|
|
||
 |
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
Journal of Clinical Oncology, Vol 24, No 10 (April 1), 2006: pp. 1620-1626 © 2006 American Society of Clinical Oncology. DOI: 10.1200/JCO.2005.04.7159 Randomized Trial of Peer Counseling on Reproductive Health in African American Breast Cancer Survivors
From The University of Texas M.D. Anderson Cancer Center and Sisters Network Inc, National Headquarters, Houston, TX. Address reprint requests to Leslie R. Schover, PhD, Department of Behavioral Science–Unit 1330, UT M.D. Anderson Cancer Center, P.O. Box 301439, Houston, TX 77230-1439; e-mail: Lschover{at}mdanderson.org
PURPOSE: We designed a peer counseling program to improve sexual function, increase knowledge about reproductive health, and decrease menopausal symptoms and infertility-related distress for African American breast cancer survivors. PATIENTS AND METHODS: Women were randomly assigned to immediate counseling or a 3-month waitlist. Three peer counselors conducted a 3-session intervention using a detailed workbook. Questionnaires at baseline, after the waitlist period, at posttreatment, and at 3-month follow-up assessed spirituality, sexual function, menopause symptoms, emotional distress, relationship satisfaction, fertility concerns, and knowledge about reproductive health and breast cancer. At the postcounseling assessment, women rated the workbook, their counselor, and the program. RESULTS: Of 93 women screened, 60 women (65%) enrolled in the study. Women who completed counseling (80%; N = 48) had a mean age of 49 years (standard deviation [SD], 8 years) and a mean follow-up of 4.5 years (SD, 3.8 years) since cancer diagnosis. Almost all rated the workbook as very easy to understand (94%) and their counselor as very knowledgeable (96%) and very skillful (98%). Eighty-one percent rated the program as "very useful to me." Immediate counseling and waitlist groups did not differ at baseline in psychologic adjustment, nor did scores change during the waitlist period. Therefore, the groups were combined in analyzing outcomes. Knowledge of reproductive issues improved significantly from baseline to 3-month follow-up (P < .0001), as did emotional distress (P = .0047) and menopause symptoms (P = .0128). Sexually dysfunctional women became less distressed (P = .0167). CONCLUSION: Women valued the Sisters Peer Intervention in Reproductive Issues After Treatment program highly and found it relevant. The program had positive effects on knowledge and target symptoms.
Reproductive health problems are common and persistent after treatment for breast cancer. Women diagnosed before menopause are at particularly high risk for sexual dysfunction, hot flashes, and distress about interrupted childbearing.1-4 African American women have the highest risk of breast cancer before age 45 of any ethnicity,5 but only a few studies have examined the relationship of ethnicity to reproductive health after breast cancer. Surveys including mostly middle-class, well-educated, African American survivors find no major differences by ethnicity6 or even slightly superior well-being among African American survivors.7,8 In a qualitative study, African American survivors rated their sex lives as less disrupted by cancer compared with white women.9 However, African American survivors may have had lower expectations about sex, since they were less open about nudity with a partner than white women and less likely to report self-stimulation or oral sex.9 In studies of women unselected for health, with socioeconomic status (SES) controlled, African American women report higher rates of problems with sexual desire and pleasure than white women,10,11 score higher on sex guilt,12 and are less likely to engage in noncoital sexual practices.9,11,13 Findings vary on whether African American women report more menopause symptoms than women of other ethnicities.7,14,15 They cite family members as their primary information source about menopause, however, demonstrating less knowledge about the subject than white women, who learn from media sources.16 Women often experience strong distress if breast cancer interrupts childbearing.4 Although ethnicity and distress about cancer-related infertility have not been studied, in general, African American women have less access to infertility treatment.17,18 Interventions for reproductive health problems after breast cancer have rarely been studied, but a randomized trial showed relief of menopause symptoms for 76 postmenopausal breast cancer survivors, after a brief nursing intervention, compared with usual care.19 An intervention targeting African American breast cancer survivors needs to overcome barriers of distrust of the medical establishment, not only because of the infamous Tuskegee experiment,20 but also given past practices of involuntary sterilization21 and racist stereotyping of women of color as hypersexual.22 Female friends and family members are a crucial source of health information and support for African American women.16,22 Therefore, we designed an intervention training peer counselors to improve women's reproductive health after breast cancer. We partnered with the National Office of Sisters Network, Inc, an advocacy group with the mission of increasing local and national attention to the devastating impact of breast cancer in the African American community.
This project was fully approved by the institutional review board of The University of Texas M.D. Anderson Cancer Center (Houston, TX) and informed consent was elicited. The intervention was designed to improve sexual function, decrease menopausal symptoms, decrease infertility-related distress, and increase women's knowledge about reproductive health after breast cancer. Participants were assigned to immediate counseling versus 3-month waitlist, using minimization, a form of adaptive randomization,23 on the basis of current age (  45 years v > 45 years), marital status (married versus not married), menopausal status at diagnosis and currently (premenopausal versus postmenopausal). The waitlist group was intended to control for any impact of time on symptoms. Postwaitlist, women received the peer counseling intervention. We hypothesized that peer counseling would produce significant improvements in the targeted symptoms beyond any changes observed during a 3-month waitlist period.
Peer Counselors
Participants
Recruitment
Intervention Each chapter started with a list of subtopics (eg, learning more about ways to overcome vaginal dryness). The survivor rated the personal importance of each on a 4-point Likert scale to alert the counselor to the most relevant issues for each session. At the end of each chapter, topics were summarized with an index of proposed solutions (self-help options versus medical options). The counselors had a detailed treatment manual. Each participant also received a resource list including books, Web sites, crisis hotlines, and local clinics offering genetic counseling or low-cost mammography. After a participant completed the informed consent and questionnaires, she met with her counselor for three sessions, each focused on a workbook chapter. Sessions typically lasted 60 to 90 minutes and took place in the participant's home or in a convenient community setting that afforded privacy. A counseling room at our cancer center was also available. Parking or bus transportation costs were reimbursed. Counselors were matched as much as possible in age to participants.
Assessment of Outcome The questionnaires included items on demographic and medical history, the Spiritual Well-Being Subscale of the Functional Assessment of Chronic Illness Therapy (FACIT-Sp),24 the Female Sexual Function Inventory (FSFI),25 seven items from the Breast Cancer Prevention Trial Menopause Symptom Checklist that form three subscales (ie, urinary incontinence, vaginal irritation, and hot flashes) with high internal consistency as well as a summary scale,1,19 the Brief Symptom Inventory-18 (BSI-18),26 and the abbreviated form of the Dyadic Adjustment Scale27,28 for participants in committed relationships; or for women who were single, the five-item dating subscale of the Cancer Rehabilitation Evaluation System.29 Participants completed five items assessing concerns about fertility, pregnancy, and health of offspring30 and a 25-item true/false knowledge test on the basis of the workbook. Follow-up questionnaires omitted some demographic and medical items. The posttreatment packet also included a program assessment form, rating the workbook, counselor, and the content of SPIRIT. An extra, stamped return envelope was provided so women could return the assessment form anonymously, apart from their other questionnaires.
Statistical Considerations
Demographic and Medical Characteristics of Participants Ninety-three women who contacted us by telephone met entry criteria. Nine women declined to participate and 24 women did not return baseline questionnaires despite reminders. Sixty women enrolled in the study (65% of women screened). By 3-month follow-up, three women dropped out of the initial counseling group (10%) and nine women dropped out of the waitlist group (29%; P = .11, two-sided Fisher's exact test). Drop-outs had lower levels of education (44% high school degree or less v 8% of completers; Fisher's exact test, P = .02). Table 1 presents medical and demographic data for the 48 women who completed the study. Their ages at baseline ranged from 30 to 77 years (mean, 49.29 years; standard deviation [SD], 8.39 years), with 27% younger than 45 years of age and 9% younger than 40 years of age. The mean years since cancer diagnosis was 4.52 (SD, 3.84 years).
Baseline Psychosocial Adjustment of Participants Table 2 compares the SPIRIT sample's mean baseline scores with normative data on standardized questionnaires measuring psychosocial adjustment. Twenty-five percent of women met the definition of being distressed on the BSI-18 (Global Severity Index [GSI]  13) at baseline, a rate identical to that in the normative sample of female oncology patients and similar to norms for healthy women.26 Distress was twice as prevalent among women who dropped out of the study (42% v 21%). This difference was not statistically significant given the small sample size.
Women in our sample had fewer problems with hot flashes and vaginal dryness but had a similar rate of urinary symptoms compared with 72 breast cancer survivors in Ganz et al's 19 intervention for menopause symptoms. Ganz et al's group was predominantly white and had a shorter follow-up (mean, 2.5 years) than the current sample. However, the SPIRIT cohort's scores on the FSFI were worse than norms for women with multiple female sexual dysfunctions,25 especially on scales measuring sexual arousal, vaginal lubrication, and pain. At baseline, 65% of women had an FSFI total score 26.55, the cutoff indicating sexual dysfunction.25 Sexual dysfunction was equally prevalent among women who dropped out and women who completed counseling. However, women who had a sexual partner (N = 25 or 53% of the sample) reported much better sexual function (mean FSFI total, 22.89; SD, 8.35) than those without a partner (mean FSFI total, 12.13; SD, 12.20; P = .0011). Spiritual well-being as measured on the FACIT-Sp was high (mean, 54.3) even compared with norms for other breast cancer patients (mean, 39.9) or African American cancer patients (mean, 39.7).24 The FACIT-SP scores were also less variable than in normative samples. Abbreviated Dyadic Adjustment Scale scores suggested slightly below average marital satisfaction, but single women reported less impact of cancer on their comfort with dating than predominantly white breast cancer survivors.1
Comparison of Psychosocial Adjustment and Demographic Factors at Baseline in the Waitlist and Immediate Counseling Groups
Change in Psychosocial Adjustment From Baseline to Postwaitlist
Impact of the Program on Target Symptoms
Significant improvements were seen in women's knowledge about reproductive health after breast cancer (true/false test scores), overall psychologic distress (BSI-18 GSI), sexual function/satisfaction (FSFI total score), and menopause symptoms (Breast Cancer Prevention Trial Menopause Checklist). Posthoc analyses indicated that most change in emotional distress, menopause symptoms, and knowledge took place between baseline and postcounseling. The hot flash subscale accounted for the improvement in overall menopause symptoms (P = .0033).
We divided the sample into women initially distressed on the BSI-18 (GSI
Although statistically significant improvement occurred on the FSFI between the posttreatment and 3-month follow-up assessments (Tukey-Kramer Test, P = .0230), the clinical significance is questionable given the small effect size (0.14). Figure 2 illustrates that women dysfunctional at baseline did not improve over time.
Although distress about infertility did not decrease significantly, most of the six items were not of concern to our SPIRIT participants. The one item frequently endorsed (33%) was worry that children would have an increased lifetime cancer risk. Only five women (10%) had wanted children at the time of cancer diagnosis and four of them reported that having cancer decreased their desire for children. The average current age of these five women was 36.00 years (SD, 5.43 years) compared with 51.08 years (SD, 7.56 years) for all other participants.
Correlates of Positive Outcomes
Program Assessment All 48 participants completed program assessment forms. Eighty-one percent rated the SPIRIT program as "very useful to me," and 19% as "somewhat useful to me." The workbook was rated as very easy to understand by 94%. Workbook content was rated as "very useful to me" by 85%, and 90% rated the workbook as doing a very good job of covering special concerns of African American breast cancer survivors. All three counselors received high marks, with 96% of women rating their counselor as knowing the facts very well and 98% as very skillful. Ninety-four percent of women rated the meeting place for the program as very convenient. Half of women had sessions at their homes, 19% at our institution, and 44% at another location (some women's sessions occurred in more than one location). Seventy-five percent of participants felt that three sessions was about right; 19% would have preferred fewer sessions; and 6% would have liked more. All but one participant had talked to others about things she learned in SPIRIT.
The SPIRIT intervention improved hot flashes, increased women's knowledge about reproductive health issues after breast cancer, and was highly rated in terms of its relevance and impact. An unanticipated benefit was the improvement over time in emotional distress. Women, who were sexually dysfunctional at baseline, had high distress about loss of fertility, and lower levels of spiritual well-being, had more improvement in distress levels. This study adds to a growing body of evidence that brief, structured, psychoeducational counseling can enhance quality of life after cancer.19,32-34 Emotional support from peers is not enough to produce better outcomes, however. The intervention needs to give health information and to teach coping skills.32,33 The higher prevalence of sexual dysfunction than in predominantly white breast cancer survivors1,2,7,19 may reflect the fact that 47% of our participants lacked a current sexual partner, resulting in lower FSFI scores. However, counselors reported that the open discussions of sexuality in the sessions were unique in their experience. Even if sexual satisfaction did not change, women who were sexually dysfunctional at baseline improved in overall psychologic distress, perhaps reflecting increased knowledge and self-efficacy. Hot flashes have been linked to anxiety,14,15 but GSI scores in this sample did not correlate with menopause symptoms. Women did experience a significant decrease in vasomotor symptoms, perhaps as a result of implementing strategies in the workbook such as seeking medication or trying relaxation techniques. As is typical for interventions utilizing peer support along with educational materials, we do not know which component is responsible for the positive outcomes.31,33,35 Given the expense and effort in training peer counselors and maintaining quality control, we would like to identify whether it is primarily new information that is beneficial, or if the empathy and support of the peer counselor enhances the intervention's success. We are currently conducting a national, randomized trial comparing the full three-session program including workbook versus workbook plus a phone card that gives participants the option of calling their counselor for up to 30 minutes. The minimal contact condition is similar to services provided by advocacy groups, ie, written brochures and a phone hotline. Recruiting truly underserved participants remains a challenge, even when partnering with a community-based organization like Sisters Network, Inc. The limited sample size and high SES of our participants may have masked any impact of SES on outcome. It also appeared that women who dropped out, especially during the waitlist period, had less education and higher distress, further reducing our sample's diversity. The length of the baseline questionnaire and sexual nature of some items reportedly discouraged many women screened by phone from entering the study. Given that we had our best outcomes with more distressed women, the significance of our findings may have been attenuated. If the SPIRIT program was implemented in the community without outcome assessment, however, questionnaires would not be a barrier to uptake. In conclusion, our pilot data suggest that women value the SPIRIT program highly and find it personally useful. Their knowledge of reproductive health increases significantly and gains are maintained at 3 months. The program also decreased hot flashes and overall emotional distress. The specific aspects of the intervention contributing to these gains remain to be defined.
The authors indicated no potential conflicts of interest.
We thank our peer counselors for their dedication and contributions: Helen George, Laereen Lyght, and Janice Workcuff.
Supported by two grants from the National Cancer Institute, Bethesda, MD, Grants No. 3P30 CA16672-2652 Cancer Center Supplement Grant and No. 1RO1 CA 102097-01 SPIRIT: Reproductive Counseling for Breast Cancer. Presented in part at the 9th Biennial Symposium on Minorities, the Medically Underserved, and Cancer, Intercultural Cancer Council, Washington, DC, March 24-28, 2004. Authors' disclosures of potential conflicts of interest and author contributions are found at the end of this article.
This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2006 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
2 or Fisher's exact tests for categoric variables and one-way analysis of variance or nonparametric tests for continuous variables. The only significant finding was that women in the waitlist group had higher incomes (P = .03). Income was used as a covariate in further analyses of group differences. 
