This material is protected by U.S. copyright law. Unauthorized reproduction is prohibited. To purchase quantity reprints,
please e-mail reprints@ons.org or to request permission to reproduce multiple copies, please e-mail pubpermissions@ons.org.
Hot Flashes and Related Outcomes in
Breast Cancer Survivors and Matched
Comparison Women
Janet S. Carpenter, PhD, RN, David H. Johnson, MD, Lois J. Wagner, MSN, RN,
and Michael A. Andrykowski, PhD
Key Points . . .
Purpose/Objectives: To compare the hot flash symptom
experience and related outcomes between breast cancer
survivors and healthy women.
Women with breast cancer may be predisposed to hot
Design: Descriptive, cross-sectional, comparative study.
flashes as a result of chemotherapy, tamoxifen use, potential
Setting: Southeastern university medical center.
disruptions in circadian rhythms of hormones and body tem-
Sample: 69 of 207 breast cancer survivors contacted via
perature, and contraindications against the use of hormone
a tumor registry and 63 age-matched healthy female vol-
unteers. Survivors were a mean of 57 years and a mean of
replacement therapy.
39 months postdiagnosis.
Breast cancer survivors had hot flashes that were signifi-
Methods: Mailed survey included a demographic, dis-
cantly more frequent, distressing, and of greater duration
ease, and treatment information form; a gynecologic his-
than healthy women of the same age. Naturally postmeno-
tory form; a two-day, prospective, hot flash diary; a de-
pausal breast cancer survivors reported a more severe symp-
tailed hot flash questionnaire; mood and affect scales; and
the Hot Flash-Related Daily Interference Scale.
tom pattern than naturally menopausal healthy women.
Main Research Variables: Hot flashes, mood, affect, in-
No clear temporal pattern of hot flashes was noted in either
terference with daily activities, and overall quality of life.
Findings: Breast cancer survivors had hot flashes that
were significantly more frequent, severe, distressing, and of
Breast cancer survivors with severe hot flashes reported
greater duration. Breast cancer survivors were less likely to
more problems in other areas of functioning, including
be using hormone replacement and more likely to have
mood, affect, daily activities, and quality of life.
tried nonhormonal prescription interventions in the past,
but reported significantly less effectiveness from hot flash
treatments. Breast cancer survivors with severe hot flashes
reported significantly greater mood disturbance; higher
ot flashes are sudden episodes of flushing, sweating,
negative affect; more interference with daily activities, in-
and a sensation of heat often preceded or followed by
cluding sleep, concentration, and sexuality; and poorer
chills (Kronenberg, 1994). Although hot flashes rapidly
overall quality of life in comparison to breast cancer survi-
are becoming recognized as a frequent, severe, and bothersome
vors with no hot flashes to mild hot flashes. Hot flash qual-
ity and triggers were not significantly different between
groups. No clear temporal pattern of hot flashes emerged.
Janet S. Carpenter, PhD, RN, is an assistant professor in the School
Conclusions: Hot flashes are a significant problem for
of Nursing, David H. Johnson, MD, is a professor in the School of
breast cancer survivors, even for those who are naturally
Medicine, and Lois J. Wagner, MSN, RN, is a doctoral student in the
postmenopausal (i.e., did not undergo menopause as a
School of Nursing, all at Vanderbilt University in Nashville, TN.
result of surgery or the effects of chemotherapy). Hot
Michael A. Andrykowski, PhD, is a professor in the Department of
flashes remained fairly stable over time and did not dimin-
Behavioral Science at the University of Kentucky in Lexington. (Sub-
ish in frequency, severity, or associated distress.
mitted July 2001. Accepted for publication November 12, 2001.) This
Implications for Nursing: The findings guide the assess-
research was supported by the 1998­2000 ONS Foundation/Sigma
ment of the uniqueness of the problem of hot flashes expe-
Theta Tau International Research Grant, the Vanderbilt University
rienced by breast cancer survivors and help define out-
School of Nursing Joint Center for Nursing Research, and the
comes to address in clinical practice or include in future
Vanderbilt-Ingram Cancer Center.
hot flash intervention research.
Digital Object Identifier: 10.1188/02.ONF.E16-E25
ONF ­ VOL 29, NO 3, 2002
cancer survivors to age-matched healthy women. Detailed
symptom among breast cancer survivors (Carpenter, 2000;
data were gathered, including hot flash frequency, severity,
Carpenter et al., 1998), a detailed study of hot flashes in breast
bother (distress), duration, quality, aggravating factors, alle-
cancer survivors with comparison to healthy women has not
viating factors, temporal pattern, and impact on mood, affect,
been published previously. Because several unique factors
daily activities, and overall QOL.
predispose breast cancer survivors to hot flashes, existing data
describing hot flashes in healthy women (Kronenberg, 1990,
Conceptual Framework
1994) may not generalize to breast cancer survivors. These
predisposing factors include ovarian disruption caused by
The University of California, San Francisco (UCSF) mo-
chemotherapy and subsequent early and artificial menopause
del of symptom management (Dodd, Miaskowski, & Paul,
(Carpenter et al., 1998; Reichman & Green, 1994), side ef-
2001; UCSF School of Nursing Symptom Management Fac-
fects of the antiestrogen tamoxifen and the selective estrogen
ulty Group, 1994) guided this study. The UCSF model de-
receptor modulator raloxifene (Carpenter et al., 1998; Love,
picts three interrelated dimensions: symptom experience,
1989; Love, Cameron, Connell, & Levanthal, 1991; Pasacreta
symptom outcomes, and symptom management strategies.
& McCorkle, 1998), potential disruptions in circadian
The symptom experience includes an individual's perception
rhythms of hormones (Bartsch et al., 1989; Mormont & Levi,
and evaluation of internal cues as a symptom (i.e., feeling
1997) and body temperature (Carpenter, Gautam, Andry-
suddenly hot and flushed is perceived as a hot flash; ratings
kowski, & Freedman, 2001), and contraindications to the use
of severity and distress). Symptom outcomes include other
of hormone replacement therapy (HRT) because of concerns
aspects of the individual and her life that may be affected by
about stimulating cancer recurrence (Brzezinski, 1995;
a symptom, including QOL and emotional and functional
Runowicz, 1996). These factors may act singly or in combi-
status. Symptom management strategies include various
nation to alter hot flashes in breast cancer survivors when
components of interventions, such as dose and route. This
compared to healthy women. Although the exact physiology
model considers the thorough assessment of a given symp-
of hot flashes is unknown, several of these factors are associ-
tom within the context of related outcomes as necessary for
ated with estrogen withdrawal, which has been implicated as
choosing, implementing, and evaluating appropriate symp-
a trigger of hot flashes. Thus, data previously and exclusively
tom management interventions. Thus, the investigators
gathered among healthy women may not reflect the hot flash
sought to describe the hot flash symptom experience (fre-
experience of breast cancer survivors.
quency, severity, bother or distress, duration, quality, aggra-
Data specifically assessing hot flashes in breast cancer
vating factors, alleviating factors, and temporal pattern) and
survivors are limited to only a handful of published studies
potentially related symptom outcomes (mood, affect, inter-
(Carpenter et al., 1998; Carpenter & Andrykowski, 1999;
ference with daily activities, and overall QOL) to better ad-
Couzi, Helzlsouer, & Fetting, 1995; Finck, Barton, Loprinzi,
dress symptom management needs.
Quella, & Sloan, 1998). Hot flashes have been found to af-
fect 65% of breast cancer survivors (Carpenter et al., 1998;
Couzi et al.) and have been described as both severe and dis-
tressing (Carpenter et al., 1998; Carpenter & Andrykowski;
Couzi et al.; Finck et al.). These studies are limited further
A cross-sectional, descriptive, comparative study was used
in that they only have included qualitative data from a small
to address the study purposes.
sample of women with breast cancer (Finck et al.); assessed
hot flash frequency, severity, or distress without providing
additional descriptive data (Carpenter et al., 1998; Carpen-
ter & Andrykowski; Couzi et al.); or did not include any
Inclusion criteria for breast cancer survivors were (a) first
comparison groups (Carpenter et al., 1998; Carpenter &
time diagnosis of breast cancer (i.e., had not experienced re-
Andrykowski; Couzi et al.; Finck et al.). For example, poten-
currence), (b) no other history of cancer, (c) disease free at the
tial differences in the quality and temporal pattern of hot
time of study participation, (d) at least three months post-
flashes between breast cancer survivors and healthy women
completion of surgery, radiation, or chemotherapy, and (e)
have not been studied previously. In addition, the majority
less than or equal to six years postdiagnosis. The latter crite-
of published reports focus on evaluating hot flash treatment
ria reflected the availability of women in the cancer registry.
options and typically limit description and measurement of
Inclusion criteria for healthy women were (a) no history of
outcomes to hot flash frequency counts and severity or dis-
cancer, (b) naturally pre-, peri-, or postmenopausal (i.e., intact
tress ratings (Barton et al., 1998; Goldberg et al., 1994;
uterus and ovaries), and (c) age matched to a breast cancer
Loprinzi et al., 1994, 1998; Quella et al., 1998). Thus, data
survivor within two years. In addition, all women were over
describing the problematic nature of hot flashes generally
age 18 and able to speak, read, and write English.
are limited in these intervention studies. In addition, out-
comes other than hot flashes, such as mood, affect, daily
activities, and overall quality of life (QOL) are not assessed
Following Institutional Review Board approval, eligible
routinely. A need exists to better understand the impact of
breast cancer survivors were identified through a university-
hot flashes on such outcomes to determine which, if any,
based medical center tumor registry located in the southeast-
might be appropriate to address in clinical practice and in-
ern United States. A list of potentially eligible women was
clude in future hot flash intervention studies.
released to their physicians, who then sent a letter introducing
The purposes of this study were to obtain a detailed, mul-
the principal investigator and describing the purpose and na-
tidimensional assessment of hot flashes in breast cancer sur-
ture of the study. The purpose of the study as stated in the
vivors and compare the experience of hot flashes in breast
letter was "to learn about hot flashes in breast cancer survivors
CARPENTER ­ VOL 29, NO 3, 2002
and how they compare to hot flashes experienced by women
Hot flash questionnaire: The hot flash questionnaire was
without cancer." To avoid response bias (i.e., only women
designed to provide detailed information about current hot
with hot flashes agreeing to participate), the letter also stated
flashes. Questions were modeled on previous work by
the following in bold typeface: "We are interested in receiv-
Kronenberg (1994) and assessed hot flash severity, bother,
ing information from you even if you do not have hot
quality, aggravating factors, alleviating factors, and temporal
flashes." The letter included a consent form for women to sign
pattern. Current overall severity and bother (distress) were
and return to the principal investigator indicating their inter-
rated using two separate 0 (not at all) to 10 (extremely) nu-
est. Those who did not return a signed consent received a sec-
meric rating scales. Current hot flash quality was assessed
ond mailing six weeks later.
using a list of 19 separate descriptors. Women were asked to
Study materials were mailed to all breast cancer survivors
check all phrases that coincided with feelings they encoun-
who returned signed consent forms. Instructions for complet-
tered during a hot flash. Additional items also asked women
ing the materials were detailed in a cover letter. Women were
to describe whether and where they perspired during hot
asked to return their completed study materials in a postage-
flashes. Aggravating factors were assessed using a list of
paid envelope provided. Women who did not return com-
seven potential hot flash triggers; women were asked to check
pleted questionnaires received a telephone call about two to
all items they felt triggered their hot flashes. Alleviating fac-
four weeks following the mailing to assess their interest in
tors were presented as a list of five behaviors that women
participating. Women who completed and returned study
might perform or use to relieve their hot flashes, and 10 po-
materials received a thank you letter and $5 for their partici-
tential hot flash treatment strategies, including HRT, prescrip-
pation. Study staff reviewed medical records of all breast can-
tion medications, vitamins, herbs, diet, exercise, behavioral
cer survivor participants to obtain accurate information about
methods (e.g., relaxation), acupuncture, massage, and "other,"
cancer diagnosis and treatment. Permission to abstract medi-
were listed. Women indicated whether they currently were
cal records was included in the consent.
using each treatment (yes or no) and whether they had tried
Recruitment of healthy women began about three months
each treatment in the past (yes or no with no time frame speci-
after the initial invitation letter was sent to the breast cancer
fied). In addition, women were asked to rate the overall level
survivors. This time allowed for a large enough pool of breast
of effectiveness of any current treatment strategies using a 0
cancer survivors to be accrued so that age matching of healthy
(not at all effective) to 10 (extremely effective) numeric rat-
women could be accomplished. Comparison women were
ing scale. The temporal pattern of hot flashes was assessed by
recruited via word-of-mouth and newspaper advertisements.
asking women to indicate the time of day and season of the
Advertisements asked interested women to phone a research
year hot flashes were most frequent, severe, and bothersome.
project office. Women who called the office were screened to
In addition, women were asked whether their hot flashes had
determine eligibility. If eligible, they were mailed a packet of
increased, decreased, or stayed the same in frequency, sever-
study materials, including a consent form, to sign and return
ity, bother, and duration during the prior six months. Re-
along with their completed questionnaires. Healthy women
sponses were intended to be analyzed as individual items
nonresponders were contacted in the same manner as breast
rather than as a summed score. Thus, Cronbach's alpha coef-
cancer survivor nonresponders.
ficient was not assessed. Given the changing nature of hot
flashes, test-retest reliability also was not assessed, as it was
expected to be low. The questionnaire was pilot-tested with
five participants to assess whether questions were understand-
Seven measures were included in the mailed study packets.
able and contained appropriate descriptors.
Demographic and disease and treatment information: A
Hot flash diary: The hot flash diary allowed for a detailed,
form was used to assess demographic information, including
prospective assessment of hot flashes occurring over a 48-
birth date, race, education, marital status, employment status,
hour period. Women were asked to record the date and time
and household income. Women also were asked to record
of each hot flash, rate the severity using a 0 (not at all) to 10
their height and weight, which was used to calculate body
(extremely) numeric rating scale, and record the duration in
mass index. Routine information regarding date of diagnosis,
minutes. Women were asked to complete the diaries around
stage of disease, and types and dates of cancer treatments was
the same time they completed the hot flash questionnaire. Pre-
obtained from medical records.
and perimenopausal women were not instructed to complete
Gynecologic and reproductive history form: This ques-
the diaries at a particular time of their menstrual cycle. Hot
tionnaire was designed to obtain information regarding meno-
flash frequencies were calculated as the number of hot flashes
pausal status, date of last menstrual period, and dates of any
recorded per day. Mean severity ratings were calculated by
hysterectomy or oophorectomy. This form included questions
summing individual severity ratings and dividing the sum by
adapted from the Massachusetts Women's Health study that
the number of hot flashes experienced. Mean duration ratings
were designed to assess menopausal status using the follow-
were calculated similarly. Frequency counts and mean sever-
ing definitions: premenopausal--regular menstrual cycles
ity and duration ratings were calculated separately for days
during the last three months, perimenopausal--3­11 months
one and two. This type of diary methodology has been used
of amenorrhea or increased menstrual irregularity if still cy-
previously (Carpenter, Andrykowski, Freedman, & Munn,
cling, and postmenopausal--12 or more months of amenor-
1999; Barton et al., 1998; Goldberg et al., 1994) and is con-
rhea. Women were classified as chemically menopausal if
sidered the gold standard for assessing hot flash frequency and
their last menstrual period occurred during the months chemo-
severity when objective sternal skin conductance monitoring
therapy treatments were received or occurred less than or
is not feasible (Carpenter et al., 1999).
equal to three months after the last chemotherapy treatment
Profile of Mood States-Short Form (POMS-SF): The
(Avis, Brambilla, McKinlay, & Vass, 1994; Brambilla,
POMS-SF is a measure of current (i.e., during the past two
McKinlay, & Johannes, 1994).
ONF ­ VOL 29, NO 3, 2002
weeks) mood disturbance consisting of 37 items (Shacham,
signed consent form indicating their willingness to participate
1983) from the original 65-item POMS (McNair, Lorr, &
(46%). The remaining 112 women (54%) did not respond to
Droppelman, 1981). A total mood disturbance score (TMD)
either invitation to participate. Completed packets were re-
is computed, as are scores for six subscales (depression, ten-
ceived from 71 of the 95 consenting women (75% of consent-
sion, anger, confusion, vigor, and fatigue). Among patients
ing breast cancer survivors; 34% of women who received the
with cancer, the POMS-SF possesses reliability and validity
mailing). However, upon reviewing completed packets, the
equal to that of the full-length POMS (Curran, Andrykowski,
investigators found two women to be ineligible. Therefore, the
& Studts, 1995). In this study, Cronbach's alphas were greater
final number of eligible breast cancer survivor responders was
than 0.80 for each of the six subscales and equal to 0.94 for
69, representing 73% of the consenting women and 33% of
the 207 women contacted. To ensure that a representative
Positive and Negative Affect Scale (PANAS): The
sample was gathered despite the lower than anticipated re-
PANAS is a 20-item, adjective list of feelings and emotions
sponse rate, the investigators compared the 69 breast cancer
that yields subscale scores for positive and negative affect
survivor responders to the 112 breast cancer survivors who
(Watson, Clark, & Tellegen, 1998). Participants respond to
did not respond to either invitation to participate on the only
each item from 1 (very slightly or not at all) to 5 (extremely)
variables available for comparison. No group differences were
to indicate how they have been feeling during the past week.
found for age at breast cancer diagnosis (p = 0.64) or year of
The PANAS has been used in prior research to assess the re-
diagnosis (p = 0.65). Thus, the breast cancer survivor sample
lationship between negative affect and symptom reporting in
appeared representative of the larger registry population.
patients with cancer (Koller et al., 1996). In this study,
Comparisons on stage at diagnosis and treatment were not fea-
Cronbach's alpha coefficient was 0.90 for positive affect and
sible because of the lack of registry information for 24% of
0.89 for negative affect.
the nonresponders. A total of 83 packets were mailed to inter-
ested healthy women. Completed materials were returned by
Hot Flash-Related Daily Interference Scale (HFRDIS):
The HFRDIS is a 10-item scale measuring the degree to which
63 healthy women (76%).
hot flashes interfere with nine daily activities. The tenth item
Participants were primarily Caucasian, postmenopausal,
measures the degree that hot flashes interfere with overall
middle-aged, of moderate educational level, and with mixed
QOL. The HFRDIS was modeled on the Brief Pain Inventory
incomes (see Table 1). Breast cancer survivors were a mean
(Daut, Cleeland, & Flanery, 1983) and the Fatigue Symptom
of slightly more than three years postdiagnosis and treatment.
Inventory (Hann, Winter, & Jacobsen, 1999), both of which
Stage at diagnosis was 0­I (59%), IIA­IIB (38%), and IIIA
invite respondents to rate the degree to which pain or fatigue
(3%). Breast cancer treatments received included surgery
interferes with various daily activities and overall enjoyment
(28%), surgery plus radiation therapy (25%), surgery plus
or QOL. The HFRDIS was developed to include daily life ac-
chemotherapy (29%), and surgery, radiation, and chemo-
tivities specific to the impact of hot flashes. Psychometric
therapy (19%). In addition, 50% of breast cancer survivors
analysis supports validity and reliability of the HFRDIS for
were using tamoxifen at the time of the survey.
use with breast cancer survivors and healthy women with a
Breast cancer survivors and healthy women groups were
Cronbach's alpha of 0.96, strong correlations with other hot
not significantly different in age, race, employment status,
flash variables, and demonstrated sensitivity to change over
income, education, or body mass index, but significantly more
time (Carpenter, 2001). Participants rate the degree hot flashes
breast cancer survivors were married (see Table 1). Marital
have interfered with each of these items during the previous
status was not related significantly to any of the primary out-
week using a 0 (do not interfere) to 10 (completely interfere)
come variables and, therefore, was not included as a covariate
scale. A total score is computed by summing items. Higher
in any of the subsequent analyses. As expected, groups dif-
scores indicate higher interference by hot flashes and, thus,
fered in menopausal status, with 79% of breast cancer survi-
greater impact on QOL. Women without hot flashes are asked
vors being postmenopausal as compared to only 57% of age-
to mark 0 for each item.
matched healthy women. In addition, given inclusion and
exclusion criteria, all healthy women were naturally meno-
Data Analysis
pausal, whereas breast cancer survivors were naturally, surgi-
cally, or chemically postmenopausal.
Demographic, disease, and treatment data were analyzed
using frequencies and descriptive statistics. Potential differ-
Detailed Assessment of Hot Flashes in
ences between women responding to the survey and those not
Breast Cancer Survivors With Comparison
responding to the survey were assessed using chi-square tests
to Healthy Women
or t-tests. Group differences in hot flashes and other outcome
Frequency, severity, bother, and duration: Using hot
variables also were assessed using chi-square tests or t-tests.
flash questionnaire data, overall severity and bother ratings
Correlations were assessed using Pearson correlation coeffi-
were strongly correlated among breast cancer survivors (r =
0.81, p < 0.001) and healthy women (r = 0.90, p < 0.001).
Significantly higher hot flash severity and bother were noted
among breast cancer survivors in comparison to healthy
women (see Table 2). Mean breast cancer survivors' severity
and bother ratings were about three times that of healthy
A total of 226 eligible women were identified through the
women. Using frequency distributions, 73% of healthy
cancer registry and sent a letter of invitation for the study
women provided overall severity and bother ratings of 0,
between January and March 1999. Of these 226 women, 2
whereas 48% of breast cancer survivors provided severity
were deceased and 17 had moved with no forwarding address.
ratings greater than or equal to 5, and 46% of breast cancer
Of the 207 women who received the mailing, 95 returned a
CARPENTER ­ VOL 29, NO 3, 2002
Table 1. Sample Characteristics
(22 breast cancer survivors versus 35 healthy women). These
naturally postmenopausal groups were not significantly differ-
ent in age, employment status, income, education, or body
mass index (p > 0.25). However, groups differed significantly
in race and marital status (p < 0.05), with more breast cancer
(n = 63)
(n = 69)
survivors being Caucasian and married. However, because
neither marital status nor race was significantly related to any
of the primary outcome variables, neither was included as a
covariate. Table 3 presents group differences in hot flash vari-
ables among the subset of naturally postmenopausal women.
Marital status
With the exception of frequency on diary day two, hot flashes
were significantly more frequent, severe, bothersome, and of
greater duration among naturally postmenopausal breast can-
cer survivors in comparison to naturally postmenopausal
healthy women (p < 0.05).
Quality: Breast cancer survivors used a mean of 5.22
phrases (SD = 2.85) to describe what they felt during a hot
flash from the list of 19 phrases provided. This number was
not significantly different from healthy women (p = 0.96).
Twenty-seven percent of breast cancer survivors used 3 or
Menopausal status
fewer descriptors, 50% used 4­6 descriptors, 18% used 7­10
Naturally pre-
descriptors, and 5% used more than 10 descriptors. Percent-
ages of breast cancer survivors endorsing each descriptor were
Naturally peri-
as follows: 93% heat, 89% sweating or perspiring, 77%
flushed, 46% clammy, 32% irritated or annoyed, 27% chills,
Naturally post-
27% burning sensation, 23% anxious, 18% frustrated, 16%
Surgically post-
pressure in head, 16% embarrassed, 11% depressed, 11%
change in heart rate, 9% ill or nauseous, 9% feeling of suffo-
Chemically post-
cation, 7% pressure in chest, 7% panicked, and 7% change in
breathing rate. No responder reported feeling suicidal. These
percentages were not significantly different from the healthy
Household income
women group (p > 0.12), indicating that the quality of hot
< $20,000
flashes was similar between groups. Additional items regarding
> $60,000
Do not care to
Table 2. Overall Group Differences in Hot Flashes
Age at interview
57.23 11.00
55.69 17.41
(n = 63)
(n = 69)
Years of education
14.87 13.39
Body mass index
27.03 16.86
Overall hot flash
Age at diagnosis
53.96 11.20
Overall hot flash
Months post-
39.25 21.62
Diary day 1
Months postsurgery
38.69 21.83
Months post-treat-
35.88 21.45
Diary day 2
survivors provided bother ratings greater than or equal to 5.
Diary data supported hot flash questionnaire data. Using dia-
ries, significantly more breast cancer survivors reported daily
hot flashes (65%) in comparison to healthy women (16%) (p
Number of hot flashes per day
< 0.001). Breast cancer survivors also reported significantly
Mean severity of all hot flashes experienced in a given day
more frequent and severe hot flashes with greater duration on
Mean number of minutes of each hot flash experienced in a
diary days one and two in comparison to healthy women.
given day
Group differences in hot flashes also were analyzed using
Note. Overall severity and bother were rated using 0­10 nu-
meric rating scales.
only the subset of women who naturally were postmenopausal
ONF ­ VOL 29, NO 3, 2002
Table 3. Group Differences in Hot Flashes Among the
= 1.75) in the past. The mean number of treatments used cur-
Subset of Naturally Postmenopausal Participants
rently or in the past was not significantly different from
healthy women (p > 0.50). The use of only two types of treat-
ments varied significantly between the groups. HRT was be-
ing used by significantly fewer breast cancer survivors at the
time of the survey (4% breast cancer survivors versus 27%
(n = 35)
(n = 22)
healthy women, p < 0.001) but had been used by significantly
more breast cancer survivors in the past (51% breast cancer
survivors versus 34% healthy women, p < 0.05). In addition,
Overall hot flash
significantly more breast cancer survivors had tried nonhor-
monal prescription medications (e.g., clonidine) in the past
(15% breast cancer survivors versus 3% healthy women, p <
Overall hot flash
0.05). Additional strategies used by breast cancer survivors at
the time of the survey included exercise (57%), vitamins
Diary day 1
(56%), diet (30%), behavioral methods such as relaxation
(24%), herbs (9%), massage (6%), and other (unspecified)
(11%). None of the breast cancer survivors or healthy women
reported using acupuncture for hot flashes. Interestingly, de-
Diary day 2
spite the similarity in the types and numbers of current treat-
ments (with the exception of HRT), breast cancer survivors
reported their current hot flash treatments to be significantly
less effective than did healthy women. Using a 0­10 numeric
rating scale, breast cancer survivors reported moderate effec-
Number of hot flashes per day
tiveness (X = 4.71, SD = 3.30) whereas healthy women re-
Mean severity of all hot flashes experienced in a given day
ported extreme effectiveness (X = 9.76, SD = 1.55) (p < 0.05).
Mean number of minutes of each hot flash experienced in a
Temporal pattern: As shown in Table 4, no differences
given day
in the temporal pattern of hot flashes were noted between
Note. Overall severity and bother were rated using 0­10 nu-
breast cancer survivors and healthy women (p > 0.05). The
meric rating scales.
majority of breast cancer survivors reported no particular
time of day that hot flashes were most frequent or severe,
perspiration/sweating revealed that 51% of breast cancer sur-
whereas nighttime was endorsed as the time of day hot
vivors reported always perspiring during hot flashes, 42%
flashes were most bothersome. Healthy women endorsed
reported sometimes perspiring, and 7% reported never per-
nighttime as the time of day hot flashes were most frequent,
spiring with hot flashes. These findings also were not signifi-
severe, and bothersome. The majority of breast cancer sur-
cantly different from healthy women (p = 0.53). For the breast
vivors and healthy women also stated that there was no par-
cancer survivors who reported perspiration, 34% stated their
ticular season of the year when hot flashes were most fre-
whole upper and lower body perspired during hot flashes,
quent, severe, or bothersome. In addition, when asked how
34% upper body only, 10% face only, and 22% had varied
hot flashes had changed over the previous six months, no
sites of perspiration.
group differences were seen for frequency, severity, bother,
Aggravating factors: Breast cancer survivors endorsed a
or duration. The majority of women reported no changes in
mean of 2.20 aggravating factors (SD = 1.19) from a list of 7
frequency (57% breast cancer survivors versus 59% healthy
potential hot flash triggers. This number was not significantly
women), severity (67% breast cancer survivors versus 65%
different from healthy women (p = 0.58). The majority of
healthy women), bother (64% breast cancer survivors versus
breast cancer survivors (71%) endorsed one to two triggers,
71% healthy women), or duration (68% breast cancer survi-
23% endorsed three to four triggers, and 6% endorsed five to
vors versus 77% healthy women). None of the differences
six triggers. The percentages of breast cancer survivors en-
were statistically significant.
dorsing each trigger were as follows: 59% no trigger (i.e., hot
flashes just seem to happen), 52% stressful or emotional situ-
Impact on Mood, Affect, Daily Activities,
ations, 46% external heat sources (e.g., stove, sunshine, hot
and Overall Quality of Life
room), 23% alcoholic beverages, 18% confined spaces, 14%
Breast cancer survivors and healthy women were compared
caffeine, and 9% other (not specified). Only stressful or emo-
in terms of mood, affect, activity, and overall QOL (see Table
tional situations were endorsed more frequently by breast
5). Using the POMS-SF, no significant group differences in
cancer survivors (52%) than healthy women (12%) as a hot
mood scores were noted. However, mean scores indicated
flash trigger (p < 0.01).
higher mood disturbance on all POMS-SF subscales among
Alleviating factors: When hot flashes occurred, breast
breast cancer survivors in comparison to healthy women.
cancer survivors reported doing significantly more things to
Using the PANAS, breast cancer survivors reported signifi-
alleviate their hot flashes in comparison to healthy women (p
cantly higher negative affect in comparison to healthy women
< 0.05). Breast cancer survivors reported fanning themselves
(p < 0.05). In addition, breast cancer survivors reported that
(71%), removing clothing to cool off (64%), moving to a
hot flashes had a significantly greater interference with daily
cooler environment (64%), doing nothing (21%), and doing
activities and overall QOL (p < 0.01). Four activities on the
other unspecified behaviors (23%). Breast cancer survivors
HFRDIS appeared to be particularly affected by hot flashes.
reported using a mean of 2.04 hot flash treatments (SD = 1.66)
Breast cancer survivors who reported daily hot flashes reported
at the time of the survey and a mean of 2.23 treatments (SD
CARPENTER ­ VOL 29, NO 3, 2002
Table 4. Temporal Pattern of Daily Hot Flashes in Breast Cancer Survivors and Healthy Women
Breast Cancer Survivors (n = 45)
Healthy Women (n = 17)
Most Frequent
Most Severe
Most Bothersome
Most Frequent
Most Severe
Most Bothersome
Time of day
Season of the year
menopausal healthy women have not been reported previ-
moderate to severe interference (ratings of 5 or greater) with
ously. Our data suggest that the natural menopausal experi-
sleep (40%), concentration (33%), mood (29%), and sexual-
ence of breast cancer survivors is significantly different than
ity (28%).
the natural menopausal experience of healthy women, pre-
Mood, affect, activity, and overall QOL also were com-
sumably because of the effects of chemotherapy and tamox-
pared between 36 breast cancer survivors with none to mild
ifen and inability to take HRT. Thus, healthcare providers in
hot flashes (severity less than 5) and 33 breast cancer survi-
clinical practice must assess all breast cancer survivors, not
vors reporting moderate to severe hot flashes (severity of 5 or
just those experiencing a chemotherapy-induced menopause.
greater). Significant group differences were seen on the
These results support the assertion that breast cancer survi-
POMS-SF, PANAS, and HFRDIS, with more severe hot
vors represent a unique population whose hot flash experi-
flashes being associated with greater mood disturbance, more
ence differs from an age-matched group of healthy women
negative affect, greater interference with daily activities, and
poorer overall QOL (see Table 6).
Table 5. Group Differences in Mood, Affect, Daily
Activities, and Overall Quality of Life
To the best of the researchers' knowledge, this study pro-
vides the most comprehensive assessment of the symptom ex-
perience of hot flashes in breast cancer survivors available in
the literature. In addition, this study is the only one to include
(n = 63)
(n = 69)
a comparison group of healthy women. The healthy women
sample in this study was limited to naturally pre-, peri-, or
postmenopausal women to provide a consistent baseline for
Profile of Mood
comparison. The investigators were interested in comparing
States-Short Form
breast cancer survivors to other naturally postmenopausal
women of the same age because the majority of menopausal
­ 0.41
survey research, and thus knowledge used in clinical practice,
is limited to naturally postmenopausal women (Dennerstein et
al., 1993; Koster, 1991; Kronenberg, 1990, 1994).
12.28 25.46
13.43 2 5.49
Overall, the study findings suggest that hot flashes are a
14.34 23.93
13.10 23.35
significant symptom management problem in women fol-
lowing treatment for breast cancer. In comparison to natu-
Total mood dis-
39.78 26.10
33.21 22.39
rally menopausal healthy women of the same age, breast
cancer survivors experienced hot flashes that were signifi-
Positive and Nega-
cantly more frequent, severe, bothersome, and of greater
tive Affect Scale
duration. Group differences could not be attributed to differ-
Positive affect
32.28 28.60
34.73 2 6.95
ences in menopausal status alone because they also occurred
Negative affect
18.41 28.22
15.53 2 5.65
when comparing the subset of naturally postmenopausal
Hot Flash-Related
breast cancer survivors and healthy women, all with intact
Daily Interference
uterus and ovaries. Although women undergoing artificially
induced menopause have been noted previously to be more
Activity items
10.51 2 1.56
­4.35 0.001
12.07 22.53
symptomatic than those undergoing a natural menopause
Overall quality-of- 11.58 22.31
10.46 21.69
­3.22 0.002
(Kronenberg, 1990, 1994), differences between naturally
life item
postmenopausal breast cancer survivors and naturally post-
ONF ­ VOL 29, NO 3, 2002
ment strategies, with the exception of HRT. As expected,
Table 6. Differences in Mood, Affect, Daily Activities,
HRT was used by significantly more healthy women than
and Overall Quality of Life by Hot Flash Severity Among
breast cancer survivors.
Breast Cancer Survivors
The finding that 4% of breast cancer survivors were using
HRT is consistent with previous reports that less than 5% of
Moderate to
None to
Mildly Severea
breast cancer survivors take HRT following diagnosis (Swain,
(n = 33)
(n = 36)
Santen, Burger, & Pritchard, 1999). Interestingly, breast cancer
survivors rated the overall effectiveness of their treatment strat-
egies to be about half as effective as did the healthy women.
Thus, these findings suggest that currently available nonhor-
Profile of Mood
monal hot flash treatments may be ineffective, underutilized, or
States-Short Form
not acceptable to breast cancer survivors. In particular, although
five times as many breast cancer survivors reported using non-
hormonal prescription medications (e.g., clonidine) in the past
compared to healthy women, few breast cancer survivors were
using these medications at the time of the survey despite fre-
13.20 14.60
11.20 26.22
­1.49 0.142
quent, severe, and bothersome hot flashes. This finding implies
23.11 12.90
25.77 24.52
­2.86 0.006
that although women may have tried these medications in the
past, they may have discontinued the medications for reasons
Total mood dis-
29.09 16.46
54.76 29.83
­3.90 0.001
such as failure to obtain relief, cost, or side effects. Although
many prescription medications are effective in reducing hot
Positive and Nega-
flashes, they can be associated with significant side effects
tive Affect Scale
(Goldberg et al., 1994; Quella et al., 1998). Further research is
Positive affect
33.83 16.87
30.41 10.12
­ 1.55 0.129
needed to understand why these options may not be acceptable
Negative affect  14.77 14.22
22.27 29.62
­4.12 0.001
to breast cancer survivors. In addition, exercise, vitamins, and
Hot Flash-Related
diet were cited as the three treatments used most commonly by
Daily Interference
breast cancer survivors despite a lack of empirical evidence
supporting their effectiveness (Barton et al., 1998; Carpenter,
Activity items
20.53 11.19
23.64 22.47
­6.56 0.001
2000; Ivarsson, Spetz, & Hammar, 1998; Mayer & Linscott,
Overall quality-
20.44 11.13
22.73 22.60
­4.65 0.001
1995; Miller, 1992). These findings support the importance of
of-life item
thoroughly assessing the types of strategies women are using to
alleviate their hot flashes and effectively guiding breast cancer
None to mildly severe hot flashes are those women with overall
severity ratings < 5.
survivors toward additional interventions when previous strat-
M oderate to severe hot flashes are those women with overall
egies have proven ineffective.
severity ratings > 5.
Interestingly, the quality, aggravating factors, and tempo-
ral pattern of hot flashes did not differ between groups. Both
breast cancer survivors and healthy women reported a simi-
undergoing a natural menopause. Although hot flashes in
lar experience in terms of the number of sensations endorsed
healthy women typically are considered a "natural" part of
as occurring during a hot flash and the associated perspiration,
the normal aging process (Matthews, 1992), the data suggest
the number and types of factors aggravating hot flashes, and
that hot flashes are an "unnatural" symptom in need of inter-
changes in hot flashes regarding time of day, season of the
vention among breast cancer survivors.
year, and occurrence over the last six months. Data presented
Data suggesting that 65% of breast cancer survivors expe-
in this study can be used to educate breast cancer survivors
rience hot flashes, with the majority of women reporting the
regarding what to expect in terms of the hot flash symptom
symptom as severe, are consistent with previous work (Car-
penter et al., 1998; Couzi et al., 1995). In a survey of 190
Several items supported the unpredictable nature of hot
women with breast cancer ages 40­65, 65% reported experi-
flashes. Over half of the breast cancer survivor group stated that
encing hot flashes during the prior month (Couzi et al.). Se-
hot flashes "just seem to happen" with no triggering event. With
verity ratings from those reporting hot flashes were 29% mild,
the exception of hot flashes being most bothersome at night, a
37% moderate, and 34% severe. In another survey, 65% of
majority of breast cancer survivors reported no particular time
114 postmenopausal women with breast cancer ages 36­83
of day or season of the year when hot flashes were most prob-
reported experiencing hot flashes during the previous two
lematic. Time of day data support previous research results.
weeks (Carpenter et al., 1998). Severity ratings for those re-
When 21 breast cancer survivors were monitored using objec-
porting hot flashes were 23% moderately severe, 28% quite
tive hot flash measurement methods, hot flashes did not peak in
severe, and 31% extremely severe. These findings suggest that
frequency at any particular time of day among breast cancer
clinicians should question all breast cancer survivors regard-
survivors and instead occurred fairly regularly throughout a 24-
ing hot flashes, as the majority of survivors will experience
hour period (Carpenter et al., 2001). Again, these data can be
this symptom following diagnosis.
used to educate breast cancer survivors so that they may antici-
Frequency, severity, bother, and duration ratings of hot
pate the unpredictable nature of hot flashes.
flashes are of particular importance given the additional re-
Although previous research using other measures of QOL
sults regarding use of hot flash treatments. In terms of current
showed only a marginally negative association with hot
hot flash treatments, breast cancer survivors and healthy
flashes (Carpenter et al., 1998), results from this study clearly
women endorsed using similar types and numbers of treat-
CARPENTER ­ VOL 29, NO 3, 2002
suggest hot flashes have a negative impact on mood, affect,
women enrolled and the limited resources. Because previous
daily activities, and overall QOL. Findings concerning con-
research has shown that breast cancer survivors subjectively
centration are consistent with research showing decreased
tend to underestimate hot flash frequency when compared to
cerebral blood flow during hot flashes in healthy women
objective assessment methods (Carpenter et al., 1999), hot
(Greene, 2000). Additional research is needed to describe the
flash frequency data reported in this study may underesti-
types of cognitive changes that women with hot flashes may
mate the number of hot flashes experienced each day by
experience. In addition, findings concerning negative affect
breast cancer survivors.
are consistent with previous research. In a study of 60 surgi-
The use of self-reported menopausal status should not be
cal patients with cancer, somatic symptoms were correlated
viewed as a limitation of this research. Although serum follicle-
significantly with negative affect using a German translation
stimulating hormone (FSH) levels with or without estradiol fre-
of the PANAS (r = 0.65, p < 0.01) (Koller et al., 1996). Indi-
quently are used as a clinical indicator of menopause, data sug-
viduals with higher negative affect (e.g., anger, depression)
gest that these measures may not be reliable among healthy
were more introspective, apprehensive, negativistic, and vigi-
women (Burger, 1994; Stellato, Crawford, McKinlay, &
lant and, thus, may be more likely to attend to and report
Longcope, 1998) or breast cancer survivors (Kostoglou-
physical symptoms (Watson & Pennebaker, 1989). Alterna-
Athanassiou et al., 1995). FSH does not reliably distinguish
tively, unrelieved physical symptoms, such as hot flashes,
among pre-, peri-, or postmenopausal healthy women (Burger;
may result in more frustration and negative affect. In either
Stellato et al.), and FSH significantly is increased with
case, findings suggest that women with unrelieved hot flashes
tamoxifen use (Kostoglou-Athanassiou et al.). In addition, es-
suffer negative psychosocial consequences, and interventions
tradiol levels were 239% higher after two years of tamoxifen
that alleviate hot flashes also may improve mood; affect; daily
therapy in comparison to pretamoxifen baseline levels (p <
activities, including sleep, concentration, and sexuality; and
0.05) (Lum, Woltering, Fletcher, & Pommier, 1997). Thus,
overall QOL. Thus, incorporating these outcomes in future
because FSH, with or without estradiol, is not likely to be a
studies will be important. In addition, assessing hot flashes in
reliable indicator of menopausal status among breast cancer
women who present in clinical practice with mood distur-
survivors, the use of self-reported menopausal status is reason-
bances or problems with sleep, concentration, or sexuality is
able. Furthermore, definitions of pre-, peri-, and postmeno-
important because alleviating hot flashes may help to improve
pausal used in this study are being used in large-scale epidemio-
these other outcomes.
logic research (Avis et al., 1994; Brambilla et al., 1994).
Findings from this study should be considered in light of
In summary, this report has provided a detailed description
limitations. First, although the investigators attempted to
of hot flashes in breast cancer survivors with comparison to
sample the population of breast cancer survivors identified
healthy women. Hot flashes appeared to be significantly more
within their cancer registry, the response rate (33%) was
problematic in breast cancer survivors in comparison to
low. However, responders appeared representative of the
healthy women of the same age. In addition, hot flashes ap-
larger pool of breast cancer survivors in terms of age at di-
peared to be associated with disruptions in mood; affect; daily
agnosis and year of diagnosis. Second, healthy women were
activities, including sleep, concentration, and sexuality; and
recruited from the community-at-large and, thus, do not rep-
overall QOL. Findings support the need for a comprehensive
resent a true case-control population. Therefore, these results
assessment of the hot flash experience and additional inter-
should be generalized with caution. A third limitation was
vention research to alleviate this symptom.
the use of self-report data only. Although objective measure-
ment of hot flashes has been shown to be valid and reliable
Author Contact: Janet S. Carpenter, PhD, RN, can be reached at
(Carpenter et al., 2001, 1999; Freedman, 1989), this method
janet.s.carpenter@vanderbilt.edu, with copy to editor at rose_
was not feasible to use in this study given the number of
Avis, N.E., Brambilla, D., McKinlay, S.M., & Vass, K. (1994). A longitu-
Carpenter, J.S. (2000). Hot flashes and their management in breast cancer.
Seminars in Oncology Nursing, 16, 214­225.
dinal analysis of the association between menopause and depression: Re-
sults of the Massachusetts women's health study. Annals of Epidemiol-
Carpenter, J.S. (2001). The hot flash-related daily interference scale: A tool
ogy, 4, 214­220.
for assessing the impact of hot flashes on quality of life following breast
cancer. Journal of Pain and Symptom Management, 22, 979­989.
Barton, D.L., Loprinzi, C.L., Quella, S.K., Sloan, J.A., Veeder, M.H.,
Egner, J.R., et al. (1998). Prospective evaluation of vitamin E for hot
Carpenter, J.S., & Andrykowski, M.A. (1999). Menopausal symptoms in
flashes in breast cancer survivors. Journal of Clinical Oncology, 16,
breast cancer survivors. Oncology Nursing Forum, 26, 1311­1317.
Carpenter, J.S., Andrykowski, M.A., Freedman, R.R., & Munn, R. (1999).
Bartsch, C., Bartsch, H., Fuchs, U., Lippert, T.H., Bellman, O., & Gutpa, D.
Feasibility and psychometrics of an ambulatory hot flash monitoring de-
vice. Menopause: The Journal of the American Menopause Society, 6,
(1989). Stage-dependent depression of melatonin in patients with primary
breast cancer: Correlation with prolactin, thyroid stimulating hormone,
and steroid receptors. Cancer, 64, 426­433.
Carpenter, J.S., Andrykowski, M.A., Studts, J.L., Cordova, M.J., Cun-
Brambilla, D.J., McKinlay, S.M., & Johannes, C.B. (1994). Defining the
ningham, L., McGrath, P.C., et al. (1998). Hot flashes in postmeno-
perimenopause for application in epidemiologic investigations. American
pausal women treated for breast cancer: Prevalence, severity, corre-
Journal of Epidemiology, 140, 1091­1095.
l a t e s , management, and relation to quality of life. Cancer, 82,
Brzezinski, A. (1995). Management of menopause in women with breast
cancer. Israel Journal of Medical Science, 3, 163­168.
Carpenter, J.S., Gautam, S., Andrykowski, M., & Freedman, R.R. (2001).
Burger, H.G. (1994). The menopause: When it is all over or is it? Australian
Circadian rhythm of objectively recorded hot flashes in postmenopausal
New Zealand Journal of Obstetrics and Gynecology, 34, 293­295.
breast cancer survivors. Menopause, 8, 181­188.
ONF ­ VOL 29, NO 3, 2002
Matthews, K.A. (1992). Myths and realities of the menopause. Psychoso-
Couzi, R.J., Helzlsouer, K.J., & Fetting, J.H. (1995). Prevalence of meno-
matic Medicine, 54, 1­9.
pausal symptoms among women with a history of breast cancer and at-
titudes toward estrogen replacement therapy. Journal of Clinical Oncol-
Mayer, D.K., & Linscott, E. (1995). Information for women: Management
of menopausal symptoms. Oncology Nursing Forum, 22, 1567­1570.
ogy, 13, 2737­2744.
McNair, P.M., Lorr, M., & Droppelman, L. (1981). POMS manual (2nd ed.).
Curran, S.L., Andrykowski, M.A., & Studts, J.L. (1995). Short form of the
Profile of Mood States (POMS-SF): Psychometric information. Psycho-
San Diego: Educational and Testing Service.
logical Assessment, 7(1), 80­83.
Miller, K.L. (1992). Alternatives to estrogen for menopausal symptoms.
Clinical Obstetrics and Gynecology, 35, 884­893.
Daut, R.L., Cleeland, C.S., & Flanery, R.C. (1983). Development of the
Mormont, M.C., & Levi, F. (1997). Circadian-system alterations during cancer
Wisconsin Brief Pain Inventory to assess pain in cancer and other dis-
processes: A review. International Journal of Cancer, 70, 241­247.
eases. Pain, 17, 197­210.
Pasacreta, J.V., & McCorkle, R. (1998). Providing accurate information
Dennerstein, L., Smith, A.M., Morse, C., Burger, H., Green, A., Hopper, J.,
et al. (1993). Menopausal symptoms in Australian women. Medical Jour-
to women about tamoxifen therapy for breast cancer: Current indica-
tions, effects, and controversies. Oncology Nursing Forum, 25, 1577­
nal of Australia, 159, 232­236.
Dodd, M.J., Miaskowski, C., & Paul, S.M. (2001). Symptom clusters and
their effect on the functional status of patients with cancer. Oncology
Quella, S., Loprinzi, C.L., Sloan, J., Vaught, N., Dekrey, W., Fischer, T., et
Nursing Forum, 28, 465­470.
al. (1998). Long-term use of megestrol acetate by cancer survivors for the
treatment of hot flashes. Cancer, 82, 1784­1788.
Finck, G., Barton, D.L., Loprinzi, C.L., Quella, S.K., & Sloan, J.A. (1998).
Definitions of hot flashes in breast cancer survivors. Journal of Pain and
Reichman, B.S., & Green, K.B. (1994). Breast cancer in young women:
Symptom Management, 16, 327­333.
Effect of chemotherapy on ovarian function, fertility, and birth de-
fects. Journal of the National Cancer Institute Monographs, 16, 125­
Freedman, R.R. (1989). Laboratory and ambulatory monitoring or meno-
pausal hot flashes. Psychophysiology, 26, 573­579.
Runowicz, C.D. (1996). Hormone replacement therapy in cancer survivors:
Goldberg, R.M., Loprinzi, C.L., O'Fallon, J.R., Veeder, M.H., Miser, A.W.,
A con opinion. CA: Cancer Journal for Clinicians, 46, 365­373.
Maillard, J.A., et al. (1994). Transdermal clonidine for ameliorating
tamoxifen-induced hot flashes. Journal of Clinical Oncology, 12, 155­158.
Shacham, S. (1983). A shortened version of the Profile of Mood States.
Journal of Personality Assessment, 47, 305­306.
Greene, R.A. (2000). Cerebral blood flow. Fertility and Sterility, 73, 143.
Stellato, R.K., Crawford, S.L., McKinlay, S.M., & Longcope, C. (1998).
Hann, D., Winter, K., & Jacobsen, P. (1999). Measurement of depressive
Can follicle-stimulating hormone be used to define menopausal status?
symptoms in cancer patients: Evaluation of the Center for Epidemiologi-
Endocrine Practice, 4, 137­141.
cal Studies Depression Scale (CES-D). Journal of Psychosomatic Re-
search, 46, 437­443.
Swain, S., Santen, R., Burger, H., & Pritchard, K. (1999). Treatment of es-
trogen deficiency symptoms in women surviving breast cancer. Part 4:
Ivarsson, T., Spetz, A.C., & Hammar, M. (1998). Physical exercise and va-
somotor symptoms in postmenopausal women. Maturitas, 29, 139­146.
Urogenital atrophy, vasomotor instability, sleep disorders, and related
symptoms. Oncology, 13, 551­575.
Koller, M., Kussman, J., Lorenz, W., Jenkins, M., Voss, M., Arens, E., et al.
University of California, San Francisco School of Nursing Symptom Man-
(1996). Symptom reporting in cancer patients: The role of negative affect
agement Faculty Group. (1994). A model for symptom management. Im-
and experienced social stigma. Cancer, 77, 983­995.
age: Journal of Nursing Scholarship, 26, 272­276.
Koster, A. (1991). Change-of-life anticipations, attitudes, and experiences
among middle-aged Danish women. Health Care for Women Interna-
Watson, D., Clark, L.A., & Tellegen, A. (1998). Development and valida-
tional, 12(1), 1­13.
tion of brief measures of positive and negative affect: The PANAS scales.
Journal of Personality and Social Psychology, 54, 1063­1070.
Kostoglou-Athanassiou, I., Ntalles, K., Gogas, J., Markopoulos, C.,
Watson, D., & Pennebaker, J.W. (1989). Health complaints, stress, and dis-
Alevizou-Terzaki, V., Athanassiou, P., et al. (1995). Sex hormones in
tress: Exploring the central role of negative affectivity. Psychological
postmenopausal women with breast cancer on tamoxifen. Hormone Re-
Review, 96, 234­254.
search, 47, 116­120.
Kronenberg, F. (1990). Hot flashes: Epidemiology and physiology. Annals
of the New York Academy of Sciences, 592, 52­86.
Kronenberg, F. (1994). Hot flashes: Phenomenology, quality of life, and
search for treatment options. Experimental Gerontology, 29, 319­336.
For more information . . .
Loprinzi, C.L., Michalak, J.C., Quella, S.K., O'Fallon, J.R., Hatfield, A.K.,
Nelimark, R.A., et al. (1994). Megestrol acetate for the prevention of hot
flashes. New England Journal of Medicine, 33, 347­352.
American Cancer Society­Breast Cancer
Loprinzi, C.L., Pisansky, T.M., Foneseca, R., Sloan, J.A., Zahasky, K.M.,
Quella, S.K., et al. (1998). Pilot evaluation of venlafaxine hydrochloride
National Alliance of Breast Cancer Organizations
for the therapy of hot flashes in cancer survivors. Journal of Clinical
Oncology, 16, 2377­2381.
Love, R.R. (1989). Tamoxifen therapy in primary breast cancer: Biology,
Living Beyond Breast Cancer
efficacy, and side effects. Journal of Clinical Oncology, 7, 803­815.
Love, R.R., Cameron, L., Connell, B.L., & Levanthal, H. (1991). Symptoms
associated with tamoxifen treatment in postmenopausal women. Archives
These Web sites are provided for information only. The hosts are
of Internal Medicine, 151, 1842­1847.
responsible for their own content and availability. Links can be
Lum, S.S., Woltering, E.A., Fletcher, W.S., & Pommier, R.F. (1997).
found using ONS Online at www.ons.org.
Changes in serum estrogen levels in women during tamoxifen therapy.
American Journal of Surgery, 173, 399­402.
CARPENTER ­ VOL 29, NO 3, 2002