, Sugawara Y, et al. Major depression, adjustment disorders, and post-traumatic stress disorder in terminally ill cancer patients: associated and predictive factors. J Clin Oncol 2004;22: 1957e1965. 27. Baker F, Denniston M, Zabora J, Polland A, Dudley WN. A POMS short form for cancer patients: psychometric and structural evaluation. Psychooncology 2000;11:273e281. 28. Cella DF, Tulsky DS, Gray G, et al. The Functional Assessment of Cancer Therapy Scale: development and validation of the general measure. J Clin Oncol 1993;11:570e579. 29. Kazis LE. The Veterans SF-36 health status questionnaire: development and application in the Veterans Health Administration. Med Outcomes Monit 2000;5:1e18. 30. Pitman RK, Lanes DM, Williston SK, et al. Psychophysiologic assessment of posttraumatic stress disorder in breast cancer patients. Psychosomatics 2001;42:133e140. 31. Cordova MJ, Cunningham LL, Carlson CR, Andrykowski MA. Posttraumatic growth following breast cancer: a controlled comparison study. Health Psychol 2001;20:176e185. 32. Deimling GT, Kahana B, Bowman KF, Schaefer ML. Cancer survivorship and psychological distress in later life. Psychooncology 2002;11: 479e494. 33. Tjemsland L, Soreide JA, Malt UF. Posttraumatic distress symptoms in operable breast cancer III: status one year after surgery. Breast Cancer Res Treat 1998;47:141e151. 34. Steginga SK, Pinnock C, Gardner M, Gardiner RA, Dunn J. Evaluating peer support for prostate cancer: the prostate cancer peer support inventory. BJU Int 2005;95:46e50. 35. Smith SK, Zimmerman S, Williams CS, Preisser JS, Clipp EC. Post-traumatic stress outcomes in nonHodgkin’s lymphoma survivors. J Clin Oncol 2008; 26:934e941. 36. Kornblith AB, Herndon JE 2nd, Weiss RB, et al. Long-term adjustment of survivors of early-stage breast carcinoma, 20 years after adjuvant chemotherapy. Cancer 2003;98:679e689. 37. Kessler RC, Birnbaum H, Demler O, et al. The prevalence and correlates of nonaffective psychosis in the National Comorbidity Survey Replication (NCS-R). Biol Psychiatry 2005;58:668e676.
24. Andrykowski MA, Cordova MJ, McGrath PC, Sloan DA, Kenady DE. Stability and change in posttraumatic stress disorder symptoms following breast cancer treatment: a 1-year follow-up. Psychooncology 2000;9:69e78.
38. Liedl A, O’Donnell M, Creamer M, et al. Support for the mutual maintenance of pain and posttraumatic stress disorder symptoms. Psychol Med 2010;40:1215e1223.
25. Katz S, Nevid JS. Risk factors associated with posttraumatic stress disorder symptomatology in HIV-infected women. AIDS Patient Care STDS 2005;19:110e120.
39. Palyo SA, Beck JG. Post-traumatic stress disorder symptoms, pain, and perceived life control: associations with psychosocial and physical functioning. Pain 2005;117:121e127.
Vol. 44 No. 4 October 2012 PTSD, Mood States, Functional Status, and QOL in Oncology Outpatients
40. Clapp JD, Masci J, Bennett SA, Beck JG. Physical and psychosocial functioning following motor vehicle trauma: relationships with chronic pain, posttraumatic stress, and medication use. Eur J Pain 2010;14:418e425. 41. Corry NH, Klick B, Fauerbach JA. Posttraumatic stress disorder and pain impact functioning and disability after major burn injury. J Burn Care Res 2010;31:13e25. 42. Barry DT, Beitel M, Cutter CJ, et al. Exploring relations among traumatic, posttraumatic, and physical pain experiences in methadone-maintained patients. J Pain 2011;12:22e28. 43. Brucker PS, Yost K, Casy J, Webster K, Cella D. General population and cancer patients norms for the Functional Assessment of Cancer Therapy-General (FACT-G). Eval Health Prof 2005;28:192e211. 44. Kazis LE, Ren XS, Lee A, et al. Health status in VA patients: results from the veterans health study. Am J Med Qual 1999;14:28e38.
531
45. Green BL, Krupnick JL, Rowland JH, et al. Trauma history as a predictor of psychologic symptoms in women with breast cancer. J Clin Oncol 2000;18:1084e1093. 46. Stuber ML, Kazak AE, Meeske K, et al. Predictors of posttraumatic stress symptoms in childhood cancer survivors. Pediatrics 1997;100:958e964. 47. Bostock L, Sheikh AI, Barton S. Posttraumatic growth and optimism in health-related trauma: a systematic review. J Clin Psychol Med Settings 2009;16: 281e296. 48. Chan MWC, Ho SMY, Tedeschi RG, Leung CWL. The valence of attentional bias and cancer-related rumination in posttraumatic stress and posttraumatic growth among women with breast cancer. Psychooncology 2011;20:544e552. 49. Kleim B, Ehlers A. Evidence for a curvilinear relationship between posttraumatic growth and posttrauma depression and PTSD in assault survivors. J Trauma Stress 2009;22:45e52.
Journal of Pain and Symptom Management
Vol. 44 No. 4 October 2012
Original Article
The Relationship Between Posttraumatic Stress Disorder, Mood States, Functional Status, and Quality of Life in Oncology Outpatients Jeffrey I. Gold, PhD, Marilyn ‘‘Marty’’ K. Douglas, DNSc, RN, FAAN, Mary Laudon Thomas, RN, MS, AOCN, Janette E. Elliott, RN-BC, MSN, AOCN, Stephen M. Rao, PhD, and Christine Miaskowski, RN, PhD, FAAN Keck School of Medicine (J.I.G.), University of Southern California, and Pain Management and Palliative Care Program, Department of Anesthesiology Critical Care Medicine, Children’s Hospital Los Angeles, Los Angeles; School of Nursing (M.K.D.), University of California, San Francisco, San Francisco; Veterans Affairs Palo Alto Health Care System (M.K.D., M.L.T., J.E.E.), Palo Alto; San Francisco Veterans Affairs Medical Center (S.M.R.), San Francisco; and Department of Physiological Nursing (C.M.), University of California, San Francisco, San Francisco, California, USA
Abstract Context. Oncology patients are at risk for developing posttraumatic stress disorder (PTSD) and other comorbid mood states, which are associated with decreases in functional status and quality of life (QOL). However, few studies have investigated the relationship between PTSD, other mood states, functional status, and QOL in oncology outpatients. Objectives. This study had four aims: 1) determine the percentages of patients with PTSD and partial PTSD; 2) evaluate for differences in demographic and clinical characteristics among patients with PTSD, partial PTSD, and no PTSD; 3) evaluate for differences in mood states, functional status, and QOL among the three PTSD groups; and 4) evaluate whether demographic and disease characteristics were predictors of PTSD. Methods. As part of a larger clinical trial that evaluated the effects of a cognitivebehavioral intervention on cancer pain management, 289 adult oncology patients (Mage ¼ 61.3, SD ¼ 11.6) completed self-report measures that assessed PTSD, other mood states, functional status, and QOL. Results. Forty-five percent of the sample met the diagnostic criteria for PTSD (n ¼ 78) and partial PTSD (n ¼ 53) and were younger than those with no PTSD. Patients with PTSD had a significantly lower Karnofsky Performance Status, shorter time since diagnosis, higher ratings of mood disturbance, lower ratings of functional status, and lower QOL than patients with no PTSD. A lower Karnofsky Performance Status, fewer months since diagnosis, and presence of bone metastases predicted a higher likelihood of being classified as having PTSD.
Address correspondence to: Jeffrey I. Gold, PhD, Keck School of Medicine, University of Southern California, Children’s Hospital Los Angeles, 4650 Sunset Ó 2012 U.S. Cancer Pain Relief Committee. Published by Elsevier Inc. All rights reserved.
Boulevard, MS #12, Los Angeles, CA 90027-6062, USA. E-mail: [email protected] Accepted for publication: October 27, 2011. 0885-3924/$ - see front matter doi:10.1016/j.jpainsymman.2011.10.014
Vol. 44 No. 4 October 2012 PTSD, Mood States, Functional Status, and QOL in Oncology Outpatients
521
Conclusion. Early identification and interventions may mediate the onset of psychological distress and improve health outcomes. J Pain Symptom Manage 2012;44:520e531. Ó 2012 U.S. Cancer Pain Relief Committee. Published by Elsevier Inc. All rights reserved. Key Words Cancer, posttraumatic stress disorder (PTSD), mood states, functional status, quality of life
Introduction Patients diagnosed and treated for cancer are at increased risk for psychological distress, such as depressive symptoms and posttraumatic stress disorder (PTSD). PTSD is an anxiety disorder that develops following exposure to an extreme traumatic stressor in which the person experienced, witnessed, or was confronted with an event or events that involved actual or threatened death or serious injury or a threat to the physical integrity of self or others (Criterion A1 from the Diagnostic and Statistical Manual of Mental Disorders, 4th ed [DSM-IV]).1 The person’s response to the event must include intense fear, helplessness, or horror (Criterion A2). Symptoms that can result from the traumatic event include persistent reexperiencing of the traumatic event (Criterion B), persistent avoidance of the stimuli associated with the trauma and numbing of general responsiveness (Criterion C), and persistent symptoms of hyperarousal (Criterion D). Because the DSM-IV criteria for PTSD were introduced as a nosological category,1 they have been seen in research as restrictive criteria for individuals who experience PTSD symptoms but do not completely fulfill the DSM-IV criteria.2 It is important to consider individuals who experience the distressing symptoms of PTSD, as an individual’s level of functioning may be impaired because of these symptoms. For this reason, partial PTSD is defined as meeting Criteria A1 and A2, as well as two of the three remaining PTSD criteria (i.e., reexperiencing, avoidance/numbing, or hyperarousal).3 Moreover, partial PTSD captures some of the debilitating symptoms that are experienced in PTSD. Findings from four reviews4e7 suggest that, in some patients, the diagnosis and treatment of cancer lead to the persistence of partial or fullblown PTSD symptoms. Although some cancers are not life threatening, the initial diagnosis of
cancer is interpreted as a life-threatening experience and may increase the risk of PTSD. In one study, 32% of Hodgkin’s disease survivors had either partial or full PTSD and reported significantly lower quality of life (QOL) scores than those without PTSD.8 Current levels of cancer symptoms and the PTSD subdimension of hyperarousal were identified as the strongest predictors of depression. Amir and Ramati9 found a significantly higher rate of PTSD and partial PTSD in 39 long-term survivors of breast cancer compared with 39 matched healthy women. Survivors reported significantly higher levels of emotional distress and lower physical and psychological QOL scores and exhibited poorer coping styles compared with the control group. Multiple regression analysis showed that chemotherapy and disease stage as well as the interaction between chemotherapy and disease stage were significant predictors of hyperarousal. A review on PTSD and cancer reported that individuals with poor premorbid physical health are at greater risk for having PTSD.6 Therefore, it is important to examine the impact of comorbidities on PTSD, given that PTSD in oncology patients is highly variable (i.e., full or partial PTSD), to provide appropriate interventions for potential impairments. In the cancer population, most studies of PTSD are focused on women with breast cancer. Within the veteran population, PTSD is studied within the context of combat-related PTSD and how it relates to substance abuse10,11 and other neuropsychiatric conditions12,13 (e.g., dementia, Korsakoff’s syndrome). However, no studies have evaluated the relationships between PTSD symptoms and other outcomes in a sample of oncology patients, the majority of whom received their care at Veterans Affairs (VA) health care settings. Although studies have noted a wide array of PTSD prevalence rates in patients with cancer
522
Gold et al.
and have reported on the impact of PTSD symptoms in cancer survivors,14 less is known about the occurrence and impact of PTSD symptoms on physical and psychological function. In one study, adults with lymphoma (n ¼ 44) who met criteria for full or partial PTSD reported lower physical activity/QOL.8 Given the paucity of research on PTSD in oncology patients, the purposes of this study among a mostly heterogeneous, male veteran, oncology outpatient population were to 1) determine the percentages of patients who met diagnostic criteria for PTSD, partial PTSD, and no PTSD (i.e., PTSD groups); 2) evaluate for differences in demographic (e.g., age, gender) and clinical (e.g., time since diagnosis, treatment) characteristics among the PTSD groups; 3) evaluate for differences in mood states, QOL, and functional status among the three PTSD groups; and 4) evaluate whether specific demographic and disease characteristics were predictors of PTSD group membership. It was hypothesized that patients with PTSD or partial PTSD would have significantly more mood disturbance, lower functional status, and QOL scores than patients with no PTSD. Additionally, it was hypothesized that demographic and disease characteristics would predict PTSD group membership.
Methods Sample and Settings This study was part of a larger randomized clinical trial that evaluated the effects of a cognitive-behavioral intervention for cancer pain management.15,16 A total of 289 patients participated in this study, of which 85% were recruited from VA medical centers. Patients were eligible if they were older than 18 years; had a diagnosis of cancer; had cancer pain; were able to read, write, and understand English; and had a minimum life expectancy of six months. Patients were excluded if they had a current cognitive or psychiatric condition that could jeopardize adherence to the protocol or had a comorbid pain syndrome that caused more severe pain than their cancer pain. Eligible patients were recruited from oncology and hematology clinics at four U.S. Department of Veterans Affairs medical centers, one public hospital, and one community group practice. Of the 406 patients approached, 84 declined to participate. A total
Vol. 44 No. 4 October 2012
of 322 patients responded and provided written informed consent. Four participants dropped out of the study before randomization. Twentynine participants dropped out of the study for a variety of reasons, including death. The institutional review boards at each of the study sites approved this study. The data presented in this article were obtained before the start of the intervention.
Study Procedures Oncologists or nursing staff identified potential participants to a research assistant who then approached the patient during an oncology clinic appointment and determined if they met the study’s inclusion criteria. These patients were invited to participate and if they agreed, written informed consent was obtained. At the time of enrollment, patients completed the baseline questionnaires. Patients’ medical records were reviewed for disease and treatment information.
Instruments Demographics. A demographic questionnaire obtained information on a variety of demographic characteristics: age, education, gender, ethnicity, marital status, and living arrangement. Disease and Treatment Characteristics. Disease and treatment characteristics were collected from medical records, including cancer diagnosis, stage of disease, treatment history, date of cancer diagnosis, and PTSD diagnosis. Patients were asked if they had been diagnosed with PTSD (but not entered into their medical record). Patients completed the Karnofsky Performance Status (KPS) scale, a measure widely used to subjectively estimate the functional status of patients with cancer.17 KPS scores are based on intervals of 10, ranging from normal functioning (100) to moribund (10), with a midpoint of requiring considerable assistance and frequent medical care (50). The KPS has acceptable testretest and interrater reliability as well as validity.17e19 PTSD ChecklistdCivilian Version. The PTSD ChecklistdCivilian Version (PCL-C) is a 17-item self-report inventory that corresponds to the DSM-IV symptoms of PTSD20 and is widely used for measuring PTSD symptoms or establishing the diagnosis of PTSD in adults.
Vol. 44 No. 4 October 2012 PTSD, Mood States, Functional Status, and QOL in Oncology Outpatients
It is a well-established, reliable, and valid measure of PTSD developed and validated by the National Center for PTSD.20,21 The PCL-C yields a total PTSD score and subscale scores for reexperiencing symptoms (Cluster B, five items), avoidance and numbing symptoms (Cluster C, seven items), and hyperarousal symptoms (Cluster D, five items). Although cancer does not qualify as a PTSD diagnosis in the International Classification of Diseases, Tenth Revision, the PCL-C was used to identify a subset of oncology patients who met PTSD criteria. The PCL-C was modified to specifically reflect symptoms related to the diagnosis and treatment of cancer. This modification was done to decrease attention on other potentially confounding traumatic events (e.g., combatrelated PTSD). Patients responded to each of the items in the following manner: ‘‘Please indicate how distressing each difficulty has been for you during the past 30 days with respect to being diagnosed and treated for cancer.’’ The patients responded to each item using a five-point Likert scale that ranged from 1 (not at all) to 5 (extremely). The PCL-C has been used with veterans20,22and chronically ill adults.23e25 Studies have established high correlations (i.e., greater than 0.85) between the PCL-C and an interviewestablished PTSD diagnosis. These studies have demonstrated strong test-retest reliability and strong internal consistency, with Cronbach’s alpha values that ranged from 0.92 to 0.97.20e22 In this study, the Cronbach’s alpha for the total score was 0.91. The DSM-IV diagnostic criteria also were used to create a diagnosis of PTSD or partial PTSD. To be placed into the PTSD group, the patient needed to endorse (i.e., have an item score of 3 or more on the PCL-C) in the following cluster symptom pattern: one item from Cluster B (reexperiencing), three items from Cluster C (avoidance/numbing), and two items from Cluster D (hyperarousal). Partial PTSD was met if the patient endorsed two of the three symptom clusters. These criteria were used to categorize patients into one of the three PTSD groups (i.e., PTSD, partial PTSD, or no PTSD) instead of cutoff scores because this approach reflects symptom endorsement from all necessary clusters.25,26 Profile of Mood StatesdShort Form. The Profile of Mood StatesdShort Form (POMS-SF) is
523
a 30-item instrument that evaluates six mood states (tension-anxiety, depression-dejection, anger-hostility, vigor-activity, fatigue-inertia, and confusion-bewilderment).27 Participants responded on a five-point Likert scale that ranged from 0 (not at all) to 4 (extremely). Subscale scores and a total mood disturbance (TMD) score are calculated with vigor weighted negatively. In a sample of 428 patients with cancer, Cronbach’s alpha values ranged from 0.78 to 0.91.27 Support for strong convergent and discriminate validity for the POMS-SF was found in correlations with the Center for Epidemiologic Studies Depression Scale, the KPS, the Medical Outcomes Study Short Form-20 Physical Functioning Scale, and the Bradburn Positive and Negative Affect Scales.27 In the present study, Cronbach’s alpha values ranged from 0.84 to 0.92 for the mood subscales and was 0.91 for the TMD score. Functional Assessment of Cancer Therapyd General. The Functional Assessment of Cancer TherapydGeneral (FACT-G) is a 27-item self-administered instrument that evaluates four domains of health-related QOL in patients receiving cancer treatment (physical, social, emotional, and functional well-being).28 Each item is rated on a five-point Likert scale that ranged from 0 (not al all) to 4 (very much). Items are summed to obtain subscale scores. The sum of all the subscales comprises the total score, which can range from 0 to 108, with a higher score indicating better healthrelated QOL. The validity and reliability of the FACT-G are well established in patients with cancer.28 In this study, the Cronbach’s alpha was 0.51 for the total score, 0.79 for the physical well-being subscale, 0.78 for the social/family well-being subscale, 0.63 for the emotional well-being subscale, and 0.80 for the functional well-being subscale. Functional Health StatusdShort Form-36 for Veterans. The Short Form-36 for Veterans (SF-36V) measures eight health status concepts (physical functioning, role limitations because of physical health problems, bodily pain, social functioning, role limitations because of emotional health problems, general mental health, vitality, and perception of general health).29 Scales are summed and linearly transformed on a 0e100 scale, with higher
524
Gold et al.
scores reflecting better health states. The validity and reliability of the SF-36V are well established.29 The Cronbach’s alpha values for the subscales ranged from 0.70 to 0.91.
Statistical Analyses Data were analyzed using the SPSS version 15.0 software for Windows (SPSS Inc., Chicago, IL). Descriptive statistics were generated for all demographic, clinical, and outcome variables. One-way analyses of variance and Chisquare analyses were performed to evaluate for differences in demographic, clinical, and outcome variables among the three PTSD groups (PTSD, partial PTSD, and no PTSD). All calculations used actual values. Adjustments were not made for missing data. One percent of the data on KPS scores was missing, 1.4% of the POMS-SF data was missing, and the subscales for SF-36V had missing data that ranged from zero to two subscales missing 1.4%. Data were missing for FACT-G (n ¼ 7) and the PCL-C (n ¼ 5). Therefore, the cohort for each analysis was dependent on the largest set of data across PTSD groups. If the overall analyses indicated differences among the three PTSD groups, Bonferroni correction post hoc analyses were conducted to determine specific group differences. All P-values were adjusted so that values of P < 0.05 were considered statistically significant. Binary logistical regression analyses were used to evaluate the unique contributions of demographic and clinical characteristics to PTSD group membership (i.e., no PTSD vs. partial/full PTSD). For these analyses, P < 0.05 was considered statistically significant.
Results Percentage of Patients With PTSD and Partial PTSD Based on the prespecified criteria, 27% (n ¼ 78) of the 289 patients met the criteria for PTSD and 18% (n ¼ 53) met the criteria for partial PTSD.
Demographic Characteristics As shown in Table 1, the sample was mostly Caucasian (60%), male (88%), not married (58%), living with someone (77%), not working (92%), and receiving care at a VA facility (85%). The average age was 61.3 years (SD ¼ 11.6), and
Vol. 44 No. 4 October 2012
patients had an average of 13 years of education (SD ¼ 2.8). No differences were found among the PTSD groups in any demographic characteristic except age. Patients in the PTSD and partial PTSD groups were significantly younger than those in the no PTSD group (F(3, 286) ¼ 4.2, P < 0.05).
Disease and Treatment Characteristics As shown in Table 1, 20% of the patients had received prior cancer treatment, and 32% were receiving current cancer treatment. Twentythree percent of the patients were diagnosed with lung cancer, 16% with prostate cancer, 12% with head and neck cancer, 8% with breast cancer (22 cases: 21 females and one male), 6% with myeloma, 5% with rectal cancer, and another 5% with colon cancer. The remaining sample (25%) had a variety of cancer diagnoses. Patients in the PTSD group had significantly lower KPS scores, and a significantly shorter amount of time had elapsed since their cancer diagnoses when compared with the patients with no PTSD (P < 0.01). Significant differences were noted between the groups based on the presence of bone metastases (P < 0.01), with a larger proportion of patients in the no PTSD group having evidence of bone metastases compared with the partial PTSD and PTSD groups (Table 1). Forty-one patients (14%) had a self-reported or medical record-documented diagnosis of non-cancer-related PTSD. No significant differences in total PTSD symptom score or other health-related outcomes were found between patients with and without non-cancer-related PTSD. Therefore, all patients (n ¼ 289) were included in the subsequent analyses.
Differences in Mood States As shown in Fig. 1, significant differences were found in all the POMS-SF subscale scores, as well as the TMD score, among the three PTSD groups (all P < 0.01). Post hoc contrasts determined that patients in the PTSD group scored significantly higher on all the POMSSF subscales, except vigor-activity, than patients in the partial PTSD group, who in turn scored higher than patients in the no PTSD group. Conversely, patients in the PTSD group scored significantly lower in the vigor-activity subscale than individuals in the partial PTSD group,
Vol. 44 No. 4 October 2012 PTSD, Mood States, Functional Status, and QOL in Oncology Outpatients
525
Table 1 Differences in Demographic, Disease, and Treatment Characteristics Among the PTSD Groups
Characteristic Age, years Education, years Karnofsky score Months since cancer diagnosis
Total Sample (N ¼ 289), Mean (SD) 61.3 13.2 75.0 33
(11.6) (2.8) (12.9) (47.3)
No PTSD (n ¼ 158), Mean (SD) 63.1 13.5 77.5 77.5
(11.3) (2.8) (12.7) (12.7)
Partial PTSD (n ¼ 53), Mean (SD) 59.4 12.6 74.2 74.2
(12.6) (2.7) (12.9) (12.9)
PTSD (n ¼ 78), Mean (SD) 59.5 13.0 70.8 70.8
(12.6) (2.7) (12.2) (12.2)
n (%)
n (%)
n (%)
n (%)
Type of care VA care No VA care
247 (85) 42 (15)
136 (86) 22 (14)
46 (87) 7 (13)
65 (83) 13 (17)
Gender Male Female
254 (88) 42 (15)
139 (88) 22 (14)
47 (89) 7 (13)
68 (87) 13 (17)
Ethnicity Caucasian Ethnic minority
173 (60) 116 (40)
96 (61) 62 (39)
33 (62) 20 (38)
44 (56) 34 (44)
Marital status Married Not married
119 (42) 168 (58)
65 (41) 93 (59)
21 (40) 32 (60)
33 (42) 45 (58)
Living arrangement Alone With somebody
65 (23) 224 (78)
35 (22) 123 (78)
9 (17) 44 (83)
21 (27) 57 (73)
Employment Working Not working
23 (8) 263 (92)
13 (8) 145 (92)
3 (6) 49 (93)
7 (9) 69 (89)
Cancer diagnosis Lungc Other
67 (23) 222 (77)
36 (23) 122 (77)
10 (19) 43 (81)
21 (27) 57 (73)
a
4.2 2.0 7.5b 7.5b
3<1 3<1 3<1 c2 (df) 1.01 (2)
0.07 (2)
0.81 (2)
0.05 (2)
1.81 (2)
0.51 (2)
1.81 (2)
12.56b (2)
Evidence of bone metastases Yes No
99 (34) 189 (66)
68 (43) 89 (56)
11 (21) 42 (79)
20 (26) 58 (74)
Prior cancer therapy Yes No
59 (20) 230 (80)
35 (22) 123 (78)
10 (19) 43 (81)
14 (18) 64 (82)
Current cancer therapy Yes No
Post Hoc
F
0.66 (2)
5.68 (2) 91 (32) 196 (68)
59 (37) 98 (62)
12 (23) 41 (77)
20 (26) 57 (73)
PTSD ¼ posttraumatic stress disorder; VA ¼ Veterans Affairs. a P < 0.05. b P < 0.01. c Most frequent type of cancer diagnosis in this sample.
who scored lower than patients in the no PTSD group.
Differences in FACT-G and SF-36V All measures of health-related QOL were significantly different among the three PTSD groups (Table 2), with higher subscale and total scores in the no PTSD group than in the PTSD group. All measures of function were significantly different among the three PTSD
groups, except for the SF-36V physical function score and its associated component score (Table 2). Again, patients in the no PTSD group reported better function than those in the PTSD group.
Predictors of PTSD Group Membership A binary logistic regression was performed, with age and gender entered first as control variables and with KPS score, months since
526
Gold et al.
Vol. 44 No. 4 October 2012
Fig. 1. Differences in Profile of Mood States subscale (a) and total (b) scores among the three PTSD groups. All values are plotted as means and SDs. For all subscales and total mood disturbance scores except vigor: full PTSD group > partial PTSD group > no PTSD group (all P < 0.05). For vigor, full PTSD group < partial PTSD group < no PTSD group (all P < 0.05). PTSD ¼ posttraumatic stress disorder.
diagnosis, cancer type (lung vs. other), and presence of bone metastases entered as predictors of PTSD diagnosis (no PTSD vs. partial/ full PTSD). Although no significant differences were found for gender and cancer type, they were controlled for because previous research demonstrated their association with PTSD.7,30 The final model with all predictors was significant and explained only 12.7% of the variance in PTSD diagnosis (Cox and Snell R-square). The nonsignificant Hosmer and Lemeshow Chi-square statistic demonstrated a good model fit. The overall classification rate for the final model with all predictors included was 65.5%, with 70.4% of the patients without PTSD and 59.7% of the patients with PTSD (partial or full PTSD) correctly classified (Table 3).
As shown in Table 3, independent of age and gender, KPS score, months since diagnosis, and presence of bone metastases were significantly related to PTSD diagnosis. For every 1% decrease in KPS score, patients were 4% more likely to be classified as having PTSD (partial or full). For every additional month since diagnosis, patients were 1% less likely to be classified as having PTSD (partial or full). Finally, patients with bone metastases were 2.2 times more likely to be classified as having PTSD (partial or full) than those without bone metastases.
Discussion This study is the first cross-sectional examination of PTSD, comorbid mood states,
Vol. 44 No. 4 October 2012 PTSD, Mood States, Functional Status, and QOL in Oncology Outpatients
527
Table 2 Differences in FACT-G and SF-36V Scores Among the PTSD Groups (n ¼ 289) Instruments
Total Sample, Mean (SD)
No PTSD (n ¼ 154), Mean (SD)
Partial PTSD (n ¼ 53), Mean (SD)
PTSD (n ¼ 78), Mean (SD)
F
Post Hoc
22.7 (5.4)
15.9 (5.9)
18.5 (4.9)
13.6 (5.8)
12.2 (5.3)
44.1b
2 and 3 < 1
19.1 (6.8)
20.2 (5.8)
21.5 (5.6)
20.2 (5.8)
17.8 (5.6)
10.8b
3<1
b
Normative Data,a Mean (SD)
FACT-G subscales Physical wellbeing Social family wellbeing Emotional wellbeing Functional wellbeing FACT total score SF-36V subscales Physical Functioning Role Physical Bodily Pain General Health Vitality Social Functioning Role Emotional Mental Health Physical component score Mental component score
19.9 (4.8)
16.6 (5.0)
19.0 (3.9)
15.4 (4.4)
12.7 (4.7)
61.2
3<2<1
18.5 (6.8)
13.1 (5.8)
15.4 (5.6)
11.8 (5.0)
9.4 (4.3)
37.5b
3<2<1
b
3<2<1
80.1 (18.1)
65.9 (16.3)
74.4 (13.2)
61.0 (14.0)
52.1 (11.9)
79.9
56.8 (28.9)
40.7 (26.5)
42.7 (27.1)
42.7 (26.8)
35.4 (24.5)
2.2
39.7 55.4 51.4 52.6 67.4
33.7 34.2 43.6 36.5 50.1
38.3 36.7 50.4 40.3 57.9
30.6 32.6 42.7 35.4 45.8
26.5 30.4 30.2 29.8 37.2
(38.4) (27.2) (23.2) (23.25) (28.2)
(25.8) (18.1) (22.1) (20.5) (27.2)
(28.1) (19.0) (22.1) (22.6) (27.9)
(23.3) (18.2) (21.7) (16.9) (27.7)
(20.4) (15.3) (15.8) (16.2) (18.6)
6.2b 3.4c 25.4b 7.2b 17.9b
1>3 1>3 1 and 2 > 3 1>3 1 and 2 > 3
59.6 (37.5) 69.0 (22.8) 37.1 (11.9)
56.6 (32.4) 64.5 (20.1) 31.3 (8.4)
68.5 (31.6) 73.6 (18.1) 31.3 (8.4)
47.0 (28.5) 58.5 (18.7) 32.5 (7.8)
39.3 (26.2) 50.3 (14.3) 30.6 (7.4)
28.6b 51.4b 0.9
1 and 2 > 3 1>2>3
47.8 (12.2)
42.5 (12.4)
48.2 (11.8)
38.3 (10.4)
33.9 (8.1)
51.4b
1 and 2 > 3
FACT-G ¼ Functional Assessment of Cancer TherapydGeneral; SF-36V ¼ Short Form-36V; PTSD ¼ posttraumatic stress disorder; FACT ¼ Functional Assessment of Cancer Therapy. a Normative data compared with cancer patients.43,44 b P < 0.01. c P < 0.05.
functional status, and QOL in a heterogeneous sample of patients with cancer. The first aim of the study was to determine the percentage of patients who met diagnostic criteria for PTSD. Approximately 45% of these patients had PTSD symptoms associated with their cancer experience. This percentage is higher than previous reports in oncology patients that ranged from 2% to 22%.8,9,14,23,24,30e36 A
number of reasons may account for this higher percentage of patients having PTSD or partial PTSD, including the fact that the majority were veterans. In general, veterans, with or without combat exposure, have a higher incidence of lifetime PTSD (30% in men and 27% in women) compared with the lifetime prevalence of PTSD among men (3.6%) and women (9.7%) in the general population.37 However,
Table 3 Results of Binary Logistic Regression Analysis Predicting PTSD Diagnosis (No PTSD vs. Partial/Full; n ¼ 281) Characteristics Age Men (Ref.) Karnofsky score Months since diagnosis Lung cancer (Ref.) Bone metastases (Ref.)
B
SE
OR
95% CI
Wald Statistic
0.20 0.16 0.40 0.01 0.25 0.80
0.01 0.42 0.01 0.01 0.31 0.29
0.98 0.85 0.96 0.99 1.28 2.24
0.96e1.00 0.38e1.92 0.94e0.98 0.99e1.00 0.70e2.36 1.27e3.94
2.61 0.15 14.67c 5.69a 0.63 7.76b
PTSD ¼ posttraumatic stress disorder; OR ¼ odds ratio; Ref. ¼ reference group. a P < 0.05. b P < 0.01. c P < 0.001.
528
Gold et al.
a validated measure focused specifically on symptoms associated with the cancer experience was used in an attempt to decrease symptoms associated with non-cancer-related events. Additionally, it should be noted that when the distributions of PTSD, partial PTSD, and no PTSD were examined between the patients from non-VA facilities and VA facilities, no significant differences were found. Given that patients with previous chart- or self-reported PTSD were not significantly different from patients without previously documented PTSD, these results suggest that PTSD or partial PTSD related to the cancer experience may account for differences in mood, functional outcomes, and QOL in patients with cancer. A cancer-specific PTSD measure and the inclusion and analysis of patients with partial PTSD symptoms may account for the increased occurrence of PTSD rates. An examination of partial PTSD provided critical information about the patients, who would have been classified as not having PTSD. An evaluation of the oncology patients’ military experience, regardless of care setting, may provide important information to clinicians. Further research is warranted in this area to evaluate the rates of PTSD and partial PTSD in VA and non-VA facilities to facilitate the identification and treatment of these high-risk patients. The second aim of this study was to evaluate for differences in demographic and clinical characteristics among the PTSD groups. No differences were found among the PTSD groups in any demographic characteristic except age. Specifically, patients in the PTSD and partial PTSD groups were significantly younger than those in the no PTSD group. All patients enrolled in this study had pain associated with their cancer experience. Therefore, the relationship between pain and functional status among the three PTSD groups was examined and significant differences emerged among the PTSD and the no PTSD groups. Patients with PTSD demonstrated poor levels of physical functioning in a group of patients with relatively low SF-36V scores. This study confirmed the relationship between increased pain and PTSD. This finding is consistent with recent theories that, in medical conditions such as cancer or other unintentional injuries, pain may trigger PTSD symptoms or pain may be a cooccurring condition.38,39 Furthermore, findings
Vol. 44 No. 4 October 2012
from studies in patients with chronic pain following motor vehicle injures,40 in major burn victims,41 and in methadone-maintained patients42 support an association between pain and PTSD. As hypothesized, patients with full PTSD or partial PTSD reported higher levels of mood disturbance than those without PTSD. As expected, the partial PTSD and PTSD groups had less vigor-activity compared with the no PTSD group. In addition, patients with PTSD or partial PTSD reported a higher number of comorbid conditions. All the between-group differences in POMS-SF subscale and total scores were both statistically significant and clinically meaningful as reflected by effect sizes that ranged from 0.14 to 1.44. Similarly, mood disturbance scores in patients with partial PTSD and PTSD, when compared with other published POMS scores in patients with cancer,27,32 were found to be consistently higher, with effect sizes ranging from 0.15 to 0.80. These results suggest that patients with partial and full PTSD not only have the disturbance of PTSD symptoms, but also experience higher levels of mood disturbance that may be related to their diagnosis of cancer. The hypothesis that patients with PTSD and partial PTSD would have decreases in functional status and QOL was confirmed. All the between-group differences in functional status and QOL outcomes represent both clinically and statistically meaningful differences (i.e., effect sizes ranged from 0.22 to 1.37). In addition, compared with normative values for the FACT-G and SF-36V,43,44 patients with PTSD or partial PTSD had significantly lower scores than the patients with no PTSD. However, patients with no PTSD were similar to the normative group on the SF-36V and the FAGT-G.43,44 Whereas the SF-36V is a generic measure of function, the FACT-G is a QOL measure that was specifically developed and normed for patients with cancer. Findings from this study suggest that patients with PTSD and partial PTSD have extremely poor functional status and QOL. Lastly, it was hypothesized that demographic and disease characteristics would predict PTSD group membership. Given that the present study found that time since diagnosis, KPS score, and the presence of bone metastases explained only a small portion of the variance in PTSD status
Vol. 44 No. 4 October 2012 PTSD, Mood States, Functional Status, and QOL in Oncology Outpatients
(i.e., 13%), additional research is warranted to determine what other demographic and clinical factors place patients with cancer at higher risk for PTSD. Additionally, a higher percentage of patients in the no PTSD group (43%) had evidence of bone metastases compared with the partial (21%) and full (26%) PTSD groups. This finding suggests that although having bone metastases was shown to be a risk factor, having bone metastases does not necessarily mean that the patient will develop PTSD. Therefore, future research should examine possible mechanisms that link the association between bone metastases and the development of PTSD. Previous research has demonstrated that factors such as social support, being non-Caucasian, living alone, and lower education are predictors of PTSD in patients living with breast cancer23,31,45 and in childhood cancer survivors.46 In addition to identifying factors predictive of functional decline in this population, the identification of protective and/or posttraumatic growth needs to be considered.47e49 Several study limitations need to be acknowledged. Findings on the prevalence of PTSD warrant replication in larger samples of patients with cancer who receive care within and outside the VA health care system. In addition, veteran status should be evaluated in all patients with cancer to determine its impact on PTSD occurrence rates. The findings are generalizable to oncology patients in the U.S. However, because the sample was primarily male, the findings may not be generalizable to women with cancer, particularly women with nonbreast cancer. Because the study was cross sectional, additional investigation is warranted on how PTSD symptoms change over time. Additionally, the reliability for the total score (0.51) and the subscale of emotional well-being (0.63) of the FACT-G was lower than expected. It is unclear why the internal consistency was lower, given that the FACT-G was shown to be reliable and valid in patients with cancer. Lastly, the identification of other relevant predictors of PTSD may require a larger sample. Despite these limitations, findings from this study suggest that PTSD symptoms are prevalent in patients with cancer and that these symptoms have a negative impact on patients’ mood, functional status, and QOL. These findings warrant replication in larger and more heterogeneous samples, including different care settings, so
529
that specific and tailored interventions can be designed and tested in future investigations.
Disclosures and Acknowledgments This research was supported by the Department of Veterans Affairs, Veterans Health Administration, Health Services Research and Development Service (project no. NRI-97026). The views expressed in this article are those of the authors and do not necessarily represent the views of the Department of Veterans Affairs. The authors declare no conflicts of interest.
References 1. American Psychiatric Association. Diagnostic and statistical manual of mental disorders, 4th ed. Washington, DC: American Psychiatry Association, 1994. 2. Mylle J, Maes M. Partial posttraumatic stress disorder revisited. J Affect Disord 2004;78:37e48. 3. Carlier IV, Gersons BP. Partial posttraumatic stress disorder (PTSD): the issue of psychological scars and the occurrence of PTSD symptoms. J Nerv Ment Dis 1995;183:107e109. 4. Gurevich M, Devins GM, Rodin GM. Stress response syndromes and cancer: conceptual and assessment issues. Psychosomatics 2002;43:259e281. 5. Neel ML. Posttraumatic stress symptomatology and cancer. Int J Emerg Ment Health 2000;2:85e94. 6. Smith MY, Redd WH, Peyser C, Vogl D. Posttraumatic stress disorder in cancer: a review. Psychooncology 1999;8:521e537. 7. Kangas M, Henry JL, Bryant RA. Posttraumatic stress disorder following cancer. A conceptual and empirical review. Clin Psychol Rev 2002;22:499e524. 8. Geffen DB, Blaustein A, Amir MC, Cohen Y. Post-traumatic stress disorder and quality of life in long-term survivors of Hodgkin’s disease and nonHodgkin’s lymphoma in Israel. Leuk Lymphoma 2003;44:1925e1929. 9. Amir M, Ramati A. Post-traumatic symptoms, emotional distress and quality of life in long-term survivors of breast cancer: a preliminary research. J Anxiety Disord 2002;16:195e206. 10. Bremner JD, Southwick M, Darnell A, Charney DS. Chronic PTSD in Vietnam combat veterans: course of illness and substance abuse. Am J Psychiatry 1996;153:369. 11. Breslau N, Davis GC, Schultz LR. Posttraumatic stress disorder and the incidence of nicotine, alcohol, and other drug disorders in persons who have experienced trauma. Arch Gen Psychiatry 2003;60: 289e294.
530
Gold et al.
12. Koenen KC, Driver KL, Oscar-Berman M, et al. Measures of prefrontal system dysfunction in posttraumatic stress disorder. Brain Cogn 2001;45: 64e78. 13. Yaffe K, Vittinghoff E, Lindquist K, et al. Posttraumatic stress disorder and risk of dementia among US veterans. Arch Gen Psychiatry 2010;67: 608e613. 14. Black EK, White CA. Fear of recurrence, sense of coherence and posttraumatic stress disorder in haematological cancer survivors. Psychooncology 2005;14:510e515. 15. Fahey KF, Rao SM, Douglas MK, et al. Nurse coaching to explore and modify patient attitudinal barriers interfering with effective cancer pain management. Oncol Nurs Forum 2008;35:233e240. 16. Thomas ML, Elliott JE, Rao S, et al. A randomized clinical trial of education or motivational-interviewingbased coaching compared to usual care to improve cancer pain management. Oncol Nurs Forum 2012; 39:39e49. 17. Conhill C, Verger E, Salamero M. Performance status assessment in cancer patients. Cancer 1990; 65:1864e1866. 18. Grieco A, Long CJ. Investigation of the Karnofsky performance status as a measure of quality of life. Health Psychol 1984;3:129e142. 19. Schag CC, Heinrich RL, Ganz PA. Karnofsky performance status revisited: reliability, validity, and guidelines. J Clin Oncol 1984;2:187e193. 20. Weathers FW, Litz BT, Huska JA, Keane TM. The PTSD Checklist-Civilian Version (PCL-C). Boston, MA: National Center for PTSD, Boston Veterans Affairs Medical Center, 1994. 21. Berlant J, van Kammen DP. Open-label topiramate as primary or adjunctive therapy in chronic civilian posttraumatic stress disorder: a preliminary report. J Clin Psychiatry 2002;63:15e20. 22. Dobie DJ, Kivlahan DR, Maynard C, et al. Posttraumatic stress disorder in female veterans: association with self-reported health problems and functional impairment. Arch Intern Med 2004;164: 394e400. 23. Andrykowski MA, Cordova MJ. Factors associated with PTSD symptoms following treatment for breast cancer: test of the Andersen model. J Trauma Stress 1998;11:189e203.
Vol. 44 No. 4 October 2012
26. Akechi T, Okuyama T, Sugawara Y, et al. Major depression, adjustment disorders, and post-traumatic stress disorder in terminally ill cancer patients: associated and predictive factors. J Clin Oncol 2004;22: 1957e1965. 27. Baker F, Denniston M, Zabora J, Polland A, Dudley WN. A POMS short form for cancer patients: psychometric and structural evaluation. Psychooncology 2000;11:273e281. 28. Cella DF, Tulsky DS, Gray G, et al. The Functional Assessment of Cancer Therapy Scale: development and validation of the general measure. J Clin Oncol 1993;11:570e579. 29. Kazis LE. The Veterans SF-36 health status questionnaire: development and application in the Veterans Health Administration. Med Outcomes Monit 2000;5:1e18. 30. Pitman RK, Lanes DM, Williston SK, et al. Psychophysiologic assessment of posttraumatic stress disorder in breast cancer patients. Psychosomatics 2001;42:133e140. 31. Cordova MJ, Cunningham LL, Carlson CR, Andrykowski MA. Posttraumatic growth following breast cancer: a controlled comparison study. Health Psychol 2001;20:176e185. 32. Deimling GT, Kahana B, Bowman KF, Schaefer ML. Cancer survivorship and psychological distress in later life. Psychooncology 2002;11: 479e494. 33. Tjemsland L, Soreide JA, Malt UF. Posttraumatic distress symptoms in operable breast cancer III: status one year after surgery. Breast Cancer Res Treat 1998;47:141e151. 34. Steginga SK, Pinnock C, Gardner M, Gardiner RA, Dunn J. Evaluating peer support for prostate cancer: the prostate cancer peer support inventory. BJU Int 2005;95:46e50. 35. Smith SK, Zimmerman S, Williams CS, Preisser JS, Clipp EC. Post-traumatic stress outcomes in nonHodgkin’s lymphoma survivors. J Clin Oncol 2008; 26:934e941. 36. Kornblith AB, Herndon JE 2nd, Weiss RB, et al. Long-term adjustment of survivors of early-stage breast carcinoma, 20 years after adjuvant chemotherapy. Cancer 2003;98:679e689. 37. Kessler RC, Birnbaum H, Demler O, et al. The prevalence and correlates of nonaffective psychosis in the National Comorbidity Survey Replication (NCS-R). Biol Psychiatry 2005;58:668e676.
24. Andrykowski MA, Cordova MJ, McGrath PC, Sloan DA, Kenady DE. Stability and change in posttraumatic stress disorder symptoms following breast cancer treatment: a 1-year follow-up. Psychooncology 2000;9:69e78.
38. Liedl A, O’Donnell M, Creamer M, et al. Support for the mutual maintenance of pain and posttraumatic stress disorder symptoms. Psychol Med 2010;40:1215e1223.
25. Katz S, Nevid JS. Risk factors associated with posttraumatic stress disorder symptomatology in HIV-infected women. AIDS Patient Care STDS 2005;19:110e120.
39. Palyo SA, Beck JG. Post-traumatic stress disorder symptoms, pain, and perceived life control: associations with psychosocial and physical functioning. Pain 2005;117:121e127.
Vol. 44 No. 4 October 2012 PTSD, Mood States, Functional Status, and QOL in Oncology Outpatients
40. Clapp JD, Masci J, Bennett SA, Beck JG. Physical and psychosocial functioning following motor vehicle trauma: relationships with chronic pain, posttraumatic stress, and medication use. Eur J Pain 2010;14:418e425. 41. Corry NH, Klick B, Fauerbach JA. Posttraumatic stress disorder and pain impact functioning and disability after major burn injury. J Burn Care Res 2010;31:13e25. 42. Barry DT, Beitel M, Cutter CJ, et al. Exploring relations among traumatic, posttraumatic, and physical pain experiences in methadone-maintained patients. J Pain 2011;12:22e28. 43. Brucker PS, Yost K, Casy J, Webster K, Cella D. General population and cancer patients norms for the Functional Assessment of Cancer Therapy-General (FACT-G). Eval Health Prof 2005;28:192e211. 44. Kazis LE, Ren XS, Lee A, et al. Health status in VA patients: results from the veterans health study. Am J Med Qual 1999;14:28e38.
531
45. Green BL, Krupnick JL, Rowland JH, et al. Trauma history as a predictor of psychologic symptoms in women with breast cancer. J Clin Oncol 2000;18:1084e1093. 46. Stuber ML, Kazak AE, Meeske K, et al. Predictors of posttraumatic stress symptoms in childhood cancer survivors. Pediatrics 1997;100:958e964. 47. Bostock L, Sheikh AI, Barton S. Posttraumatic growth and optimism in health-related trauma: a systematic review. J Clin Psychol Med Settings 2009;16: 281e296. 48. Chan MWC, Ho SMY, Tedeschi RG, Leung CWL. The valence of attentional bias and cancer-related rumination in posttraumatic stress and posttraumatic growth among women with breast cancer. Psychooncology 2011;20:544e552. 49. Kleim B, Ehlers A. Evidence for a curvilinear relationship between posttraumatic growth and posttrauma depression and PTSD in assault survivors. J Trauma Stress 2009;22:45e52.