Female sexual dysfunction is very common after cancer treatment. Two-thirds of the 7 million female cancer survivors in the United States were treated for breast, gynecologic, bladder, or colorectal malignancies,1 and at least 50% experience long-term, severe sexual problems.2-4 The most common dysfunctions are vaginal dryness, pain, and decreased sexual desire.2,5,6 The risk of sexual dysfunction is increased by abrupt ovarian failure,5,6 severe vaginal atrophy from using aromatase inhibitors,7 direct genital damage from pelvic radiation therapy,8-10 and genital graft-versus-host disease.11 Urinary and fecal incontinence often lead to avoidance of sexual contact.12
The recent NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines) for Survivorship advocate systematic assessment of female sexual function and referral for multidisciplinary treatment (to view the most recent version of these guidelines, visit NCCN. org).13 Unfortunately, women with cancer report few satisfying discussions about sexuality.14,15 Fewer than 20% of sexually dysfunctional women treated for cancer seek professional help,3,16,17 and distress over sexual dysfunction ranks high in surveys of unmet needs of cancer survivors.18,19 Only a few gynecologists and mental health professionals have expertise in managing relevant physical symptoms of sexual dysfunction20,21 or in providing evidence-based cognitive behavioral treatment.22-24 Furthermore, insurance coverage is poor, especially for mental health services.
An Internet-based intervention may be a cost-effective way for oncology settings to comply with the new guidelines. The authors recently showed that an Internet-based intervention for couples after prostate cancer treatment using e-mail contact with a therapist was as effective in improving sexual function as 3 in-person sessions of cognitive-behavioral therapy.25 Pilot studies with Internet-based interventions for female sexual dysfunction have shown promise for healthy women26 and in gynecologic cancer survivors.27,28 The authors created a Web site, Tendrils: Sexual Renewal for Women After Cancer, and tested a prototype in a randomized trial, comparing use on a self-help basis or supplemented with sexual counseling. The authors hypothesized that both groups would improve on self-report measures of sexual function and satisfaction, but that the counseled group would have a significantly larger gain.
Materials and Methods
The research protocol, including recruitment materials and Web site content, was approved by The University of Texas MD Anderson Institutional Review Board. All participants provided informed consent. No adverse events were reported. Eligible women were 1 to 7 years postdiagnosis of localized breast or gynecologic cancer, and off active treatment other than hormonal therapy. They scored as sexually dysfunctional (<26.5) on the Female Sexual Function Index (FSFI),29 had been in a sexual relationship for at least 6 months, and had a partner willing to participate in behavioral homework. They lived close enough to attend 3 in-person counseling sessions, could read English, and had Internet access.
Recruitment
The study recruited for 16 months, and introductory letters and flyers were sent to 1123 women from the university’s tumor registry who met eligibility criteria for cancer type, stage, and date of diagnosis. Flyers were given to the breast and gynecologic outpatient clinics, and some women were approached during outpatient clinic appointments. The study was also listed on ClinicalTrials.gov. Of 117 women screened for eligibility, 22 (19%) declined participation and 23 (20%) were ineligible.
Study Design
All women used the Web site for a 12-week treatment period. Half were adaptively randomized, using minimization,30 to have 3 supplemental in-person counseling sessions. Minimization balanced treatment groups on the following factors: education (≥4-year college degree vs no college degree), age (≤49 years vs ≥50 years), current menopausal status, and cancer site (breast vs gynecologic).
Women completed questionnaires on the Web site at baseline, at the end of treatment, and at 3- and 6-month follow-up. Participants received a $20 gift card on completing questionnaires at each follow-up. Items assessed background and medical history. The FSFI was the primary outcome measure.29 The FSFI is a 19-item, multiple-choice questionnaire with excellent internal consistency, discriminant validity, and test-retest reliability, and has been validated with female patients with cancer.31 Subscales measure sexual desire, arousal, lubrication, orgasm, satisfaction, and pain. The total score reflects both function and satisfaction. One limitation is that scores are negatively biased if women have not been sexually active with a partner in the past 4 weeks.31 The authors also included the Menopausal Sexual Interest Questionnaire (MSIQ), a 10-item scale with excellent internal consistency and test-retest reliability, with subscales measuring desire, responsiveness (pleasure and orgasm), and satisfaction.32 The Brief Symptom Inventory-18 (BSI-18) assessed emotional distress with a Global Severity Index (GSI) summary score.33 Norms are available for community samples and oncology patients. The Quality of Life in Adult Cancer Survivors (QLACS) scale yielded a summary score from its 47 items measuring global quality of life.34,35 At posttreatment, women rated the Tendrils program on 12 Likert scales.
Description of the Intervention
The password-protected Tendrils Web site included text, graphics, animations, and multicultural photographs and clipart. Instructions suggested an order for using the site, but women could navigate from the home page to sections describing the sexual and fertility consequences of their type of cancer and treatment; genital anatomy, including an interactive, vulvar self-portrait with pain and pleasure mapping; sex after menopause; managing vaginal dryness and pain (with detailed advice on vaginal moisturizers, lubricants, pelvic floor exercises, and dilators); causes and treatment options for loss of desire or orgasm problems; ways to improve body image; resuming sex comfortably using sensate focus exercises; sexual issues related to ostomies or incontinence; communication with sexual partners and health professionals; dating; lesbian relationships; and sex after childhood and adolescent cancer. Videos included 11 interviews with women cancer survivors and vignettes played by actors illustrating common problems and coping strategies.
A therapist manual provided overall guidance and content checklists for each of the 3 counseling sessions. Two master’s-level mental health professionals provided counseling and were supervised weekly by the first author (LRS). Counselors guided women through the Web site and discussed behavioral homework.
Statistical Analyses
Demographic and clinical characteristics were summarized with means, standard deviations, ranges, and frequencies, and compared between intervention groups by Fisher test, t test, or a Wilcoxon rank sum test depending on the data distribution. Questionnaires were not scored if more than an allowed number of items were missing. Linear mixed models (LMM) were conducted to assess within and between group score changes over time for each outcome.36 LMM is widely used in analyzing correlated longitudinal data because it accounts for missing data through incorporating random effects characterizing heterogeneity among subjects. Each outcome score was regressed onto time period, treatment group, and a time by treatment interaction. Post hoc analyses evaluated the changes across time points within and between groups. Similar LMM models analyzed the relationship of Web site use and outcomes.
Results
Attrition Over Time
Figure 1 summarizes the number of women who entered the study and attrition. Seventy-two women provided informed consent and were minimized to a treatment group. However, 14% in the self-help group and 25% in the counseled group dropped out without completing baseline questionnaires (Fisher exact test; P=.372). For the 58 women who completed the baseline, attrition during the treatment period was similar in both groups (22%), but 34% did not complete 6-month follow-up questionnaires. To determine whether dropping out (defined as the last point at which a patient was missing data for all 4 questionnaires) was associated with specific participant characteristics, a discrete survival time model was conducted with terms for treatment group, each baseline questionnaire score, age, cancer site (breast vs gynecologic), education (≥4-year college degree vs no college degree), and time points. The area under the receiver operating characteristic curve was 0.885. The only significant factor was that younger women were more likely to drop out (odds ratio, 0.91; P=.034). Even though groups were balanced with regard to age, it was used as a covariate in outcome analyses.
Demographic and Medical Factors
Table 1 presents demographic and medical factors characterizing the 58 women who completed baseline questionnaires. The self-help and counseled groups did not differ significantly on any variable. The sample was reasonably diverse (21% minority ethnicity) but well educated; 80% were treated for breast cancer.
Impact of Intervention on Questionnaire Scores
Our hypothesis was confirmed for sexual outcomes. When groups were combined, gains were significant from baseline to posttreatment on the FSFI (effect, 3.41; P<.001) and MSIQ (effect, 6.54; P<.001). Table 2 summarizes the results of LMMs analyzing within-group effects for total scores on each questionnaire. Figure 2 illustrates trends over time graphically for the 2 groups on each outcome measure. At posttreatment, total FSFI scores improved significantly (P<.001) for the counseled group, with a trend (P=.054) seen in the self-help group (between-group difference; P=.024). Although gains remained significant at 6-month follow-up, most women did not attain the 26.6 score considered to mark “normal sexual function.”29 The authors also examined FSFI results excluding women at each assessment who were not sexually active.31 Although 40 fewer scores were included, the LMM analysis again revealed a significant treatment effect in the counseling group that was maintained over time. Figures 2A and B show FSFI total scores across time for all women versus only sexually active women.
Attrition across time.
Citation: Journal of the National Comprehensive Cancer Network J Natl Compr Canc Netw 11, 11; 10.6004/jnccn.2013.0162
Demographic and Medical Characteristics of the Sample
Summary of Linear Mixed Model Analyses of Outcomes Within Treatment Groups Across Time
For the MSIQ, within-group treatment effects were highly significant for counseled women (P<.001) but fell short of significance for the self-help group (P=.082; between-group difference, P=.011). However, improvement for counseled women regressed at 6-month follow-up, whereas the self-help group improved slightly over time.
On nonsexual outcomes, distress (BSI-18 GSI scores) improved significantly across time (effect, -2.96; P=.001), as did the QLACS total score (effect, -13.73; P<.001). Table 2 shows that changes at posttreatment were only significant within the self-help group, however (GSI, P=.011; QLACS summary score, P=.008), with gains maintained at 6 months.
Use of the Web Site
Web site use was electronically recorded, excluding time spent completing questionnaires. For each participant, total minutes of use were calculated during the 12-week treatment period, between posttreatment and 6-month follow-up assessments, and across the entire study period. The treatment groups did not differ significantly on use during the treatment period (self-help mean [SD], 108.6 [141.9]; counseled, 143.4 [134.8]). Use across the entire study was very similar between groups, with the mean (SD) for the combined sample at 149.0 (157.1). However, between posttreatment and 6-month follow-up, the self-help group spent significantly more minutes on the Web site (self-help, 38.6 [60.9]; counseled, 7.6 [17.7]; P=.015).
The authors performed linear regression analysis to determine whether more time on the site was associated with better outcomes on the FSFI and MSIQ. Models included terms for baseline or posttreatment score, use time, intervention group, and their interaction. Use time during a particular period was not associated with improvement on the FSFI or MSIQ. However, a trend was seen (P=.065) for use time across the entire study period to be associated with improvement in the 6-month MSIQ. Greater Web site use in the self-help group posttreatment was also consistent with the improvement seen in their scores during that time, in contrast to backsliding seen in the counseling group.
Posttreatment Program Ratings
Table 3 presents mean ratings by treatment group for 12 aspects of the intervention program. Ratings were in the positive range on all scales. The only significant difference between groups was that counseled women rated the intervention more positively on addressing emotional concerns. Women rated the counselors, in particular, as helpful and empathic.
(A-E) Mean outcomes scores at each assessment point for all treatment groups.
Abbreviations: BSI-18 GSI, Brief Symptom Inventory-18 Global Severity Index; FSFI, Female Sexual Function Index; FU, follow-up; MSIQ, Menopausal Sexual Interest Questionnaire; QLACS, Quality of Life for Adult Cancer Survivors.
Citation: Journal of the National Comprehensive Cancer Network J Natl Compr Canc Netw 11, 11; 10.6004/jnccn.2013.0162
Discussion
This randomized trial suggests that an Internet-based intervention can significantly improve sexual function and satisfaction in women with sexual dysfunction several years after treatment for breast or gynecologic cancer. Although supplemental in-person sexual counseling was associated with larger improvements during the treatment period, women in the self-help group were more likely to persist in using the Web site during the next 6 months. Their outcome measures improved slowly during that period, in contrast with some back-sliding for counseled women, particularly on the MSIQ. The intervention also reduced emotional distress and improved ratings of overall quality of life at posttreatment, particularly within the self-help group. Younger women were more likely to drop out of the study. Although younger patients with cancer are more emotionally distressed,37 baseline BSI-18 was not predictive of dropping out. Cancer disrupts more life roles for younger women, which may have interfered with taking time for the intervention.37 Because of the sample size, the authors had limited power to identify demographic or medical factors that may influence the efficacy of the intervention.
Questionnaire scores indicate that our participants had severe and pervasive sexual problems, even compared to similar cohorts.29,31,38,39 Although sexual function improved, most women did not achieve criteria for normalcy on the FSFI or MSIQ. However, the magnitude of improvement in the counseled group was similar to results associated with a 3-session, individual intervention using cognitive-behavioral sex therapy techniques with 31 survivors of gynecologic cancer.38 As in that study, some backsliding occurred by 6 months, suggesting a need for relapse prevention.
Future research should focus on facilitating adoption of the intervention. Accrual for this study was disappointing. The intervention was difficult to publicize in a large cancer center. Sending letters to potential women was the most successful recruitment strategy, but more than half of women older than 50 years are sexually inactive.40 With the proliferation of patient advocacy groups on the Internet social media may help publicize the intervention within a cancer center and in the community.41 The choice of an inperson counseling format may also have discouraged some women who expressed reluctance to come to the cancer center for extra appointments, probably accounting for the greater drop-out rate in the counseled group after women found out their assigned treatment arm. Counseling may be more appealing if delivered by phone42 or e-mail,25 or through bulletin board27,28 or real-time43 Internet groups.
Attrition also needs to improve. Some women do not want to complete questionnaires because of the time involved or the sensitive nature of some items, accounting for drop-outs among women who gave informed consent but then failed to complete the baseline questionnaires, and women who finished the intervention but never completed a follow-up.44 Nonadherence rates of 15% to 30% are common in randomized trials of interventions targeting psychological problems, with little evidence that attrition is worse in Internet-based formats.44,45 The next version of Tendrils will require fewer questionnaires and will use them interactively, prompting users to set short-term goals and track their progress with self-report instruments.45
Web site usability also may have been a limitation. Despite overall positive evaluation ratings, the authors used a beta version of software that, when encrypted, no longer allowed searches using keywords or allowed direct links from one part of the text to another. Usability testing at the NCI laboratory revealed that the home page and navigation need reorganization. Text must be presented in shorter sections using more bullet lists.
Posttreatment Evaluation Ratings of the Intervention Programa
Future research needs to identify the best formats for supplemental counseling. Adding human support enhances adherence to a range of eHealth interventions aimed at changing behavior, with self-guided interventions helping 15% or fewer of those who try them.45 However, low-intensity Internet interventions attract people who do not seek professional help, sometimes stimulating them to get care.46 In future effectiveness trials, the authors intend to test a sequential multiple assignment randomized trial (SMART) model,47 starting with self-help use and adding medical referrals or more-intensive behavioral treatment on patient request, when women fail to use the Web site, or when outcome scores are poor. Although the Web site is designed for patients, oncology health professionals can learn from it and use the therapist manual to become comfortable providing assessment, brief counseling, and referrals for multidisciplinary care. More-extensive professional education can also be developed to implement the sexuality survivorship guideline.
This research was funded by a grant from the National Cancer Institute, CA4R4212932 (Schover, Principal Investigator) and was also supported in part by The University of Texas MD Anderson Cancer Center Support Grant, CA016672. Drs. Schover and Martinetti may receive compensation from future commercialization of the intervention. Preliminary results were presented at the 6th Biennial Cancer Survivorship Research Conference, June, 2012, Arlington, VA. The remaining authors have disclosed that they have no financial interests, arrangements, affiliations, or commercial interests with the manufacturers of any products discussed in this article or their competitors.
References
- 1.↑
American Cancer Society. Cancer Treatment and Survivorship Facts and Figures, 2012-2013. Atlanta, GA: American Cancer Society; 2012.
- 2.↑
Sadovsky R, Basson R, Krychman M et al.. Cancer and sexual problems. J Sex Med 2010;7(Suppl 1, Pt 2):S349–373.
- 3.↑
Hill EK, Sandbo S, Abramsohn E et al.. Assessing gynecologic and breast cancer survivors’ sexual health care needs. Cancer 2011;117:2643–2651.
- 4.↑
Kedde H, van de Wiel HB, Weijmar Schultz WC et al.. Sexual dysfunction in young women with breast cancer. Support Care Cancer 2013;21:271–280.
- 5.↑
Schover LR. Premature ovarian failure and its consequences: vasomotor symptoms, sexuality, and fertility. J Clin Oncol 2008;26:753–758.
- 6.↑
Carter J, Goldfrank D, Schover LR. Simple strategies for vaginal health promotion in cancer survivors. J Sex Med 2011;8:549–559.
- 7.↑
Baumgart J, Nilsson K, Evers AS et al.. Sexual dysfunction in women on adjuvant endocrine therapy after breast cancer. Menopause 2013;20:162–168.
- 8.↑
Lind H, Waldenström AC, Dunberger G et al.. Late symptoms in long-term gynaecological cancer survivors after radiation therapy: a population-based cohort study. Br J Cancer 2011;104:737–745.
- 9.
Milbury K, Cohen L, Jenkins R et al.. The association between psychosocial and medical factors with long-term sexual dysfunction after treatment for colorectal cancer. Support Care Cancer 2013;21:793–802.
- 10.↑
Provencher S, Oehler C, Lavertu S et al.. Quality of life and tumor control after short split-course chemoradiation for anal canal carcinoma. Radiat Oncol 2010;5:41–49.
- 11.↑
Hirsch P, Leclerc M, Rybojad M et al.. Female genital chronic graft-versus-host disease: importance of early diagnosis to avoid severe complications. Transplantation 2012;93:1265–1269.
- 12.↑
Dunberger G, Lind H, Steineck G et al.. Self-reported symptoms of faecal incontinence among long-term gynaecological cancer survivors and population-based controls. Eur J Cancer 2010;46:606–615.
- 14.↑
Scanlon M, Blaes A, Geller M et al.. Patient satisfaction with physician discussions of treatment impact on fertility, menopause and sexual health among pre-menopausal women with cancer. J Cancer 2012;3:217–225.
- 15.↑
Flynn KE, Reese JB, Jeffery D et al.. Patient experiences with communication and sex during and after treatment of cancer. Psychooncol 2012;21:594–601.
- 16.↑
Shifren JL, Johannes CB, Monz BU et al.. Help-seeking behavior of women with self-reported distressing sexual problems. J Womens Health (Larchmt) 2009;18:461–468.
- 17.↑
Huyghe E, Sui D, Odensky E et al.. Needs assessment survey to justify establishing a reproductive health clinic at a comprehensive cancer center. J Sex Med 2009;6:149–163.
- 18.↑
Holm LV, Hansen DG, Johansen C et al.. Participation in cancer rehabilitation and unmet needs: a population-based cohort study. Support Care Cancer 2012;20:2913–2924.
- 19.↑
Zebrack BJ, Block R, Hayes-Lattin B et al.. Psychosocial service use and unmet need among recently diagnosed adolescent and young adult cancer patients. Cancer 2013;119:201–214.
- 20.↑
Sobecki JN, Curlin FA, Rasinski KA et al.. What we don’t talk about when we don’t talk about sex: results of a national survey of U.S. obstetrician/gynecologists. J Sex Med 2012;9:1285–1294.
- 21.↑
Goldfarb SB, Abramsohn E, Andersen BL et al.. A national network to advance the field of cancer and female sexuality. J Sex Med 2013;10:319–325.
- 22.↑
Scott JL, Kayser K. A review of couple-based interventions for enhancing women’s sexual adjustment and body image after cancer. Cancer J 2009;15:48–56.
- 23.
Brotto LA, Yule M, Breckon E. Psychological interventions for the sexual sequelae of cancer: a review of the literature. J Cancer Surviv 2010;4:346–360.
- 24.↑
Taylor SA, Harley C, Ziegler L et al.. Interventions for sexual problems following treatment for breast cancer: a systematic review. Breast Cancer Res Treat 2011;130:711–724.
- 25.↑
Schover LR, Canada AL, Yuan Y et al.. A randomized trial of internet-based versus traditional sexual counseling for couples after localized prostate cancer treatment. Cancer 2012;118:500–509.
- 26.↑
Jones LM, McCabe MP. The effectiveness of an internet-based psychological treatment program for female sexual dysfunction. J Sex Med 2011;8:2781–2792.
- 27.↑
Classen CC, Chivers ML, Urowitz S et al.. Psychosexual distress in women with gynecologic cancer: a feasibility study of an online support group. Psychooncol 2013;22:930–935.
- 28.↑
Wiljer D, Urowitz S, Barbera L et al.. A qualitative study of an internet-based support group for women with sexual distress due to gynecologic cancer. J Cancer Educ 2011;26:451–458.
- 29.↑
Wiegel M, Meston C, Rosen R. The female sexual function index (FSFI): cross-validation and development of clinical cutoff scores. J Sex Marital Ther 2005;31:1–20.
- 30.↑
Perry M, Faes M, Reelick MF et al.. Study wise minimization: a treatment allocation method that improves balance among treatment groups and makes allocation unpredictable. J Clin Epidemiol 2010;63:1118–1122.
- 31.↑
Baser RE, Li Y, Carter J. Psychometric validation of the Female Sexual Function Index (FSFI) in cancer survivors. Cancer 2012;118:4606–4618.
- 32.↑
Rosen RC, Lobo RA, Block BA et al.. Menopausal Sexual Interest Questionnaire (MSIQ): a unidimensional scale for the assessment of sexual interest in postmenopausal women. J Sex Marital Ther 2004;30:235–250.
- 33.↑
Zabora J, Brintzenhofe-Szoc K, Jacobsen P et al.. A new psychosocial screening instrument for use with cancer patients. Psychosomatics 2001;42:241–246.
- 34.↑
Avis NE, Smith KW, McGraw S et al.. Assessing quality of life in adult cancer survivors (QLACS). Qual Life Res 2005;14:1007–1023.
- 35.↑
Avis NE, Ip E, Foley KL. Evaluation of the Quality of Life in Adult Cancer Survivors (QLACS) scale for long-term cancer survivors in a sample of breast cancer survivors. Health Qual Life Outcomes 2006;4:92–103.
- 36.↑
Stroup WW. Generalized Linear Mixed Models: Modern Concepts, Methods and Applications. Boca Raton, FL: CRC Press; 2012.
- 37.↑
Avis NE, Levine B, Naughton MJ et al.. Explaining age-related differences in depression following breast cancer diagnosis and treatment. Breast Cancer Res Treat 2012;136:581–591.
- 38.↑
Brotto LA, Erskine Y, Carey M et al.. A brief mindfulness-based cognitive behavioral intervention improves sexual functioning versus wait-list control in women treated for gynecologic cancer. Gynecol Oncol 2012;125:320–325.
- 39.↑
Schover LR, Rhodes MM, Baum G et al.. SPIRIT: a peer counseling program to improve reproductive health among African American breast cancer survivors. Cancer 2011;117:4983–4992.
- 40.↑
Schick V, Herbenick D, Reece M et al.. Sexual behaviors, condom use, and sexual health of Americans over 50: implications for sexual health promotion for older adults. J Sex Med 2010;7(Suppl 5):315–329.
- 41.↑
Morgan AJ, Jorm AF, Mackinnon AJ. Internet-based recruitment to a depression prevention intervention: lessons from the Mood Memos Study. J Med Internet Res 2013;15:e31.
- 42.↑
Marcus AC, Garrett KM, Cella D et al.. Can telephone counseling post-treatment improve psychosocial outcomes among early stage breast cancer survivors? Psychooncol 2010;19:923–932.
- 43.↑
Stephen JE, Christie G, Flood K et al.. Facilitating online support groups for cancer patients: the learning experience of psychooncology clinicians. Psychooncol 2011;20:832–840.
- 45.↑
Mohr DC, Guijpers P, Lehman K. Supportive accountability: a model for providing human support to enhance adherence to eHealth interventions. J Med Internet Res 2011;13:e30.
- 46.↑
McKellar J, Austin J, Moos R. Building the first step: a review of low-intensity interventions for stepped care. Addict Sci Clin Pract 2012;7:26.
- 47.↑
Nahum-Shani I, Qian M, Almirall D et al.. Experimental design and primary data analysis methods for comparing adaptive interventions. Psychol Methods 2012;17:457–477.
