Self-management after prostate cancer treatment: evaluating the feasibility of providing a cognitive and behavioural programme for lower urinary tract symptoms



Study Type – Therapy (case series)

Level of Evidence 4

What’s known on the subject? and What does the study add?

Self-management intervention has been shown to be effective for lower urinary tract symptoms (LUTS) in benign prostate disease.

The evidence base for self-management of LUTS after treatment for prostate cancer is small and characterised by heterogeneous samples in terms of disease stage, treatment profiles, timings and intervention content.

This pilot study provides data suggesting that a narrowly targeted, cognitive and behavioural self-management intervention can improve LUTS in men who have had radiotherapy treatment for prostate cancer.

The intervention could be applied in routine practice and further testing in an adequately powered randomized controlled trial is required.


• To test the feasibility of a self-management intervention to help men cope with lower urinary tract symptoms as a result of radiotherapy for prostate cancer.


• A quasi-experimental design was used incorporating a pre-post-test evaluation. In total, a population of 71 men were screened for moderate to severe urinary symptoms 3 months or longer post-radiotherapy. Of these mean, 22 were recruited into the intervention from an eligible population of 43 symptomatic men.

• Urinary symptoms were measured before the intervention and again after 4 months of follow-up through International Prostate Symptom Scores (IPSS) and bladder diaries.

• Health-related quality of life was measured in relation to cancer per se and prostate cancer specifically, and confidence to cope was measured by a self-efficacy questionnaire.

• The self-management intervention comprised pelvic floor muscle exercises, bladder retraining, patient education and problem solving and coping strategies


• Lower urinary tract symptoms, as measured by the IPSS, showed a significant improvement, with a median score change of 5 (P < 0.005).

• This was supported by objective changes in median bladder void volume of +7.5 mL (P < 0.05) and the median number of daily voids of −1 (P < 0.005).

• In addition, decreases in emotional distress and problems associated with urinary function suggest that the intervention had a positive impact on health-related quality of life.


• The provision of such an intervention was feasible within the clinical setting and provided benefits for men.

• Symptom change vs those of normative recovery values for IPSS showed an intervention effect.

• This intervention could be applied in routine practice and further testing is required in a randomized controlled trial.


European Organisation for the Research and Treatment of Cancer


health-related quality of life.


Surviving prostate cancer is becoming more common as a result of earlier diagnosis and improved treatment of the disease [1]. In Europe, the relative mean 5-year survival rate for all men diagnosed with prostate cancer is 76%[2] and numbers of survivors are predicted to rise over the next decade. Although men with prostate cancer now live longer, there is a growing recognition that the persisting impact of chronic symptoms associated with the disease and its treatment also needs to be considered [3–6]. Although behavioural, psychosocial and educational programmes have been developed to address coping with a prostate cancer diagnosis and illness [5,7–12] and symptoms as a result of cancer treatment [13–23], the evidence base for managing long-term and enduring side-effects is still small. In addition, the economic consequences of addressing long-term symptoms have an impact not only for the individual, but also for health services, yet measures for assessing these are not readily available.

The effects of prostate cancer treatment can have a long latency period [24,25]. Symptoms such as urinary frequency, incontinence, changes in bowel function, sexual impairment and fatigue are commonly experienced by men after their initial treatment [26] and, over time, these symptoms can have various aetiologies: they may reduce, become a lasting adverse effect, or increase as men age. The incidence of moderate to severe urinary symptoms, as defined by the IPSS at 6 months is 37% for hormone therapy, 39% for radiotherapy and 28% for surgery [27]. Longitudinal studies of the latest sophisticated radiotherapy techniques such as intensity modulated radiotherapy indicate that symptoms decrease substantially at 8 years, although a significant number of men continue to experience symptoms: 9% LUTS, 1.6% rectal symptoms and 49% of men report ongoing sexual dysfunction [28]. In addition, population-based needs assessments consistently show that symptoms persist after prostate cancer treatment [29,30] and are associated with psychosocial concerns [6,31]. A recent UK study exploring supportive care needs confirmed this, finding that 97% (n= 727) of men treated for prostate cancer reported LUTS of frequency, nocturia, urgency and dysuria [32]. Over half of these men reported their symptoms as moderate to severe. Furthermore, those with greater psychological distress had greater levels of disease and greater severity of symptoms. The high prevalence of enduring symptoms in men treated with prostate cancer may be indicative of the stigma associated with urinary dysfunction, especially incontinence, together with the lack of efficacy of medication and assessment for longer-term management [33,34]. Reticence on the part of patients and clinicians alike raises significant barriers to diagnosis and treatment [35,36]. Logically, men’s post-treatment experience will be optimized if the enduring side effects and supportive care needs are taken into account [15].

Self-management as an intervention is increasingly being used for managing chronic symptoms and has been shown to be successful in managing benign prostate disease [37]. This approach involves the patient being actively involved in their care. It is clinician-led and encompasses an interactive process whereby individual responses and behaviour are addressed and shaped with the intention of managing the physical and psychosocial consequences of disease. By sensitively targeting men’s needs and promoting men’s motivation to engage and succeed, self-management can offer the means to ameliorate the effects of prostate cancer and its treatment [38].

The present study is a Phase II quasi-experimental, pre-post study [39] aimed to test the delivery of a self-management intervention, and to determine whether such an approach could be effective in reducing LUTS for men who had received radiotherapy for prostate cancer. Recruitment concentrated on men post-radiotherapy because profiles of treatment side-effects vary across treatment types [27,40–42], and radiotherapy is currently under-researched in terms of side-effects management. The present study also aimed to develop a feasible protocol for subsequent trial comparison. Two hypotheses were tested: (i) compared to baseline scores, participants in the intervention programme would report a significant reduction in urinary tract symptoms as defined by the IPSS; (ii) those men receiving the intervention would report improved health-related quality of life (HRQL) over baseline and greater perceived ability to cope with their illness and its associated problems.


The present study had local research and ethics approval. The intervention was carried out at a cancer centre over a period of 18 months. Eligibility was assessed by screening to ensure that delivery was only to those men with defined urinary problems. Study participants were identified and approached by clinicians and informed consent was taken. Men were eligible to participate if they had locally confined prostate cancer disease (up to stage T3bNO) and were at least 3 months post-completion of radiotherapy. Participants were screened using the IPSS tool and were invited into the study if they had an IPSS score ≥ 8, which is defined as moderate to severe symptoms.

The self-management intervention was led by a specialist prostate cancer nurse trained in cognitive-behavioural techniques. Delivery was in three individual sessions of 60 mins and one group 90 mins session, all of which took place at fortnightly intervals over a period of 2 months. Participants went through the intervention in cohorts of five. The programme included two components: (i) a cognitive component involving problem solving, skill-building, coping strategies for symptom management, recognizing urinary problems, information provision and emotional support; (i) a behavioural component involving self-monitoring of symptoms and bladder retraining techniques, including pelvic floor exercises and biofeedback (Table 1). Pelvic floor exercises are traditionally used to address incontinence post-radical prostatectomy, although there is a growing acknowledgement that pelvic floor exercises can help ameliorate LUTS associated with the urge to void [37] and these were therefore included as part of a range of techniques.

Table 1. Intervention delivery
Week 1Individual session
Detailed assessment, problem identification, skill-building and coping strategies for symptoms, goal-setting, individual coaching
Week 3Group session
Pelvic floor exercise, information provision
Week 5Individual session
Self-monitoring of symptoms bladder retraining techniques and biofeedback
Week 7Individual session
Reviewing urinary problems and goal achievement. Redefining goals for the longer-term, emotional support

After completion of the intervention, participants were monitored by the specialist nurse via telephone at 1, 2 and 4 months. To ensure consistency of delivery, the sessions and follow-up calls were recorded by use of assessment sheets, which covered baseline characteristics, current situation, desired goals, learning and progress of behavioural techniques, assessment of change and review for the future. The 4-month follow up meant that each individual participant was involved with the programme for a total of 6 months.

Evaluation of participants was carried out separately by the researcher using a range of self-completion assessments applied at baseline (immediately before the intervention) and study end (immediately after the final telephone follow-up session, 6 months after baseline). The IPSS questionnaire was used as the primary outcome measure. This tool is used extensively for assessing LUTS. Its base questionnaire, the AUA-7, has a Cronbach’s α of 0.86, suggesting the tool has good internal validity, with excellent test-retest reliability (r= 0.92) [43,44]. The scale contains seven questions addressing a range of urinary problems. Each answer range is in six categories (0–5) resulting in a maximum value of 35. A score < 8 indicates mild symptoms, a score of 8–19 indicates moderate symptoms and a score of ≥20 indicates severe symptoms. To provide objective data, participants were also asked to complete a bladder diary over the course of 7 days at each of the two assessment points. This involved keeping a log of frequency and volume of passing urine and any incidence of incontinence. Bladder diaries have been utilized clinically to assess LUTS, as well as to give an indication as to the extent of frequency and volume of urine passed [45].

Secondary outcomes were measured using the European Organisation for the Research and Treatment of Cancer (EORTC) QLQ-C30 and QLQ-PR25. The EORTC QLQ-C30 is a 30-item HRQL instrument developed to assess HRQL in patients with cancer. There are five functional scales (physical, role, emotional, cognitive and social functioning) along with global quality of life and symptom scales. Questions are scored on a four-point scale (seven points for global quality of life) and a linear transformation is applied to standardize the raw scores and produce an Item Score in respect of each functional scale. Item scores are in the range 0–100, with a higher score representing a higher level of functioning [46]. This was used in conjunction with the EORTC QLQ-PR25, which is designed for use amongst patients with localized and metastatic prostate cancer. It includes subscales assessing urinary symptoms (nine items), bowel symptoms (four items), treatment-related symptoms (six items) and sexual functioning (six items).

Self-efficacy, which is seen as a key indicator of the ability to self-manage and effect behaviour change, was also measured [47]. For the present study, the six-item scale for managing chronic disease developed by the Stanford Patient Education Research Center was used [48]. Data published for the six-item scale show an internal consistency reliability of 0.91. The scale covers several domains that are common across chronic disease, including symptom control, role function, emotional functioning and communicating with physicians. Scores are in the range 1–10, where 10 represents being very certain of being able to engage in an activity and 1 represents being very uncertain.

Descriptive statistical analysis was used in respect of the population and sample base and non-parametric Wilcoxon signed rank tests were applied to explore the difference between the outcome measures pre- and post-intervention.



From a population of 71 men who had received radiotherapy for prostate cancer, 43 men (43/71; 60%) had IPSS scores of ≥8. Over half the men (22/43; 51%) agreed to participate in the study with 21 (21/43; 49%) refusing through a variety of reasons (from not wishing to pay for additional parking costs, to co-morbidity restricting mobility). Those not recruited tended to be further from the end of their treatment (median of 39 months not recruited/4 months recruited). All the men approached had received neo-adjuvant hormone treatment before and during radiotherapy. Dosage data were recorded for 68 of the men, of whom 35 (35/68; 51%) had a treatment dose of ≥70 Gy.

Of the 22 men who were recruited to the study, seven (7/22; 32%) dropped out during the course of the programme and 15 (15/22; 68%) participated in all stages of the intervention and assessments. Of the seven that dropped out of the intervention, one died of unrelated causes, three did not commence the intervention (one because of travel commitments and two for no reason) and three dropped out during the course of the sessions (one because of co-morbidity, one failed to submit their final questionnaires and one for no reason) (Fig. 1). The profile of those who completed vs those who dropped out was very similar in relation to treatment dose and time out of treatment (median respective radiotherapy dose 70/70 Gy, median respective time from treatment 4/3 months).

Figure 1.

Recruitment flow.


Baseline IPSS scores for the 15 participants who completed the intervention was in the range 8–26, with a median of 11, and the distribution was positively skewed. Some 14 participants had moderate symptoms scoring in the range 8–20, and one had severe symptoms scoring 26 on the IPSS. The bladder diary was kept correctly by 14 of the participants. One participant did not record actual urine levels, but only numbers of voids, and his volume data were therefore not included. Overall median void volume across the 14 participants was in the range 80–260 mL per void and median void frequency per day for all 15 participants was in the range 6–12 voids.

HRQL baseline scores on functioning items and global quality of life as measured by the EORTC QLQ-C30 were equal to, or slightly higher than, EORTC reference values for patients with prostate cancer with T1–T4 tumours [49]. These slightly higher than expected HRQL scores are reflective of the high proportion of patients in the present study who were further out of treatment, and also indicate that all the men had disease up to stage T3bNO. By contrast, the normative data covers patients with disease up to stage T4, which would include men who are more symptomatic than this sample. EORTC reference values are only available for mean scores. Rounded mean baseline scores and reference values were: emotional functioning, 86/82; physical functioning, 87/78; role functioning, 77/75; cognitive functioning, 85/80; social functioning, 83/83; and global quality of life, 71/69. Baseline median scores, which are a more appropriate measure for this small sample size, confirm the above (Table 2). The median (range) baseline item score for urinary symptoms HRQL, as measured by the EORTC QLQ-PR25, was 29 (13–50).

Table 2. European Organisation for the Research and Treatment of Cancer (EORTC) QLQ-C30 baseline scores and reference value scores
ItemBaseline score*EORTC QLQ-C30 reference values for prostate cancer T1–T4
  • *

    Item scores are in the range 0–100, with a higher score representing a higher level of functioning.

Emotional functioning83.385.682.3
Physical functioning86.787.177.6
Role functioning83.376.775.0
Cognitive functioning83.385.279.8
Social functioning100.083.383.2
Global quality of life66.771.168.6

The median baseline score for self-efficacy was 7.66 For this sample, baseline levels of self-efficacy were above the normative values for subjects with chronic disease, as reported by the test authors [48]. By comparison, mean ± SD (range) scores, respectively, were 7.90 ± 1.50 (5.66 to 10.00) and 5.17 ± 2.22 (1–10).


After completion of the intervention, urinary symptoms measured by the IPSS were significantly reduced. For the 15 men who completed the intervention median scores dropped by 5 points (from 11 to 6) and the range reduced from 8–26 to 2–14 (z=−3.048; P < 0.005). All the men except one showed some reduction in symptoms. A comparison of pre- and post-IPSS scores is given in Fig. 2. An intention-to-treat analysis for the 22 men recruited to the present study confirms the reduction in symptoms (median = 11.5/9.5). These improvements in perceived symptoms were supported by significant improvements in actual void frequency and volume measures as recorded in the bladder diaries. Median volume per void increased from 150 mL to 157.5 mL (z=−1.995; P < 0.05) and the median number of voids per day decreased from 9 to 8 (z=−3.048; P < 0.005).

Figure 2.

IPSS scores pre and post the intervention.

HRQL scores also showed change. After the intervention, the participating men were less emotionally distressed than before the intervention: median item scores increased from 83.33 to 100 (z=−2.375; P < 0.05). Men’s urinary symptom HRQL was also improved: median item scores reduced from 29.17 to 22.22 (z=−2.832; P < 0.05). Other HRQL dimensions were stable, with little difference in physical, social, role or cognitive function or global quality of life. Pre- and post-intervention medians and significance levels are shown in Table 3.

Table 3. European Organisation for the Research and Treatment of Cancer (EORTC) QLQ-30 and EORTC QLQ-25 pre- and post-item scores
ItemMeanMedianWilcoxon signed ranks test
Pre-item scoresPost-item scoresPre-item scoresPost-item scores z P
EORTC QLQ C30      
 Emotional functioning85.692.883.3100.0−2.3750.018
 Physical functioning87.188.086.793.3−0.5970.551
 Role functioning76.783.383.3100.0−1.3720.170
 Cognitive functioning85.283.383.383.3−0.5520.581
 Social functioning83.390.0100.0100.0−1.4910.136
 Global quality of life71.173.966.775.0−1.2700.204
EORTC QLQ PR25      
 Urinary symptoms30.821.329.222.2−2.8310.005

The men in the present study sample started from a very high level of confidence in coping and controlling disease symptoms and its effects, as measured by the self-efficacy questionnaire. There was no significant improvement in scores as a result of the intervention: median scores moved from 7.66 to 8.33 (z=−0.702; P= 0.483).


The present study hypothesis was that the provision of a self-management intervention based on cognitive and behavioural approaches would reduce urinary symptoms and improve HRQL, enhancing men’s ability to cope more effectively with enduring effects of prostate cancer and its treatment. The observed improvements in IPSS scores indicate that this hypothesis was met and that the intervention did have a positive, physical effect. This outcome was reinforced by the improvements in volume and frequency of voiding, as recorded in the bladder diaries kept by the men. In addition, decreases in emotional distress and problems associated with urinary symptoms reinforced the finding and suggest that the intervention had a positive impact on HRQL as well. Considering the effect of the intervention, the quasi-experimental nature of the present study should be acknowledged in that participants were recovering from therapy and the improvements observed may not entirely be a result of the intervention process. However, clinical studies examining IPSS scores suggest that a 1.34 point mean recovery, between 3 months and 18 months post-radiotherapy, can be expected naturally [50]. The greater change found in the present study (mean change 4.73/median change 5) would suggest an intervention effect. These findings would be comparable to the intervention effect found in men with benign prostatic disease [51].

To understand this observed effect, the narrowly targeted sample was influential. Studies providing self-management strategies and lifestyle change in men with prostate cancer, such as patient education and coping interventions, have traditionally focused on multiple symptoms and have recruited heterogeneous populations in terms of treatment received, time subsequent to treatment, disease stage and other individual characteristics. These have proved less effective in changing physical outcomes [7,11] although they have shown some benefits in improving satisfaction and HRQL [5,9,10]. In the present study the focus of the intervention was on urinary symptoms and its delivery was targeted at specific problems that men were experiencing and screened for before recruitment. As a study feature, the more homogenous sample would contribute to the positive intervention outcomes observed and would have clinical resonance in targeting self-management intervention to those who would most benefit [38].

The present study did not indicate an improvement in the mediating variable of self-efficacy, which has been seen in other behavioural and psychosocial intervention studies in prostate cancer [5,21] and would be expected in light of the significant improvement in physical symptoms found in the present study. At baseline, participants exhibited higher levels of self-efficacy than would be expected [48] and, from this high base, did not improve significantly over the intervention period. This could be a feature of the measure used. Of necessity at the time of commencing the present study, the scale used focused on measuring self-efficacy in relation to chronic disease, which may have less validity when applied to prostate cancer [13]. Self-efficacy is not a fixed personality characteristic but changes in relation to the task or life event at hand. For these men, prostate cancer symptoms and issues may be very different from chronic disease issues and future research may benefit from incorporating a prostate cancer-specific self-efficacy measure.

The present study also sought to identify the most meaningful and useful elements of the intervention in terms of participant benefit. Although the intervention was delivered in mostly individual sessions, it was apparent from the feedback that the group sessions, conducted in an open discussion format, provided a more supportive, empathetic and social environment, facilitating learning and adherence to the programme. Several studies have delivered cognitive behavioural interventions in group sessions and report benefits [5,8,37]. Carmact Taylor et al. [8] found that group sessions were positively received by participants and yielded good attendance and retention. However, in the present study, there were some men who preferred and valued the availability of individual sessions, mainly for the discussion of more sensitive issues, where they lacked the confidence to speak in a group setting. Individual needs are crucial to address [15] and an opportunity for these to be confidentially elicited should be available. For subsequent trial comparison, this intervention should be delivered across three or four small group sessions and include the additional opportunity for one-to-one sessions between the participant and the practitioner delivering the programme.

Guidelines emerged in terms of employing ways of enhancing intervention uptake and completion. Discussion in the sessions revealed that the nearer to treatment a self-management intervention takes place, the better the reported outcome could be. This echoes the identification of the ‘teachable moment’ by Denmark-Wahnefried et al. [52]. In that study, health behaviours amongst breast and prostate patients were explored and it was recommended that interventions are most valuable as near to diagnosis and treatment as possible when the receptivity and determination of patients is at its strongest. Further research needs to consider recruiting men very soon after the completion of their treatment.

Another finding to emerge from discussion was the importance of the clinician to the recruitment process. Clinician endorsement of the intervention played an important role in giving patients the confidence to participate [52,53]. In the present study, clinician’s views were found to work both negatively and positively. A confident appraisal of an intervention by a clinician has the effect of encouraging take-up, whereas a clinician that does not believe in, or does not know about the possibilities of intervention, can lead a patient to accept the status quo and not seek to do anything about his symptoms.

Identifying problems at clinical follow-up remains an issue to be examined. Sixty percent of the men screened for the present study had moderate to severe LUTS three months or more post-radiotherapy treatment. Despite the high prevalence identified in the present study, it is the case that LUTS are still considered by men and clinicians alike to be an inevitable consequence of prostate cancer therapy [54]. This is supported by surveys of men’s unmet needs, which describe an incidence of LUTS of 97% and identify the subsequent affect on HRQL [32]. The under-reporting of such adverse effects may be as a result of a lack of detailed LUTS assessment throughout the cancer diagnosis and treatment, or assessment techniques focusing on toxicity of treatment alone rather than underlying morbidity. Furthermore, men are unlikely to complain of such symptoms and this is reflected in the intervention assessment notes by only a minority of men in the present study having received advice in relation to symptoms. This further reflects the EpiLUTS study by Sexton et al. [53] in which less than a third of those suffering LUTS in the normal population reported visiting a healthcare professional for symptoms. Sexton et al. [53] go on to suggest that, when symptom assessment is not initiated by the clinician, patients believe that the symptoms are an inevitable part of the ageing process.

Although the high rates of LUTS reported in the present study indicate an unmet need as a result of prostate cancer treatment, this should also be considered in the context of levels of LUTS seen in healthy, age-matched populations. The EpiLUTS study group found that at least half of all men surveyed reported some of the symptoms defined as LUTS [53]. Common conditions such as benign prostate hyperplasia and overactive bladder result in storage, voiding and post-micturition symptoms that were seen more commonly in older age groups. These symptoms can have a relationship with sexual dysfunction, heart disease, metabolic syndrome, diabetes, depression and sleep disorders [55]. There is also evidence of overlap of storage symptoms, such as overactive bladder and post-micturition symptoms, and that a prostate, or prostate treatment, aetiology may not always be the simplistic cause of LUTS [53,56]. This would suggest that existing pharmacological management strategies fail to address symptoms in a differentiated manner [56]. There is clearly a need to create individualized treatment plans based on detailed assessment and a range of intervention strategies. In considering outcomes, future studies need to assess baseline scores before prostate cancer therapy. In addition, the positive results of the intervention reported in the present study suggest that, clinically, men require and can benefit from formal assessment, and that the use of the IPSS is beneficial in defining that need.

Understanding the long-term impact of cancer treatment requires an appreciation of symptom clusters and how these relate not only to bladder damage, but also to other existing co-morbidities. Radiation bladder toxicity is usually identified as a self-limiting overactive bladder; however, subsequent changes to pelvic blood flow, fibrosis and co-morbidity may all have an impact on urinary and pelvic symptoms [57–59].

This intervention may also be relevant to patient groups who have had other prostate or pelvic treatments because it is not necessarily focused on the physiology of side effects but on the coping strategies undertaken by the individual.

In conclusion, this intervention shows promise but requires further testing as part of a randomized controlled trial to better define its potential benefit. Self-management interventions may be one possible management strategy for late effects and a way of meeting men’s needs that could be integrated into clinical practice as part of a nurse-led, follow-up strategy or within primary care.


Funding is acknowledged for this study as being provided by the CR-UK as part of its small grants programme (Ref: CRUK/06/040).


None declared.