Breast cancer (BC) is a major public health problem, accounting for over 2 million cases worldwide each year [1]. In addition to population-based educational and public health policy interventions to minimize exposure to modifiable BC risk factors and optimize cancer screening, identifying women at increased risk and implementing risk-stratified, evidence-based prevention and intensified screening strategies for them is a priority [2]. Health care providers often have difficulty assessing and communicating BC risk as well as the absolute benefits and disadvantages of risk management interventions such as risk-reducing medication, surgery, and cancer screening [3,4].

Several tools exist to estimate BC risk based on personal risk factors, but none provides risk-adapted, individually-tailored, risk management information [5,6]. iPrevent was designed to help women and their health care providers, including primary care physicians (PCP), breast surgeons (BS), and genetics clinicians (GC), to assess and manage BC risk collaboratively [7]. It integrates BC risk estimation, using either the International Breast Cancer Intervention Study model or the Breast and Ovarian Analysis of Disease Incidence and Carrier Estimation Algorithm model (as appropriate for the woman’s risk factors), with tailored risk management information [8-10].

iPrevent users are first given a qualitative risk estimate according to Cancer Australia definitions: average or slightly above average risk (<1.5 times population risk at that age), moderately increased risk (1.5-3 times population risk), or high risk (>3 times population risk) [11]. Women can then choose to see their risk information displayed as a percentage, a pictogram, and a graph. Women are also provided with a menu of risk management strategies appropriate to their risk category, based on Australian National Guidelines [11], with more detailed optional information about each strategy, including estimates of the absolute (rather than relative) risk reductions for each medical and surgical intervention and tailored lifestyle advice.

The aims of this pilot study of patients and their clinicians were to assess the iPrevent prototype with regard to its clinical usability and the acceptability of its content and layout and to identify potential barriers to its implementation. Exploratory aims included assessing its potential impact on patient risk perception, anxiety, BC worry, and BC prevention knowledge.

This pilot study of the iPrevent prototype has found good usability and acceptability without evidence of an adverse impact on anxiety or BC worry. The observation that the 8 PCP recruited only 3 patients between them in the required 3-month period suggests that implementation of iPrevent into primary care might be substantially more challenging than implementation into the specialist setting, where recruitment of patients was much higher. Another interpretation is that the study requirements (eg, obtaining written informed consent) were onerous, especially for PCP in busy practices, and thus, the low recruitment by PCP in this study might not reflect the uptake of iPrevent in routine practice. However, as earlier focus groups had highlighted that PCP generally do not see BC risk assessment and management as being in their domain, iPrevent might be able to contribute to overcoming provider unfamiliarity and lack of confidence for this group of clinicians [3].

The prototype was considered too long by a majority of clinicians and some patients, indicating another potential barrier to implementation. Patients took a median of 15 minutes and up to 60 minutes to enter their risk factor data, and the subsequent median time taken for the clinician consultation using the iPrevent output was 20 minutes. To address this issue, we have now incorporated changes to streamline the data entry for family history. This study also highlighted the need for patients to be able to enter their data into iPrevent at home prior to a consultation.

iPrevent may improve BC risk perception given an additional 46% (16/35) patients accurately estimated their BC risk category after using iPrevent. As a higher perceived risk of BC is associated with considering medical prevention and risk-reducing surgery among high-risk women [24-26], iPrevent could become a potential behavior-modifying tool. While this pilot study provides no information about the uptake of risk management strategies after using iPrevent, this issue will be an important endpoint for future larger studies. Other studies have found that women who have access to more thorough information from genetic counselors, combined with support to make decisions, have a higher uptake of risk reduction methods [27-29]; thus, we hypothesize that iPrevent might have a similar impact.

Use of iPrevent did not appear to increase patient worry or anxiety, consistent with the literature that has found that decreased anxiety and better psychological outcomes are associated with improved accuracy of perceived risk [24,30,31]. Use of iPrevent seemed to improve BC knowledge, a recognized critical first step in helping individuals understand screening options, weigh potential benefits and risks for risk-reducing measures, and make informed decisions [32-34]. In addition, 89% (33/37) patients indicated that some or most of the information contained in iPrevent was new to them (Table 3).

This pilot had several limitations. First, the sample was small and the study did not achieve its target patient recruitment. The majority of patients were young and highly educated, so the acceptability and usability of iPrevent might differ in the general community where computer literacy might be lower. Similarly, clinicians who chose to participate could have been more highly engaged with BC risk assessment and risk management than nonparticipant clinicians. Finally, only short-term outcomes were measured, and the impact on long-term satisfaction and uptake of BC risk-reducing measures could not be determined. As a result of this study, enhancements have been made to iPrevent with the aim of further increasing acceptability and usability.