My Road Ahead study protocol: a randomised controlled trial of an online psychological intervention for men following treatment for localised prostate cancer

The efficacy of psychosocial support has been repeatedly reported in the general cancer setting. Unfortunately, access to evidence-based and timely psychosocial support is often limited, particularly in rural and remote regions of Australia but also in public hospitals across the country and, indeed, worldwide. It has been reported that only 14 to 21% of Australian women with cancer attend face-to-face support [16, 17]. Men are lower utilisers of health services in general [18] and international reports indicate that mental health service utilisation by men is lower than women [19].

A recent systematic review of psychosocial interventions for men with prostate cancer and their partners highlighted the poor quality of current research in this field [20]. The authors concluded that group cognitive-behavioural and psycho-educational interventions with a specific focus on cancer specific distress show the most robust research evidence for efficacy, however, further research is required [20]. Another review explored psychosocial interventions addressing sexual or relationship functioning in men treated for prostate cancer [21]. This review concluded that there is some evidence to support the efficacy of psychosocial interventions in improving sexual functioning especially when delivered face-to-face or through utilisation of complex interventions that specifically target sexuality and the relationship. However, this review also concluded that further research is required [21]. Related to sexuality and intimacy, masculinity is emerging as an important area [22] and yet it is not routinely addressed and integrated into interventions.

Mental and behavioural health promotion, prevention, treatment and management-oriented interventions that are delivered via the internet or other electronic technologies, with or without human support [23], often referred to as “e-Interventions”, can overcome many barriers of access that are commonly encountered in our healthcare system. Whilst the use of e-Interventions has grown in both the mental health and general health setting [24‐26] it has received limited attention in the cancer setting.

In the cancer setting, most online research has focused on developing and evaluating online peer support interventions with mixed outcomes in female cancer samples [27‐29]. van den Brink [30] developed an online information and support intervention for head and neck cancer patients in the Netherlands with the aim of overcoming the communication and information bottlenecks in supportive head and neck cancer care. The authors reported that patient use and satisfaction with the intervention was very high despite patients having recently undergone intensive surgical treatment. Patient age was not a barrier (mean age 59 years and range 38-78) and, despite 56% of participants having limited computer experience prior to the use of the intervention, consistently positive feedback was received. This online support intervention showed improved QoL outcomes in participants as compared to those in a control group who did not access the intervention [30]. However, the intervention focused on psycho-education and support rather than skills in coping with the experience of cancer diagnosis, treatment and recovery.

Schover et al. [31] developed and assessed the efficacy of an online sexual counselling intervention in comparison to traditional face-to-face sexual counselling for couples following treatment for localised prostate cancer. Sexual counselling involved three face-to-face sessions with a therapist while the online intervention delivered the therapy via email contact with a therapist. This study found that the delivery of the intervention via the internet was as effective as the traditional face-to-face intervention in producing enduring improvements in sexual outcomes [31]. While this study highlighted the potential utility of online interventions in this population, to date, no online self-directed intervention with minimal support and a focus on a range of identified problem areas has been developed and evaluated for localised prostate cancer.

A recent review by Leykin et al. [32] explored the role of online interventions in supporting people with cancer. This review highlighted that e-Interventions can provide the majority of people with an acceptable alternative support modality in a systematic and scalable way. Research has found that telephone and internet-based support is particularly important for men with cancer, predominately prostate cancer, living outside metropolitan regions in Australia [33]. Psycho-education and self-directed therapeutic techniques can easily be transferred to an online environment and may also break down the barriers that many people face in terms of accessing support, including lack of available resources and perceived and personal stigma, fear and uncertainty in accessing mental health services [32, 34]. Leykin et al. [25] criticised current online resources for being predominately information and support based, with a need for evidenced-based treatment interventions.

The current study involves a randomised controlled trial (RCT) assessing the effectiveness of three online interventions: (i) access to an online intervention program (My Road Ahead-only); (ii) access to the online intervention plus to a moderated online forum (My Road Ahead-plus-forum); and (iii) access to the online form only (forum-only). This study design was chosen to evaluate the online intervention in comparison to a more basic level of care (some peer support) and to separate the effect of the online intervention and the peer support forum on treatment outcomes. Past RCTs have used a variety of control conditions from wait-list control groups, care as usual, or some basic information provision. Cuijpers et al. [35] have criticised past internet-based RCTs for failing to use control groups beyond a wait-list control, since there may be an overestimation of intervention efficacy when those in a wait-list control condition are unlikely to make self-directed steps to improving their situation if they are “waiting”. A web-based comparative condition has been recommended to overcome these problems [36]. Basic moderated forums have been used in previous studies as a control group as a means to control for spontaneous improvement or to control for the general attention and the possible beneficial effects from sharing one’s experience with others (e.g. [37, 38]). In addition online support groups may themselves have positive effects for people with cancer (e.g., [28, 29]), and the use of three groups would allow for exploration of the relative benefits of the online psychological intervention and peer support (forum).

Primary outcome:

Psychological distress as measured by the combined total score of the DASS-21 will be used as the primary outcome measure.

It is hypothesised that:

Secondary outcomes:

It is hypothesised that:

Mediation and moderation effects:

It is hypothesised that:

This study is registered with the Australian New Zealand Clinical Trials Registry; Identifier: ACTRN12611000278932. Ethical approval to conduct the study has been obtained from Melbourne Health, Swinburne University of Technology, Deakin University and Peter MacCallum Cancer Centre Human Research Ethics Committee’s. CONSORT [39] requirements for RCTs will be adhered to throughout the study.

Participant consent will be obtained using an online consent process. The consent form is provided to the participant in PDF format. Participants are asked to confirm that they understand the study conditions, that they have received a copy of the participant information and that they consent to take part in the study by selecting three check-boxes during the consent process. Participants will then move on to create a personal, password protected account on the My Road Ahead website and complete the baseline questionnaires. After completing the baseline questionnaires participants will then be randomly assigned using a computer generated sequential 1-to1- to 1 allocation across the three (3) groups. Participants and researchers involved in the management of the project are not blinded to the randomisation outcomes. Group 1 participants will receive access to My Road Ahead, the 6 module online program. Group 2 participants will get access to My Road Ahead plus access to the moderated forum. Group 3 participants will receive access to the moderated forum only.

Participants in all groups will receive a weekly email reminder to return to the program and/or forum and to encourage them to continue to participate in the research over the course of the 10 weeks of the intervention phase. Participants will also receive three reminder emails to complete the questionnaires at each time point.

Participants in all groups will undertake the online assessment at five time points; baseline, week 5, post-intervention (week 10 for group 3) and 12 weeks post intervention (week 22 for group 3), and 6 months post-intervention (see Figure 1).

A medium effect size is estimated for the primary outcome measure, total distress, as measured by the DASS-21 total score between groups across the five assessment points. To power the study at the 80% level, based on an alpha level of .05 with an estimated medium effect size (f = .25) and with three groups completing five assessment points, 125 complete responses will be required. To account for an estimated 20% drop-out rate a total of 150 participants will be required; 50 per group.

Data will be analysed using repeated measures Multivariate Analysis of Variance (MANOVA) with between and within subjects factors examined while controlling for age. To examine any bias that may arise as a consequence of attrition, baseline data will be examined to explore if there are any statistically significant factors associated with participants who drop out compared to those who do not. This will allow an analysis for attrition bias in the final results based on the predicted probability of attrition as recommended by Rubin [49] and Heckman [50]. An intention to treat analysis and then longitudinal hierarchical linear model analyses will be performed [51, 52], avoiding the need for the imputation of missing data.