Scolaris Content Display Scolaris Content Display

Cochrane Database of Systematic Reviews Review - Intervention

Interventions for sexual dysfunction following treatments for cancer in women

Authors' declarations of interest

Version published: 02 February 2016 Version history

https://doi.org/10.1002/14651858.CD005540.pub3
Collapse all Expand all

Abstract

available in

Background

The proportion of people living with and surviving cancer is growing. This has led to increased awareness of the importance of quality of life, including sexual function, in those affected by cancer. Sexual dysfunction is a potential long‐term complication of many cancer treatments. This includes treatments that have a direct impact on the pelvic area and genitals, and also treatments that have a more generalised (systemic) impact on sexual function.

This is an update of the original Cochrane review published in Issue 4, 2007, on interventions for treating sexual dysfunction following treatments for cancer for men and women. Since publication in 2007, there has been an increase in the number of trials for both men and women and this current review critiques only those for women. A review in press will present those for men.

Objectives

To evaluate the effectiveness of interventions for treating sexual dysfunction in women following treatments for cancer. To assess adverse events associated with interventions.

Search methods

We searched the Cochrane Central Register of Controlled Trials (CENTRAL 2015, Issue 9), MEDLINE, EMBASE, PsycINFO, AMED, CINAHL, Dissertation Abstracts and the NHS Research Register. The searches were originally run in January 2007 and we updated these to September 2015.

Selection criteria

We included randomised controlled trials (RCTs) that assessed the effectiveness of a treatment for sexual dysfunction. The trial participants were women who had developed sexual dysfunction as a consequence of a cancer treatment. We sought evaluations of interventions that were pharmaceutical, mechanical, psychotherapeutic, complementary or that involved physical exercise.

Data collection and analysis

Two review authors independently extracted the data and assessed trial quality. We considered meta‐analysis for trials with comparable key characteristics.

Main results

Since the original version of this review we have identified 11 new studies in women. The one study identified in the earlier version of this review was excluded in this update as it did not meet our narrower inclusion criteria to include only interventions for the treatment, not prevention, of sexual dysfunction.

In total 1509 female participants were randomised across 11 trials. All trials explored interventions following treatment either for gynaecological or breast cancer. Eight trials evaluated a psychotherapeutic or psycho‐educational intervention. Two trials evaluated a pharmaceutical intervention and one pelvic floor exercises. All involved heterosexual women. Eight studies were at a high risk of bias as they involved a sample of fewer than 50 participants per trial arm. The trials varied not only in intervention content but in outcome measurements, thereby restricting combined analysis. In the trials evaluating a psychotherapeutic intervention the effect on sexual dysfunction was mixed; in three trials benefit was found for some measures of sexual function and in five trials no benefit was found. Evidence from the other three trials, two on different pharmaceutical applications and one on exercise, differed and was limited by small sample sizes. Only the trial of a pH‐balanced vaginal gel found significant improvements in sexual function. The trials of pharmaceutical interventions measured harm: neither reported any. Only one psychological intervention trial reported that no harm occurred because of the intervention; the other trials of psychological support did not measure harm.

Authors' conclusions

Since the last version of this review, the new studies do not provide clear information on the impact of interventions for sexual dysfunction following treatments for cancer in women. The sexual dysfunction interventions in this review are not representative of the range that is available for women, or of the wider range of cancers in which treatments are known to increase the risk of sexual problems. Further evaluations are needed.

PICOs

Population
Intervention
Comparison
Outcome

The PICO model is widely used and taught in evidence-based health care as a strategy for formulating questions and search strategies and for characterizing clinical studies or meta-analyses. PICO stands for four different potential components of a clinical question: Patient, Population or Problem; Intervention; Comparison; Outcome.

See more on using PICO in the Cochrane Handbook.

Plain language summary

available in

Interventions for sexual dysfunction following treatments for cancer

Background

In women sexual dysfunction is a potential complication of many types of cancer treatments.

This review evaluated the effectiveness of treatments (interventions) of any kind, for example drugs or exercise, for treating sexual dysfunction in women following cancer treatment.

The review is an update of one published in 2007 that assessed the effectiveness of interventions for men and women. We decided to present this revised review separately for women because of the increase in the number of trials. Another review for men is underway.

Study characteristics

We identified 11 new trials on interventions for women in September 2015. We excluded one trial that was included in the earlier version of this review because it assessed treatment for preventing sexual dysfunction and was no longer relevant to this review. Interventions differed in their content and how the researchers measured benefit. Eight of the interventions involved psychological support such as counselling on sexual matters, or peer support. One of the others was of a testosterone cream, another tested a vaginal pH‐balanced gel and the other was of pelvic floor exercise. The findings from six of the trials are weak because they involved small numbers of women.

Key results

Across the trials the impact on sexual function was different. This makes it difficult to derive clear conclusions. For instance, in those that evaluated a psychological support treatment, four studies found that it improved some measures of sexual function but not others, but five found that it did not improve sexual function according to any of the measures used. For the other interventions tested, only the trial of the vaginal gel found improvements in sexual function and no side effects were reported. Only one of the psychological interventions reported that no harm occurred because of the intervention. The other trials of psychological support did not assess harm. This is an important gap as some women may find it distressing to discuss personal sexual problems as part of their treatment.

Further evaluations are needed for all interventions. Current studies have only explored effectiveness in women with gynaecological and breast cancers, but there is a risk of sexual problems after treatments for other cancers. New evaluations need to involve larger numbers of participants.

Authors' conclusions

available in

Implications for practice

This update found insufficient evidence for the effectiveness of topical pharmacological treatments, psychotherapeutic interventions and pelvic floor exercises. Apart from the pharmacological treatment we found limited evaluations of any potential harms these interventions may incur.

Implications for research

Improved diagnosis and treatments for cancer mean that more people will be living with and surviving cancer, and of these a significant proportion may have sexual dysfunction. Further evidence is needed on treatments for sexual dysfunction from high‐quality trials with large samples that fully report key methodological characteristics and harms. It is difficult to suggest which interventions might be most useful for further evaluation, as the trials we evaluated were at risk of biased results. Across the trials there was also minimal overlap in how the intervention was evaluated. Future trials need to consider preparatory stages in intervention development, to increase appropriateness and need, such as in the case of psychotherapeutic interventions, for example those recommended in the Medical Research Council (MRC) guidelines on developing and evaluating complex interventions (Craig 2008). They also need to fully report details of the intervention. Consideration is also needed of the choice of outcome measures that are appropriate to the patient population and the nature of the intervention, such as measures of sexual function, quality of life and mental health.

The trials evaluated in this review varied in design; some included a comparative arm with no additional support for sexual dysfunction and others compared the intervention arm with an active control. Future research should consider the appropriateness of a control group that is given no support despite sexual dysfunction. Perhaps at the very least trialists should consider using a comparative arm for which the provision of usual care for sexual dysfunction is provided.

Sexual dysfunction is difficult to treat in this population as it may result from multiple causes, not all of which are a direct outcome of the cancer treatment. These causes include the emotional stress of having the diagnosis and its impact on daily living, as well as natural aging. Thus it may be a particularly difficult disorder to treat.

Background

available in

This review is an update of a previously published review in the Cochrane Database of Systematic Reviews (Issue 4, 2007) on 'Interventions for sexual dysfunction following treatments for cancer' (Miles 2007). This review presents the findings on the effectiveness and safety of interventions for sexual dysfunction in women following treatments for cancer. In its earlier version the review involved interventions for both men and women following treatments for cancer, and preventative treatments for sexual dysfunction. A separate review, in press, will report on interventions for men. Interventions for the prevention of sexual dysfunction are not included in either of these reviews.

Description of the condition

Sexual dysfunctions are characterised by a disturbance in sexual desire and in the psychological and physiological changes that characterise the sexual response (Diagnostic and Statistical Manual of Mental Disorders 5th edition (DSM‐5) (DSM‐5 2013)).

The DSM‐5 recognises three sexual dysfunctions in women:

  • sexual interest/arousal disorder;

  • genitopelvic pain/penetration disorder;

  • female orgasmic disorder.

Sexual dysfunction poses challenges to one's social, mental, emotional and physical wellbeing. In cancer, it is an important indicator of quality of life both in people undergoing treatments and in those surviving the disease (Ozyilkan 1995). It may occur as a result of the nature of a cancer, such as a cancer that affects the mechanics or hormonal pathways in sexual function, as a result of the treatment for cancer such as surgery, chemotherapy or radiotherapy that may have local or systemic and hormonal effects, and it may occur from the emotional trauma of receiving a diagnosis or indeed living with cancer for the patient as well as their sexual partner. The causes of sexual dysfunction following cancer treatments may be multi‐factorial but it may often result from the direct effects of the treatment. A hysterectomy, for example, may disrupt the pelvic anatomy and the local nerve supplying the pelvic floor, which are part of the trunk stability mechanism involved in sexual arousal, and this may be accompanied by loss of vaginal lubrication and sensation (Jackson 2006). Other cancer treatments may affect sexual function, for example chemotherapy may induce the symptoms of menopause, in particular vaginal dryness. Sexual dysfunction following a cancer treatment may also relate to lowered physical fitness, symptoms of depression or anxiety, and changes to self esteem and feelings of sexuality (Greenwald 2008; Pignata 2001; Stead 2007; Tabano 2002; Wiggins 2008). Although some treatment‐related sexual dysfunction effects are short‐term, sexual dysfunction is one of the more prevalent long‐term complications following many types of life‐saving cancer treatment (Ananth 2003; Ganz 1998; Syrjala 1998).

Sexual dysfunction following cancer treatments is becoming an increasingly important issue to tackle. With an aging global population, improved diagnosis and better cancer treatments, more people are living with and surviving cancer. For example, the US Centers for Disease Control and Prevention estimated that the number of cancer survivors in the US increased from 3 million in 1971 to 11.6 million in 2007, and of those in 2007 the largest group of survivors in women was those with breast cancer (CDC 2011). Sexual dysfunction is a known problem for people receiving various cancer treatments, including those for cancer of the bladder (Bhatt 2006; Zippe 2004), head and neck (Low 2009; Singer 2008), and rectum (Hendren 2005; Ho 2011). However, much of the research to date on the prevalence and treatments for sexual dysfunction following cancer treatments has focused on cancer treatments that have a direct impact on the genitals or other areas of the body involved in sexual functioning. It has been estimated that 50% of women who undergo gynaecological cancer treatments suffer long‐term sexual dysfunction as a result (National Cancer Institute 2012). There is also evidence for a high prevalence of sexual dysfunction following treatments for breast cancer, where in one survey it was found in 45% women (Kedde 2013), and in another survey 76% had sexual dysfunction (Goldfarb 2009). A review found that the incidence of sexual dysfunction in women following radical rectal cancer resection varied from 19% to 62% (Cowan 2013).

Despite recognition of the problem, sexual morbidity remains under‐treated in this patient group (Hill 2011). This is, in part, because of the embarrassment associated with sexual dysfunction not only from the patient's point‐of‐view but from that of clinicians (Dening 2013). In addition, sexual interest/arousal disorders in general may be under‐recognised (Montgomery 2008).

Description of the intervention

There are various types of interventions available for the treatment of sexual dysfunction in women, these include:

  • complementary and alternative medicine interventions such as the herbal therapies ginkgo and ginseng;

  • exercises to strengthen the pelvic floor;

  • mechanical interventions such as lubricating gels and clitoral therapy devices;

  • pharmacological interventions including hormonal therapies such as testosterone or oestrogen (these may be applied topically or taken orally);

  • psychotherapeutic and psycho‐educational interventions such as counselling or psychotherapy, which aim to reduce the sexual dysfunction as well as improve the patient's communication skills with their sexual partner. Psychotherapeutic interventions may be combined with education to facilitate an understanding of treatment and it effects.

The range of interventions is in part explained by the range of causes of sexual dysfunction, but also because it is known that the appropriateness, effectiveness and suitability of each intervention vary between individual patients.

How the intervention might work

The various types of interventions work in different as well as complex ways, with some mechanisms of action poorly understood. We briefly consider the mechanisms of action according to the following groupings.

Complementary and alternative medicine

There are various complementary and alternative medicines that are used to try to enhance sexual function. Their mechanisms of action are often unknown. Maca is an Andean plant belonging to the mustard family and it has been used for centuries to enhance fertility (Shin 2010). Maca is also reported to improve sexual function in healthy human populations (Gonzales 2003). Ginkgo biloba has been evaluated for depression‐induced sexual dysfunction (Cohen 1998).

Exercises to strengthen the pelvic floor

Pelvic floor muscles give structural support to the pelvic organs (urethra, vagina and rectum). Pelvic floor dysfunction may be a direct outcome of gynaecological cancers and various cancer treatments. For example, radical hysterectomy and radiotherapy of the pelvic area may disrupt the anatomy and the local nerve supply to the pelvic floor muscles (Jackson 2006). This disruption may lead to urinary continence and a lack of sexual arousal. Pelvic floor exercises may reduce these problems by increasing muscle tone.

Mechanical

There are various mechanical devices for sexual dysfunction. In women a clitoral therapy device such as a hand‐held battery powered vacuum creates a gentle suction to engorge the clitoris blood flow and thereby increase sensation (Schroder 2005).

Pharmacological

The relationship between endogenous testosterone levels and sexual function in women has not been clearly established. However, testosterone therapy has been found to improve the signs and symptoms of hypoactive sexual desire (Hubayter 2008). Exogenous oestrogen will increase vaginal blood flow and lubrication, and oestrogen therapy has been shown to improve clitoral sensitivity and the ability to reach orgasm (Berman 1999).

Psychotherapeutic and psycho‐educational

Although evidence for the effectiveness of psychotherapeutic interventions is limited, the role of psychological processes in sexual function is acknowledged (Kazdin 2009). Normal sexual function is a biopsychosocial process that relies on the co‐ordination of not only endocrine, vascular and neurological factors but also psychological factors (Allahdadi 2009). Treatment of sexual dysfunction in women with a psychotherapeutic intervention is a common approach in clinical practice (Berner 2012). There are a number of psychotherapy approaches including cognitive behaviour therapy, counselling and relationship therapy. Some interventions may include educational aspects, such as a discussion on the value of intimate communication with sexual partner and advice on vaginal lubrication.

Why it is important to do this review

Trials evaluating the effectiveness of treatments for sexual dysfunction often exclude patients with other major health problems, such as cancer. As cancer and its treatments disrupt physiological and anatomical integrity, it is not possible to simply extrapolate the results of these trials into the cancer field.

With the world's aging population, and improved diagnosis and better cancer treatments, more people are living with and surviving cancer. It is recognised that sexual morbidity is a major cause of poor quality of life. A growing number of treatments are being developed and evaluated; it is important to review their efficacy, safety and acceptability. This review is an update of an earlier Cochrane review on interventions for sexual dysfunction following treatments for cancer in men and women (Miles 2007). Since its publication many more trials have been undertaken. In light of this growth we have split this review. In this current update we evaluate only interventions for women. The update on interventions for men will be reported elsewhere in a new review.

Objectives

available in

  • To evaluate the effectiveness of interventions for treating sexual dysfunction in women following treatments for cancer.

  • To assess adverse events associated with interventions.

Methods

available in

Criteria for considering studies for this review

Types of studies

We only included randomised controlled trials (RCTs). Published trials that were provided in full rather than in abstract only (such as a conference abstract) were eligible. There was no language restriction. In the case of cross‐over trials, we only reported the first period results (if available) in this review in order to exclude any carry‐over effect. Trials undertaken in any care setting, including hospital and community, were eligible.

Types of participants

Trial participants were women (aged 16 years or over), receiving any active treatment for any cancer or who had previously received any treatment for cancer. Participants must have been functioning sexually with a partner prior to cancer treatment. Participants at the start of the trial needed to have experienced any type of sexual dysfunction (however identified) subsequent to cancer treatment. If participants were not required for trial inclusion to have sexual dysfunction subsequent to cancer treatment, we included the study if it reported:

  • 50% or more having sexual dysfunction following treatment;

  • subgroup analysis on those with documented sexual dysfunction subsequent to cancer treatment;

  • mean/average score in both trial arms at early measurement post‐treatment indicating a level of sexual dysfunction.

We did not use the DSM‐5 criterion that the duration of the sexual dysfunction had to be at least six months. We felt that this may not be relevant in this patient group as a significant proportion may have advanced disease with a prognosis of less than six months.

Types of interventions

We included evaluations of any type of intervention for treating sexual dysfunction following a treatment for cancer. These included pharmacological, mechanical, psychotherapeutic and psycho‐educational, complementary medicine or exercise. The intervention for sexual dysfunction could be compared with a placebo, usual care or another active treatment.

We did not include trials of preventative strategies to refine the cancer treatment, such as peritoneal vaginoplasty in combination with a radical hysterectomy.

For this update, we only included studies of interventions for the treatment of sexual dysfunction in women, not the prevention of sexual dysfunction.

Types of outcome measures

Primary outcomes

The primary outcomes of interest were efficacy in regards to improved sexual function, quality of life and safety. We evaluated efficacy using the following outcomes:

  • the proportion of participants within each trial arm with improved sexual function;

  • mean scores on standardised sexual function questionnaires; these included the Female Sexual Function Index (FSFI), which involves 19 items yielding an overall sexual function score in addition to subscale scores for sexual desire, arousal, lubrication, orgasm and satisfaction (Rosen 2000);

  • mean scores on standardised quality of life measures including the Quality of Marriage Index, which involves six items of marital quality (Norton 1983).

We evaluated safety using the following outcomes:

  • number and type of adverse effects;

  • number of participants who dropped out due to adverse effects.

Secondary outcomes

The secondary outcomes of interest were efficacy in regards to psychological and physiological functioning, and symptoms of disease.

We reported measured outcomes on sexual dysfunction relating to the patient's sexual partner.

Exclusion criteria

We did not include data from observational cohort or cross‐sectional studies, nor RCTs evaluating the effectiveness of preventative measures such as breast reconstruction, or avoidance of one particular therapy. We did not include studies comprising of healthy volunteers or patients who reported sexual dysfunction subsequent to non‐cancer treatments.

Search methods for identification of studies

Electronic searches

In the original review we searched for interventions for both men and women (Miles 2007). In this review we also planned to review interventions for both men and women. It was the high number of relevant studies found from these update searches which led us, to permit timely completion, to split the review into one on interventions for sexual dysfunction following treatments for cancer in women, and one on interventions for sexual dysfunction following treatments for cancer in men.

We searched the following electronic databases irrespective of language and publication status for the 2015 update:

  • Cochrane Central Register of Controlled Trials (CENTRAL 2015, Issue 9);

  • MEDLINE (1966 to August 2015);

  • EMBASE (1980 to September 2015);

  • PsycINFO (1966 to September 2015);

  • AMED (1985 to August 2015);

  • CINAHL (1966 to September 2015);

  • National Health Service Research Register (containing the Medical Research Council Directory) (1990 to January 2007);

  • metaRegister of controlled trials (mRCT) (www.controlled-trials.com/mrct), ClinicalTrials.gov (www.clinicaltrials.gov) and the World Health Organization (WHO) International Clinical Trials Registry Platform (ICTRP) (http://apps.who.int/trialsearch/) to September 2015.

Please see Appendix 1 for search strategies.

Searching other resources

We checked for further relevant trials via the reference lists and undertook forward citation tracking for included trials. We also checked for further relevant trials via the reference lists of any related reviews.

Data collection and analysis

Two authors (BC, LJ) screened the keywords and abstracts of electronic citations. Where they appeared to be relevant, we obtained the full texts.

Selection of studies

Following screening, two review authors (BC, LJ) assessed the full text of potentially eligible citations for inclusion. If differences of opinion had arisen we planned to discuss this with the other review authors (VV, MK and AT) and if resolution had been difficult, we planned to attempt to contact the study authors for clarification. We documented studies excluded after full‐text assessment, giving reasons for exclusion.

Data extraction and management

We designed a data extraction form for this review. Where possible, we extracted the following information for each trial.

  • Methods: trial design, duration, allocation method, masking and care setting.

  • Aim and inclusion criteria.

  • The number of patients eligible, the number randomised and reasons why any patients were not included in the trial.

  • The number of participants evaluated at follow‐up(s), reasons for loss to follow‐up and how the trials, if stated, handled deviations from randomised allocation and missing response.

  • Participant characteristics: their age, gender, cancer and treatment, and measure of sexual function at baseline.

  • Content of the intervention including who delivered it, duration and number of sessions and the mode of delivery (including whether it was conducted with individuals or in a group setting). We also report whether the content of the intervention was standardised by the use of a manual.

  • Comparison intervention including content, duration and mode of delivery.

  • Outcome data at the end of treatment and at the end of follow‐up, including how it was measured. We extracted details on all outcomes on which authors collected measurements but only provide details in our results of those relevant to this review, specifically on sexual function, quality of life and safety, and of psychological and physiological functioning and symptoms of disease. We extracted, as appropriate, all outcome data on our outcomes of interest if they were reported in the trial papers. These included baseline scores, change scores and scores at follow‐up between the trial arms.

  • We also planned to extract any qualitative evidence in the included studies, such as analysis of participants' views on the value of the intervention.

Where information was lacking, we attempted to contact the trial authors or trial sponsors.

Two review authors (BC/VV) independently extracted data. One author (BC) entered the extracted data into Review Manager 2014 and a second author checked the data; specifically LJ checked entries on trial description and VV checked entries on trial findings. If there had been any discrepancies, the other review authors would have been consulted and discrepancy resolved by consensus.

Assessment of risk of bias in included studies

Two authors independently assessed risk of bias for each study using the criteria outlined in the Cochrane Handbook for Systematic Reviews of Interventions, resolving any disagreements by discussion (Higgins 2011). We completed a 'Risk of bias' table for each included study. We assessed the following for each study.

  • Random sequence generation (checking for possible selection bias). We assessed the method used to generate the allocation sequence as: low risk of bias (any truly random process: random number table; computer random number generator); and unclear risk of bias (method used to generate sequence not clearly stated). We excluded studies using a non‐random process, which were therefore at high risk of bias (odd or even date of birth; hospital or clinic record number).

  • Allocation concealment (checking for possible selection bias). The method used to conceal allocation to interventions before assignment determines whether the intervention allocation could have been foreseen in advance of, or during, recruitment, or changed after assignment. We assessed the methods as: low risk of bias (telephone or central randomisation; consecutively numbered, sealed, opaque envelopes); and unclear risk of bias if the method was not clearly stated. We excluded studies that did not conceal allocation, which were therefore at high risk of bias (open list).

  • Blinding of participants and personnel (performance bias). We assessed the methods used to blind study participants and outcome assessors from knowledge of which intervention a participant received. We assessed the methods as: low risk of bias if the study stated that it was blinded and described the method used to achieve blinding: identical tablets, matched in appearance and smell; and unclear risk of bias if the study stated that it was blinded but did not provide an adequate description of how blinding was achieved. We judged a study as high risk if there was no blinding or incomplete blinding, and the outcome was likely to have been influenced by lack of blinding. We also judged a study as high risk if blinding was attempted but it was likely that the blinding could have been broken and the outcome was likely to be influenced by lack of blinding.

  • Incomplete outcome data (attrition bias). We assessed whether there was attrition bias due to the amount, nature or handling of incomplete outcome data. We judged the study as having low risk of attrition bias if there were no missing outcome data or the reasons for missing data were unlikely to be related to true outcome, or missing data and reasons for it were similar across trial arms, or the missing data had been imputed using appropriate methods. We judged the study as high risk if the reason for missing outcome data was likely to be related to the outcome, with either imbalance across trial arms in numbers of reasons for missing data and if an inappropriate application of simple imputation was potentially used. We judged the study as unclear risk if there was insufficient reporting of attrition to permit judgement of low or high risk.

  • Sample size (checking for possible biases confounded by small size). Small studies have been shown to overestimate treatment effects, probably because the conduct of small studies is more likely to be less rigorous, allowing critical criteria to be compromised (Zhang 2013). We considered studies to be at low risk of bias if they had 200 participants or more, at unclear risk if they had 50 to 200 participants, and at high risk if they had fewer than 50 participants.

  • Selective outcome reporting (checking if there was a selection of a subset of the original variables recorded on the basis of the results). We assessed selective outcome reporting, if a protocol was available, by comparing outcomes in the protocol and published report. If they were the same we assessed it as low risk in this domain, if they differed we considered it as high risk. If a protocol was not available, then we compared the outcomes listed in the methods section of an article with the outcomes for which results were reported. If they differed we considered the study as high risk. If a protocol was not available and even though the outcomes listed in the methods section and the results section were the same, we considered the study as having an unclear risk of bias in this domain.

We incorporated the results of the 'Risk of bias' assessment into the review through systematic narrative description and commentary about each item.

Measures of treatment effect

Treatment effects were measured using dichotomous data or ordinal rating scales.

Dichotomous data

Where dichotomous data were reported (if data were available) we planned to extract or generate odds ratios (ORs) and their 95% confidence intervals (CIs) where appropriate.

Continuous data

We assessed effect measures for ordinal data as continuous data. Where continuous data were reported, we planned to extract or generate the mean difference (MD) from the means and standard deviations.

Unit of analysis issues

For any identified cluster‐randomised controlled trials we planned to check for errors in the unit of analysis and, if errors were found and sufficient data were available, to recalculate the results using the appropriate unit of analysis (Higgins 2011). For data arising from RCTs with a cross‐over design, if available, we planned to use in any combined analysis only data from the first comparative phase prior to cross‐over. This decision was based on the possibility of a 'carry‐over' of treatment effect from the experimental or comparative treatment.

Dealing with missing data

Missing studies can result from an inadequate search for data or from publication bias in that papers with negative findings are less likely to be published. How we dealt with this is detailed in Search methods for identification of studies and in Assessment of reporting biases.

We anticipated finding a significant amount of loss to follow‐up in this review. This was due either to the patient's declining health and the caregiver's need for more time with their loved one, or because of the death of the patient. We report attrition rates, per trial, in the 'Risk of bias' tables (see Characteristics of included studies). This included, if available, per trial arm reasons for attrition, and whether the trial analysis entailed any re‐inclusions. We did not undertake any imputation for missing participant data.

A common item missing in outcome data is the standard deviation for continuous outcomes. Where data were not reported, but might be available, we attempted to contact the study authors. We attempted contact with them up to two times. If contact with the author was not possible, we planned to calculate or impute this using relevant data, only if a minority of the trials (to be combined in a meta‐analysis) had a missing standard deviation (Higgins 2008). If we had undertaken such imputation we planned to perform sensitivity analyses to assess its impact on combined analysis.

We did not exclude trials on the basis of missing data. In the Discussion section we address the potential impact of missing data on the findings of the review.

Assessment of heterogeneity

If meta‐analysis had been possible, we would have assessed statistical heterogeneity between trials using the Chi2 test and I2 statistic (a Chi2 P value of less than 0.05 or an I2 value equal to or more than 50% is considered to indicate substantial heterogeneity). If substantial heterogeneity was identified, we planned to undertake subgroup analyses to investigate its possible sources.

Assessment of reporting biases

If meta‐analysis had been possible we would have sought to explore publication bias visually using funnel plots. In our interpretation of the plots we planned to use the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011).

Data synthesis

For any combined analysis that had been possible, as the patient populations were quite variable in age and treatments (as were the interventions), we would have employed random‐effects meta‐analyses.

Subgroup analysis and investigation of heterogeneity

Subgroup analysis explores whether the overall effect varies with different trial populations and with the nature and content of the interventions. In this update we planned the following subgroup analysis:

  • trials of participants with severe sexual dysfunction;

  • trials of participants with advanced cancer.

Sensitivity analysis

We planned to perform sensitivity analyses in order to explore (by excluding trials) the influence of the following factors:

  • unpublished trials;

  • trials with high risk of bias.

Results

Description of studies

Results of the search

In this 2015 update we searched for both interventions for men and women; two review authors independently screened 85% of 4218 citations identified. Thirty studies evaluating interventions for women appeared to be relevant at citation assessment. At full‐text assessment, 11 studies were eligible (Aktas 2015; Barton 2007; Baucom 2009; Classen 2013; Lee 2011; Marcus 2010; Rowland 2009; Schover 2011; Schover 2013; Svensk 2009; Yang 2012). We found two further articles at full‐text assessment that related to studies in progress (DuHamel 2013; Schofield 2013) (see Characteristics of ongoing studies) and we excluded seven further articles because they did not fulfil the inclusion criteria, as they did not report sexual outcomes or were not a RCT. Details of these studies and those excluded in the original review (and involving populations of women) are in the table Characteristics of excluded studies. We found seven studies that are awaiting classification (see Characteristics of studies awaiting classification). The interventions for men identified in this updated search are reviewed in another Cochrane review, which is in progress. The search process is documented in Figure 1.

Figure 1
Study flow diagram.

Study flow diagram.

We excluded one RCT identified in the original review in this update as it no longer fitted our tighter inclusion criteria, as it is a preventative intervention (Pitkin 1971).

Included studies

The 11 completed trials randomised 1509 female participants (Aktas 2015; Barton 2007; Baucom 2009; Classen 2013; Lee 2011; Marcus 2010; Rowland 2009; Schover 2011; Schover 2013; Svensk 2009; Yang 2012).

The trials were undertaken in populations from four countries. Most were undertaken in US populations (Aktas 2015; Barton 2007; Baucom 2009; Marcus 2010; Rowland 2009; Schover 2011; Schover 2013). Others were undertaken in Canada (Classen 2013), South Korea (Lee 2011; Yang 2012), and Sweden (Svensk 2009).

One of the trials was of cross‐over design (Barton 2007); the remainder were of parallel design. Three of the studies were feasibility or pilot trials (Baucom 2009; Classen 2013; Yang 2012). Two studies compared different forms of delivery of an intervention (Schover 2011; Schover 2013), and two used wait‐list controls (Classen 2013; Schover 2011). Four trials involved samples of fewer than 50 participants (Baucom 2009; Classen 2013; Svensk 2009; Yang 2012). Four had samples of over 100 participants (Barton 2007; Marcus 2010; Rowland 2009; Schover 2011). None of the trials reported drug, staff or project pharmaceutical funding or affiliation.

All trials involved participants that had breast or gynaecological cancer. Where reported, the mean age of participants ranged across trials from 44 to 57 years. The treatments the participants underwent for cancer were surgery, radiation, endocrine therapy and/or chemotherapy.

As per the inclusion criteria participants in the majority of trials had documented poor sexual functioning following a cancer treatment (Barton 2007; Baucom 2009; Classen 2013; Lee 2011; Marcus 2010; Rowland 2009; Schover 2011; Schover 2013; Svensk 2009;Yang 2012). We included another trial as the authors were explicit in assuming that for a high proportion of the participants the cancer treatment would result in sexual dysfunction (Aktas 2015). Across the studies how sexual dysfunction was measured varied. This included self report and scales specifically developed for the project. Validated tools used included the Changes of Sexual Functioning Questionnaire (Clayton 1997), Derogatis Inventory of Sexual Functioning (Derogatis 1979), the Female Sexual Distress Revised Scale (Derogatis 2008), Female Sexual Function Index (Rosen 2000), and the sexual function subscale of the Australian Pelvic Floor Questionnaire (Baessler 2009). Only one trial specifically set out to only include those with severe sexual dysfunction (Classen 2013), although in two other trials the scores obtained on the scales used for sexual dysfunction suggested overall a significant problem (Lee 2011; Schover 2011). One trial evaluated sexual outcomes in both the women and their partners (Baucom 2009).

All involved participants were in a couple relationship, although only two of the trials documented that participants were sexually active prior to the cancer treatment (Aktas 2015; Classen 2013). In three trials the intervention was started during active treatment for cancer of surgery, radiography or chemotherapy (Aktas 2015; Baucom 2009; Svensk 2009); for one of these studies the treatment, radiology, was ongoing (Svensk 2009). In one other trial the intervention was started just after the treatment for cancer (Marcus 2010). In another the mean number of months since treatment for cancer ended was 24.3 in the intervention group and 31.3 in the control group (Classen 2013). In the remaining six trials time since cancer treatment was not reported (Barton 2007; Lee 2011; Rowland 2009; Schover 2011; Schover 2013; Yang 2012).

In three trials the effect of the intervention on sexual dysfunction was the primary focus (Aktas 2015; Barton 2007; Schover 2013). In five of the other trials there was no declared primary outcome; in four of these sexual dysfunction was one of several measures of wellbeing, including quality of life, mental health and distress, body image, spiritual wellbeing and personal growth (Baucom 2009; Marcus 2010; Schover 2011; Svensk 2009). In the fifth trial the other measures related to the impact of an exercise intervention on pelvic floor muscles (Yang 2012). In the other two included studies sexual dysfunction was a secondary outcome; in one the primary outcome was feasibility of the intervention (Classen 2013), and in the other mental health (Rowland 2009).

Further details on these trials are documented in the table Characteristics of included studies.

Trial interventions

The interventions evaluated were pharmacological, psychotherapeutic, psycho‐educational or exercise. Most trials (8/11) evaluated an intervention involving psychotherapeutic techniques. Within these interventions there was heterogeneity in content.

Pharmacological interventions

Two types of pharmacological interventions were tested:

  • Topical testosterone cream;

  • Vaginal pH‐balanced gel.

Psychotherapeutic/psycho‐educational

The focus and delivery of the psychotherapeutic and psycho‐educational interventions varied and included:

  • Art therapy;

  • Counselling one to one, in group or by telephone. This included professional counselling or peer support. Topics included discussion of sexual matters, relationship functioning, quality of life and information on treatment;

  • Relationship enhancement;

  • Web‐based support and information.

Exercise

  • Pelvic floor exercises with and without biofeedback.

Further descriptions of the included trials can be found in the Characteristics of included studies table.

Outcomes evaluated

There was variation across the trials in how the effect of the intervention on sexual function was measured, as well as for other types of outcomes, including quality of life and psychological functioning. Where there was overlap, the trials differed in how they presented their results.

Excluded studies

We excluded 49 studies in the previous version of this review and 19 at this update.

The main reasons for exclusion of studies were that the report was a discussion paper, not a RCT, the study did not include cancer patients or that the evaluation had no sexual function outcomes. Reasons for exclusion for all excluded studies are listed in the Characteristics of excluded studies table.

Ongoing studies

There are seven relevant trials in progress (Davis 2015; DuHamel 2013; Gessler 2015; Hummel 2015; NCT00459134 2015; NCT02091765 2015; Schofield 2013). The trials vary. For example, one of the studies in progress is testing a psychotherapeutic intervention in female survivors of anal or rectal cancer (DuHamel 2013); another is testing a psychotherapeutic intervention in women following treatments for gynaecological cancer (Schofield 2013). Further details on these trials are documented in the table Characteristics of ongoing studies.

Risk of bias in included studies

All trials were vulnerable to a number of biases, most commonly selection bias. See Figure 2 and Figure 3.

Figure 2
'Risk of bias' graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.

'Risk of bias' graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.

Figure 3
'Risk of bias' summary: review authors' judgements about each risk of bias item for each included study.

'Risk of bias' summary: review authors' judgements about each risk of bias item for each included study.

Allocation

Most trials (n = 8) did not report the method of randomisation sequence generation (Aktas 2015; Barton 2007; Baucom 2009; Classen 2013; Marcus 2010; Rowland 2009; Schover 2011; Schover 2013). The method concealment of allocation was described adequately in three of the trials (Barton 2007; Baucom 2009; Svensk 2009).

Blinding

Two trials were at a low risk of performance bias (Barton 2007; Baucom 2009).

In the other trials the risk was either high, as the authors stated that the trial was single‐blinded, or the risk was unclear as they provided no details.

Incomplete outcome data

The risk of attrition bias was low in most trials (n = 6) (Barton 2007; Baucom 2009; Classen 2013; Marcus 2010; Svensk 2009; Yang 2012).

Selective reporting

The risk of selective reporting was unclear in most studies (n = 8) as they did not reference a protocol or clearly state primary outcomes.

Other potential sources of bias

Seven trials were at a high risk of bias as they involved fewer than 50 participants in each trial arm (Aktas 2015; Baucom 2009; Classen 2013; Lee 2011; Schover 2013; Svensk 2009; Yang 2012).

Effects of interventions

The effects of the interventions are reported as fully as possible. Some trials did not report for all outcomes assessed the actual scores for or number of participants experiencing an outcome.

Pharmaceutical interventions

Two trials evaluated a pharmaceutical intervention.

Testosterone cream versus placebo

One cross‐over trial of 150 women with cancer evaluated the effect of testosterone 2% in Vanicream versus plain Vanicream in the control group (Barton 2007). As per the inclusion criteria, at baseline all participants had a decrease in sexual desire. This was defined as a score of less than 8 on a 0‐ to 10‐point scale with 10 being highest interest (actual scores not provided). Participants were instructed to apply a spoonful of the cream (intervention treatment or control) onto a large area of their abdomen or thighs. The trial reported cross‐over periods separately and so according to our protocol we only report the pre‐cross‐over results.

Sexual function

Outcomes for sexual desire/interest and frequency of desire were captured using the Changes in Sexual Functioning Questionnaire (CSFQ) (Clayton 1997). The overall score for the scale runs from 0 to 100, with higher scores representing poorer outcome.

Using the CSFQ for the summed score of desire/interest and frequency of desire it was found that there was no significant difference in effect between the trial arms. The first cross‐over mean score in the intervention group was 48.53 and in the control group it was 44.41. The mean difference (MD) in change from baseline was 1.08 (95% confidence interval (CI) ‐2.78 to 4.94). Neither was there a significant difference in the CSFQ subscale on feelings of sexual pleasure (MD 4.67, 95% CI ‐0.10 to 9.44) (see Table 1 for the actual scores).

Open in table viewer
Table 1. Barton 2007 scores for sexual functioning

Measure

Arm

Mean score (95% CI) Week 4*

Mean score (95% CI) Week 8**

Changes in Sexual

Functioning Questionnaire

Rx/placebo

Placebo/Rx

5.69 (4.13 to 10.63)

3.40 (2.08 to 6.75)

6.87 (6.42 to 11.38)

8.17 (7.36 to 13.82)

Changes in Sexual

Functioning Questionnaire,

Pleasure subscale

Rx/placebo

Placebo/Rx

9.38 (7.00 to 11.23)

4.71 (0.43 to 8.98)

9.33 (6.84 to 16.49)

11.04 (9.03 to 18.59)

Profile of Mood States

Rx/placebo

Placebo/Rx

‐0.55 (‐3.77 to 2.67)

3.02 (‐0.27 to 6.31)

1.07 (‐1.89 to 4.03)

5.11 (2.16 to 8.07)

Profile of Mood States,

Vitality subscale

Rx/placebo

Placebo/Rx

0.56 (‐2.58 to 3.71)

2.76 (‐0.52 to 6.04)

1.82 (‐1.74 to 5.39)

4.68 (2.01 to 7.34)

CI: confidence interval
Rx: intervention

*Pre‐cross‐over; **Post‐cross‐over

Quality of life

Quality of life was not measured.

Adverse effects

The trial reported no significant difference between trial arms in adverse effects of acne, voice deepening, abnormal hair loss or growth, peripheral oedema or headache. The actual numbers of participants suffering these events were not reported.

Secondary and other outcomes

The Profile of Mood States (POMS) was used to measure psychological function. They also reported the findings of a subscale on vitality. There was no significant difference between trial arms for both outcomes (MD ‐3.37, 95% CI ‐8.15 to 1.01; MD ‐2.20, 95% CI ‐6.84 to 2.44) (see Table 1 for the actual scores). Impact on physical symptoms of cancer were not reported.

pH‐balanced vaginal gel versus placebo

In one trial, 98 women with breast cancer were randomised to either apply three times a week a pH‐balanced vaginal gel or a placebo gel (Lee 2011). The women had experienced menopause after chemotherapy or endocrine therapy. In both trial arms, participants at baseline reported a high degree of dyspareunia. On a scale of 0 to 10 with 10 being the highest score for dyspareunia, the baseline mean in the intervention group was 8.23 (standard deviation (SD) 0.99) and in the control group 8.20 (SD 0.95). Likewise the score (using the same scoring system) for vaginal dryness with pain was high: in the intervention group it was 8.20 (SD 0.83) and in the control group 7.92 (SD 0.89).

Sexual function

Effectiveness of the gel was measured by whether it reduced dyspareunia and vaginal dryness. This was measured using a visual analogue scale of 0 to 10 with higher scores reflecting poorer function/more severe symptoms.

At 12 weeks of treatment there was a significant difference favouring those in the intervention group in reduced dyspareunia and other measures of vaginal health, including reduced vaginal dryness (mean dyspareunia 5.48 (SD 1.06) in the intervention group, mean dyspareunia 6.11 (SD 1.42) in the control group; MD ‐0.63, 95% CI ‐1.13 to ‐0.13); mean vaginal dryness 4.23 (SD 1.40) in the intervention group and 6.51 (SD 1.51) in the control group; MD ‐2.28, 95% CI ‐2.85 to ‐1.71).

Quality of life

Quality of life was not measured.

Adverse effects

The most common adverse effect was the sensation of vaginal burning or irritation; there was no significant difference between the trial arms in the proportions of women experiencing this (18/49 in the intervention group versus 13/49 in the control group; odds ratio (OR) 1.61, 95% CI 0.68 to 3.80). There were no reported severe adverse effects.

Secondary and other outcomes

Impact on symptoms of cancer and psychological functioning were not reported.

Psychotherapeutic and psycho‐educational

Eight trials assessed the effectiveness of a psychotherapeutic intervention or psycho‐educational intervention (Aktas 2015; Baucom 2009; Classen 2013; Marcus 2010; Rowland 2009; Schover 2011; Schover 2013; Svensk 2009). The content of four interventions was specific to sexual matters (Aktas 2015; Baucom 2009; Classen 2013; Schover 2013). Three were broader in topic, covering in addition to sexual matters other quality of life issues such as physical, psychological, social and spiritual wellbeing (Marcus 2010; Rowland 2009; Schover 2011). The intervention in the other trial was art therapy and it did not explore sexual matters (Svensk 2009). Instead it provided participants with a medium for expression and reflection. It was included as sexual functioning was a specific outcome. The interventions not only varied in content, but in how they were delivered and in their comparison. In brief:

  • Specialist nurse counselling on sexual matters (Aktas 2015). The comparison group received usual post‐cancer treatment care;

  • Face‐to‐face counselling on relationship enhancement compared with an active control of provision of a list of support services (Baucom 2009);

  • Online support group to address the psychosexual impact of cancer compared with a wait‐list control (Classen 2013);

  • Telephone counselling on living with physical change, sexuality, relationships and economic change compared with an active control of a list of support services (Marcus 2010);

  • Sexual relationship enhancement intervention compared to an active control of an educational leaflet (Rowland 2009);

  • Peer counselling aiming to improve knowledge and reduce symptoms relating to sexual dysfunction compared to participant‐initiated call to receive brief telephone counselling on sexual dysfunction (Schover 2011);

  • Online intervention on sexual and fertility consequences of cancer and treatment (Schover 2013). Participants' use of the website was guided by a counsellor in one trial arm and not in the other arm;

  • Art therapy intervention compared with a control group who had usual care (Svensk 2009).

Detailed information regarding treatment arms and outcomes is provided in the Characteristics of included studies.

This diversity and complexity of interventions precluded any taxonomy of simple labels that would accurately summarise efficacy per intervention type. It also precluded classification into general themes and groups. Moreover the trials varied in how they measured impact on sexual functioning, with five not reporting their findings in a way that was appropriate for combined analysis (Baucom 2009; Marcus 2010; Rowland 2009; Schover 2011; Schover 2013). Therefore for these reasons their results are presented individually.

Results

Three studies found that a psychotherapeutic or psycho‐educational intervention had a beneficial effect on some measures of sexual function (Aktas 2015; Baucom 2009; Marcus 2010). In others the benefit of the intervention was either unclear (Rowland 2009; Schover 2013), or no difference was found in sexual function in comparison with the control intervention (Classen 2013; Schover 2011; Svensk 2009). Two trials reported a significant improvement in quality of life in the intervention group compared to the control group (Baucom 2009; Svensk 2009), but in another no difference was found (Schover 2011). Only one study specifically set out to measure adverse events (Baucom 2009). Four studies evaluated psychological health, such as symptoms of depression or anxiety (Baucom 2009; Marcus 2010; Rowland 2009; Schover 2011), and four evaluated physical functioning, impact on symptoms of cancer or impact of treatment on menopausal symptoms (Baucom 2009; Schover 2011; Schover 2013; Svensk 2009).

Aktas 2014

Aktas 2015 tested whether a specialist nurse counselling service compared to usual care would improve sexual satisfaction in 70 women following a hysterectomy for gynaecological cancer. Sexual dysfunction was not measured at baseline, however the authors assumed that most of the women would develop sexual problems because of their cancer treatments. The nurses visited the women in the intervention group three times a week until hospital discharge, and subsequently at home at one and 12 weeks. The nurse discussed sexual problems with the women including body image, self confidence, the fear of being rejected, loss of desire, sexual intercourse, vaginal sensitivity, vaginismus, dyspareunia, inability to reach orgasm, shortening of the vagina and incomplete penis penetration.

Sexual function

Sexual outcomes were measured using the Golombok Rusk Inventory of Sexual Satisfaction (GRISS), which involves 28 items graded on a Likert scale (Rust 1985). It provides a total score on satisfaction with sexual function and scores for 12 subscales of which seven were assessed in this study. The lower the scores the greater the sexual function.

At 12 weeks, using the total score for the GRISS, there was a statistically significant difference favouring the intervention group (mean 7.15 in the intervention group and 9.46 in the control group (MD ‐1.40, 95% CI ‐2.27 to ‐0.53).

There were mixed outcomes on the GRISS subscales. There was significantly more improvement in the intervention group in:

  • sexual non‐communication (MD ‐1.63, 95% CI ‐2.52 to ‐0.74);

  • anorgasmia (difficulty reaching orgasm) (MD ‐1.73, 95% CI ‐2.56 to ‐0.90);

  • avoidance (MD ‐2.11, 95% CI ‐3.65 to ‐0.57);

  • non‐sensuality (MD ‐1.93, 95% CI ‐3.45 to ‐0.41).

There was no significant difference between the trial arms in:

  • frequency of sexual contact (MD ‐0.11, 95% CI ‐1.01 to 0.79);

  • vaginismus (MD ‐1.27, 95% CI ‐2.91 to 0.37);

  • sexual dissatisfaction (MD 0.43, 95% CI ‐0.66 to 1.52).

  • the proportion of women who had resumed sexual relations by 12 weeks (7/35 in the intervention group and 10/35 in the control group; OR 2.36, 95% CI 0.88 to 6.34).

Actual scores on the subscales are reported in Table 2.

Open in table viewer
Table 2. Aktas 2014 follow‐up subscale scores for sexual function

Measure*

Mean score at follow‐up in intervention group

Mean score at follow‐up in control group

Sexual non‐communication

5.14 (SD 2.07)

6.94 (SD 1.28)

Anorgasmia

8.83 (SD 1.79)

11.26 (SD 1.42)

Avoidance

6.34 (SD 3.13)

9.60 (SD 2.94)

Non‐sensuality

7.43 (SD 3.09)

10.20 (SD 2.90)

Infrequency of sexual contact

6.03 (SD 1.58)

6.43 (SD 1.77)

Sexual dissatisfaction

7.0 (SD 2.82)

9.63 (SD 2.45)

Vaginismus

9.26 (SD 3.01)

12.17 (SD 3.31)

SD: standard deviation

Quality of life and adverse effects

Quality of life and adverse effects were not reported on.

Secondary and other outcomes

No other outcomes were reported.

Baucom 2009

Baucom 2009 conducted a pilot study of 14 couples, in which the wife had early‐stage breast cancer. Mean baseline sexual functioning in both trial arms using the sexual drive and relationship subscale of the Derogatis Inventory of Sexual Functioning (DISF‐SR) suggested a level of sexual dysfunction. Higher scores suggested greater sexual function, drive and satisfaction, and better relationship. For both trials arms the score was below mid‐score (16) on this subscale, with a mean score of 11.50 in the intervention group and a mean score of 10.33 in the comparison group. All women had received cancer treatment and in most cases it involved surgery. Couples were randomised to either counselling for relationship enhancement or an active control of provision of a list of community support services. The intervention used a cognitive behavioural approach and consisted of six sessions in which the therapist discussed with the couple how to share feelings and thoughts and to reach decisions jointly. The focus was on cancer‐related topics including fear, mortality, sexuality and body image.

The small number of participants meant that inferential statistics were not used. Instead the researchers looked at between‐group effect sizes, taking into account in their analysis the change scores for each participant. They reported that overall there was greater improvement in couples in the intervention group in all outcomes in relation to relationship and sexual functioning.

Sexual function

The outcomes were sexual functioning and drive using a four‐item domain of the DISF‐SR of which the mean in the intervention group 13.00 (SD 3.11) and in the control group 9.80 (SD 5.93);

Quality of life

Relationship satisfaction was assessed using the Quality of Marriage Index (QMI), with higher scores indicating greater quality. QMI mean in the intervention group 39.71 (SD 3.45) and in the control group 40.20 (SD 5.07).

Adverse effects

No adverse effects were reported.

Secondary and other outcomes

They reported fewer overall physical symptoms secondary to cancer treatment, such as nausea and dizziness, in the intervention group compared to the control group. The study did not report any general measures of psychological functioning. See Table 3 for the actual scores for all outcomes reported.

Open in table viewer
Table 3. Baucom 2009 follow‐up scores for sexual functioning

Measure

Mean score (SD) at follow‐up in intervention group

Mean score (SD) at follow‐up in control group

Sexual drive and relationship

13.00 (3.11)

9.80 (5.93)

Relationship satisfaction

39.71 (3.45)

40.20 (5.07)

Brief Symptom Inventory

6.71 (5.77)

15.80 (20.91)

Post‐traumatic Growth Inventory

62.00 (29.10)

66.20 (35.73)

Functional Well‐being FACT‐B

3.22 (0.34)

2.89 (0.91)

Self‐image Scale ‐ self acceptance

21.29 (4.75)

19.20 (9.09)

Self‐image Scale ‐ partner acceptance

21.96 (2.27)

21.50 (4.79)

Brief Fatigue Inventory

2.98 (0.96)

3.22 (2.25)

Usual pain

2.29 (1.50)

2.40 (1.67)

Rotterdam Symptom Inventory

18.71 (2.36)

23.80 (9.81)

SD: standard deviation

Classen 2013

Classen 2013, in a feasibility study, included 27 women who were sexually distressed following gynaecological cancer treatment. In this trial they sought to identify women whose distress was "above and beyond what might ordinarily occur for women with gynaecologic cancer". This was defined as scoring at least 24 on the Female Sexual Distress Revised Scale. This is higher than the recommended cutoff of 11 for sexual distress. Actual baseline scores were not provided.

The women had received, on average two to three years earlier, surgery, chemotherapy or radiotherapy for gynaecological cancer and were randomised in the study to an online support group, GyneGals, or a wait‐list control. The aim of the 12‐week online support group was to address the psychosexual impact of cancer and its treatments. It involved a closed group discussion forum moderated daily by two psychologists, who posted weekly a message on the forum to introduce discussion topics such as coping with emotional challenges or exploring sexuality. A dedicated website provided psycho‐educational material to support the discussion.

Sexual function

Sexual function was measured using the Female Sexual Distress Scale (FSDS).

No significant difference was found on sexual distress using the FSDS between those who received psychosexual support compared to the wait‐list control group in an intention‐to‐treat analysis or in an analysis of only those who actively used the forum (MD 2.28, 95% CI ‐2.15 to 6.71; MD 3.56, 95% CI ‐2.26 to 9.38, respectively). See Table 4 for the actual mean scores.

Open in table viewer
Table 4. Classen 2013 follow‐up subscale scores for sexual function

Measure

Mean change from baseline score at follow‐up in intervention group

Mean change from baseline score at follow‐up in control group

Sexual function intention‐to‐treat*

2.54 (SD 9.59)

0.26 (SD 3.19)

Sexual function actively used*

3.82 (SD 9.43)

0.26 (SD 3.19)

Anxiety and depression** intention‐to‐treat

0.63 (SD 4.39)

0.50 (SD 2.91)

Anxiety and depression** actively used

1.73 (SD 3.93)

0.50 (SD 2.91)

Intimacy*** intention‐to‐treat

0.19 (SD 1.33)

‐0.17 (SD 1.21)

Relationship*** intention‐to‐treat

‐0.04 (SD 0.68)

‐0.01 (SD 0.82)

Instrumental*** intention‐to‐treat

‐0.22 (SD 0.85)

‐0.10 (SD 1.26)

Intimacy***

0.75 (SD 1.01)

‐0.17 (SD 1.21)

Relationship***

‐0.15 (SD 0.50)

‐0.01 (SD 0.82)

Instrumental**

*‐0.21 (SD 0.71)

‐0.10 (SD 1.26)

* Measured using the Female Sexual Distress Scale.** Measured using the Hospital Anxiety and Depression Scale. ***Measured using the Illness Intrusiveness Ratings Scale.

SD: standard deviation

Quality of life and adverse effects

Quality of life and adverse effects were not measured

Secondary and other outcomes

The main outcome of this study was feasibility and they found that the majority of women found the website content to be appropriate (15/21) and felt comfortable about discussing their sexual concerns (14/21) in a forum. Using the Hospital Anxiety and Depression Scale (HADS) and the Illness Intrusion Rating Scale there was no significant difference in psychological functioning (for example, for HADS intention‐to‐treat analysis: MD 0.13, 95% CI ‐2.44 to 2.70). See Table 4. Physiological functioning and impact on physical symptoms of disease were not reported.

Marcus 2010

Marcus 2010, in a multi‐centre study, included 304 women who had just completed treatment for breast cancer, which in the majority of cases (74%) involved chemotherapy. Sexual function was measured by using a 25‐item scale, which was composed of questions developed specifically for the project. Scores ranged from 0 to 100 with higher scores indicating greater sexual dysfunction. At baseline participants on average had a level of sexual dysfunction, as suggested by mean scores between 40 and 50 (actual scores not provided). The 152 participants randomised to the intervention received 16‐ to 45‐minute telephone counselling sessions on topics regarding living with physical change, sexuality, relationships and economic change. Those randomised to the comparison group received a mailed booklet listing services in the community relevant to breast cancer.

Their findings were from a before and after comparison in each trial arm. They were not a direct comparison of effects between the trial arms.

Sexual function was measured at follow‐up as it was for baseline. The Impact of Events Scale was used to measure cancer‐specific distress and the Center for Epidemiologic Studies Depression Scale to measure depression.

Sexual function

For sexual function they reported an overall significant improvement in the intervention group in mean changes scores from baseline at 12 and 18 months follow‐up (at 18 months the effect size was 0.23, P value = 0.04), and likewise for "personal growth", described as deriving benefit from the cancer experience at 12 and 18 months (at 18 months the effect size was 0.22, P value = 0.03). The comparison group showed no change from baseline. The actual scores on these scales per trial arm were not reported.

Quality of life and adverse effects

Adverse effects were not reported and neither were outcomes for quality of life.

Secondary and other outcomes

In both trial arms there was an improvement when comparing mean scores for cancer‐specific distress and depression. Outcomes on symptoms of disease were not reported. The actual scores on these scales per trial arm were not reported.

Rowland 2009

Rowland 2009 studied 155 women treated for breast cancer who were randomised to either a psycho‐educational programme or an active control involving an educational leaflet entitled 'Facing forward: a guide for cancer survivors'. All who were invited onto the trial had reported that they had problems with body image, sexuality and intimacy and/or communication with a partner. At baseline on a scale of 1 to 6, with higher scores indicating a better outcome, pain with sex was 4.1 in the intervention group and 4.3 in the control group. The intervention consisted of six two‐hour weekly group meetings, which aimed to enhance participants' communication skills, reduce anxiety in intimate situations and provide information on sexual anatomy, menopause and sexual dysfunction.

Sexual function

The study's secondary outcomes were sexual function, via Likert scale (scores range from 1 to 6) items on satisfaction, pain and comfort, and relationship functioning using the Revised Dyadic Adjustment Scale (scores range from 0 to 69). For both outcomes the higher the scores the better the function. Baseline relationship functioning was 50.4 in the intervention group and 50.8 in the control group.

No significant differences were found in mean change scores from baseline to four‐month follow‐up between the trial arms in sexual satisfaction in regards to variety of sex and sexual relationship outcomes (MD 0.13, 95% CI ‐0.24 to 0.50; MD 0.10, 95% CI ‐0.30 to 0.50, respectively). Neither were there significant differences for most other outcomes relating to sex and relationships:

  • pain interfering with pleasure (MD 0.30, 95% CI ‐0.12 to 0.72);

  • improved comfort with sexuality (MD 0.20, 95% CI 0.00 to 0.40);

  • comfort about being touched (MD ‐0.20, 95% CI ‐0.23 to 0.63);

  • comfort in undressing (MD 0.40, 95% CI ‐0.21 to 1.01);

  • comfort in being nude (MD 0.40, 95% CI ‐0.06 to 0.86);

  • impact of cancer on sex (MD ‐0.10, 95% CI ‐0.37 to 0.17).

However, in the case of pain during sexual activity, impact on relationship and communication there were significant differences favouring those in the intervention group (MD 0.80, 95% CI 0.20 to 1.40; MD 2.40, 95% CI 1.02 to 3.78, MD 0.30, 95% CI 0.07 to 0.53, respectively). See Table 5 for the actual mean scores.

Open in table viewer
Table 5. Rowland 2009 follow‐up subscale scores for sexual and psychological function

Measure

Mean baseline (SD) and change score (SD) at follow‐up in intervention group

Mean baseline (SD) and change score (SD) at follow‐up in control group

Satisfaction with variety of sex

4.0 (1.1), 0.1 (1.2)

4.0 (1.0), ‐0.03 (1.0)

Satisfaction with sexual relationship

4.8 (1.1), 0.0 (1.5)

4.7 (1.4), ‐0.3 (1.3)*

Pain with sex

3.8 (2.0), 0.7 (1.9)

4.1 (1.9), ‐0.1 (1.7)

Pain interfering with pleasure

4.4 (1.7), 0.3 (1.4)

4.4 (1.7), 0.0 (1.1)

Improved comfort with sexuality*

3.3 (0.6)

3.1 (0.6)

Comfort about being touched

4.7 (1.6), 0.1 (1.4)

4.6 (1.6), ‐0.1(1.2)

Comfort undressing

4.7 (1.7), 0.2 (1.9)

5.0 (1.4), ‐0.2 (1.8)

Comfort being nude

5.3 (1.4), 0.3 (1.3)

5.2 (1.4), ‐0.1 (1.6)

Impact of cancer on sex

2.6 (1.0), 0.1 (0.8)

2.4 (1.0), 0.2 (0.9)

Impact on relationship

50.9 (5.7), 1.1 (4.0)

51.0 (5.6), ‐1.3 (4.6)

Improved communication*

3.5 (0.7)

3.2 (0.7)

Mental health

80.8 (11.6), ‐0.7 (10.4)

82.6 (9.2), ‐3.8 (9.4)

* Measured at follow‐up only

SD: standard deviation

Qaulity of life and adverse effects

Adverse events were not reported and neither were outcomes for quality of life.

Secondary and other outcomes

The study's primary outcome was mental health using the Mental Health Index‐32. The index provides a total score with higher scores indicating better mental health. There was no significant difference in mean change in score from baseline to follow‐up in those in the intervention group compared to those in the control group (MD 3.10, 95% CI ‐0.18 to 6.38). See Table 5 for the actual mean scores.

Outcomes on impact on symptoms of disease were not reported.

Schover 2011

Schover 2011 randomised 300 African American women who were breast cancer survivors to either receiving a workbook and regular peer counselling or a workbook and participant‐initiated telephone counselling. Participants at baseline had sexual dysfunction as defined on the Female Sexual Function Index (FSFI), with a mean score across the sample of 18.2 (the recommended score for indicating sexual dysfunction is less than 26.5 (Conaglen 2010)). The intervention workbook was designed specifically for African American women to improve knowledge and reduce symptoms relating to sexual dysfunction, menopause and distress about infertility. In the peer counsellor group participants met the counsellor individually three times a week for six weeks. Each session focused on a chapter of the workbook. The telephone counselling group were given the workbook and the counsellor's contact details and they were encouraged to call her.

The trial measured a range of outcomes at six and 12 months, which included spiritual wellbeing using the Functional Assessment of Cancer, relationship functioning using the Dyadic Adjustment Scale, emotional distress using the Brief Symptom Inventory (BSI), mental distress using the Global Severity Index and sexual function using the FSFI.

Sexual function

Using mixed model analysis they found overall no significant change across time in either group in sexual function.The exact scores for this outcome are not provided by the authors.

Quality of life and adverse events

Adverse effects were not reported and neither were outcomes for quality of life.

Secondary and other outcomes

Using mixed model analysis they found overall no significant change across time in either group in relationship functioning, in total score for menopausal symptoms or childbearing distress. They found depression decreased after treatment in the peer‐counselled group to less than 2.5 but increased in the telephone‐counselled group to nearly 4, but by 12 months depressive symptoms in the two groups were very similar, around a score of 3. The exact scores for any of these outcomes are not provided by the authors.

Outcomes on physiological functioning were not reported.

Schover 2013

Schover 2013 randomised 58 women with breast or gynaecological cancer to either receive counselling and use of a password‐protected website or "self‐help", which entailed access to the website without counselling support. All participants had some indication at baseline of sexual dysfunction using the FSFI as indicated by a score of less than 15 (scores for this scale ranged from 2 to 36, with the lower the score the poorer the sexual function). Actual scores for the participants were not provided. The intervention website provided information on the sexual and fertility consequences of their cancer and treatment, on management of vaginal dryness and pain, sexual issues and pelvic floor exercises. Those in the counselling group were given three sessions to guide them through the website and to discuss behavioural homework. At three and six months post‐treatment sexual outcomes were measured using the FSFI and Menopausal Sexual Interest Questionnaire (MSIQ). Emotional distress was measured using the BSI and quality of life using the Quality of Life in Adult Cancer Survivors scale. The treatment effect was explored by using linear mixed models regression.

Sexual function

Based on the FSFI scale they found a significant treatment effect in improved sexual function in those who received counselling (mean change of 7.44 from baseline to post‐treatment (no SD provided)). However, this was not apparent in a subgroup analysis of those who were sexually active; the authors did not provide actual data on this. Using the MSIQ scale there was also a significant effect in those in the counselling group (mean change score 13.22 from baseline to post‐treatment, SD not provided), but this was only apparent to three months.

Quality of life

Quality of life was found to improve significantly in the self help group only. Full data on this outcome was not provided by the authors.

Adverse effects

Adverse effects were not reported.

Secondary and other outcomes

Emotional distress using the BSI at follow‐up was reduced in the self help group by 3.73, and in the counselled group by 2.63, but this was not significant. Full data on this outcome was not provided by the authors.

Outcomes on physiological functioning and symptoms of disease were not reported.

Svensk 2009

Svensk 2009, in a trial of 41 women with breast cancer, tested the effect of five weeks of individual art therapy sessions compared with a control group who did not receive art therapy. Participants had a reduced level of sexual function and enjoyment at the first assessment following breast cancer treatments. This was indicated by the lower baseline scores for the European Organization for Research and Treatment of Cancer Quality of Life module (EORTC QLQ‐BR23) subscale on sexual functioning at baseline in both trial arms (mean 25.44 in the intervention group and 25.00 in the control group, compared to 38.60 in the intervention group and 30.16 in the control group at first follow‐up). The therapy, based on phenomenological methods, aimed to provide time and space for expression and reflection, to give support in the process of restoring body image and to act as a supporting agency.

Sexual function

The EORTC QLQ‐BR23 instrument was used to measure sexual outcomes, and the impact of treatment and cancer on symptoms and body image. It is comprised of 23 items. A higher score for sexual functioning indicates a better level of functioning.

After radiation treatment there was no significant difference between trial arms in sexual function at two months (MD 8.44, 95% CI‐8.01 to 24.89) and at six months (MD 5.88, 95% CI ‐10.15 to 21.91). In the intervention group the baseline mean was 25.44 (SD 19.54) and at six months it was 43.21 (SD 24.52). In the control group the mean at baseline was 25.00 (SD 21.97) and at six months it was 28.33 (SD 26.55).

There was no significant difference in sexual enjoyment at two months (MD 2.40, 95% CI ‐16.61 to 21.41) or at six months (MD ‐3.17, 95% CI ‐23.43 to 17.09). See Table 6.

Open in table viewer
Table 6. Svensk 2009 follow‐up scores

Measure

Mean score (SD) at follow‐up in intervention group

Mean score (SD) at follow‐up in control group

Sexual function at 2 months

38.60 (24.88)

30.16 (28.20)

Sexual function at 6 months

34.21 (24.52)

28.33 (26.55)

Sexual enjoyment at 2 months

56.25 (26.44)

53.85 (25.60)

Sexual enjoyment at 6 months

69.05 (24.34)

72.22 (27.83)

Overall quality of life at 2 months

78.75 (14.68)

66.67 (22.82)

Overall quality of life at 6 months

85.00 (12.57)

67.50 (20.03)

Physical health at 2 months

68.75 (13.51)

61.39 (19.71)

Physical health at 6 months

74.82 (13.33)

63.93 (19.80)

Psychological health at 2 months

70.83 (15.29)

63.69 (15.87)

Psychological health at 6 months

73.96 (10.28)

69.38 (13.13)

Social relationships at 2 months

74.58 (14.43)

69.44 (15.66)

Social relationships at 6 months

77.50 (12.99)

71.67 (16.31)

Body image at 2 months

81.67 (24.57)

80.95 (20.77)

Body image at 6 months

91.25 (12.82)

83.33 (27.64)

Adverse effects at 2 months

14.76 (11.13)

20.33 (12.23)

Adverse effects at 6 months

10.24 (7.61)

14.97 (12.71)

Breast symptoms at 2 months

20.42 (16.10)

21.03 (14.58)

Breast symptoms at 6 months

17.08 (11.30)

20.63 (17.80)

Arm symptoms at 2 months

16.11 (21.77)

26.46 (29.71)

Arm symptoms at 6 months

15.56 (20.52)

18.52 (17.33)

Upset by hair loss at 2 months

11.11 (19.25)

50.00 (23.57)

Upset by hair loss at 6 months

25.00 (16.67)

50.00 (70.71)

SD: standard deviation

Quality of life

Quality of life was measured using the Swedish version of the World Health Organization (WHO) instrument WHOQOL‐BREF. At two and six months there was a significant difference favouring the intervention in quality of life (MD 12.08, 95% CI 0.39 to 23.77; MD 17.50, 95% CI 7.14 to 27.86, respectively).

Adverse effects

Adverse effects were not reported.

Secondary and other outcomes

Using the European Organization for Research and Treatment of Cancer (EORTC) QLQ‐BR23 there was no significant difference between the trial arms at two and six months follow‐up in body image, cancer therapy side effects, 'breast symptoms', 'arm symptoms', being upset by hair loss, physiological health, psychological heath and quality of social relationships. This is apart from being upset about hair loss at two months, where there was a significant difference favouring those in the intervention group. See Table 6. Physiological functioning were not reported.

Exercise

One trial of 28 women evaluated the effect of exercise on sexual function (Yang 2012). The women had pelvic floor dysfunction following a radical hysterectomy and pelvic lymph node dissection for gynaecological cancer. The trial compared four weeks of pelvic floor muscle training with a control group receiving no training. At the start of the intervention in both trial arms the overall mean for sexual function using the sexual function subscale of the Pelvic Floor Questionnaire indicated sexual dysfunction. Total scores on the scale ranged from 0 to 20. Eight items on the scale covered sufficient lubrication, vaginismus, coital incontinence, vaginal laxity, dyspareunia and 'sexual bother'. A zero score meant that there was no problem in any of the items. In the intervention group the mean score was 6.02 and in the control group 4.62.

Sexual function

Sexual function was measured at four weeks via self report (yes or no), the Australian Pelvic Floor Questionnaire (APFQ) using a four‐point scoring system (0 = no problem to 3 = severe problem) and the cervical cancer‐specific EORTC QLQ CX24, which assessed function using three items. Single‐item measures on the EORTC QLQ CX24 were used in this study to assess sexual worry, activity and enjoyment. The trial had mixed outcomes for sexual function.

The trial reported no significant difference between trial arms in the number of participants at follow‐up who were sexually functioning (8/12 in the intervention group compared with 5/12 in the control group; OR 2.80, 95% CI 0.53 to 14.74). In contrast, the mean score using the EORTC QLQ CX24 was significantly different between trial arms, favouring those who received the intervention (MD 10.30, 95% CI 1.48 to 19.12), as was the mean change score from baseline to follow‐up for sexual function using APFQ (MD ‐3.20, 95% CI ‐5.55 to ‐0.85).

Using the EORTC QLQ CX24 the authors found no significant difference between trials arms in sexual worry or enjoyment but they found a difference in sexual activities favouring the intervention group (MD ‐10.00, 95% CI ‐22.25 to 2.25; MD 2.70, 95% CI ‐9.45 to 14.85; MD 18.40, 95% CI 5.18 to 31.62, respectively). Actual mean scores per trial arm were not reported.

Quality of life

Quality of life outcomes were not reported.

Adverse effects

The authors stated a physical therapist and a trained evaluator monitored adverse effects and any participants experiencing an aggravation of pelvic floor symptoms or difficulty in continuing the exercises was excluded from the final assessment. They did not provide the numbers of those excluded for these events.

Secondary and other outcomes

Psychological and physiological functioning, and symptoms of disease were not reported.

Discussion

available in

This is the first update of the 2007 Cochrane systematic review on interventions for sexual dysfunction following treatment for cancer in women (Miles 2007). The original publication included interventions for men and for women. It identified only one trial of an intervention for women. This trial was not included in this update as it no longer fitted the tighter inclusion criteria, which excluded trials on preventative treatments. Since the original review, as demonstrated in our identification of 11 new trials and two in progress, there has been much more research interest in interventions for women. Three‐quarters of the trials evaluated interventions that involved a psychotherapeutic element; the others evaluated a pharmaceutical gel, testosterone cream and a pelvic floor exercise. The trials involved women with gynaecological or breast cancer. Overall, the trials were mostly not of high quality as they under‐reported design features. Three were pilot or feasibility studies involving small samples and in which the primary aim was not to test effectiveness. We did not identify any randomised controlled trials (RCTs) that evaluated a complementary or alternative medicine intervention.

Summary of main results

The variation in interventions evaluated, including between those that involved a psychotherapeutic element, is the main reason why there is no combined analysis in this review. Psychotherapeutic interventions involved sexual counselling, art therapy, peer support or relationship therapy; overall they showed a mixed effect on sexual functioning. The evidence that testosterone cream, pH‐balanced vaginal gel or pelvic floor exercise improves sexual function is weak. The evidence for each of these interventions is limited to one study and involved in three of the four studies fewer than 50 participants per arm. Evidence on potential harms was under‐reported in all studies. This is apart from the trial on a topical cream and the one on a pH‐balanced vaginal gel, which reported no significant increase in adverse effects in those in the intervention group compared to those allocated a placebo. Surprisingly few trials assessed quality of life, mental health or physiological functioning, therefore it is difficult to derive any conclusions on the effect of these interventions on these outcomes. The two trials of pharmaceutical interventions measured harm and neither reported any. Only one of the psychological intervention trials reported that no harm occurred because of the intervention; the other trials did not measure harm.

Overall completeness and applicability of evidence

We searched widely for evidence using 11 citation databases and using search terms to help identify as wide a range of interventions as possible. We found some notable gaps in the evidence. We did not identify any RCTs evaluating a complementary therapy or alternative medicine, despite trials of these interventions, such as Ginkgo biloba, being tested in other populations. However, a limitation is that we did not search foreign language databases, therefore we may have missed some studies, for example Chinese herbal medicines and acupuncture. In any further update we recommend that these need to be considered.

We also identified no completed trials of interventions that were evaluated in populations with cancers other than breast or gynaecological cancer. However, one of the trials in progress is assessing an intervention following treatment for rectal or anal cancer. This is an important omission as colorectal cancer is one of the most commonly diagnosed cancers and the treatment, such as surgery and radiation, is likely to affect sexual function (Donovan 2010). Moreover, it is known that head and neck cancer survivors struggle with disfigurement and changes in body image, and have increased social isolation, all of which may lead to sexual difficulties (Low 2009; Singer 2008). Another limitation is that all trials were undertaken in advanced western societies, with the majority being undertaken in US populations. Therefore our findings may not be applicable to other populations, in particular where cultural practices in regards to sexual matters may be different.

Quality of the evidence

We could not adequately assess risk of bias in that we found all trials to have at least one methodological item under‐reported, including blinding or randomisation sequence generation. The quality of the evidence is also weakened as eight of the trials had samples of fewer than 50 participants per trial arm. Some of these were also pilot studies, the focus of which was on feasibility rather than effectiveness.

Potential biases in the review process

We undertook a comprehensive search of 11 databases; the last search was completed in September 2015. More recently published eligible studies will be captured at the next update. We identified seven potentially eligible studies in progress from trial registers and conference abstracts. In further updates of this review we recommend consideration of searching foreign language databases.

At least two authors undertook all steps of this review. This limited the risk of errors in determining study eligibility, data extraction, 'Risk of bias' assessment and data synthesis.

Agreements and disagreements with other studies or reviews

There are two recent relevant systematic reviews of psychotherapy in general populations. Both are non‐Cochrane reviews evaluating psychotherapeutic interventions in both men and women (Berner 2012; Fruhauf 2013). The review by Berner 2012 included 20 trials but only two of these included women with sexual dysfunction. They found no benefit of the interventions for women. The review by Fruhauf 2013 included 34 trials that compared the psychosocial intervention with a comparison group. They found in combined analysis of five trials an improvement in sexual function in those who received sexual skills training. However, it is important to note that these reviews were undertaken in non‐cancer populations and therefore their findings are not necessarily transferable to the populations considered in our review.

Study flow diagram.

Figures and Tables -
Figure 1

Study flow diagram.

Navigate to figure in ReviewOpen in new tab

'Risk of bias' graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.

Figures and Tables -
Figure 2

'Risk of bias' graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.

Navigate to figure in ReviewOpen in new tab

'Risk of bias' summary: review authors' judgements about each risk of bias item for each included study.

Figures and Tables -
Figure 3

'Risk of bias' summary: review authors' judgements about each risk of bias item for each included study.

Navigate to figure in ReviewOpen in new tab
Table 1. Barton 2007 scores for sexual functioning

Measure

Arm

Mean score (95% CI) Week 4*

Mean score (95% CI) Week 8**

Changes in Sexual

Functioning Questionnaire

Rx/placebo

Placebo/Rx

5.69 (4.13 to 10.63)

3.40 (2.08 to 6.75)

6.87 (6.42 to 11.38)

8.17 (7.36 to 13.82)

Changes in Sexual

Functioning Questionnaire,

Pleasure subscale

Rx/placebo

Placebo/Rx

9.38 (7.00 to 11.23)

4.71 (0.43 to 8.98)

9.33 (6.84 to 16.49)

11.04 (9.03 to 18.59)

Profile of Mood States

Rx/placebo

Placebo/Rx

‐0.55 (‐3.77 to 2.67)

3.02 (‐0.27 to 6.31)

1.07 (‐1.89 to 4.03)

5.11 (2.16 to 8.07)

Profile of Mood States,

Vitality subscale

Rx/placebo

Placebo/Rx

0.56 (‐2.58 to 3.71)

2.76 (‐0.52 to 6.04)

1.82 (‐1.74 to 5.39)

4.68 (2.01 to 7.34)

CI: confidence interval
Rx: intervention

*Pre‐cross‐over; **Post‐cross‐over
Figures and Tables -
Table 1. Barton 2007 scores for sexual functioning
Navigate to table in Review
Table 2. Aktas 2014 follow‐up subscale scores for sexual function

Measure*

Mean score at follow‐up in intervention group

Mean score at follow‐up in control group

Sexual non‐communication

5.14 (SD 2.07)

6.94 (SD 1.28)

Anorgasmia

8.83 (SD 1.79)

11.26 (SD 1.42)

Avoidance

6.34 (SD 3.13)

9.60 (SD 2.94)

Non‐sensuality

7.43 (SD 3.09)

10.20 (SD 2.90)

Infrequency of sexual contact

6.03 (SD 1.58)

6.43 (SD 1.77)

Sexual dissatisfaction

7.0 (SD 2.82)

9.63 (SD 2.45)

Vaginismus

9.26 (SD 3.01)

12.17 (SD 3.31)

SD: standard deviation
Figures and Tables -
Table 2. Aktas 2014 follow‐up subscale scores for sexual function
Navigate to table in Review
Table 3. Baucom 2009 follow‐up scores for sexual functioning

Measure

Mean score (SD) at follow‐up in intervention group

Mean score (SD) at follow‐up in control group

Sexual drive and relationship

13.00 (3.11)

9.80 (5.93)

Relationship satisfaction

39.71 (3.45)

40.20 (5.07)

Brief Symptom Inventory

6.71 (5.77)

15.80 (20.91)

Post‐traumatic Growth Inventory

62.00 (29.10)

66.20 (35.73)

Functional Well‐being FACT‐B

3.22 (0.34)

2.89 (0.91)

Self‐image Scale ‐ self acceptance

21.29 (4.75)

19.20 (9.09)

Self‐image Scale ‐ partner acceptance

21.96 (2.27)

21.50 (4.79)

Brief Fatigue Inventory

2.98 (0.96)

3.22 (2.25)

Usual pain

2.29 (1.50)

2.40 (1.67)

Rotterdam Symptom Inventory

18.71 (2.36)

23.80 (9.81)

SD: standard deviation
Figures and Tables -
Table 3. Baucom 2009 follow‐up scores for sexual functioning
Navigate to table in Review
Table 4. Classen 2013 follow‐up subscale scores for sexual function

Measure

Mean change from baseline score at follow‐up in intervention group

Mean change from baseline score at follow‐up in control group

Sexual function intention‐to‐treat*

2.54 (SD 9.59)

0.26 (SD 3.19)

Sexual function actively used*

3.82 (SD 9.43)

0.26 (SD 3.19)

Anxiety and depression** intention‐to‐treat

0.63 (SD 4.39)

0.50 (SD 2.91)

Anxiety and depression** actively used

1.73 (SD 3.93)

0.50 (SD 2.91)

Intimacy*** intention‐to‐treat

0.19 (SD 1.33)

‐0.17 (SD 1.21)

Relationship*** intention‐to‐treat

‐0.04 (SD 0.68)

‐0.01 (SD 0.82)

Instrumental*** intention‐to‐treat

‐0.22 (SD 0.85)

‐0.10 (SD 1.26)

Intimacy***

0.75 (SD 1.01)

‐0.17 (SD 1.21)

Relationship***

‐0.15 (SD 0.50)

‐0.01 (SD 0.82)

Instrumental**

*‐0.21 (SD 0.71)

‐0.10 (SD 1.26)

* Measured using the Female Sexual Distress Scale.** Measured using the Hospital Anxiety and Depression Scale. ***Measured using the Illness Intrusiveness Ratings Scale.

SD: standard deviation
Figures and Tables -
Table 4. Classen 2013 follow‐up subscale scores for sexual function
Navigate to table in Review
Table 5. Rowland 2009 follow‐up subscale scores for sexual and psychological function

Measure

Mean baseline (SD) and change score (SD) at follow‐up in intervention group

Mean baseline (SD) and change score (SD) at follow‐up in control group

Satisfaction with variety of sex

4.0 (1.1), 0.1 (1.2)

4.0 (1.0), ‐0.03 (1.0)

Satisfaction with sexual relationship

4.8 (1.1), 0.0 (1.5)

4.7 (1.4), ‐0.3 (1.3)*

Pain with sex

3.8 (2.0), 0.7 (1.9)

4.1 (1.9), ‐0.1 (1.7)

Pain interfering with pleasure

4.4 (1.7), 0.3 (1.4)

4.4 (1.7), 0.0 (1.1)

Improved comfort with sexuality*

3.3 (0.6)

3.1 (0.6)

Comfort about being touched

4.7 (1.6), 0.1 (1.4)

4.6 (1.6), ‐0.1(1.2)

Comfort undressing

4.7 (1.7), 0.2 (1.9)

5.0 (1.4), ‐0.2 (1.8)

Comfort being nude

5.3 (1.4), 0.3 (1.3)

5.2 (1.4), ‐0.1 (1.6)

Impact of cancer on sex

2.6 (1.0), 0.1 (0.8)

2.4 (1.0), 0.2 (0.9)

Impact on relationship

50.9 (5.7), 1.1 (4.0)

51.0 (5.6), ‐1.3 (4.6)

Improved communication*

3.5 (0.7)

3.2 (0.7)

Mental health

80.8 (11.6), ‐0.7 (10.4)

82.6 (9.2), ‐3.8 (9.4)

* Measured at follow‐up only

SD: standard deviation
Figures and Tables -
Table 5. Rowland 2009 follow‐up subscale scores for sexual and psychological function
Navigate to table in Review
Table 6. Svensk 2009 follow‐up scores

Measure

Mean score (SD) at follow‐up in intervention group

Mean score (SD) at follow‐up in control group

Sexual function at 2 months

38.60 (24.88)

30.16 (28.20)

Sexual function at 6 months

34.21 (24.52)

28.33 (26.55)

Sexual enjoyment at 2 months

56.25 (26.44)

53.85 (25.60)

Sexual enjoyment at 6 months

69.05 (24.34)

72.22 (27.83)

Overall quality of life at 2 months

78.75 (14.68)

66.67 (22.82)

Overall quality of life at 6 months

85.00 (12.57)

67.50 (20.03)

Physical health at 2 months

68.75 (13.51)

61.39 (19.71)

Physical health at 6 months

74.82 (13.33)

63.93 (19.80)

Psychological health at 2 months

70.83 (15.29)

63.69 (15.87)

Psychological health at 6 months

73.96 (10.28)

69.38 (13.13)

Social relationships at 2 months

74.58 (14.43)

69.44 (15.66)

Social relationships at 6 months

77.50 (12.99)

71.67 (16.31)

Body image at 2 months

81.67 (24.57)

80.95 (20.77)

Body image at 6 months

91.25 (12.82)

83.33 (27.64)

Adverse effects at 2 months

14.76 (11.13)

20.33 (12.23)

Adverse effects at 6 months

10.24 (7.61)

14.97 (12.71)

Breast symptoms at 2 months

20.42 (16.10)

21.03 (14.58)

Breast symptoms at 6 months

17.08 (11.30)

20.63 (17.80)

Arm symptoms at 2 months

16.11 (21.77)

26.46 (29.71)

Arm symptoms at 6 months

15.56 (20.52)

18.52 (17.33)

Upset by hair loss at 2 months

11.11 (19.25)

50.00 (23.57)

Upset by hair loss at 6 months

25.00 (16.67)

50.00 (70.71)

SD: standard deviation
Figures and Tables -
Table 6. Svensk 2009 follow‐up scores
Navigate to table in Review