Sexual, psychological and dyadic qualities of the prostate cancer ‘couple’


Mark S. Soloway, PO Box 016960 (M-814), Miami, Florida 33101, USA.



To examine the levels of sexual, psychological and dyadic functioning of the prostate cancer ‘couple’ (as studies have shown that spouses/partners play an integral role in the patient's adjustment to prostate cancer treatment), to encourage the creation of innovative psychosexual interventions to be used in the outpatient setting, and to offer insights into a novel area of prostate cancer research.


In all, 103 men newly diagnosed with prostate cancer, and their partners, were assessed in an academic outpatient setting using instruments measuring sexual function, depressed mood, psychological distress and dyadic adjustment.


The partners’ mean scores on sexual function questions were 55.75, significantly higher than those of the patients (51.7, P = 0.018), showing that partners perceived their sexual performance at a better level. Partners’ mean scores on the depression and distress measures were also significantly higher. On those items that monitored the accuracy of the patients’ perceptions of their sexual function, partners rated the patients significantly lower in ability to gain erections (patient/partner means 2.67/4.52; P < 0.001) and to perform sexually (patient/partner means 1.38/4.68; P < 0.001) than they rated themselves.


Information from this study could be useful in constructing interventions that allow the physician and the prostate cancer ‘couple’ to reflect on issues of sexual function and psychological distress that might once have been considered taboo. The results characterize the disparities between patients with prostate cancer and their partners on self-reported questionnaires, and underscore how important it is to hear the voice of the ‘couple’.


The Brief Index of Sexual Function for Women


The Brief Sexual Function Questionnaire for Men


The Sexual Adjustment Questionnaire


The Beck Depression Inventory


Profile of Mood States


Visual Analogue Scales of Distress


The Dyadic Adjustment Scale


The Sociodemographic Questionnaire


health-related quality of life


Major challenges confront men with prostate cancer and their partners. Because most men with prostate cancer are asymptomatic, it is not surprising that couples are not prepared for the diagnosis of prostate cancer and the difficult treatment decisions they face.

Quality-of-life concerns and patient values are important in determining treatment choice [1]. Patients might choose a treatment with a lower long-term survival rate to increase the possibility of remaining sexually potent [2]. For the partner, survival is paramount [3]. Although most partners wish to be an active participant in the patient's treatment decisions, they are much less concerned about the patient's sexual morbidity [3,4]. This challenges a traditional notion that sexual intercourse is equally important to the patient and his partner.

Nonetheless, erectile dysfunction, the most common long-term side effect of prostate cancer treatment, significantly affects the marital relationship [4]. Changes in role during the treatment and recovery periods may lead to emotional distancing, making sexual interactions difficult as the patient and his partner attempt to protect each other's dignity [5]. Often, the only discussion between the patient and his partner related to sexual function comes when couples are presented with treatment options. If there is a loss of sexual function after treatment, communication related to sexual function is likely to stop [6].

Two pilot studies from Memorial-Sloan Kettering Cancer Center, examining psychological distress in men with prostate cancer and their partners, suggested that the partner plays an integral role in the patient's adjustment to treatment [7,8]. In the few other studies where both the patient and partner were considered, sexual activity before diagnosis, and social and psychological resources, were predictors of sexual adjustment and as antecedent conditions that affected the couple's coping skills [5,9]. Building on these findings, we hypothesized that prostate cancer is a couple's disease and, consequently, at different points during treatment it might be the ‘couple’ that should be counselled as the ‘patient’.

This study was conducted to define ‘the couple’ with newly diagnosed prostate cancer. Variables thought to be important were the levels of sexual, psychological and dyadic functioning, and sociodemographic factors. By examining differences within individual couples we hoped to encourage the creation of innovative psychosexual interventions for the outpatient setting, as well as to offer insights into a novel area of prostate cancer research.


Data from 103 newly diagnosed, untreated men with prostate cancer and their partners were analysed in a study approved by the Institutional Review Board at the University of Miami. Homosexual men were excluded from the study because too few partnered homosexual men were seen in the urology outpatient clinic for effective analysis. Following methods of convenience sampling, consecutive untreated referrals (patients, married or in committed relationships, and their partners) were identified in the academic outpatient setting and familiarized with the variables of the study by the urologist. The patient and his partner, after reading and agreeing to the terms of the informed consent, independently completed the questionnaires in the outpatient office. Thirty-five couples declined to participate.

The following instruments were selected: the Brief Index of Sexual Function for Women (BISF-W) [10] and The Brief Sexual Function Questionnaire for Men (BSFQ) [11] measured sexual satisfaction, desire, and activity. An appendix to the BSFQ, questions validating the accuracy of the man's responses, was added to the BISF-W. Ad hoc questions from the Sexual Adjustment Questionnaire (SAQ), an instrument designed to assess sexual function in alcoholics and their spouses, were added to both inventories [12]. The Beck Depression Inventory (BDI) [13] examined attitudes and symptoms frequently shown by depressed patients. The Profile of Mood States (POMS) [14] assessed overall distress and six mood states (tension-anxiety, depression-dejection, anger-hostility, vigour-activity, fatigue-inertia, and confusion-bewilderment). Visual analogue scales of distress (VAS) [15] were two horizontal VAS with equally spaced unlabelled intervals (0–10) characterizing the level of distress at the diagnostic evaluation and before treatment. The Dyadic Adjustment Scale (DAS) [16] evaluated dyadic satisfaction, dyadic cohesion, dyadic consensus, and the expression of affection. The Sociodemographic Questionnaire (SQ) provided demographic and medical information.

Descriptive statistical analysis, Pearson product-moment correlation, one-way anova, independent t-tests, paired samples t-tests and multiple linear regression analysis were used to assess the results.


The sociodemographic characteristics are shown in Table 1; 85% of the study participants were 45–70 years old, with the mean ages of the partners and patients being similar. Of the couples, 95% were married and 83% were in relationships of >10 years. The study population was highly educated; nearly 43% of the participants had completed at least a bachelor's degree. The participants were also ethnically diverse (Table 1) and more than half were employed.

Table 1.  The demographic data
Sociodemographic factorsPartnerPatient
Mean (median, range) age, years58 (60, 34–78)62 (62,43–80)
Education level:
Mean duration at college 3 year, Bachelor degree 
Frequency of ≥ 2 years at college, %3864
Racial/ethnic balance, %
Black 410
White Non-Hispanic5448
White Hispanic3738
Asian 1 1
Employed in last 6 months, %
Health problems, %
No health problems5954
High blood pressure2534
Diabetes 3 7
Heart disease 3 3
Other cancer 9 1
Satisfaction with care, %
‘Somewhat’ to ‘very’ satisfied7173
‘Somewhat’ to ‘very’ dissatisfied1713
Chances of being diagnosed with cancer, %
Not likely3751
Somewhat unlikely2727
‘Highly’ to ‘extremely’ likely 9 6
Stress during evaluation, %
‘Moderate’ to ‘very’5841
Stress at diagnosis, %
‘Moderate’ to ‘very’6644
PSA, ng/mL, at diagnosis, %
Gleason score at diagnosis, %
8 or 9 6

Most of the men presented at diagnosis with what was thought to be organ-confined prostate cancer, most having a PSA level of ≤ 10 ng/mL and more than half with a Gleason score of ≤ 6; 85% of the patients had been diagnosed within the last 3 months. During the evaluation, half of the patients felt it was ‘not likely’ they would be given a diagnosis of prostate cancer; only 37% of their partners were that optimistic. Because the patients were relatively young it was not surprising that 54% had no comorbid conditions.

Pearson product-moment correlation of the sexual function measures indicated that the SAQ total score was linearly related to the selected core sexual function questions from the BISF-W and BSFQ (P < 0.001). anova indicated there were no significant mean differences of sexual function within groups of the study sample for ethnicity (P = 0.095), educational level (P = 0.440), patient's PSA at diagnosis (P = 0.124), and patient's Gleason score at diagnosis (P = 0.306).

Student's t-tests were used to confirm that there were no significant mean differences of key variables within the two largest ethnic sample populations (non-Hispanic whites and Hispanics). There were no significant mean differences for depression, sexual function and dyadic relationship, but there were (P = 0.039) for psychological distress (POMS score), with Hispanics showing a higher mean score.


The data were initially analysed using independent t-tests to determine if there were any significant differences in the individual men (patient) and women (partner) populations. As the term ‘patient’ was redefined for this study to include the prostate cancer ‘couple’, paired-sample t-tests were then used on all the data, consistent with similar studies [8,17], to ascertain whether the patient and his partner scored significantly different from each other.

For sexual function (Table 2), compared with the mean scores of normal controls and of depressed and impotent men examined with the BSFQ, the men in the current study appeared to present at diagnosis with several sexual function issues [11]. In the areas of sexual satisfaction, interest in sexual activity and physiological competence, the men's mean scores were similar to depressed and/or impotent populations rather than normal controls (Table 2). For example, 65% were dissatisfied with their sex life, 55% expressed decreased frequency of sexual drive and sexual thoughts and 60% reported erections that were insufficient for penetration.

Table 2.  Partner and patient sexual function means
Questionnaire variablesScores
  • *

    Lower mean score suggests fewer problems affecting sexual function. In all other variables higher mean score suggests higher functioning.

  • †Lower mean score suggests higher physiological competence.

Female dimensions (BISF-W)[18]Healthy with partnersMenopause surgicallyPresent study 
D1: Thoughts and desires 5.31 2.5 3.67 
D2: Arousal 6.21 3.85 3.61 
D3: Frequency of sex activity 3.9 1.83 3.9 
D4: Receptivity/initiation 8.85 6.39 9.96 
D5: Pleasure/orgasm 4.91 2.26 2.74 
D6: Relationship satisfaction 8.9 7.07 5.24 
D7: Problems affecting sex function* 4.47 5.55 5.36 
Composite score    
(D1 + D2 + D3 + D4 + D5 + D6-D7)33.618.3524.1 
Male dimensions (BSFQ)[11]ControlsDepressedImpotentPresent study
Sexual activity/performance36.817.224.136.58
Satisfaction13.7 6.6 9.4 7.25
Interest 9.2 7.2 9.1 5.89
Physiological competence 3.3 4.4 11.6 6.23
‘Paired’ sexual function scoresPatient meanPartner mean P  
Total SAQ51.7055.75 0.018 
Total BSFQ and BISF-W40.5749.61<0.001 

When comparing scores from the BISF-W [18] administered to healthy women (mean age 40.4 years) with partners and with surgically menopausal women (mean age 47.1 years), the present women functioned similarly to the surgically menopausal group, especially in the areas of arousal, pleasure/orgasm and relationship satisfaction (Table 2). In only two of the seven BISF-W subtests (frequency and receptivity/initiation of sexual activity) did the present women have equivalent mean scores to healthy women with partners.

On the SAQ, the paired t-test results indicated that the partners were significantly more positive about the communication and sexual components in their relationship than were the patients (patient/partner means 51.70/55.75, P = 0.02). Although the BISF-W was modelled after the BSFQ, the actual questionnaires were not identical. Twenty questions from each measure were re-coded to allow for comparability. Paired-sample t-tests showed that the mean partner score was significantly higher than that of the patients (patient/partner means 40.57/49.61, P < 0.001), showing that partners perceived their sexual performance at a better level. Results also established that the partner was significantly more satisfied in the relationship than the patient. The patient had significantly higher mean scores on the frequency of sexual thoughts (patient/partner means 2.96/1.96, P < 0.001) and the level of pleasure felt from sexual experiences (patient/partner mean 3.88/1.85, P < 0.001). On those items that monitored the accuracy of the patients’ perceptions of their sexual function, the partners rated the patients significantly lower in ability to gain erections, with patient/partner means of 2.67/4.52 (higher number lower score; P < 0.001) and to perform sexually, with patient/partner means of 1.38/4.68 (higher number lower score; P < 0.001). These results further characterize the disparities in self-reported levels of sexual functioning and emphasize how important it is to hear the voice of the ‘couple’.

Although the mean BDI scores for patients and partners (Table 3) were significantly different, with the mean depressed mood level for partners significantly greater, these mean values were within the normal range. Nevertheless, the frequency of depression in the study population (≈30%) was (patient/partner): mild-moderate, 26.2%/19.2%; moderate-severe, 2%/10.2%; and severe, partner only, 2.2%[13]. An independent t-test assessing the patients and partners resulted in similar findings.

Table 3.  Means and P values of affective and dyadic adjustment scores
Affective and dyadic adjustment scalesMean scoresP
  1. NS, no significance difference.

BDI (lower mean score suggests less depression)
Total  5.63  8.130.006
POMS factors (lower mean score suggests less psychological distress for all variables except vigour/activity)
Tension-anxiety  9.41  11.690.006
Depression-dejection  6.05  9.030.02
Anger-hostility  5.74  7.100.043
Vigour-activity 18.91 17.37NS
Fatigue-inertia  4.39  6.170.001
Confusion-bewilderment  6.60  7.760.007
Total 51.52 59.040.024
DAS factors (higher mean score suggests healthier dyadic adjustment)
Dyadic consensus 53.64 53.99NS
Affective expression  11.64  11.70NS
Dyadic satisfaction 27.36 27.30NS
Dyadic cohesion 16.29 16.44NS
Total108.30 110.38NS
VAS factors (lower mean score suggests less stress)
Stress during evaluation  4.92  6.060.004
Stress before treatment  4.93  6.24<0.001

The mean POMS total mood disturbance score of the partner was also significantly higher than the patient’s. Of the six mood states measured, the means, all within the normal range, of the tension-anxiety, depression-dejection, anger-hostility, fatigue-inertia and confusion-bewilderment subscales were significantly greater in the partner than in the patient. These findings were similar to the scores computed for men and women independently. The VAS results confirmed that partners were significantly more distressed at evaluation and diagnosis than the patients. There were no significant differences in the means of the patient/partner responses on the DAS or its four subtests.

To assess the strength of the instruments in measuring study variables, multiple regression analysis was conducted to evaluate if the total BDI, POMS and DAS score (and/or their subtest scores), as well as any sociodemographic variable, predicted sexual function. The analysis showed that the predictors of sexual activity were dyadic cohesion (DAS subtest), ethnicity, employment status and the patient's thoughts about his chance of being diagnosed with cancer. The linear combination of these factors in the patients and partners was significantly related to sexual activity (P < 0.001). The sample correlation coefficient was 0.542, indicating that ≈ 30% of the variance of sexual function could be accounted for by these factors.

Correlations assessed between depression-dejection (POMS subtest) and the BDI, and between comparable BISF-W/BSFQ items and the SAQ questions, as well as the results from the multiple linear regression analysis, were helpful in validating that the instruments chosen for the study created an integrated and focused tool.


Since the mid-1990s there has been extensive research into health-related quality of life (HRQoL) aspects of prostate cancer, concentrating on the patient's level of functioning and emotional status [5,19] No HRQoL studies have examined sexual function, depression, psychological distress and dyadic relationship levels in the patient and his partner. The present study provides evidence that the prostate cancer ‘couple’ at diagnosis has complex and disparate needs, and shows a need for psychosexual interventions that augment current treatments for sexual and psychological problems arising from a prostate cancer diagnosis.


Researchers have discovered that partners are often the communication conduit between the patient and his physician, in addition to serving as key advocates during diagnosis and treatment [9]. However, studies have reported that the partner often feels ignored by the patient's physician [9]. Sneeuw et al.[17] advocated that spouses serve as proxy ‘raters’ when assessing HRQoL of the patient. Therefore, the physician might wish to include the partner in the consultation process, with informational and emotional support offered to both [17,20]. There might also be poor agreement between the professional and patient assessment of distress, with physicians underestimating severity [20]. Using questionnaires before treatment, as in the present study, seems to be informative to the physician and therapeutic for couples, as it stimulates discussion about sexual and emotional issues relevant to the couple's concerns [20].


In the 1994 Memorial-Sloan Kettering Cancer Center study, researchers examined the effect of disease stage and treatment regimen on HRQoL in patients with prostate cancer and in their partners [8]. The authors concluded, as in the present study, that partners had significantly higher levels of psychological distress, and needed to be assessed regularly to identify how that distress might affect the patient's adaptation to his prostate cancer. Patient and partner depression were measured in a later study [7], and 31% of evaluable patients and their partners were referred for psychological assessment. The present study confirmed that, at diagnosis, 30% of both patients and partners might be appropriate for referral for further evaluation and treatment, and that partners of patients with prostate cancer have significantly higher depression scores than the patients. The question remains as to whether patients and partners are being evaluated and/or referred by their physician for their depression, a diagnosis that might ultimately affect adjustment to treatment, and the outcome.

The POMS total mood disturbance score and the subtests of depression-dejection, tension-anxiety, fatigue-inertia, confusion/bewilderment and anger/hostility were also significant, suggesting that the partners are more psychologically distressed and, in fact, might be distracted (higher confusion/bewilderment score) by their distressed mood. In addition, the Hispanic patient and his partner were significantly more distressed than the white non-Hispanic couples. Together with the fact that the partner's mean score on each VAS was significantly higher than the patient’s, we concluded that partners are more likely to be psychologically distressed than patients, and this should be carefully considered, as the partner's psychological distress might predict a greater HRQoL problem index for men with prostate cancer [1].


The return of sexual function after treatment for prostate cancer depends on patient age, the clinical and pathological stage of his disease, the treatment method and, in the surgical patient, whether the neurovascular bundles have been preserved [21]. In the present study, 69 of 103 patients elected to have radical surgery; 46 (66%) of those patients reported normal erectile function to the urologist at diagnosis. Interestingly, in the same study population, 75 men (78%) admitted having erections that were not sufficient for penetration, when questioned more specifically. At best, 66% of all patients, irrespective of age, are potent after radical surgery [21,22]. Patients undergoing external beam radiation therapy might, because of age and stage of disease, have poorer sexual function outcomes [4]. Therefore up to half of all patients treated for prostate cancer might have erectile dysfunction 12 months after treatment. Moreover, partners reported that patients had lower levels of sexual performance and poorer quality erections than study patients themselves reported. Thus, if partners were asked to assess the patient's erectile function after treatment, it might be even less than half of all patients treated that have erections adequate for sexual activity. The partner's decreased level of sexual function must also be included in the equation. Given the emphasis that many patients place on sexual function, physicians might find it helpful to address the challenging problems facing 50% of men with erectile dysfunction that might not be resolved by current medical therapies.

While technological assistance for erectile dysfunction (e.g. injections, prostheses) might help some patients adjust to what they believe is a loss of virility, partners might find these methods unappealing [23]. To confound matters, aids to promote erections might make patients feel vulnerable and awkward and, ultimately, produce barriers to seeking any kind of help for sexual problems [4]. It has been reported that physicians avoid discussions on the impact of sexual problems because of lack of knowledge or comfort with issues of sexual intimacy [1,9,23]. This leaves couples to ‘grieve’ in silence over the loss of an integral part of their marriage [1,9]. Thus, some investigators, including the present authors, suggest that counselling should be expanded from primarily helping patients to choose an appropriate prostate cancer treatment, to facilitating the couples’ successful adjustment to treatment outcomes [1].

Partners are extremely reticent about addressing sexual issues for fear of further increasing patient anxiety about the diagnosis of prostate cancer [5]. Tacitly they wish for patients to have erections, understanding that erections are how men define their masculinity [6]. Patients, on the other hand, are demoralised by their changing role and the potential for impaired performance, and have reported that they assume, based on their partner's silence, that the loss of sexual relations has little effect on the partner [1,9]. Left to adjust to the loss of their sexual relationship in silence, both the patient and partner often feel isolated [9]. For couples, this isolation can lead to distress, and distress might place patients with prostate cancer at risk of poor adjustment to their disease. Recognizing from the present study that the partners are already distressed over the diagnosis and the treatment choices, the additional distress caused by sexual isolation could be even more detrimental to patient outcome. As perhaps only half of all patients who report they are potent at diagnosis will have an acceptable level of potency after treatment, it is critical to develop new psychosexual educational interventions as alternatives for pharmacological or technological options. Even before any decision is made about treatment, discussion of these issues could help to address differences about issues of sexual activity, psychological distress and intimacy and, ultimately, enhance sexual satisfaction after treatment.

Most of the couples studied were in long-term relationships. Moreover, the scores on the DAS describe a sample of couples that showed dyadic cohesion, consensus, affection and satisfaction. In a study examining sexuality and marital life, Trudel [24] concluded that there was a statistically significant relationship between marital functioning and sexual behaviour. As dyadic cohesion was a predictor of sexual function in the present study, there is obvious potential for prostate cancer ‘couples’ to adapt to the sexual function outcomes of treatment. That adjustment is possible if couples continue to communicate during the diagnosis, treatment and the recovery process on critical sexual issues, and if partners can be encouraged to be active in the decision-making process regarding treatment and its sexual consequences.

Because of the nature of the present population, the study design had at least one limitation. A convenience-sampling model was used to recruit patients seeking treatment at an academic outpatient clinic, but the sample might not be representative of the entire population of patients with prostate cancer and their partners. Nevertheless, the information from this study could be very useful in constructing a receptive environment (i.e. sexual, psychological and dyadic assessment; psychosexual educational interventions) where the newly diagnosed patient with prostate cancer and his partner can, with their physician, begin to reflect on the WHO definition of sexual heath, the ‘integration of somatic, emotional, intellectual and social aspects’ of being sexual [3], in ways that might once have been considered taboo.


This study was supported by the Spector Family Foundation donation awarded to Mark S. Soloway, MD, to use at his discretion to fund prostate cancer research.


None declared. Source of funding: private donation from Spector Family Foundation.