Ij

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.
ONLINE EXCLUSIVE
Improving Cancer Screening Among Lesbians
Over 50: Results of a Pilot Study
Suzanne L. Dibble, RN, DNSc, and Stephanie A. Roberts, MD
Key Points . . .
Purpose/Objectives: To explore the impact of two one-hour lesbian-
specific educational interventions by a lesbian physician on the cancer
screening behaviors of lesbians.
The risk profile of lesbians for developing cancer is worse
Design: A pilot pre- and post-test intervention study.
than for heterosexual women.
Setting: Two lesbian, gay, bisexual, and transgendered senior orga-
nizations in the San Francisco Bay Area (one urban, one suburban).
They may not be known as such, but lesbians probably are at
--
Sample: 36 participants aged 5081 (X = 60.2, SD = 6.48). The ma-
most work sites either as coworkers or patients.
jority were Caucasian (86%), single (61%), living in urban areas (67%),
Many screening programs do not have appropriate materials
--
employed (56%), and educated beyond high school (X = 15.47 years, SD
or programs designed for lesbian audiences.
= 2.90, range 921). Eleven percent (n = 4) did not have any health in-
surance and were not on Medicaid or Medicare.
The hostility, fear, and discomfort experienced by lesbians in
Methods: A lesbian physician led a one-hour, didactic, lesbian-
the healthcare system should not be underestimated.
specific educational program on cancer screening, including a review of
current research findings with regard to lesbians' risk for cancer and 45
minutes of information on recommended cancer screening, followed by
a 15-minute question-and-answer period. Participants completed a pre-
to be higher among lesbians, as would be predicted by stud-
and postintervention survey.
ies suggesting that lesbians have increased risk for develop-
Findings: Follow-up data were available for 22 women. Of the six
ing cancer (Cochran et al., 2001; Dibble, Roberts, Robertson,
women (27%) who had not focused their attention on breast screening
behaviors for two years or more, one-third had obtained mammograms
& Paul, 2002; Roberts, Dibble, Scanlon, Paul, & Davids,
and half began performing monthly breast self-examinations. Of the four
1998).
women (18%) who had not undergone a pelvic examination for three
Research into the health concerns of sexual minorities is
years or more, one obtained a pelvic examination. The women reported
becoming more prevalent and sophisticated. Scientists are
no changes in colorectal cancer screening behaviors.
beginning to treat sexual orientation as a demographic vari-
Conclusions: Some of these difficult-to-reach women changed their
able similar to religion or ethnicity (Sell, 1997). In research
behavior in a very short period of time, supporting the need for a larger
studies, lesbian and bisexual women may be grouped together
study to confirm these findings.
because of overlapping sexual behaviors or separated at other
Implications for Nursing: A need exists to develop appropriate inter-
times when their sexual behaviors differ (Johnson, Smith, &
ventions for the underserved population of lesbians older than 50.
Guenther, 1994). Studies, however, hint that lesbian and bi-
sexual women may differ in more than sexual behaviors, such
as in their interaction with the healthcare system (Koh, 2000;
n 2003, 658,800 women are estimated to be diagnosed
I
Smith, Johnson, & Guenther, 1985), their degree of social iso-
with cancer and 270,600 will die from the disease (Jemal
et al., 2003). Approximately 211,300 women will be di-
agnosed with breast cancers, 77,000 will be diagnosed with
Suzanne L. Dibble, RN, DNSc, is a professor in the Lesbian Health
colorectal cancers, and 83,700 will be diagnosed with gyne-
Research Center of the Institute for Health and Aging in the School
cologic cancers (Jemal et al.). Some of the women included in
of Nursing at the University of California, San Francisco; and
these statistics are lesbians. The actual number of female
Stephanie A. Roberts, MD, is a doctor in private practice in Walnut
sexual minorities (lesbian, bisexual, or transgendered women)
Creek, CA. This study was funded by the San Francisco Bay Area
is unknown; thus, any attempt to report the distribution of
Community of Women's "A Fund of Our Own" and administered by
sexual orientation in women is subject to some bias and dis-
the Horizons Foundation and the Lesbian Health Research Center at
tortion (Solarz, 1999). In the National Health and Social Life
the University of California, San Francisco. (Submitted September
Survey, 6.2% of women reported same sex behavior or desire
2002. Accepted for publication January 25, 2003.) (Mention of spe-
cific products and opinions related to those products do not indicate
(Laumann, Gagnon, Michael, & Michaels, 1994). If this is
or imply endorsement by the Oncology Nursing Forum or the Oncol-
true, then an estimated 40,846 lesbians in the United States
ogy Nursing Society.)
will be diagnosed with cancer in 2003. These figures may be
conservative if the rates of some cancers eventually are found
Digital Object Identifier: 10.1188/03.ONF.E71-E79
ONCOLOGY NURSING FORUM VOL 30, NO 4, 2003
E71
lished group was assigned randomly to an experimental (breast
lation (Norman, Perry, Stevenson, Kelly, & Roffman, 1996),
and cervical cancer early-detection information and coaching)
and their cancer risk (Dibble, Roberts, Davids, Paul, &
or control group (another program). At the end of the study, the
Scanlon, 1999).
cancer intervention groups (n = 274) practiced more breast self-
The term "lesbian" describes "not only sexual orientation,
examination (BSE) (p < 0.001) and had significantly more
but also an identity based on psychological responses, cultural
mammograms (p = 0.029) than the control group.
values, societal expectations, and a woman's own choices in
The Witness Project is an intervention designed for African
identity formation" (White & Levinson, 1995, p. 463). Bi-
American women in Arkansas that provides culturally sensi-
sexual women have the potential for attraction to both men
tive messages promoting early detection from African Ameri-
and women; they are attracted to individuals rather than to a
can breast cancer survivors in churches and community orga-
person of a particular gender or biologic sex (Tucker, High-
nizations (Erwin, Spatz, Stotts, & Hollenberg, 1999). Of the
leyman, & Kaplan, 1995). Lesbians are diverse and represent
204 participants in the Witness Project, a significant increase
all religious, ethnic, economic, age, and cultural groups. Be-
was found in the practice of BSE (p < 0.0005) and mammog-
cause same-sex behavior is stigmatized and lesbians often
raphy (p < 0.005) pre- and postintervention using a McNemar
defy stereotypes, they may remain a hidden population in their
chi-square test. No comparable increase existed in the control
interactions with researchers and healthcare providers. The
group (n = 206) screening behaviors.
assumption of heterosexuality is so prevalent (Denenberg,
ENCOREplus, a national program from the YWCA de-
1995; Rankow, 1995) that healthcare providers and researchers
may perpetuate the invisibility of lesbians within the healthcare
signed to reach underserved women from all ethnicities, was
system.
tested in 27,494 women (Fernandez, DeBor, Candreia, Wagner,
Not only are lesbians often invisible within the healthcare
& Stewart, 1999). The program activities included outreach,
system, they also are less likely than heterosexual women to
education, enabling, support services, and provider networking
use preventive cancer-related screening services. A meta-
and linkage. Of the participants older than 40 who were
analysis of seven large surveys completed from 19871996 (N
nonadherent to breast cancer screening guidelines at baseline
= 11,876) demonstrated that lesbian and bisexual women were
(70%), 58% received mammograms in the six months follow-
less likely than heterosexual women to undergo routine
ing the intervention. Of the participants older than 18 who were
screening procedures such as mammograms and gynecologic
nonadherent to cervical cancer screening guidelines at baseline
examinations (Cochran et al., 2001). Additional surveys and
(69%), 37% received Pap tests in the six months following the
clinic sampling performed in subsequent years have con-
intervention. Another program designed to serve women from
firmed this finding (Koh, 2000; Lauver, Karon, & Egan, 1999;
all ethnicities used formal and informal meetings to disperse
Rankow & Tessaro, 1998a, 1998b; White & Dull, 1997).
written materials, show videos, and generally educate women,
Cancer-screening behaviors are important because early
all in their native language, about breast and cervical cancer and
detection of cancer often can result in a cure and detecting
associated screening guidelines (Kernohan, 1996). Some (37%)
precancerous conditions can result in preventing cancer in the
of the study sample attended breast cancer screening six months
first place. Two population-based studies suggest that the use
after the program intervention, and an additional 20% of the
of cancer-screening services also may vary within the female
women had cervical tests. On the other hand, a program that
sexual minority population. The first study is the 1997 Los
used similar venues to disperse breast cancer screening educa-
Angeles County Health Survey of 4,697 women, of whom 51
tional materials in the population's native language and simul-
identified as lesbian and 36 as bisexual. Lesbian but not bi-
taneously launched a media campaign targeting Vietnamese
sexual women were less likely than heterosexual women to
American women did not yield a significant change in clinical
have received Pap tests or clinical breast examinations in the
breast examinations or mammograms (Nguyen, Vo, McPhee,
prior two years (Diamant, Wold, Spritzer, & Gelberg, 2000).
& Jenkins, 2001).
The second is the Women's Health Initiative, which found
Two clinical trials have aimed at improving colorectal cancer
that heterosexual women and adult lesbians were more likely
screening. In the first study (Thompson, Michnich, Gray, Fried-
than no-adult-sex, bisexual, and lifetime-lesbian women to
lander, & Gilson, 1986), the effectiveness of three interventions
have received Pap tests and mammograms in the prior year
was evaluated for improving patient participation in fecal occult
(Valanis et al., 2000). The number of self-identified lesbians
blood tests: physician or nurse discussion, a reminder phone
in these studies was low, even lower than would be expected
call, or a reminder postcard. All the interventions were better
from general population estimates.
than no intervention. The reminder postcard was the best, in-
Numerous programs have been developed during the past 20
creasing adherence by 25%. In the second study (Pignone, Har-
years to increase the cancer-screening behaviors of women,
ris, & Kinsinger, 2000), 249 patients aged 5075 in central
particularly in regard to breast cancer screening. The interven-
North Carolina who had not had any colorectal screening tests
tions have used communication from physicians and healthcare
in five years and did not have a family history of colorectal can-
plans, as well as various community-based interventions. Table
cers were randomized into two groups. The first group watched
1 offers information about some of these studies. A few of these
an 11-minute video about colorectal cancer, received an educa-
programs have been subjected to the rigors of research, but none
tional brochure about colon cancer screening, and had their
have focused on lesbians.
charts flagged indicating interest in screening. The other group
A number of U.S. studies have focused on ethnic minority
watched a video about car safety, and their charts were not dis-
women including Latinas, Vietnamese, Asians, and African
tinguished. Screening tests were completed by 37% of the inter-
Americans. The Por La Vida intervention program for the
vention group and 23% of the control group (p < 0.03).
Latina women's community was tested using a randomized
Based on the work completed by others in developing targeted
clinical trial (Navarro et al., 1998). Lay community workers
cancer-screening programs, the current study's authors decided
(N = 36) each recruited approximately 14 peers. Each estab-
to test the effect of a lesbian-specific screening program given
ONCOLOGY NURSING FORUM VOL 30, NO 4, 2003
E72
Table 1. Selected Intervention Studies Designed to Increase Breast Cancer Screening
Authors
Sample
Interventions
Results
McDermott & Marty
292 female college students
Women were randomly assigned to one of two
Significantly more members of treatment
(1984)
groups: facilitator-conducted program using
group practiced BSE at least once within three
modeling plus rehearsal (treatment) and a
months after the intervention.
group receiving pamphlets about breast cancer
and breast self-examination (BSE).
McPhee et al. (1989)
62 internal-medicine residents
Women were randomly assigned to one of the
Residents' charts were reviewed to assess per-
following groups: computer-generated remind-
formance of seven tests.
ers of patients with overdue tests at the time of
Group 1 increased performance of six of
their visit, monthly seminars about screening
seven tests.
with feedback about their performance rates, or
Group 2 increased performance of four of
a control group. Half of the medical residents in
seven tests.
each group also received patient-education ma-
Patient reminders increased performance of
terials and their patients received notices of
one of two targeted screening tests.
overdue tests.
Marcus et al. (1992)
2,000 women with abnormal
Women received either a personalized follow-
Bus passes were effective in increasing the re-
Pap tests
up letter and pamphlet, a slide-show program
turn rate for people at county hospital, whereas
on Pap tests, or transportation incentives (bus
slides and letters were effective for patients at
pass or parking permit).
other clinics.
Rothman et al. (1993)
190 women older than 40 who
Women viewed persuasive mammography
Women who viewed internal messaging were
were not adhering to mammog-
video with one of three attributional empha-
more likely than women in other two groups to
raphy-screening guidelines
sis: internal, external, or informational. Atti-
have had a mammogram 12 months later.
tudes about breast cancer and mammography
were assessed immediately and six months
later.
Skinner et al. (1994)
435 women aged 4065 who
Women received either individually tailored or
Recipients of tailored letters were more likely to
had been to a family practice
standardized mammography recommendation
have mammograms, especially those with in-
within two years
letters with follow-up phone calls eight months
comes less than $26,000 and African American
later.
women.
Herman et al. (1995)
803 women older than 65 seen
All staff received intensive training in breast
Women in the prevention group were offered
by 66 residents attending am-
cancer screening. Patients were randomized to
clinical breast examination (CBE) 32% more
bulatory clinic at public hospi-
one of three groups: control or no intervention,
frequently than other two groups (not signifi-
tals
education about breast cancer screening (edu-
cant when adjusted for race, age, comorbidity,
cation), and education plus a flow sheet on
and physician's gender and training). Women
chart to facilitate compliance (prevention).
in the prevention group without previous CBE
were offered CBE significantly more often than
other groups.
Five-year follow-up study of 91
The intervention included (a) identification by
Foley et al. (1995)
After charts were audited, the researchers
women older than 40 who had
the nurses of eligible women who were over-
found that mammogram recommendation and
participated in a nurse-initiated
due for a mammogram, (b) completion of a
completion rates increased over time for both
intervention study to improve
checklist by residents indicating whether a
the intervention and control groups. However,
mammography recommenda-
mammogram was recommended and why, and
the nurse-initiated intervention group had a sig-
tion and completion rates and
(c) a nurse-initiated reminder system for pa-
nificantly higher change over time.
189 controls
tients.
204 African American women
Culturally sensitive messages from African
Erwin et al. (1996)
Use of BSE and mammography increased sig-
in eastern Arkansas
American breast cancer survivors were pre-
nificantly (p < 0.001) after participation in pro-
sented at churches and community organiza-
gram.
tions, emphasizing the need for early detection
for survival.
Kernohan (1996)
1,628 women from minority
Two specifically trained health-promotion fa-
Significant increases in cervical tests and
ethnic groups (South Asian,
cilitators conducted formal and informal
breast cancer screening were self-reported, as
African Caribbean, Eastern Eu-
meetings in ethnically diverse areas, educat-
well as anecdotal observations of attendance at
ropean, and others)
ing women about breast and cervical cancer
local clinics.
and related screening guidelines. They con-
ducted these meetings, showed videos, and
offered written materials in the clients' native
(Continued on next page)
languages.
ONCOLOGY NURSING FORUM VOL 30, NO 4, 2003
E73
Table 1. Selected Intervention Studies Designed to Increase Breast Cancer Screening (Continued)
Sample
Authors
Results
Interventions
395 women in Philadelphia
Davis et al. (1997)
Women who received a phone call were 28%
Mammography reminders were given in one of
health maintainance organiza-
more likely to obtain a mammogram than those
three forms: birthday card, personalized letter
tions (HMOs)
who received a letter, and those who received
from medical director with written materials pro-
a birthday card were 15% more likely to obtain
moting mammography, or a multicomponent
a mammogram.
telephone call with reminder, counseling, and
scheduling appointment.
Mexican American women in
Suarez et al. (1997)
Each community saw a 6% and 7% absolute
Three-year intervention was aimed at increasing
two communities
increase in Pap tests and a 17% and 19% ab-
Pap test and mammography screening that in-
solute increase in mammograms.
cluded the presentation of role models in the
media and was reinforced with peer volunteers.
Urban women 5277 years old
Women in intervention group were nearly three
Weber & Reilly (1997)
Women received a case-management interven-
in six primary care practices
times more likely to receive a mammogram
tion from culturally sensitive community health
supported by a computerized
(95% confidence interval, 1.754.73).
educators or usual care.
clinical information system
36 Latina community workers
The screening group showed statistically sig-
Navarro et al. (1998)
Women were randomized to attend a 12-week
were recruited, as well as ap-
nificant improvement in BSE (33% versus 19%
intensive cancer screening group or "Commu-
proximately 14 peers per
for control group), mammography (21% ver-
nity Living Skills" group. Pre- and postinterven-
worker
sus 7%), and Pap test (23% versus 16%). No
tion assessment of breast and cervical cancer
difference was found for CBE.
screening was performed.
108 women 4070 years old
Street et al. (1998)
Both groups perceived importance increased
Women completed a baseline questionnaire as-
from two clinics (family prac-
after the intervention, although no significant
sessing their perception of personal importance
tice and free clinic)
difference was found between the two. Youn-
of breast cancer, knowledge, and anxiety about
ger women tended to prefer the multimedia ex-
screening. They were randomized to see educa-
perience and learned more from the interven-
tional materials either by way of an interactive
tion than older women.
multimedia program or brochure, then asked
initial questions again to assess learning.
8,385 women due for cervical
Segnan et al. (1998)
The letter signed by a general practitioner with
Women received one of the following letters:
screening and 8,069 women
a fixed appointment was more effective for
Group A: signed by a general practitioner with
due for breast cancer screening
breast and cervical screening than two of the
a fixed appointment; Group B: open-ended in-
other letters. Group B was 39% less effective,
vitation signed by general practioner; Group C:
and Group C was 14% less effective. Group D
signed by program coordinator with fixed ap-
pointment; or Group D: extended letter high-
had no difference.
lighting benefits of cancer screening signed by
general practitioner with fixed appointment.
ENCOREplus   (YWCA) is a health promotion
27,494 women, primarily eth-
Fernandez et al. (1999)
58% of women who initially were nonadherent
n i c minorities with low in-
program providing outreach, referral, and other
with Pap- and breast-screening guidelines had
comes, attending YWCA
services to facilitate breast and cervical cancer
mammograms and 37% had Pap tests.
screening. Its effectiveness was evaluated.
314 women 5074 years old
Women were randomized to two groups: one
Taylor et al. (1999)
Mammography completion within eight weeks
with at least one general medi-
received usual care and the other was educated
of clinic visit was significantly higher in the in-
cine clinic appointment at uni-
about screening guidelines and breast cancer
tervention group (49%) than control group
versity-affiliated hospital in Se-
risk by physician and nurse, saw video, and, if
(22%). Bus passes and rescheduling efforts did
attle (due for mammography)
interested, had an appointment scheduled and
not contribute to the observed increases in
was given bus passes. If the mammogram ap-
screening participation.
pointment was more than a week later, they
were called or sent a reminder. If they missed
their appointment, they were called and encour-
aged to make another appointment.
140 women older than 18 at six
Pharmacists administered the Gail model risk
Giles et al. (2001)
Monthly BSE increased from 31% to 56% (p <
community pharmacies and
assessment tool and provided education and
0.001) after six months. No significant change
two health-screening fairs
training on BSE, CBE, and mammography.
was found in obtaining CBE and mammograms
except for women 4049 years old.
Simon et al. (2001)
1,717 women at two locations
Women were randomized to one of three
No difference in mammography use occurred
of multisite inner-city health de-
groups: received letter instructing them to visit
among the three groups.
their primary care physician; received letter in-
partment in Detroit. Most
structing them to contact their clinic to sched-
women were African American,
ule a mammogram; did not receive a letter.
older than 50, and had minimal
(Continued on next page)
health insurance.
ONCOLOGY NURSING FORUM VOL 30, NO 4, 2003
E74
Table 1. Selected Intervention Studies Designed to Increase Breast Cancer Screening (Continued)
Authors
Sample
Interventions
Results
Nguyen et al. (2001)
788 and 807 Vietnamese
Women and physicians in Alameda County re-
The intervention group was no more likely to
American women and physi-
ceived interventions of continuing education,
recognize, receive, or be up-to-date with CBE or
cians in pre- and postinterven-
health fairs, educational material, and a media
mammograms than the control group, al-
tion groups in Alameda and
campaign. Women in Orange County served as
though they were more likely to have planned
Orange Counties, CA
the control.
a CBE or mammogram.
Hiatt et al. (2001)
1,599 women in an under-
Women were randomized to a community-
83% of women in intervention group obtained
served multiethnic population,
based outreach using lay health workers, an
mammograms and 95% obtained Pap tests
aged 4075 in the San Fran-
"in-reach" intervention targeted at four clinics
versus 68% and 83%, respectively, in the con-
cisco Bay Area from 1993
using provider education and computer re-
trol group.
1996
minders, or a patient-navigator intervention to
enhance follow-up and resolution of abnormal
Pap tests and mammograms.
Valanis et al. (2002)
510 female HMO members 52
Women were randomized to receive one or
32% of the combined group (p = 0.05), 39% of
6 9 years old who had no
combination of the following interventions: a
the outreach group (p = 0.006), and 26% of the
mammograms in the prior two
clinic office in-reach intervention, sequential
in-reach group obtained both services com-
years and no Pap tests in the
letter or telephone outreach intervention, or
pared to 19% of the usual-care group
past three years
usual care.
tidisciplinary panel of experts (two physicians, one epide-
by a lesbian physician. This approach takes advantage of the
miologist, and three oncology nurses) established content
power of the physician and the power of being a member of
validity of the instrument. Most of the questions, except
the targeted community--an insider. The goal of this project
those about colon cancer, had been used in other studies
was to determine whether attending a lesbian-specific screen-
conducted by the principal investigator with excellent re-
ing program fostered a behavior change in screening behav-
sults (Dibble et al., 1999, 2002; Roberts et al., 1998). Be-
iors among lesbians.
cause these questions were demographic in nature and mea-
Methods
sured multiple concepts, typical measures of reliability
(e.g., Cronbach's alpha) were not appropriate to calculate.
Design
Test-retest reliability estimates also were not appropriate to
explore because the authors were looking for change and
This pilot study was conducted using a pre- and post-test
not stability.
design to explore the impact of a lesbian-specific educational
intervention from a lesbian physician on the cancer-screening
Procedures
behaviors of lesbians older than 50.
After approval by the institutional review board, both edu-
Settings
cational programs were scheduled and advertised at an ur-
The settings for this study were two lesbian, gay, bisexual,
ban and a suburban center. Participants were asked at the be-
and transgendered senior organizations in the San Francisco
ginning of the educational program whether they would be
Bay Area. In the first program, conducted in an urban setting,
willing to participate in a research study to explore the use-
all the women (N = 7) who attended the presentation partici-
fulness of the program. An information sheet about the
pated in the study and completed both the pre- and postinterven-
study was distributed to the audience. Those who were will-
tion testing. In the second program, conducted in a suburban
ing to participate completed the pretest in about 10 minutes.
setting, 33 women attended the presentation. These women
Consent was implied by the return of the completed ques-
were very concerned about remaining anonymous. Of the 33
tionnaire. After each program, names, telephone numbers,
women, 29 completed the preintervention questionnaire. How-
and e-mail addresses for the follow-up were collected to be
ever, only 23 would give their contact information to the re-
kept (in a locked file cabinet, accessible to just one member
searchers for the postintervention follow-up. Fifteen of the 23
of the research team) only until the end of the study. The
completed the postintervention testing. Therefore, of the 40
contact information then was destroyed. To protect the
women who received the intervention, 55% (n = 22) completed
women's privacy, each questionnaire was assigned a study
both pre- and postintervention questionnaire packets.
number. Follow-up questionnaires were matched with the
initial questionnaire using age, zip code, and education
Sample
rather than names. Although these safeguards were ex-
To be included in the study, the women had to be aged 50
plained to the women, mistrust of research was evident.
Eighteen percent of the participants who answered the pre-
or older and attend a lesbian-specific educational program
about cancer screening.
test prior to the educational program refused to participate
in the post-test because of the lack of anonymity. The post-
Instruments
test data were gathered by a research assistant in a telephone
interview approximately six months postintervention (range
The pre- and postintervention surveys were created, pilot-
tested, and revised prior to being used in this study. A mul-
= 5.57 months).
ONCOLOGY NURSING FORUM VOL 30, NO 4, 2003
E75
Intervention
$30,000 per year. Most (72%) had a family history of cancer,
and 17% (n = 6) had a personal history of cancer. Twenty-two
The intervention consisted of a one-hour didactic presenta-
women (61%) completed the follow-up portion of the study.
tion from a lesbian family practice physician followed by a
Table 2 contains a comparison of demographic information by
15-minute question-and-answer session with the audience.
study completion. No significant differences were found be-
The presentation, titled "Cancer Screening Tests: What Les-
tween the two groups. Information is not available for the
bians Over Age 50 Need to Know," began with a five-minute
women who chose not to participate in the study. The sample
biographic sketch during which the physician discussed her
sizes varied by question because of missing data.
background in family medicine, lesbian health research, and
Breast Cancer Screening
lesbian community activism. She then presented a 15-minute
review of current research findings with regard to lesbians'
All of the women had obtained a mammogram sometime
risk for cancer.
during their lifetimes. However, six women (27%) had not
In this review, the physician discussed the lack of knowl-
received one in two years or more. The authors constructed
edge about the true incidence of various cancers among les-
a completion ratio of total mammograms obtained to an es-
bians because national cancer registries do not gather informa-
timate of best practice (mammograms yearly when older than
tion about sexual orientation and most researchers do not ask
50). On average, these six women had only 29% of the
questions about sexual orientation in their demographics. She
mammograms that they should have experienced during their
described what cancer risk differences have been identified
lifetime, whereas the women who had a recent mammogram
between lesbians and heterosexual women and the impact of
had a completion ratio of more than 80% (t = 2.81, p =
those differences on the potential development of cancer:
0.011). Therefore, these six women are the focus of a more
Being a lesbian does not increase a woman's risk of cancer,
detailed analysis of their breast health practices and a marker
but some differences in lifestyles between lesbians and hetero-
for the worth of the program in supporting positive breast
sexual women might. For instance, lesbians are less likely to
seek health care because of the discomfort of revealing their
sexual orientation to healthcare providers (White & Dull,
Table 2. Demographic Characteristics of Participants by
1997). In addition, lesbians are less likely to visit a doctor for
Follow-Up Status
routine gynecologic services such as birth control (Cochran et
al., 2001). Therefore, lesbians are less likely to have cancers
Completed Study Lost to Follow-Up
detected at earlier, more treatable stages. Lesbians are at
(n = 22)
(n = 14)
higher risk for breast and ovarian cancers because they are
less likely to have children by age 30, if at all, and have a
--
--
p
Characteristic
SD
SD
t
X
X
higher body mass index (weight-to-height ratio) than hetero-
0.228
Age (years)
59.0
4.5
62.0
8.6
1.25
sexual women (Dibble et al., 2002). Lesbians are more likely
0.275
Education (years)
15.1
2.7
16.1
3.2
1.11
to have smoked than heterosexual women (Roberts et al.,
1998). This information was followed by 45 minutes of infor-
p
Characteristic
n
%
n
%
c
mation on recommended cancer-screening procedures for
breast, cervical, colon, lung, ovarian, and uterine cancers, with
0.357
Ethnicity
0.302
emphasis given to colon cancer screening. The screening pro-
Caucasian
20
91
11
86
Other
02
09
03
14
cedures described followed the recommendations of the
Area lived
0.097
1.000
American Cancer Society (2001) and the U.S. Preventive
Urban
16
73
08
62
Health Services Task Force (2001). Referral information to a
Suburban
06
27
05
38
lesbian-sensitive healthcare provider also was offered.
Relationship status
1.226
0.175
Partnered
10
45
83
21
Data Management and Analyses
Other
11
55
11
79
Data were entered and verified using the SPSS Version 11
Employment
0.094
0.954
Employed
12
55
08
57
(SPSS Inc., Chicago, IL). Descriptive statistics were generated
Retired
06
27
04
29
to describe the sample characteristics. Although researchers
04
18
02
14
Other
planned to use paired t tests or McNemar chi-square analyses
1.000
Personal income
0.000
as appropriate for the level of data, the sample sizes did not
< $20,000
10
53
07
58
permit this quantitative analytic plan. Descriptive statistics
09
47
05
41
$20,000+
were used to describe the outcomes of this pilot study.
1.000
0.004
Health insurance
Yes
19
86
13
93
Results
03
14
01
07
No
Individual history of
Sample Information
0.370
cancer
0.584
--
05
23
01
07
Yes
Thirty-six women aged 5081 (X = 60.2, SD = 6.48) par-
17
77
13
93
No
ticipated in the program. The majority were Caucasian (86%),
Family history of
single (61%), living in urban areas (67%), employed (56%)
1.000
0.218
cancer
--
and educated beyond high school (X = 15.47 years;, SD =
Yes
17
77
09
64
2.90, range = 921). Eleven percent (n = 4) did not have any
No
05
23
05
36
health insurance and were not covered by Medicaid or Medi-
care. Most (60%) of the women reported incomes of less than
Note. Missing data exist for some variables.
ONCOLOGY NURSING FORUM VOL 30, NO 4, 2003
E76
Table 3. Breast Cancer Screening Behaviors for At-Risk Lesbians
Family
Years Since Last
Postintervention
Clinical Breast
Post-Intervention
Breast Self-
Postintervention
Age
History
Insurance
Mammogram
Mammogram
Examination (CBE)
CBE
Examination (BSE)
BSE
54
No
Yes
6
No
Yearly
Yes
> Monthly
Monthly
55
No
No
7
No
Yearly
No
> Monthly
Monthly
58
No
Yes
2
Yes
Yearly
No
> Monthly
Monthly
59
No
Yes
7
No
> Yearly <
No
> Monthly
> Monthly <
63
Not known
Yes
2
No
> Yearly <
No
> Monthly
Monthly
65
No
Yes
5
Yes
Yearly
No
> Monthly
> Monthly <
care screening. Table 3 indicates that, after the intervention,
minority status have not been as explicated. In fact, differ-
one-third of these women received mammograms and half
ences among sexual minorities in health beliefs, health behav-
began performing monthly BSE.
iors, health outcomes, and experiences with the healthcare in-
dustry largely are unexplored. This is the first intervention
Gynecologic Cancer Screening
study examining the impact a lesbian-targeted educational
program made to address the cancer-screening behaviors of
All of the women experienced a pelvic examination some
time during their lifetime. However, four women (18%) had not
lesbians.
In this pilot study, some of these challenging-to-reach women
had one for three years or more. Therefore, these four women
changed their behavior in a very short period of time. The suc-
are the focus of a more detailed analysis of their gynecologic
cess of this pilot program suggests that a future randomized
health practices and a marker for the worth of the program in
control trial of this intervention should be conducted in the
supporting positive gynecologic care screening. After the inter-
sometimes-hidden population of lesbians older than 50. As is
vention, one woman, whose sister had ovarian cancer, obtained
the case with all minority groups, future success may depend on
a pelvic examination (see Table 4). Prior to the intervention, she
the way in which programs are created and introduced. For
"just could not make [herself] do it."
women who may put off health care for fear of having their
Colorectal Cancer Screening
sexual orientation discovered and recorded or because they
have found hostility within the healthcare system as a result of
Of the 22 women, 12 (55%) were up-to-date with their co-
their sexual orientation, the use of specialized educational pro-
lorectal cancer screening, having had a recent sigmoidoscopy
grams is vital. For this minority group, simply identifying the
or colonoscopy. Of those 12 women, two stated that they
women who comprise it is not enough; healthcare profession-
would never have another because of the pain associated with
als, researchers, and educators must understand that some
the procedure. The 10 women who had no recent screening for
women belonging to this sexual minority may fear exposure.
colorectal cancer were the focus of a more detailed analysis of
Many lesbians in this age group have encountered hostility
their colorectal health practices. After the intervention, only
from healthcare professionals and, thus, are reluctant to seek
one of the women obtained screening (see Table 5). The
health care. In addition, many of the screening programs do not
women described three major barriers to screening: (a) lack of
have culturally appropriate materials designed to appeal to
money, (b) fear of the pain, and (c) their healthcare provider
women who are intimate with other women.
did not arrange for the test.
Strengths and Limitations
Discussion
The intervention occurred in two lesbian, gay, bisexual, and
In the United States, for most of the 20th century, minority
transgendered senior centers by a lesbian physician and was
designed to protect the confidentiality of the participants; yet,
members of society were observed to have had poorer health
outcomes than Caucasian, heterosexual, married men. For
almost 20% of the women refused to participate in follow-up
most of that century, an unchallenged assumption existed that
interviews because of the perceived lack of anonymity. This
the blame for those less-than-optimal outcomes somehow was
study was conducted in the San Francisco Bay Area, suppos-
located with the minority population. In recent years, the bi-
edly a "safe" place for nonheterosexuals; what the rate of re-
ases inherent in science and in healthcare delivery systems
fusal would have been elsewhere only can be speculated. This
finding may shed some light on the low numbers of lesbians
have been exposed, particularly related to racial, ethnic, and
cultural differences. However, differences related to sexual
responding to other large, national studies.
Table 4. Gynecologic Cancer Screening Behaviors for At-Risk Lesbians
Years Since Last
Postintervention
Age
Family History
Insurance
Hysterectomy
Pelvic Examination
Reason for Delay
Pelvic Examination
58
No
Yes
No
13
Don't like
No
59
No
Yes
No
05
Disabled--self-care issues
No
63
Not known
Yes
No
03
Plan every 5 years
No
66
Yes
Yes
No
08
Don't like
Yes
ONCOLOGY NURSING FORUM VOL 30, NO 4, 2003
E77
Table 5. Colorectal Cancer Screening Behaviors for At-Risk Lesbians
Family
Occult
Postintervention
Postintervention
Age
History
Insurance
Blood Test
Colonoscopy
Colonoscopy
Sigmoidoscopy
Sigmoidoscopy
Comments
53
No
Yes
2 years
No
No
No
No
I will have the colonoscopy in the next
six months.
54
No
Yes
Never
No
No
Yes, 7 years ago
Yes
Decided to have sigmoidoscopy be-
cause of increased risk of perforation
55
No
No
> 5 years
No
No
No
No
Afraid of the pain
57
No
Yes
1.5 years
No
No
Yes, 8 years ago
No
Will not have because of pain with first
sigmoidoscopy
57
No
Yes
Never
No
No
No
No
Afraid it will hurt
59
No
No
6 months
No
No
No
No
Too invasive, afraid of pain, no money
59
No
Yes
> 5 years
No
No
No
No
Fear of physicians and treatment
60
No
No
1.5 years
No
No
No
No
I do not have the money.
63
No
Yes
2 years
No
No
Yes, 26 years ago
No
Healthcare provider did not arrange
63
Not known
Yes
3 years
Yes, 13 years ago
No
Yes, 8 years ago
No
Afraid it will hurt
The time interval for the study was too short. Some women
meanings and barriers of the various cancer screening behav-
in the study stated that they were on a waiting list for colon
iors among lesbians and other sexual minority women.
cancer screening, and others were planning on obtaining a
Conclusion
mammogram. Whether these women were just trying to
please the investigators was not clear. Future studies should
The intervention for this study was a culturally sensitive pro-
follow women for at least two years to determine the actual
value of the intervention. Researchers also should use self-
gram by a culturally competent provider. The success of this
program in increasing cancer screening among lesbians older
report plus medical record confirmation to verify that the
than 50 suggests that a minority-specific intervention can in-
women actually have a record of a clinical breast examination,
mammogram, sigmoidoscopy, or colonoscopy. The authors
crease positive health behaviors such as screening for cancer.
As research into the health concerns of lesbians increases in size
also wondered whether a one-time intervention is enough or
and scope, so must the health care of the women who make up
whether a "screening coach" or some other ongoing reminder
system are needed.
sexual minorities become more culturally competent. That the
health or health care of women is compromised by any minor-
Although this was a community-based sample, very few
ity status is universally unacceptable.
lesbians of color were represented in this study. In fact, the
sample size for this pilot study was quite small, which may
The authors would like to thank the women who participated in the pro-
lead to unbalanced numbers in potential covariates and thus
grams; Lea Aukerman, PhD, for her help in data collection; Valerie Hayden,
MSW, and Dotty Fowler, MA, for their help in arranging the educational pro-
erroneous findings. The next step is a randomized clinical trial
grams; Karen Casey, MS, RN, for her editorial assistance; and Farley
of this intervention, which should have a sample size large
Urmston, BA, for her data management.
enough to answer the research question, diversity among the
participants, and multiple sites throughout the country to in-
Author Contact: Suzanne L. Dibble, RN, DNSc, can be reached at
crease the generalizabilty of the findings. In addition, quali-
sdibble@itsa.ucsf.edu, with copy to editor at rose_mary@earth
tative studies are necessary to understand more about the
link.net.
References
American Cancer Society. (2001). Prevention and early detection. Retrieved
Library/ons_publications/onf/2002/January_February_2002/E1-E7.pdf
November 1, 2001, from http://www.cancer.org/docroot/PED/ped_0.asp
Dibble, S.L., Roberts, S.A., Davids, H.R., Paul, S.M., & Scanlon, J.L. (1999). A
Cochran, S.D., Mays, V.M., Bowen, D., Gage, S., Bybee, D., Roberts, S.J.,
comparison of breast cancer risk factor distributions between lesbian and bi-
et al. (2001). Cancer-related risk indicators and preventive screening be-
sexual women. Medical Student JAMA. Retrieved October 7, 1999, from http:/
haviors among lesbians and bisexual women. American Journal of Public
/www.ama-assn.org/sci-pubs/msjama/articles/vol_282/no_13/cancer.htm
Health, 91, 591597.
Erwin, D.O., Spatz, T.S., Stotts, R.C., & Hollenberg, J.A. (1999). Increasing
Davis, N.A., Nash, E., Bailey, C., Lewis, M.J., Rimer, B.K., & Koplan, J.P.
mammography practice by African American women. Cancer Practice, 7,
(1997). Evaluation of three methods for improving mammography rates in
7885.
a managed care plan. American Journal of Preventive Medicine, 13, 298
Erwin, D.O., Spatz, T.S., Stotts, R.C., Hollenberg, J.A., & Deloney, L.A.
302.
(1996). Increasing mammography and breast self-examination in African
Denenberg, R. (1995). Report on lesbian health. Women's Health Interna-
American women using the Witness Project model. Journal of Cancer
tional, 5(2), 8191.
Education, 11, 210215.
Diamant, A.L., Wold, C., Spritzer, K., & Gelberg, L. (2000). Health behav-
Fernandez, M.E., DeBor, M., Candreia, M.J., Wagner, A.K., & Stewart, K.R.
iors, health status, and access to and use of health care: A population-based
(1999). Evaluation of ENCORE plus. A community-based breast and cer-
study of lesbian, bisexual, and heterosexual women. Archives of Family
vical cancer screening program. American Journal of Preventive Medicine,
Medicine, 9, 10431051.
16(Suppl. 3), 3549.
Dibble, S., Roberts, S.A., Robertson, P.A., & Paul, S.M. (2002). Risk factors for
Foley, E.C., D'Amico, F., & Merenstein, J.H. (1995). Five-year follow-up of
ovarian cancer: Lesbian and heterosexual women. Oncology Nursing Forum,
a nurse-initiated intervention to improve mammography recommendation.
29, E1E7. Retrieved January 18, 2002, from http://www.ons.org/images/
Journal of the American Board of Family Practice, 8, 452456.
ONCOLOGY NURSING FORUM VOL 30, NO 4, 2003
E78
Giles, J.T., Kennedy, D.T., Dunn, E.C., Wallace, W.L., Meadows, S.L., &
among women over forty with an internally-oriented message. Health Psy-
Cafiero, A.C. (2001). Results of a community pharmacy-based breast can-
chology, 12(1), 3745.
cer risk-assessment and education program. Pharmacotherapy, 21, 243
Segnan, N., Senore, C., Giordano, L., Ponti, A., & Ronco, G. (1998). Promot-
253.
ing participation in a population screening program for breast and cervi-
Herman, C.J., Speroff, T., & Cebul, R.D. (1995). Improving compliance with
cal cancer: A randomized trial of different invitation strategies. Tumori, 84,
breast cancer screening in older women. Results of a randomized con-
348353.
trolled trial. Archives of Internal Medicine, 155, 717722.
Sell, R.L. (1997). Defining and measuring sexual orientation: A review. Ar-
Hiatt, R.A., Pasick, R.J., Stewart, S., Bloom, J., Davis, P., Gardiner, P., et al.
chives of Sexual Behavior, 26, 64358.
(2001). Community-based cancer screening for underserved women: De-
Simon, M.S., Gimotty, P.A., Moncrease, A., Dews, P., & Burack, R.C.
sign and baseline findings from the Breast and Cervical Cancer Interven-
(2001). The effect of patient reminders on the use of screening mammog-
tion Study. Preventive Medicine, 33, 190203.
raphy in an urban health department primary care setting. Breast Cancer
Jemal, A., Murray, T., Samuels, A., Ghafoor, A., Ward, E., Thun, M.J.
Research and Treatment, 65(1), 6370.
(2003). Cancer statistics, 2003. CA: A Cancer Journal for Clinicials, 53,
Skinner, C.S., Strecher, V.J., & Hospers, H. (1994). Physicians' recommen-
526.
dations for mammography: Do tailored messages make a difference?
Johnson, S.R., Smith, E.M., & Guenther, S.M. (1994). Comparison of gyne-
American Journal of Public Health, 84, 4349.
cologic health care problems between lesbians and bisexual women: A
Smith, E.M., Johnson, S.R., & Guenther, S.M. (1985). Health care attitudes
survey of 2,345 women. Journal of Reproductive Medicine, 32, 805811.
and experiences during gynecologic care among lesbians and bisexuals.
Kernohan, E.E. (1996). Evaluation of a pilot study for breast and cervical cancer
American Journal of Public Health, 75, 10851087.
screening with Bradford's minority ethnic women: A community develop-
Solarz, A.L. (Ed.). (1999). Lesbian health: Current assessment and directions
ment approach, 199193. British Journal of Cancer, 29(Suppl.), S42S46.
for the future. Washington, DC: National Academy Press.
Koh, A. (2000). Use of preventive health behaviors by lesbian, bisexual, and
Street, R.L., Jr., Van Order, A., Bramson, R., & Manning, T. (1998). Precon-
heterosexual women: Questionnaire survey. Western Journal of Medicine,
sultation education promoting breast cancer screening: Does the choice of
172, 379384.
media make a difference? Journal of Cancer Education, 13, 152161.
Laumann, E.O., Gagnon, J.H., Michael, R.T., & Michaels, S. (1994). The
Suarez, L., Roche, R.A., Pulley, L.V., Weiss, N.S., Goldman, D., & Simpson,
social organization of sexuality: Sexual practices in the United States.
D.M. (1997). Why a peer intervention program for Mexican-American
Chicago: University of Chicago Press.
women failed to modify the secular trend in cancer screening. American
Lauver, D.R., Karon, S.L., & Egan, J. (1999). Understanding lesbians' mam-
Journal of Preventive Medicine, 13, 411417.
mography utilization. Women's Health Issues, 9, 264274.
Taylor, V., Thompson, B., Lessler, D., Yasui, Y., Montano, D., Johnson,
Marcus, A.C., Crane, L.A., Kaplan, C.P., Reading, A.E., Savage, E., Gunning,
K.M., et al. (1999). A clinic-based mammography intervention targeting
J., et al. (1992). Improving adherence to screening follow-up among
inner-city women. Journal of General Internal Medicine, 14, 104111.
women with abnormal Pap smears: Results from a large clinic-based trial
Thompson, R.S., Michnich, M.E., Gray, J., Friedlander, L., & Gilson, B.
of three intervention strategies. Medical Care, 30, 216230.
(1986). Maximizing compliance with hemoccult screening for colon can-
McDermott, R.J., & Marty, P.J. (1984). Seeking an effective strategy for pro-
cer practice. Medical Care, 24, 904914.
moting breast self-examination among women. Patient Education and
Tucker, N., Highleyman, L., & Kaplan, R. (Eds.). (1995). Bisexual politics:
Counseling, 6, 116121.
Theories, queries, and visions. Binghamton, NY: Haworth Press.
McPhee, S.J., Bird, J.A., Jenkins, C.N., & Fordham, D. (1989). Promoting
U.S. Preventive Health Services Task Force. (2001). Reviews and recommen-
cancer screening. A randomized, controlled trial of three interventions. Ar-
dations: Screening. Retrieved November 1, 2001, from http://www.ahcpr
chives of Internal Medicine, 149, 18661872.
.gov/clinic/uspstfix.htm
Navarro, A.M., Senn, K.L., McNicholas, L.J., Kaplan, R.M., Roppe, B., &
Valanis, B.G., Bowen, D.J., Bassford, T., Whitlock, E., Charney, P., & Carter,
Campo, M.C. (1998). Por La Vida model intervention enhances use of can-
R.A. (2000). Sexual orientation and health: Comparisons in the Women's
cer screening tests among Latinas. American Journal of Preventive Medi-
Health Initiative sample. Archives of Family Medicine, 9, 843853.
cine, 15, 3241.
Valanis, B.G., Glasgow, R.E., Mullooly, J., Vogt, T.M., Whitlock, E.P.,
Nguyen, T., Vo, P.H., McPhee, S.J., & Jenkins, C.N.H. (2001). Promoting
Boles, S.M., et al. (2002). Screening HMO women overdue for both mam-
early detection of breast cancer among Vietnamese-American women:
mograms and Pap tests. Preventive Medicine, 34, 4050.
Results of a controlled trial. Cancer, 91(Suppl. 1), 267273.
Weber, B.E., & Reilly, B.M. (1997). Enhancing mammography use in the
Norman, A.D., Perry, M.J., Stevenson, L.Y., Kelly, J.A., & Roffman, R.A.
inner city: A randomized trial of intensive case management. Archives of
(1996). Lesbian and bisexual women in small cities--At risk for HIV?
Internal Medicine, 157, 23452349.
HIV prevention community collaborative. Public Health Reports, 111,
White, J., & Dull, V. (1997). Health risk factors and health-seeking behavior
347352.
in lesbians. Journal of Women's Health, 6, 103112.
Pignone, M., Harris, R., & Kinsinger, L. (2000). Videotape-based decision aid
White, J., & Levinson, W. (1995). Lesbian health care: What a primary care
for colon cancer screening: A randomized, controlled trial. Annals of In-
physician needs to know. Western Journal of Medicine, 162, 463466.
ternal Medicine, 133, 761769.
Rankow, E.J. (1995). Breast and cervical cancer among lesbians. Women's
For more information . . .
Health International, 5, 123129.
Rankow, E.J., & Tessaro, I. (1998a). Cervical cancer risk and Papanicolaou
Gay and Lesbian Medical Association
screening in a sample of lesbian and bisexual women. Journal of Family
Practice, 47, 139143.
www.glma.org
Rankow, E.J., & Tessaro, I. (1998b). Mammography and risk factors for
MEDLINEplus: Gay and Lesbian Health
breast cancer in lesbian and bisexual women. American Journal of Health
www.nlm.nih.gov/medlineplus/gaylesbianhealth.html
Behaviors, 22, 403410.
OutSmart
Roberts, S.A., Dibble, S.L., Scanlon, J., Paul, S., & Davids, H.R (1998). Dif-
www.outsmartmagazine.com
ferences in risk factors for breast cancer: Lesbians and heterosexual women.
Journal of the Gay and Lesbian Medical Association, 2(3), 93101.
Links can be found using ONS Online at www.ons.org.
Rothman, A.J., Salovey, P., Turvey, C., & Fishkin, S.A. (1993). Attributions
of responsibility and persuasion: Increasing mammography utilization
ONCOLOGY NURSING FORUM VOL 30, NO 4, 2003
E79