- Email: [email protected]
Impact of Hot Flashes on Quality of Life Among Postmenopausal Women Being Treated for Breast Cancer
436
Journal of Pain and Symptom Management
Vol. 19 No. 6 June 2000
Original Article
Impact of Hot Flashes on Quality of Life Among Postmenopausal Women Being Treated for Breast Cancer Kevin D. Stein, PhD, Paul B. Jacobsen, PhD, Danette M. Hann, PhD, Harvey Greenberg, MD, and Gary Lyman, MD Psychosocial Oncology (K.D.S., P.B.J.) and Comprehensive Breast Cancer (H.G., G.L.) Programs, University of South Florida H. Lee Moffitt Cancer Center and Research Institute, Tampa, Florida, and Behavioral Research Center (D.M.H.), American Cancer Society, Atlanta, Georgia, USA
Abstract Hot flashes are among the most commonly reported symptoms among women who have completed treatment for breast cancer. Relatively little is known, however, about hot flashes among women while they are undergoing breast cancer treatment. The present study investigated the prevalence and severity of hot flashes of women during chemotherapy and radiotherapy for breast cancer. We also sought to identify the medical, demographic, and treatment correlates of hot flashes during treatment and to document the impact of hot flashes on quality of life. Seventy postmenopausal women with breast cancer completed a self-report questionnaire packet during chemotherapy and radiotherapy. Forty percent (n ⫽ 28) reported hot flashes during the week prior to assessment. Of the 28 women endorsing hot flashes, 25% (n ⫽ 7) rated them as severe, 39% (n ⫽ 11) rated them as moderate, and 36% (n ⫽ 10) rated them as mild. Women with hot flashes were significantly (p ⬍ 0.05) younger and reported significantly (p ⬍ 0.001) more fatigue, poorer sleep quality, and poorer physical health compared to women without hot flashes. Multivariate analyses revealed that, even after controlling for relevant medical, demographic, and treatment variables, the prevalence of hot flashes significantly (p ⬍ 0.05) predicted poorer sleep quality, more fatigue, and worse physical health. The results indicate that hot flashes are experienced by a sizable percentage of postmenopausal breast cancer patients as they undergo treatment. Hot flashes during cancer treatment appear to have a negative impact upon patient quality of life that may be due, in part, to fatigue and interference with sleep. Future research should seek to evaluate interventions to relieve hot flashes during breast cancer treatment as a means of improving patient quality of life. J Pain Symptom Manage 2000;19:436-445. © U.S. Cancer Pain Relief Committee, 2000. Key Words Hot flashes, breast cancer, fatigue, quality of life
Introduction Address reprint requests to: Kevin D. Stein, PhD, Behavioral Research Center, American Cancer Society, Clifton Rd. NE, Atlanta, GA 30329, USA. Accepted for publication: July 14, 1999. © U.S. Cancer Pain Relief Committee, 2000 Published by Elsevier, New York, New York
During menopause, decreasing levels of the hormone estrogen produce a range of distressing vasomotor symptoms that include night sweats, chills, and hot flashes. Of these symptoms, hot 0885-3924/00/$–see front matter PII S0885-3924(00)00142-1
Vol. 19 No. 6 June 2000
Impact of Hot Flashes
flashes are the most common and are associated with the most physical and emotional distress.1 Hot flashes may have a negative impact on a woman’s quality of life by causing decreased sleep quality and increased levels of fatigue and depression.2,3 In healthy women, hot flashes can be managed through the use of hormone replacement therapy (HRT).4 Women with breast cancer, however, often forego the use of hormone replacement therapy due to the potential for an increase in risk of disease recurrence.5,6 As a result, these women often suffer frequent and distressing hot flashes.7 Researchers have only recently begun to document the prevalence and severity of hot flashes in breast cancer patients. In one of the first studies to focus on this issue, Couzi et al.2 surveyed 190 postmenopausal women (average age ⫽ 54.9 years) who had been diagnosed with breast cancer 2–6 years previously. To be included in the study, women had to have an absence of menses for at least 6 months, be between the ages of 40 and 65, and have no recent evidence of disease recurrence or metastasis. Participants were asked about the prevalence and severity of a number of vasomotor, gynecologic, and other medical symptoms. The results indicated that hot flashes were the most frequently reported symptom, endorsed by 65% of the women. More recently, Carpenter et al.3 assessed hot flashes and quality of life in 114 postmenopausal women (average age ⫽ 58.8 years) treated previously for breast cancer. To be included in the study, women had to have had an absence of menses for at least 1 year, be at least 3 months post-treatment, and have no recent evidence of disease recurrence or metastasis. On average, these women had completed treatment 34.9 months previously (range ⫽ 4– 116 months). Consistent with Couzi et al.,2 65% of the women reported experiencing hot flashes at the time of assessment. Of those women reporting hot flashes, 59% rated their symptoms as severe. Women with hot flashes were found to be younger at diagnosis and less educated compared to women without hot flashes. Among women with hot flashes, more severe symptoms were associated with higher body mass index, younger age at diagnosis, and use of tamoxifen. Both prevalence and severity of hot flashes were marginally (p ⬍ 0.10) related to poorer mental and physical quality of life.
437
The results of these studies suggest that hot flashes are common and may be severe among women previously treated for breast cancer. The degree to which these findings also apply to women currently undergoing treatment for breast cancer remains unknown. To address this issue, the current study sought to: 1) document the prevalence and severity of hot flashes among postmenopausal women during treatment for breast cancer; 2) identify medical, demographic, and treatment correlates of hot flashes during breast cancer treatment; and 3) determine the impact of the hot flashes on breast cancer patients’ quality of life during breast cancer treatment.
Methods Participants To be eligible for the analyses described here, patients had to: a) be women 18 years of age or older; b) be diagnosed with breast cancer and receiving adjuvant chemotherapy or adjuvant radiotherapy; c) be postmenopausal (i.e., report absence of menses for at least 12 months); d) have no known untreated or unstable major medical conditions; e) have no known major psychiatric or neurological disorders that would interfere with completion of study measures; f) be able to read English; g) have no history of treatment for other types of cancer; and h) provide written informed consent to participate in a longitudinal study of the side effects of breast cancer treatment. Eighty-nine women (55 radiotherapy and 34 chemotherapy patients) met these criteria and agreed to participate in the longitudinal study. Two radiotherapy patients dropped out of the study and three chemotherapy patients became ineligible because their treatment ended prematurely or they began chemotherapy. One chemotherapy patient became ineligible because she began radiotherapy during chemotherapy treatment. In addition, eight radiotherapy patients and five chemotherapy patients were excluded due to missing data. Thus, complete longitudinal data were obtained from a total of 70 patients: 42 radiotherapy patients and 28 chemotherapy patients.
Procedure and Measures As a part of a larger longitudinal study of side effects of breast cancer treatment, partici-
438
Stein et al.
pants were recruited during an outpatient appointment prior to their first scheduled chemotherapy or radiotherapy treatment. Those patients who met eligibility criteria and provided informed consent were asked to complete a self-report questionnaire packet (described below) prior to the beginning of treatment and at several times during the course of treatment. For the purpose of the current study, all analyses were conducted on self-report data collected approximately 4–6 weeks after the start of treatment. The self-report measures administered to study participants included a demographic data form, the SF-36 Health Survey, a modified version of the Memorial Symptom Assessment Scale (MSAS), the State-Trait Anxiety Inventory (STAI), the Center for Epidemiological Studies Depression Scale (CES-D), the Profile of Mood States Fatigue Scale (POMS-F), the Fatigue Symptom Inventory (FSI), the Multidimensional Fatigue Symptom Inventory (MFSI), and the Pittsburgh Sleep Quality Index (PSQI). The order of the instruments was randomized for each individual packet to reduce the chance of order effects. Medical record reviews were conducted to obtain information about stage of disease, type of surgery, adjuvant chemotherapy agent(s), adjuvant radiation therapy, and adjuvant hormonal therapy. The SF-36 Health Survey8 is a 36-item selfreport measure of physical and emotional functioning that was adapted from the Medical Outcome Study (MOS-36-SF). The SF-36 Health Survey contains eight subscales: physical functioning, role functioning-physical, bodily pain, general health, vitality, social functioning, role functioning-emotional, and mental health. Scores on each of the eight subscales can range from 0 to 100. The MOS-36-SF yields two summary scores: physical health summary and mental health summary. These summary scores are expressed as T-scores (mean ⫽ 50; SD ⫽ 10). Consistent with the individual subscales, higher scores on the MOS-36-SF summary scales indicate better functioning. Data from over 20,000 patients in the Medical Outcome Study has demonstrated that the MOS-36-SF is a valid measure of both physical and emotional functioning.8 An item from a modified version of the Memorial Symptom Assessment Scale (MSAS)9 was used in the present study to assess the prev-
Vol. 19 No. 6 June 2000
alence, severity, and distress of hot flashes. Respondents indicated if they had experienced hot flashes (presence/absence) during the past week. If the respondent confirmed the presence of hot flashes, they rated the severity (1 ⫽ slight; 3 ⫽ severe) and level of distress (0 ⫽ not at all; 3 ⫽ very much) associated with the symptom. The State Trait Anxiety Inventory (STAI)10 contains two separate 20-item scales that measure state (current or situational) anxiety and trait (general) anxiety. For the state version, respondents rate on a 4-point Likert scale (1 ⫽ not at all; 4 ⫽ very much so) the extent to which they are currently experiencing each item. For the trait version, respondents indicate on a 4-point Likert scale (1 ⫽ almost never; 4 ⫽ almost always) how often they generally experience each item. Scores for each scale range from 0 to 80. The STAI is supported by an extensive literature regarding its reliability and validity.11 The Center for Epidemiological Studies Depression Scale (CES-D)12 is a 20-item measure of depressive symptomatology. Respondents rate how frequently they have experienced each depressive symptom during the past week on a 4-point rating scale (0 ⫽ rarely or none of the time; 3 ⫽ most or all of the time). Total scores range from 0 to 60. The reliability and validity of the CES-D have been demonstrated in use with a wide range of populations, including older adults13 and cancer patients.14 The Profile of Mood States Fatigue Scale (POMS-F)15 consists of 7 items that assess feelings of weariness and low energy. Respondents indicate the degree to which they have experienced each of these feelings during the previous week on a 5-point scale (0 ⫽ not at all; 4 ⫽ extremely). Total scores range from 0 to 28. The POMS-F has been used extensively to assess cancer-related fatigue16 and has been shown to have acceptable reliability and validity.15 The Fatigue Symptom Inventory (FSI)17 is a 14-item scale that assesses the frequency, severity, and daily pattern of fatigue as well as its interference with quality of life. Frequency is measured both by the number of days in the past week (0–7) and the percentage of each day on average (0 ⫽ not at all; 10 ⫽ entire day) that respondents felt fatigued. Severity is measured on an 11-point scale (0 ⫽ not at all fatigued; 10 ⫽ as fatigued as I could be) that as-
Vol. 19 No. 6 June 2000
Impact of Hot Flashes
sesses average fatigue during the past week. Interference of fatigue with quality of life is measured on seven separate 11-point scales (0 ⫽ no interference; 10 ⫽ extreme interference) that assess the degree to which fatigue was judged to interfere with general activity, ability to bathe and dress, normal work activity, ability to concentrate, relations with others, enjoyment of life, and mood in the past week. Responses on the seven items are summed to provide a total interference score. The reliability and validity of the FSI has been demonstrated in previous research with both breast cancer patients and healthy women.17,18 The Multidimensional Fatigue Symptom Inventory (MFSI)19 is an 83-item measure of fatigue that includes subscales assessing global, somatic, affective, cognitive, and behavioral dimensions of fatigue. All MFSI items are scored on a five-point Likert scale (0 ⫽ not at all; 4 ⫽ extremely) indicating the level of fatigue during the past week. Factor analytic techniques confirmed the five-factor structure of the MFSI. Preliminary evidence indicates that the MFSI is a reliable and valid measure of fatigue symptom domains in women with breast cancer.19 The Pittsburgh Sleep Quality Index (PSQI)20 measures sleep patterns and sleep-related difficulties. A modified version of the PSQI was used to obtain self-reports of overall sleep quality during the past week. The PSQI has been
439
shown to discriminate between normal, depressives, and individuals with sleep disorders.20
Results Demographic and Medical Characteristics Information about the demographic characteristics of the chemotherapy patients and radiotherapy patients is presented in Table 1. Chemotherapy patients ranged in age from 41 to 78 years (mean ⫽ 57.1, SD ⫽ 8.9). A majority of these women were white (93%), married (68%), and had completed some college course work (78%). Forty-six percent of the chemotherapy patients were employed full- or part-time and an additional 21% were on leave with or without pay. A majority of the women (57%) reported a household income of $40,000 or more per annum. Radiotherapy patients ranged in age from 39 to 81 years (mean ⫽ 64.0, SD ⫽ 9.4). A majority of these women were also white (88%), married (71%), and had completed some college course work (62%). Seventeen percent of the radiotherapy patients were employed full- or part-time and an additional 5% were on leave with or without pay. Twenty-four percent of the radiotherapy patients reported a household income of $40,000 or more per annum. As shown in Table 1, chemotherapy and radiotherapy patients differed significantly (p ⬍ 0.05) in terms of age, level of education, annual household income,
Table 1 Demographic Characteristics of Study Patients by Type of Treatment Variable Age, yearsa* Marital Statusb Married Other Raceb White Nonwhite Educationb* College or more High School or less Employmentb* Employed Not Employed Annual Household Incomeb* ⬍ $40,000 per annum ⬎ $40,000 per annum
Chemotherapy patients (N ⫽ 28)
Radiotherapy patients (N ⫽ 42)
57.10 ⫾ 8.94
63.98 ⫾ 9.41
19 (68) 9 (32)
30 (71) 12 (29)
24 (86) 4 (14)
39 (93) 3 (7)
22 (79) 6 (21)
26 (62) 16 (38)
13 (46) 15 (54)
7 (17) 35 (83)
12 (43) 16 (57)
32 (76) 10 (24)
expressed as mean ⫾ SD. expressed as N (%). *p ⬍ 0.05. All other comparisons were nonsignificant (p ⬎ 0.05). aValue bValue
440
Stein et al.
and employment status. Compared to radiotherapy patients, chemotherapy patients were younger, more highly educated, more likely to be employed full- or part-time, and had a higher annual household income. Disease stage for the chemotherapy patients at time of diagnosis was classified as follows: stage I, 4 patients (14%); stage II, 19 patients (68%); and stage III, 5 patients (18%). With respect to surgical treatment, 15 chemotherapy patients (54%) underwent a mastectomy, 12 patients (43%) underwent lumpectomy, and 1 patient (3%) underwent no surgical treatment other than breast biopsy prior to chemotherapy. With respect to chemotherapy regimens, 7 patients (25%) received doxorubicin, 11 patients (39%) received doxorubicin and cyclophosphamide, 9 patients (32%) received doxorubicin, cyclophosphamide, and fluorouracil, and 1 patient (4%) received mitoxantrone. None of the chemotherapy patients were receiving hormonal therapy (tamoxifen) at the time of assessment. For radiotherapy patients, disease stage at time of diagnosis was as follows: stage I, 25 patients (60%); stage II, 15 patients (36%); and stage III, 2 patients (4%). With respect to surgical treatment, 9 patients (21%) underwent a mastectomy, 31 patients (74%) underwent a lumpectomy, and 2 patients (5%) underwent a breast biopsy only. Among radiotherapy patients, 6 patients (14%) were receiving hormonal therapy (i.e., tamoxifen) at the end of assessment. In terms of their medical characteristics, chemotherapy patients and radiotherapy patients differed significantly (p ⬍ 0.05) on stage of disease and type of surgery. Compared to radiotherapy patients, chemotherapy patients had more advanced disease and were more likely to have received a mastectomy. In addition, significantly (p ⬍ 0.05) more radiotherapy patients than chemotherapy patients were receiving hormonal therapy (i.e., tamoxifen).
Prevalence and Severity of Hot Flashes Of the 70 women undergoing radiotherapy or chemotherapy treatment for breast cancer, 40% (n ⫽ 28) reported experiencing hot flashes in the week prior to data collection. Of the 28 women who reported hot flashes, 46% (n ⫽ 13) were chemotherapy patients and 54% (n ⫽ 15) were radiotherapy patients. Chemotherapy
Vol. 19 No. 6 June 2000
and radiotherapy patients did not differ significantly in the prevalence of hot flashes (p ⬎ 0.05). Among women experiencing hot flashes (n ⫽ 28), 36% (n ⫽ 10) rated their symptoms as mild, 39% (n ⫽ 11) rated them as moderate, and 25% (n ⫽ 7) rated them as severe. Consistent with the prevalence of hot flashes, the severity of hot flashes also did not differ significantly (p ⬎ 0.05) by the type of treatment (chemotherapy vs. radiotherapy). Among women reporting hot flashes (n ⫽ 28), ratings of distress associated with hot flashes were as follows: 4% (n ⫽ 1) not at all, 43% (n ⫽ 12) a little bit, 28% (n ⫽ 8) somewhat, and 25% (n ⫽ 7) very much. As with the prevalence and severity of hot flashes, the distress associated with hot flashes did not differ significantly (p ⬎ 0.05) by the type of treatment. Due to the finding that chemotherapy and radiotherapy patients did not differ in the prevalence, severity, and distress of hot flashes, the two samples were combined together in subsequent analyses.
Differences Based on Prevalence of Hot Flashes
T-tests and 2 analyses were used to compare women with hot flashes to women without hot flashes on all medical/demographic/treatment variables (see Table 2). Women with hot flashes were found to be significantly younger compared to women without hot flashes (p ⬍ 0.05). In contrast, there were no significant differences between the two groups in level of education, marital or occupational status, annual household income, stage of disease, type of surgery, type of chemotherapy (if applicable), or use of hormonal treatment (p ⬎ 0.05). As shown in Table 3, t-tests were also used to examine differences between women with hot flashes and women without hot flashes on psychosocial and quality of life outcomes including fatigue (POMS-F; FSI; MFSI), sleep problems (PSQI), depression (CES-D), anxiety (STAI-S), and physical and mental health (MOS-36-SF). Results indicated that women with hot flashes reported significantly higher levels of fatigue (p ⬍ 0.01), greater interference of fatigue with quality of life (p ⬍ 0.01), poorer sleep quality (p ⬍ 0.01), and poorer physical health (p ⬍ 0.01). The two groups did not differ in terms of their mental health or levels of anxiety and depression. To further delineate the nature of the differences in fatigue between the two groups, scores from the MFSI were evaluated (see Table 3).
Vol. 19 No. 6 June 2000
Impact of Hot Flashes
441
Table 2 Demographic Characteristics of Study Patients by Prevalence of Hot Flashes Variable Age, yearsa* Marital Statusb Married Other Raceb White Nonwhite Educationb College or more High School or less Employmentb Outside Home Other Annual Household Incomeb ⬍ $40,000 per annum ⬎ $40,000 per annum
Women with HF (N ⫽ 28)
Women without HF (N ⫽ 42)
57.64 ⫾ 10.08
63.61 ⫾ 8.89
21 (75) 7 (25)
28 (67) 14 (33)
24 (86) 4 (14)
39 (93) 3 (7)
8 (29) 20 (71)
11 (26) 31 (74)
8 (29) 20 (71)
12 (29) 30 (71)
16 (57) 12 (43)
28 (67) 14 (33)
expressed as mean ⫾ SD. expressed as N (%). *p ⬍ 0.05. All other comparisons were nonsignificant (p ⬎ 0.05).
aValue bValue
T-tests comparing mean MFSI scale scores indicated that women experiencing hot flashes reported significantly more global fatigue (p ⬍ 0.05), somatic fatigue (p ⬍ 0.05), cognitive fa-
tigue (p ⬍ 0.05), and less vigor (p ⬍ 0.05) than women who did not suffer from hot flashes. To further delineate the nature of quality of life differences between the two groups, sub-
Table 3 Mean Scores on Psychosocial and Quality of Life Outcomes by Hot Flash Prevalence
Variable POMS-Fb FSI disruptiveness indexb MFSI-SF fatigue dimensions Global fatiguea Somatic fatiguea Affective fatigue Cognitive fatiguea Vigora PSQI sleep qualityb CES-D STAI-S MOS-36-SF summary scales Physical summary scaleb Mental summary scale Physical MOS-36-SF subscale scores Physical functioningc Role physical Bodily pain General healthb Mental MOS-36-SF subscale scores Social functioninga Role emotional Vitality Mental health ⬍ 0.05. ⬍ 0.001. cp ⬍ 0.0001. HF ⫽ Hot flashes. ap bp
Women with HF (N ⫽ 28) M (SD)
Women without HF (N ⫽ 42) M (SD)
13.89 (8.13) 3.60 (2.52)
8.36 (7.10) 2.00 (1.86)
11.96 (7.58) 6.00 (6.36) 5.64 (5.79) 5.79 (6.11) 9.75 (5.26) 1.43 (0.69) 13.67 (11.56) 37.59 (13.47)
7.50 (6.52) 2.76 (4.41) 3.24 (4.65) 3.17 (3.46) 13.45 (6.29) 0.88 (0.63) 9.25 (10.19) 33.19 (13.38)
34.33 (10.77) 44.87 (10.94)
41.33 (9.14) 48.72 (12.38)
50.18 (30.14) 14.29 (31.50) 58.79 (24.08) 53.89 (22.15)
71.66 (21.49) 27.38 (38.18) 68.95 (25.09) 69.38 (20.89)
52.68 (25.77) 48.81 (45.77) 36.61 (23.34) 70.71 (18.48)
68.75 (24.42) 63.49 (44.05) 47.74 (26.09) 77.81 (20.34)
442
Stein et al.
scale scores from the MOS-36-SF were evaluated (see Table 3). T-tests comparing mean MOS-36-SF subscale scores indicated that women experiencing hot flashes reported poorer physical functioning (p ⬍ 0.0001), general health (p ⬍ 0.001), and social functioning (p ⬍ 0.05) compared to women without hot flashes. Multivariate analyses were conducted to determine the extent to which the presence of hot flashes accounted for variability in the major psychosocial and quality of life outcomes after controlling for variability attributable to medical/demographic/treatment variables. Separate hierarchical regression analyses were conducted for each of the major psychosocial and quality of life outcomes found to be related to the presence of hot flashes (physical health, sleep quality, and fatigue). Medical/demographic/treatment variables were included in the multivariate analyses if they were found to be even marginally (p ⬍ 0.10) correlated with the outcome measure. For physical health (as measured by the MOS-36-SF physical summary scale), the medical/demographic/treatment variables that met the above criterion were level of education and income. For sleep quality (as measured by the PSQI), level of education was the only variable to meet the criterion. For fatigue (as measured by the POMS-Fatigue Scale), age was the only medical/demographic/treatment variable to meet the criterion. Despite the fact that type of treatment (chemotherapy vs. radiotherapy) was not even marginally correlated (p ⬍ 0.10) with any of the psychosocial or quality of life outcome measures, it was included in all multivariate analyses. Results of the hierarchical regression analyses indicated that, for each of the psychosocial and quality of life outcomes, the prevalence of hot flashes accounted for a significant amount of variance after controlling for medical/demographic/ treatment variables included in the models. Specifically, the prevalence of hot flashes accounted for 11% additional variance in physical health (p ⬍ 0.01; see Table 4), 15% additional variance in sleep quality (p ⬍ 0.001; see Table 5), and 9% additional variance in fatigue (p ⬍ 0.05; see Table 6).
Differences Based on Severity of Hot Flashes Subsequent analyses sought to evaluate group differences as a function of hot flash severity. Women with hot flashes (n ⫽ 28) were
Vol. 19 No. 6 June 2000
divided into two groups: those who reported their hot flashes to be mild (n ⫽ 10); and those who reported their hot flashes to be moderate or severe (n ⫽ 18). T-tests and 2 analyses were used to compare these two groups on all medical/demographic/treatment variables and on each of the psychosocial and quality of life outcome measures. Results indicated no significant (p ⬍ 0.05) differences between the two groups on any of the medical/demographic/ treatment variables. Likewise, no significant (p ⬍ 0.05) differences were noted on sleep quality, mental health, and levels of anxiety or depression (see Table 7). However, as shown in Table 7, women with moderate or severe hot flashes reported significantly more fatigue (p ⬍ 0.05) and poorer physical health (p ⬍ 0.05) than women with mild hot flashes. In addition, women with moderate or severe hot flashes reported a greater impact of fatigue on quality of life (p ⬍ 0.05) and higher levels of global (p ⬍ 0.01) and somatic (p ⬍ 0.05) symptoms of fatigue. To further evaluate group differences in quality of life, t-tests comparing mean MOS-36SF subscale scores were performed. Results indicated that women who experienced moderate or severe hot flashes reported significantly greater interference with role functioning due to physical limitations (p ⬍ 0.05), greater bodily pain (p ⬍ 0.05), and lower vitality (p ⬍ 0.01) than women who experienced mild hot flashes. Due to the finding that no medical/demographic/treatment variables were even marginally significantly (p ⬍ 0.10) related to hot flash severity, multivariate analyses were not conducted.
Discussion This study represents the first investigation of the prevalence, severity, and correlates of
Table 4 Hierarchical Regression Analyses of Physical Health Variable Step 1 Type of treatment Education Income Step 2 Prevalence of hot flashes
B
r2 change
⫺0.16 0.23a 0.27a
0.15 0.15 0.15
⫺0.34a
0.11a
Multiple R ⫽ 0.51, F(4,65) ⫽ 5.76, p ⬍ 0.001. ap ⬍ 0.05.
Vol. 19 No. 6 June 2000
Impact of Hot Flashes
Table 5 Hierarchical Regression Analyses of Sleep Quality Variable Step 1 Type of treatment Education Step 2 Prevalence of hot flashes
B
r2 change
⫺0.11 0.18
0.05 0.05
⫺0.39a
0.15a
Multiple R ⫽ 0.45, F(3,66) ⫽ 5.60, p ⬍ 0.01. ap ⬍ 0.05.
hot flashes among women actively undergoing chemotherapy and radiotherapy treatment for breast cancer. The findings indicate that hot flashes are occurring in a large percentage (40%) of breast cancer patients in active treatment. Among those women reporting hot flashes, 67% rated them as either moderate or severe and 58% indicated that they were somewhat or very much distressed by their symptoms. Moreover, women with hot flashes experienced poorer sleep quality, higher levels of fatigue, and poorer physical health compared to women without hot flashes. Multivariate analyses revealed that the impact of both the prevalence and severity of hot flashes upon these psychosocial and quality of life outcomes remained significant even after controlling for relevant medical, demographic, and treatment variables. The magnitude of the impact of hot flashes on quality of life is illustrated by findings showing that, compared to women without hot flashes, women with hot flashes experienced 66% more fatigue, 63% poorer sleep quality, and 20% poorer physical health. These differences underscore the importance of further study of this highly prevalent and distressing symptom. This discussion will evaluate the findings of the present study in light of previous research in this area, consider the limitations of these findings, and suggest relevant clinical implications and directions for future research. Table 6 Hierarchical Regression Analyses of Fatigue Variable Step 1 Type of treatment Education Step 2 Prevalence of hot flashes
B
r2 change
⫺0.06 0.10
0.05 0.05
⫺0.31a
0.09a
Multiple R ⫽ 0.37, F(3,66) ⫽ 3.40, p ⬍ 0.05. ap ⬍ 0.05.
443
The prevalence of hot flashes in the sample under study (40%) was found to be markedly lower than the 65% prevalence rate found in two previous studies that assessed hot flashes in women who had completed treatment for breast cancer.2,3 This finding is particularly interesting considering the fact that the average ages of women in all three studies were roughly the same and that all women were considered postmenopausal at the time of assessment. A possible explanation for this finding is that 9% of the women in the present study were using tamoxifen at the time of assessment, versus 35% in the study by Couzi et al.2 and 47% in the study by Carpenter et al.3 Previous research has found hot flashes to be a common side effect of tamoxifen use.21 Moreover, both Couzi et al.2 and Carpenter et al.3 reported that women who used tamoxifen reported a higher prevalence and severity of hot flashes compared to women not taking tamoxifen. Thus, the somewhat higher prevalence rates of hot flashes found in these two studies may be accounted for, in part, by greater use of tamoxifen among women who have completed breast cancer treatment. The present study showed that both the prevalence and severity of hot flashes were associated with greater fatigue, poorer sleep quality, and poorer physical health among breast cancer patients during treatment. Furthermore, multivariate analyses demonstrated that the relationship between hot flashes and quality of life outcomes remained significant even after controlling for the medical, demographic, and treatment variables that were related to these outcomes. The poorer quality of life associated with hot flashes in this population underscores the need for safe and effective methods of controlling hot flashes in breast cancer patients during as well as after the completion of treatment. Due to the potential for disease reoccurrence, women with breast cancer are often advised by their physicians to avoid the use of the hormone replacement therapies that can be used to control vasomotor symptoms of menopause, including hot flashes. Currently, several other treatments for relief of hot flashes in breast cancer patients are being explored, including vitamin E, venlafaxine hydrochloride, and the progestational agent megestrol acetate.22–24 Preliminary evidence suggests that both venlafaxine hydrochloride and megestrol
444
Stein et al.
Vol. 19 No. 6 June 2000
Table 7 Mean Scores on Psychosocial and Quality of Life Outcomes by Hot Flash Severity
Variable POMS-Fa FSI disruptiveness indexa MFSI-SF fatigue dimensions Global fatigueb Somatic fatiguea Affective fatigue Cognitive fatigue Vigor PSQI sleep quality CES-D STAI-S MOS-36-SF summary scales Physical summary scalea Mental summary scale Physical MOS-36-SF subscale scores Physical Functioning Role Physicala Bodily Paina General Health Mental MOS-36-SF subscale scores Social Functioning Role Emotional Vitalityb Mental Health ap bp
Moderate-Severe HF (N ⫽ 18) M (SD)
Mild HF (N ⫽ 10) M (SD)
16.44 (8.06) 4.30 (2.47)
9.30 (6.27) 2.33 (2.18)
14.83 (7.18) 7.83 (7.15) 6.83 (6.64) 7.22 (7.11) 9.39 (4.95) 1.50 (0.78) 15.94 (13.27) 38.83 (15.16)
6.80 (5.37) 2.70 (2.45) 3.50 (3.10) 3.20 (2.20) 10.40 (6.00) 1.30 (0.48) 9.60 (6.31) 35.11 (9.53)
31.20 (7.70) 43.82 (11.25)
39.96 (13.45) 46.76 (10.65)
45.00 (28.65) 5.55 (18.30) 51.05 (20.98) 49.94 (20.88)
59.50 (32.01) 30.00 (43.78) 72.70 (23.94) 61.00 (23.69)
48.61 (25.32) 48.15 (46.05) 28.33 (19.85) 68.00 (20.16)
60.00 (26.22) 50.00 (47.79) 51.50 (22.49) 75.60 (14.66)
⬍ 0.05. ⬍ 0.01.
acetate may be helpful in relieving hot flashes among women who have completed breast cancer treatment. This line of research offers hopeful preliminary evidence that breast cancer patients will soon have non-estrogen therapies to help relieve hot flashes. However, the efficacy of these therapies in women currently undergoing treatment for breast cancer has yet to be determined. Future research should focus on the safety and efficacy of non-hormonal therapies for the relief of hot flashes among women treated for breast cancer. Although this study indicates that hot flashes are a significant problem for women as they undergo treatment for breast cancer, certain limitations should be considered when evaluating the results. First, because this is the initial study to evaluate hot flashes during treatment for breast cancer, the results should be considered preliminary and in need of replication and further study. Because the sample size was relatively small for some comparisons, findings that failed to reach statistical significance (p ⬍ 0.05) may partly reflect low statistical power. In addition, the possibility exists that one or more of the significant findings may be a reflection
of Type I error. Second, the present study used data from a single assessment during treatment to provide information about the prevalence, severity, and correlates of hot flashes. Investigation of the incidence and trajectory of hot flashes throughout the course of cancer treatment should be the focus of future studies. Third, a more thorough assessment of the experience of hot flashes is needed. In addition to evaluating the prevalence, severity, and distress of hot flashes, future research should investigate when hot flashes are most and least intense, how long they last, the average number of hot flashes during the day, and how they impact one’s ability to perform normal activities. Finally, the current sample is rather homogeneous, representing mostly Caucasian, middle-class, fairly well-educated women treated at a single National Cancer Institute-designated cancer center. Future research should also examine hot flashes in samples that are more heterogeneous in terms of their sociodemographic characteristics. In summary, the present study provides the first systematic evaluation of hot flashes among women actively undergoing treatment for breast
Vol. 19 No. 6 June 2000
Impact of Hot Flashes
cancer. The results indicate that hot flashes are experienced by nearly half of the patients during treatment and that they are experienced as severe by a quarter of the patients. Furthermore, women with hot flashes reported experiencing significantly more fatigue, poorer sleep quality, and poorer physical quality of life compared with women without hot flashes. These results indicate the need for research evaluating interventions to control hot flashes in women undergoing breast cancer treatment.
Acknowledgments This work was supported by an American Cancer Society Institutional Research Grant (#202) to the H. Lee Moffitt Cancer Center and Research Institute. We thank Lora Azzarello and Alice Hager for their assistance with the data collection and manuscript preparation.
References 1. Kronenberg F. Hot flashes: epidemiology and physiology. Ann NY Acad Sci 1990;592:52–86. 2. Couzi RJ, Helzlsouer KJ, Fetting JH. Prevalence of menopausal symptoms among women with a history of breast cancer and attitudes toward estrogen replacement therapy. J Clin Oncol 1995;13:2737–2744. 3. Carpenter JS, Andrykowski MA, Cordova M, Cunningham L, Studts J, McGrath P, et al. Hot flashes in postmenopausal women treated for breast cancer: prevalence, severity, correlates, management, and relation to quality of life. Cancer 1998;82:1682–1691. 4. Ravnikar V. Physiology and treatment of hot flashes. Obstet Gynecol 1990;75(Suppl 3):3S–7S. 5. Dupont W, Page D. Menopausal estrogen replacement therapy and breast cancer. Arch Intern Med 1991;151:67–72. 6. Steinberg KK, Thacker SB, Smith J, et al. A meta-analysis of the effect of estrogen replacement therapy and the risk of breast cancer. JAMA 1991; 265:1985–1990. 7. Ingle JN, Mailliard JA, Schaid DJ, Krook JE, Gesme DH, Windschitl HE, et al. A double-blind trial of tamoxifen plus prednisolone versus tamoxifen plus placebo in postmenopausl women with metastatic breast cancer. Cancer 1991;68:34–39.
445
10. Spielberger CD, Gorsuch RL, Lushene RD. STAI: Manual for the State-Trait Anxiety Inventory. Palo Alto, CA: Consulting Psychologists Press, 1970. 11. Spielberger CD. State-Trait Anxiety Inventory for Adults. Palo Alto, CA: Consulting Psychological Press, 1983. 12. Radloff LS. The CES-D Scale: a self-report depression scale for research in the general population. Appl Psycho Meas 1977;1:385-401. 13. Radloff LS, Teri L. Use of the Center for Epidemiologic Studies-Depression Scale with older adults. In: Brink TL, Ed. Clinical Gerontology: A Guide to Assessment and Intervention. New York: Haworth Press, 1986:119–135. 14. Beeber LS, Shea J, McCorkle R. The Center for Epidemiologic Studies Depression Scale as a measure of depressive symptoms in newly diagnosed patients. J Psychosoc Oncol 1998;16:1–20. 15. McNair DM, Lorr M, Droppleman LF. The Manual for the Profile of Mood States. San Diego, CA: Educational and Industrial Testing Service, 1981. 16. Winningham ML, Nail LM, Burke MB, et al. Fatigue and the cancer experience: The state of knowledge. Oncol Nurs Forum 1994;21:23-36. 17. Hann D, Jacobsen PB, Azzarello LM, et al. Measurement of fatigue in cancer patients: development and validation of the Fatigue Symptom Inventory. Qual Life Res 1998;7:301–310. 18. Broeckel JA, Jacobsen PB, Horton J, Balducci L, Lyman GH. Characteristics and correlates of fatigue after adjuvant chemotherapy for breast cancer. J Clin Oncol 1998;16:1689–1696. 19. Stein KD, Martin SC, Hann DM, et al. Construction of a multidimensional measure of fatigue for clinical use with cancer patients. Cancer Prac 1998;6:143–152. 20. Buysse DJ, Reynolds CF, Monk TH, et al. The Pittsburgh Sleep Quality Index: a new instrument for psychiatric practice and research. Psychiatry Res 1988;28:193–213. 21. Love RR, Cameron L, Connell BL, Leventhal H. Symptoms associated with tamoxifen treatment in postmenopausal women. Arch Intern Med 1991; 151:1842–1847. 22. Barton DL, Loprinzi CL, Quella SK, Sloan JA, Veeder MH, Egner JR, et al. Prospective evaluation of vitamin E for hot flashes in breast cancer survivors. J Clin Oncol 1998;16:495–500.
8. Ware JE, Snow KK, Kosinski M, Gandek B. SF-36 Health Survey: manual and interpretation guide. Boston, MA: New England Medical Center, 1993.
23. Loprinzi CL, Pisansky TM, Fonseca R, Sloan JA, Zahasky KM, Quella SK, et al. Pilot evaluation of venlafaxine hydrochloride for the therapy of hot flashes in cancer survivors. J Clin Oncol 1998;16: 2377–2381.
9. Portenoy RK, Thaler HT, Kornblith AB, et al. The Memorial Symptom Assessment Scale: An instrument for the evaluation of symptom prevalence, characteristics and distress. Eur J Cancer 1994;30A:1326–1336.
24. Quella SK, Loprinzi CL, Sloan JA, Vaught NL, DeKrey WL, Fischer T, et al. Long term use of megestrol acetate by cancer survivors for the treatment of hot flashes. Cancer 1998;82:1784–1788.
Journal of Pain and Symptom Management
Vol. 19 No. 6 June 2000
Original Article
Impact of Hot Flashes on Quality of Life Among Postmenopausal Women Being Treated for Breast Cancer Kevin D. Stein, PhD, Paul B. Jacobsen, PhD, Danette M. Hann, PhD, Harvey Greenberg, MD, and Gary Lyman, MD Psychosocial Oncology (K.D.S., P.B.J.) and Comprehensive Breast Cancer (H.G., G.L.) Programs, University of South Florida H. Lee Moffitt Cancer Center and Research Institute, Tampa, Florida, and Behavioral Research Center (D.M.H.), American Cancer Society, Atlanta, Georgia, USA
Abstract Hot flashes are among the most commonly reported symptoms among women who have completed treatment for breast cancer. Relatively little is known, however, about hot flashes among women while they are undergoing breast cancer treatment. The present study investigated the prevalence and severity of hot flashes of women during chemotherapy and radiotherapy for breast cancer. We also sought to identify the medical, demographic, and treatment correlates of hot flashes during treatment and to document the impact of hot flashes on quality of life. Seventy postmenopausal women with breast cancer completed a self-report questionnaire packet during chemotherapy and radiotherapy. Forty percent (n ⫽ 28) reported hot flashes during the week prior to assessment. Of the 28 women endorsing hot flashes, 25% (n ⫽ 7) rated them as severe, 39% (n ⫽ 11) rated them as moderate, and 36% (n ⫽ 10) rated them as mild. Women with hot flashes were significantly (p ⬍ 0.05) younger and reported significantly (p ⬍ 0.001) more fatigue, poorer sleep quality, and poorer physical health compared to women without hot flashes. Multivariate analyses revealed that, even after controlling for relevant medical, demographic, and treatment variables, the prevalence of hot flashes significantly (p ⬍ 0.05) predicted poorer sleep quality, more fatigue, and worse physical health. The results indicate that hot flashes are experienced by a sizable percentage of postmenopausal breast cancer patients as they undergo treatment. Hot flashes during cancer treatment appear to have a negative impact upon patient quality of life that may be due, in part, to fatigue and interference with sleep. Future research should seek to evaluate interventions to relieve hot flashes during breast cancer treatment as a means of improving patient quality of life. J Pain Symptom Manage 2000;19:436-445. © U.S. Cancer Pain Relief Committee, 2000. Key Words Hot flashes, breast cancer, fatigue, quality of life
Introduction Address reprint requests to: Kevin D. Stein, PhD, Behavioral Research Center, American Cancer Society, Clifton Rd. NE, Atlanta, GA 30329, USA. Accepted for publication: July 14, 1999. © U.S. Cancer Pain Relief Committee, 2000 Published by Elsevier, New York, New York
During menopause, decreasing levels of the hormone estrogen produce a range of distressing vasomotor symptoms that include night sweats, chills, and hot flashes. Of these symptoms, hot 0885-3924/00/$–see front matter PII S0885-3924(00)00142-1
Vol. 19 No. 6 June 2000
Impact of Hot Flashes
flashes are the most common and are associated with the most physical and emotional distress.1 Hot flashes may have a negative impact on a woman’s quality of life by causing decreased sleep quality and increased levels of fatigue and depression.2,3 In healthy women, hot flashes can be managed through the use of hormone replacement therapy (HRT).4 Women with breast cancer, however, often forego the use of hormone replacement therapy due to the potential for an increase in risk of disease recurrence.5,6 As a result, these women often suffer frequent and distressing hot flashes.7 Researchers have only recently begun to document the prevalence and severity of hot flashes in breast cancer patients. In one of the first studies to focus on this issue, Couzi et al.2 surveyed 190 postmenopausal women (average age ⫽ 54.9 years) who had been diagnosed with breast cancer 2–6 years previously. To be included in the study, women had to have an absence of menses for at least 6 months, be between the ages of 40 and 65, and have no recent evidence of disease recurrence or metastasis. Participants were asked about the prevalence and severity of a number of vasomotor, gynecologic, and other medical symptoms. The results indicated that hot flashes were the most frequently reported symptom, endorsed by 65% of the women. More recently, Carpenter et al.3 assessed hot flashes and quality of life in 114 postmenopausal women (average age ⫽ 58.8 years) treated previously for breast cancer. To be included in the study, women had to have had an absence of menses for at least 1 year, be at least 3 months post-treatment, and have no recent evidence of disease recurrence or metastasis. On average, these women had completed treatment 34.9 months previously (range ⫽ 4– 116 months). Consistent with Couzi et al.,2 65% of the women reported experiencing hot flashes at the time of assessment. Of those women reporting hot flashes, 59% rated their symptoms as severe. Women with hot flashes were found to be younger at diagnosis and less educated compared to women without hot flashes. Among women with hot flashes, more severe symptoms were associated with higher body mass index, younger age at diagnosis, and use of tamoxifen. Both prevalence and severity of hot flashes were marginally (p ⬍ 0.10) related to poorer mental and physical quality of life.
437
The results of these studies suggest that hot flashes are common and may be severe among women previously treated for breast cancer. The degree to which these findings also apply to women currently undergoing treatment for breast cancer remains unknown. To address this issue, the current study sought to: 1) document the prevalence and severity of hot flashes among postmenopausal women during treatment for breast cancer; 2) identify medical, demographic, and treatment correlates of hot flashes during breast cancer treatment; and 3) determine the impact of the hot flashes on breast cancer patients’ quality of life during breast cancer treatment.
Methods Participants To be eligible for the analyses described here, patients had to: a) be women 18 years of age or older; b) be diagnosed with breast cancer and receiving adjuvant chemotherapy or adjuvant radiotherapy; c) be postmenopausal (i.e., report absence of menses for at least 12 months); d) have no known untreated or unstable major medical conditions; e) have no known major psychiatric or neurological disorders that would interfere with completion of study measures; f) be able to read English; g) have no history of treatment for other types of cancer; and h) provide written informed consent to participate in a longitudinal study of the side effects of breast cancer treatment. Eighty-nine women (55 radiotherapy and 34 chemotherapy patients) met these criteria and agreed to participate in the longitudinal study. Two radiotherapy patients dropped out of the study and three chemotherapy patients became ineligible because their treatment ended prematurely or they began chemotherapy. One chemotherapy patient became ineligible because she began radiotherapy during chemotherapy treatment. In addition, eight radiotherapy patients and five chemotherapy patients were excluded due to missing data. Thus, complete longitudinal data were obtained from a total of 70 patients: 42 radiotherapy patients and 28 chemotherapy patients.
Procedure and Measures As a part of a larger longitudinal study of side effects of breast cancer treatment, partici-
438
Stein et al.
pants were recruited during an outpatient appointment prior to their first scheduled chemotherapy or radiotherapy treatment. Those patients who met eligibility criteria and provided informed consent were asked to complete a self-report questionnaire packet (described below) prior to the beginning of treatment and at several times during the course of treatment. For the purpose of the current study, all analyses were conducted on self-report data collected approximately 4–6 weeks after the start of treatment. The self-report measures administered to study participants included a demographic data form, the SF-36 Health Survey, a modified version of the Memorial Symptom Assessment Scale (MSAS), the State-Trait Anxiety Inventory (STAI), the Center for Epidemiological Studies Depression Scale (CES-D), the Profile of Mood States Fatigue Scale (POMS-F), the Fatigue Symptom Inventory (FSI), the Multidimensional Fatigue Symptom Inventory (MFSI), and the Pittsburgh Sleep Quality Index (PSQI). The order of the instruments was randomized for each individual packet to reduce the chance of order effects. Medical record reviews were conducted to obtain information about stage of disease, type of surgery, adjuvant chemotherapy agent(s), adjuvant radiation therapy, and adjuvant hormonal therapy. The SF-36 Health Survey8 is a 36-item selfreport measure of physical and emotional functioning that was adapted from the Medical Outcome Study (MOS-36-SF). The SF-36 Health Survey contains eight subscales: physical functioning, role functioning-physical, bodily pain, general health, vitality, social functioning, role functioning-emotional, and mental health. Scores on each of the eight subscales can range from 0 to 100. The MOS-36-SF yields two summary scores: physical health summary and mental health summary. These summary scores are expressed as T-scores (mean ⫽ 50; SD ⫽ 10). Consistent with the individual subscales, higher scores on the MOS-36-SF summary scales indicate better functioning. Data from over 20,000 patients in the Medical Outcome Study has demonstrated that the MOS-36-SF is a valid measure of both physical and emotional functioning.8 An item from a modified version of the Memorial Symptom Assessment Scale (MSAS)9 was used in the present study to assess the prev-
Vol. 19 No. 6 June 2000
alence, severity, and distress of hot flashes. Respondents indicated if they had experienced hot flashes (presence/absence) during the past week. If the respondent confirmed the presence of hot flashes, they rated the severity (1 ⫽ slight; 3 ⫽ severe) and level of distress (0 ⫽ not at all; 3 ⫽ very much) associated with the symptom. The State Trait Anxiety Inventory (STAI)10 contains two separate 20-item scales that measure state (current or situational) anxiety and trait (general) anxiety. For the state version, respondents rate on a 4-point Likert scale (1 ⫽ not at all; 4 ⫽ very much so) the extent to which they are currently experiencing each item. For the trait version, respondents indicate on a 4-point Likert scale (1 ⫽ almost never; 4 ⫽ almost always) how often they generally experience each item. Scores for each scale range from 0 to 80. The STAI is supported by an extensive literature regarding its reliability and validity.11 The Center for Epidemiological Studies Depression Scale (CES-D)12 is a 20-item measure of depressive symptomatology. Respondents rate how frequently they have experienced each depressive symptom during the past week on a 4-point rating scale (0 ⫽ rarely or none of the time; 3 ⫽ most or all of the time). Total scores range from 0 to 60. The reliability and validity of the CES-D have been demonstrated in use with a wide range of populations, including older adults13 and cancer patients.14 The Profile of Mood States Fatigue Scale (POMS-F)15 consists of 7 items that assess feelings of weariness and low energy. Respondents indicate the degree to which they have experienced each of these feelings during the previous week on a 5-point scale (0 ⫽ not at all; 4 ⫽ extremely). Total scores range from 0 to 28. The POMS-F has been used extensively to assess cancer-related fatigue16 and has been shown to have acceptable reliability and validity.15 The Fatigue Symptom Inventory (FSI)17 is a 14-item scale that assesses the frequency, severity, and daily pattern of fatigue as well as its interference with quality of life. Frequency is measured both by the number of days in the past week (0–7) and the percentage of each day on average (0 ⫽ not at all; 10 ⫽ entire day) that respondents felt fatigued. Severity is measured on an 11-point scale (0 ⫽ not at all fatigued; 10 ⫽ as fatigued as I could be) that as-
Vol. 19 No. 6 June 2000
Impact of Hot Flashes
sesses average fatigue during the past week. Interference of fatigue with quality of life is measured on seven separate 11-point scales (0 ⫽ no interference; 10 ⫽ extreme interference) that assess the degree to which fatigue was judged to interfere with general activity, ability to bathe and dress, normal work activity, ability to concentrate, relations with others, enjoyment of life, and mood in the past week. Responses on the seven items are summed to provide a total interference score. The reliability and validity of the FSI has been demonstrated in previous research with both breast cancer patients and healthy women.17,18 The Multidimensional Fatigue Symptom Inventory (MFSI)19 is an 83-item measure of fatigue that includes subscales assessing global, somatic, affective, cognitive, and behavioral dimensions of fatigue. All MFSI items are scored on a five-point Likert scale (0 ⫽ not at all; 4 ⫽ extremely) indicating the level of fatigue during the past week. Factor analytic techniques confirmed the five-factor structure of the MFSI. Preliminary evidence indicates that the MFSI is a reliable and valid measure of fatigue symptom domains in women with breast cancer.19 The Pittsburgh Sleep Quality Index (PSQI)20 measures sleep patterns and sleep-related difficulties. A modified version of the PSQI was used to obtain self-reports of overall sleep quality during the past week. The PSQI has been
439
shown to discriminate between normal, depressives, and individuals with sleep disorders.20
Results Demographic and Medical Characteristics Information about the demographic characteristics of the chemotherapy patients and radiotherapy patients is presented in Table 1. Chemotherapy patients ranged in age from 41 to 78 years (mean ⫽ 57.1, SD ⫽ 8.9). A majority of these women were white (93%), married (68%), and had completed some college course work (78%). Forty-six percent of the chemotherapy patients were employed full- or part-time and an additional 21% were on leave with or without pay. A majority of the women (57%) reported a household income of $40,000 or more per annum. Radiotherapy patients ranged in age from 39 to 81 years (mean ⫽ 64.0, SD ⫽ 9.4). A majority of these women were also white (88%), married (71%), and had completed some college course work (62%). Seventeen percent of the radiotherapy patients were employed full- or part-time and an additional 5% were on leave with or without pay. Twenty-four percent of the radiotherapy patients reported a household income of $40,000 or more per annum. As shown in Table 1, chemotherapy and radiotherapy patients differed significantly (p ⬍ 0.05) in terms of age, level of education, annual household income,
Table 1 Demographic Characteristics of Study Patients by Type of Treatment Variable Age, yearsa* Marital Statusb Married Other Raceb White Nonwhite Educationb* College or more High School or less Employmentb* Employed Not Employed Annual Household Incomeb* ⬍ $40,000 per annum ⬎ $40,000 per annum
Chemotherapy patients (N ⫽ 28)
Radiotherapy patients (N ⫽ 42)
57.10 ⫾ 8.94
63.98 ⫾ 9.41
19 (68) 9 (32)
30 (71) 12 (29)
24 (86) 4 (14)
39 (93) 3 (7)
22 (79) 6 (21)
26 (62) 16 (38)
13 (46) 15 (54)
7 (17) 35 (83)
12 (43) 16 (57)
32 (76) 10 (24)
expressed as mean ⫾ SD. expressed as N (%). *p ⬍ 0.05. All other comparisons were nonsignificant (p ⬎ 0.05). aValue bValue
440
Stein et al.
and employment status. Compared to radiotherapy patients, chemotherapy patients were younger, more highly educated, more likely to be employed full- or part-time, and had a higher annual household income. Disease stage for the chemotherapy patients at time of diagnosis was classified as follows: stage I, 4 patients (14%); stage II, 19 patients (68%); and stage III, 5 patients (18%). With respect to surgical treatment, 15 chemotherapy patients (54%) underwent a mastectomy, 12 patients (43%) underwent lumpectomy, and 1 patient (3%) underwent no surgical treatment other than breast biopsy prior to chemotherapy. With respect to chemotherapy regimens, 7 patients (25%) received doxorubicin, 11 patients (39%) received doxorubicin and cyclophosphamide, 9 patients (32%) received doxorubicin, cyclophosphamide, and fluorouracil, and 1 patient (4%) received mitoxantrone. None of the chemotherapy patients were receiving hormonal therapy (tamoxifen) at the time of assessment. For radiotherapy patients, disease stage at time of diagnosis was as follows: stage I, 25 patients (60%); stage II, 15 patients (36%); and stage III, 2 patients (4%). With respect to surgical treatment, 9 patients (21%) underwent a mastectomy, 31 patients (74%) underwent a lumpectomy, and 2 patients (5%) underwent a breast biopsy only. Among radiotherapy patients, 6 patients (14%) were receiving hormonal therapy (i.e., tamoxifen) at the end of assessment. In terms of their medical characteristics, chemotherapy patients and radiotherapy patients differed significantly (p ⬍ 0.05) on stage of disease and type of surgery. Compared to radiotherapy patients, chemotherapy patients had more advanced disease and were more likely to have received a mastectomy. In addition, significantly (p ⬍ 0.05) more radiotherapy patients than chemotherapy patients were receiving hormonal therapy (i.e., tamoxifen).
Prevalence and Severity of Hot Flashes Of the 70 women undergoing radiotherapy or chemotherapy treatment for breast cancer, 40% (n ⫽ 28) reported experiencing hot flashes in the week prior to data collection. Of the 28 women who reported hot flashes, 46% (n ⫽ 13) were chemotherapy patients and 54% (n ⫽ 15) were radiotherapy patients. Chemotherapy
Vol. 19 No. 6 June 2000
and radiotherapy patients did not differ significantly in the prevalence of hot flashes (p ⬎ 0.05). Among women experiencing hot flashes (n ⫽ 28), 36% (n ⫽ 10) rated their symptoms as mild, 39% (n ⫽ 11) rated them as moderate, and 25% (n ⫽ 7) rated them as severe. Consistent with the prevalence of hot flashes, the severity of hot flashes also did not differ significantly (p ⬎ 0.05) by the type of treatment (chemotherapy vs. radiotherapy). Among women reporting hot flashes (n ⫽ 28), ratings of distress associated with hot flashes were as follows: 4% (n ⫽ 1) not at all, 43% (n ⫽ 12) a little bit, 28% (n ⫽ 8) somewhat, and 25% (n ⫽ 7) very much. As with the prevalence and severity of hot flashes, the distress associated with hot flashes did not differ significantly (p ⬎ 0.05) by the type of treatment. Due to the finding that chemotherapy and radiotherapy patients did not differ in the prevalence, severity, and distress of hot flashes, the two samples were combined together in subsequent analyses.
Differences Based on Prevalence of Hot Flashes
T-tests and 2 analyses were used to compare women with hot flashes to women without hot flashes on all medical/demographic/treatment variables (see Table 2). Women with hot flashes were found to be significantly younger compared to women without hot flashes (p ⬍ 0.05). In contrast, there were no significant differences between the two groups in level of education, marital or occupational status, annual household income, stage of disease, type of surgery, type of chemotherapy (if applicable), or use of hormonal treatment (p ⬎ 0.05). As shown in Table 3, t-tests were also used to examine differences between women with hot flashes and women without hot flashes on psychosocial and quality of life outcomes including fatigue (POMS-F; FSI; MFSI), sleep problems (PSQI), depression (CES-D), anxiety (STAI-S), and physical and mental health (MOS-36-SF). Results indicated that women with hot flashes reported significantly higher levels of fatigue (p ⬍ 0.01), greater interference of fatigue with quality of life (p ⬍ 0.01), poorer sleep quality (p ⬍ 0.01), and poorer physical health (p ⬍ 0.01). The two groups did not differ in terms of their mental health or levels of anxiety and depression. To further delineate the nature of the differences in fatigue between the two groups, scores from the MFSI were evaluated (see Table 3).
Vol. 19 No. 6 June 2000
Impact of Hot Flashes
441
Table 2 Demographic Characteristics of Study Patients by Prevalence of Hot Flashes Variable Age, yearsa* Marital Statusb Married Other Raceb White Nonwhite Educationb College or more High School or less Employmentb Outside Home Other Annual Household Incomeb ⬍ $40,000 per annum ⬎ $40,000 per annum
Women with HF (N ⫽ 28)
Women without HF (N ⫽ 42)
57.64 ⫾ 10.08
63.61 ⫾ 8.89
21 (75) 7 (25)
28 (67) 14 (33)
24 (86) 4 (14)
39 (93) 3 (7)
8 (29) 20 (71)
11 (26) 31 (74)
8 (29) 20 (71)
12 (29) 30 (71)
16 (57) 12 (43)
28 (67) 14 (33)
expressed as mean ⫾ SD. expressed as N (%). *p ⬍ 0.05. All other comparisons were nonsignificant (p ⬎ 0.05).
aValue bValue
T-tests comparing mean MFSI scale scores indicated that women experiencing hot flashes reported significantly more global fatigue (p ⬍ 0.05), somatic fatigue (p ⬍ 0.05), cognitive fa-
tigue (p ⬍ 0.05), and less vigor (p ⬍ 0.05) than women who did not suffer from hot flashes. To further delineate the nature of quality of life differences between the two groups, sub-
Table 3 Mean Scores on Psychosocial and Quality of Life Outcomes by Hot Flash Prevalence
Variable POMS-Fb FSI disruptiveness indexb MFSI-SF fatigue dimensions Global fatiguea Somatic fatiguea Affective fatigue Cognitive fatiguea Vigora PSQI sleep qualityb CES-D STAI-S MOS-36-SF summary scales Physical summary scaleb Mental summary scale Physical MOS-36-SF subscale scores Physical functioningc Role physical Bodily pain General healthb Mental MOS-36-SF subscale scores Social functioninga Role emotional Vitality Mental health ⬍ 0.05. ⬍ 0.001. cp ⬍ 0.0001. HF ⫽ Hot flashes. ap bp
Women with HF (N ⫽ 28) M (SD)
Women without HF (N ⫽ 42) M (SD)
13.89 (8.13) 3.60 (2.52)
8.36 (7.10) 2.00 (1.86)
11.96 (7.58) 6.00 (6.36) 5.64 (5.79) 5.79 (6.11) 9.75 (5.26) 1.43 (0.69) 13.67 (11.56) 37.59 (13.47)
7.50 (6.52) 2.76 (4.41) 3.24 (4.65) 3.17 (3.46) 13.45 (6.29) 0.88 (0.63) 9.25 (10.19) 33.19 (13.38)
34.33 (10.77) 44.87 (10.94)
41.33 (9.14) 48.72 (12.38)
50.18 (30.14) 14.29 (31.50) 58.79 (24.08) 53.89 (22.15)
71.66 (21.49) 27.38 (38.18) 68.95 (25.09) 69.38 (20.89)
52.68 (25.77) 48.81 (45.77) 36.61 (23.34) 70.71 (18.48)
68.75 (24.42) 63.49 (44.05) 47.74 (26.09) 77.81 (20.34)
442
Stein et al.
scale scores from the MOS-36-SF were evaluated (see Table 3). T-tests comparing mean MOS-36-SF subscale scores indicated that women experiencing hot flashes reported poorer physical functioning (p ⬍ 0.0001), general health (p ⬍ 0.001), and social functioning (p ⬍ 0.05) compared to women without hot flashes. Multivariate analyses were conducted to determine the extent to which the presence of hot flashes accounted for variability in the major psychosocial and quality of life outcomes after controlling for variability attributable to medical/demographic/treatment variables. Separate hierarchical regression analyses were conducted for each of the major psychosocial and quality of life outcomes found to be related to the presence of hot flashes (physical health, sleep quality, and fatigue). Medical/demographic/treatment variables were included in the multivariate analyses if they were found to be even marginally (p ⬍ 0.10) correlated with the outcome measure. For physical health (as measured by the MOS-36-SF physical summary scale), the medical/demographic/treatment variables that met the above criterion were level of education and income. For sleep quality (as measured by the PSQI), level of education was the only variable to meet the criterion. For fatigue (as measured by the POMS-Fatigue Scale), age was the only medical/demographic/treatment variable to meet the criterion. Despite the fact that type of treatment (chemotherapy vs. radiotherapy) was not even marginally correlated (p ⬍ 0.10) with any of the psychosocial or quality of life outcome measures, it was included in all multivariate analyses. Results of the hierarchical regression analyses indicated that, for each of the psychosocial and quality of life outcomes, the prevalence of hot flashes accounted for a significant amount of variance after controlling for medical/demographic/ treatment variables included in the models. Specifically, the prevalence of hot flashes accounted for 11% additional variance in physical health (p ⬍ 0.01; see Table 4), 15% additional variance in sleep quality (p ⬍ 0.001; see Table 5), and 9% additional variance in fatigue (p ⬍ 0.05; see Table 6).
Differences Based on Severity of Hot Flashes Subsequent analyses sought to evaluate group differences as a function of hot flash severity. Women with hot flashes (n ⫽ 28) were
Vol. 19 No. 6 June 2000
divided into two groups: those who reported their hot flashes to be mild (n ⫽ 10); and those who reported their hot flashes to be moderate or severe (n ⫽ 18). T-tests and 2 analyses were used to compare these two groups on all medical/demographic/treatment variables and on each of the psychosocial and quality of life outcome measures. Results indicated no significant (p ⬍ 0.05) differences between the two groups on any of the medical/demographic/ treatment variables. Likewise, no significant (p ⬍ 0.05) differences were noted on sleep quality, mental health, and levels of anxiety or depression (see Table 7). However, as shown in Table 7, women with moderate or severe hot flashes reported significantly more fatigue (p ⬍ 0.05) and poorer physical health (p ⬍ 0.05) than women with mild hot flashes. In addition, women with moderate or severe hot flashes reported a greater impact of fatigue on quality of life (p ⬍ 0.05) and higher levels of global (p ⬍ 0.01) and somatic (p ⬍ 0.05) symptoms of fatigue. To further evaluate group differences in quality of life, t-tests comparing mean MOS-36SF subscale scores were performed. Results indicated that women who experienced moderate or severe hot flashes reported significantly greater interference with role functioning due to physical limitations (p ⬍ 0.05), greater bodily pain (p ⬍ 0.05), and lower vitality (p ⬍ 0.01) than women who experienced mild hot flashes. Due to the finding that no medical/demographic/treatment variables were even marginally significantly (p ⬍ 0.10) related to hot flash severity, multivariate analyses were not conducted.
Discussion This study represents the first investigation of the prevalence, severity, and correlates of
Table 4 Hierarchical Regression Analyses of Physical Health Variable Step 1 Type of treatment Education Income Step 2 Prevalence of hot flashes
B
r2 change
⫺0.16 0.23a 0.27a
0.15 0.15 0.15
⫺0.34a
0.11a
Multiple R ⫽ 0.51, F(4,65) ⫽ 5.76, p ⬍ 0.001. ap ⬍ 0.05.
Vol. 19 No. 6 June 2000
Impact of Hot Flashes
Table 5 Hierarchical Regression Analyses of Sleep Quality Variable Step 1 Type of treatment Education Step 2 Prevalence of hot flashes
B
r2 change
⫺0.11 0.18
0.05 0.05
⫺0.39a
0.15a
Multiple R ⫽ 0.45, F(3,66) ⫽ 5.60, p ⬍ 0.01. ap ⬍ 0.05.
hot flashes among women actively undergoing chemotherapy and radiotherapy treatment for breast cancer. The findings indicate that hot flashes are occurring in a large percentage (40%) of breast cancer patients in active treatment. Among those women reporting hot flashes, 67% rated them as either moderate or severe and 58% indicated that they were somewhat or very much distressed by their symptoms. Moreover, women with hot flashes experienced poorer sleep quality, higher levels of fatigue, and poorer physical health compared to women without hot flashes. Multivariate analyses revealed that the impact of both the prevalence and severity of hot flashes upon these psychosocial and quality of life outcomes remained significant even after controlling for relevant medical, demographic, and treatment variables. The magnitude of the impact of hot flashes on quality of life is illustrated by findings showing that, compared to women without hot flashes, women with hot flashes experienced 66% more fatigue, 63% poorer sleep quality, and 20% poorer physical health. These differences underscore the importance of further study of this highly prevalent and distressing symptom. This discussion will evaluate the findings of the present study in light of previous research in this area, consider the limitations of these findings, and suggest relevant clinical implications and directions for future research. Table 6 Hierarchical Regression Analyses of Fatigue Variable Step 1 Type of treatment Education Step 2 Prevalence of hot flashes
B
r2 change
⫺0.06 0.10
0.05 0.05
⫺0.31a
0.09a
Multiple R ⫽ 0.37, F(3,66) ⫽ 3.40, p ⬍ 0.05. ap ⬍ 0.05.
443
The prevalence of hot flashes in the sample under study (40%) was found to be markedly lower than the 65% prevalence rate found in two previous studies that assessed hot flashes in women who had completed treatment for breast cancer.2,3 This finding is particularly interesting considering the fact that the average ages of women in all three studies were roughly the same and that all women were considered postmenopausal at the time of assessment. A possible explanation for this finding is that 9% of the women in the present study were using tamoxifen at the time of assessment, versus 35% in the study by Couzi et al.2 and 47% in the study by Carpenter et al.3 Previous research has found hot flashes to be a common side effect of tamoxifen use.21 Moreover, both Couzi et al.2 and Carpenter et al.3 reported that women who used tamoxifen reported a higher prevalence and severity of hot flashes compared to women not taking tamoxifen. Thus, the somewhat higher prevalence rates of hot flashes found in these two studies may be accounted for, in part, by greater use of tamoxifen among women who have completed breast cancer treatment. The present study showed that both the prevalence and severity of hot flashes were associated with greater fatigue, poorer sleep quality, and poorer physical health among breast cancer patients during treatment. Furthermore, multivariate analyses demonstrated that the relationship between hot flashes and quality of life outcomes remained significant even after controlling for the medical, demographic, and treatment variables that were related to these outcomes. The poorer quality of life associated with hot flashes in this population underscores the need for safe and effective methods of controlling hot flashes in breast cancer patients during as well as after the completion of treatment. Due to the potential for disease reoccurrence, women with breast cancer are often advised by their physicians to avoid the use of the hormone replacement therapies that can be used to control vasomotor symptoms of menopause, including hot flashes. Currently, several other treatments for relief of hot flashes in breast cancer patients are being explored, including vitamin E, venlafaxine hydrochloride, and the progestational agent megestrol acetate.22–24 Preliminary evidence suggests that both venlafaxine hydrochloride and megestrol
444
Stein et al.
Vol. 19 No. 6 June 2000
Table 7 Mean Scores on Psychosocial and Quality of Life Outcomes by Hot Flash Severity
Variable POMS-Fa FSI disruptiveness indexa MFSI-SF fatigue dimensions Global fatigueb Somatic fatiguea Affective fatigue Cognitive fatigue Vigor PSQI sleep quality CES-D STAI-S MOS-36-SF summary scales Physical summary scalea Mental summary scale Physical MOS-36-SF subscale scores Physical Functioning Role Physicala Bodily Paina General Health Mental MOS-36-SF subscale scores Social Functioning Role Emotional Vitalityb Mental Health ap bp
Moderate-Severe HF (N ⫽ 18) M (SD)
Mild HF (N ⫽ 10) M (SD)
16.44 (8.06) 4.30 (2.47)
9.30 (6.27) 2.33 (2.18)
14.83 (7.18) 7.83 (7.15) 6.83 (6.64) 7.22 (7.11) 9.39 (4.95) 1.50 (0.78) 15.94 (13.27) 38.83 (15.16)
6.80 (5.37) 2.70 (2.45) 3.50 (3.10) 3.20 (2.20) 10.40 (6.00) 1.30 (0.48) 9.60 (6.31) 35.11 (9.53)
31.20 (7.70) 43.82 (11.25)
39.96 (13.45) 46.76 (10.65)
45.00 (28.65) 5.55 (18.30) 51.05 (20.98) 49.94 (20.88)
59.50 (32.01) 30.00 (43.78) 72.70 (23.94) 61.00 (23.69)
48.61 (25.32) 48.15 (46.05) 28.33 (19.85) 68.00 (20.16)
60.00 (26.22) 50.00 (47.79) 51.50 (22.49) 75.60 (14.66)
⬍ 0.05. ⬍ 0.01.
acetate may be helpful in relieving hot flashes among women who have completed breast cancer treatment. This line of research offers hopeful preliminary evidence that breast cancer patients will soon have non-estrogen therapies to help relieve hot flashes. However, the efficacy of these therapies in women currently undergoing treatment for breast cancer has yet to be determined. Future research should focus on the safety and efficacy of non-hormonal therapies for the relief of hot flashes among women treated for breast cancer. Although this study indicates that hot flashes are a significant problem for women as they undergo treatment for breast cancer, certain limitations should be considered when evaluating the results. First, because this is the initial study to evaluate hot flashes during treatment for breast cancer, the results should be considered preliminary and in need of replication and further study. Because the sample size was relatively small for some comparisons, findings that failed to reach statistical significance (p ⬍ 0.05) may partly reflect low statistical power. In addition, the possibility exists that one or more of the significant findings may be a reflection
of Type I error. Second, the present study used data from a single assessment during treatment to provide information about the prevalence, severity, and correlates of hot flashes. Investigation of the incidence and trajectory of hot flashes throughout the course of cancer treatment should be the focus of future studies. Third, a more thorough assessment of the experience of hot flashes is needed. In addition to evaluating the prevalence, severity, and distress of hot flashes, future research should investigate when hot flashes are most and least intense, how long they last, the average number of hot flashes during the day, and how they impact one’s ability to perform normal activities. Finally, the current sample is rather homogeneous, representing mostly Caucasian, middle-class, fairly well-educated women treated at a single National Cancer Institute-designated cancer center. Future research should also examine hot flashes in samples that are more heterogeneous in terms of their sociodemographic characteristics. In summary, the present study provides the first systematic evaluation of hot flashes among women actively undergoing treatment for breast
Vol. 19 No. 6 June 2000
Impact of Hot Flashes
cancer. The results indicate that hot flashes are experienced by nearly half of the patients during treatment and that they are experienced as severe by a quarter of the patients. Furthermore, women with hot flashes reported experiencing significantly more fatigue, poorer sleep quality, and poorer physical quality of life compared with women without hot flashes. These results indicate the need for research evaluating interventions to control hot flashes in women undergoing breast cancer treatment.
Acknowledgments This work was supported by an American Cancer Society Institutional Research Grant (#202) to the H. Lee Moffitt Cancer Center and Research Institute. We thank Lora Azzarello and Alice Hager for their assistance with the data collection and manuscript preparation.
References 1. Kronenberg F. Hot flashes: epidemiology and physiology. Ann NY Acad Sci 1990;592:52–86. 2. Couzi RJ, Helzlsouer KJ, Fetting JH. Prevalence of menopausal symptoms among women with a history of breast cancer and attitudes toward estrogen replacement therapy. J Clin Oncol 1995;13:2737–2744. 3. Carpenter JS, Andrykowski MA, Cordova M, Cunningham L, Studts J, McGrath P, et al. Hot flashes in postmenopausal women treated for breast cancer: prevalence, severity, correlates, management, and relation to quality of life. Cancer 1998;82:1682–1691. 4. Ravnikar V. Physiology and treatment of hot flashes. Obstet Gynecol 1990;75(Suppl 3):3S–7S. 5. Dupont W, Page D. Menopausal estrogen replacement therapy and breast cancer. Arch Intern Med 1991;151:67–72. 6. Steinberg KK, Thacker SB, Smith J, et al. A meta-analysis of the effect of estrogen replacement therapy and the risk of breast cancer. JAMA 1991; 265:1985–1990. 7. Ingle JN, Mailliard JA, Schaid DJ, Krook JE, Gesme DH, Windschitl HE, et al. A double-blind trial of tamoxifen plus prednisolone versus tamoxifen plus placebo in postmenopausl women with metastatic breast cancer. Cancer 1991;68:34–39.
445
10. Spielberger CD, Gorsuch RL, Lushene RD. STAI: Manual for the State-Trait Anxiety Inventory. Palo Alto, CA: Consulting Psychologists Press, 1970. 11. Spielberger CD. State-Trait Anxiety Inventory for Adults. Palo Alto, CA: Consulting Psychological Press, 1983. 12. Radloff LS. The CES-D Scale: a self-report depression scale for research in the general population. Appl Psycho Meas 1977;1:385-401. 13. Radloff LS, Teri L. Use of the Center for Epidemiologic Studies-Depression Scale with older adults. In: Brink TL, Ed. Clinical Gerontology: A Guide to Assessment and Intervention. New York: Haworth Press, 1986:119–135. 14. Beeber LS, Shea J, McCorkle R. The Center for Epidemiologic Studies Depression Scale as a measure of depressive symptoms in newly diagnosed patients. J Psychosoc Oncol 1998;16:1–20. 15. McNair DM, Lorr M, Droppleman LF. The Manual for the Profile of Mood States. San Diego, CA: Educational and Industrial Testing Service, 1981. 16. Winningham ML, Nail LM, Burke MB, et al. Fatigue and the cancer experience: The state of knowledge. Oncol Nurs Forum 1994;21:23-36. 17. Hann D, Jacobsen PB, Azzarello LM, et al. Measurement of fatigue in cancer patients: development and validation of the Fatigue Symptom Inventory. Qual Life Res 1998;7:301–310. 18. Broeckel JA, Jacobsen PB, Horton J, Balducci L, Lyman GH. Characteristics and correlates of fatigue after adjuvant chemotherapy for breast cancer. J Clin Oncol 1998;16:1689–1696. 19. Stein KD, Martin SC, Hann DM, et al. Construction of a multidimensional measure of fatigue for clinical use with cancer patients. Cancer Prac 1998;6:143–152. 20. Buysse DJ, Reynolds CF, Monk TH, et al. The Pittsburgh Sleep Quality Index: a new instrument for psychiatric practice and research. Psychiatry Res 1988;28:193–213. 21. Love RR, Cameron L, Connell BL, Leventhal H. Symptoms associated with tamoxifen treatment in postmenopausal women. Arch Intern Med 1991; 151:1842–1847. 22. Barton DL, Loprinzi CL, Quella SK, Sloan JA, Veeder MH, Egner JR, et al. Prospective evaluation of vitamin E for hot flashes in breast cancer survivors. J Clin Oncol 1998;16:495–500.
8. Ware JE, Snow KK, Kosinski M, Gandek B. SF-36 Health Survey: manual and interpretation guide. Boston, MA: New England Medical Center, 1993.
23. Loprinzi CL, Pisansky TM, Fonseca R, Sloan JA, Zahasky KM, Quella SK, et al. Pilot evaluation of venlafaxine hydrochloride for the therapy of hot flashes in cancer survivors. J Clin Oncol 1998;16: 2377–2381.
9. Portenoy RK, Thaler HT, Kornblith AB, et al. The Memorial Symptom Assessment Scale: An instrument for the evaluation of symptom prevalence, characteristics and distress. Eur J Cancer 1994;30A:1326–1336.
24. Quella SK, Loprinzi CL, Sloan JA, Vaught NL, DeKrey WL, Fischer T, et al. Long term use of megestrol acetate by cancer survivors for the treatment of hot flashes. Cancer 1998;82:1784–1788.