Breast cancer is the most common cancer in the UK and a leading cause of death in women with approximately 55,000 women a year diagnosed with the disease. Of these around 11,400 will die from the disease. To try and combat the disease, every year nearly 2 million women in England are screened in the National Health Service Breast Screening Programme (NHSBSP).
It is estimated that only about 1 in 6 women who are at high-risk of breast cancer have been identified by attending Family History, Risk and Prevention (FHRP) Clinics. If more high-risk women could be identified there may be fewer breast cancers due to chemoprevention, and a reduced number of deaths by detecting more breast cancers at an earlier and more treatable stage.
Prof. Gareth Evans approached Research IT to develop a system that collects information on risk factors such as mammographic density and self-reported information on family history and hormone-related factors via questionnaire. Risk estimation can be offered in real-time to women invited for breast screening via an online web system to allow consent and self-report measures to be provided. All women are invited to have a discussion about their risk. Those who are identified as high or moderate risk who meet certain criteria are then referred to the FHRPC to discuss their options. The criteria are also determined by age so not all moderate/high risk women will be referred.
The software produced by Andrew connected all the data collection tools such as the imaging machines and survey tools, and provided a method of monitoring the progress of a patient through the prediction process. The software also produced the final risk letter to the patient and GP at the end of the process, so involved managing sensitive data in consultation with the Trust’s Information Governance office.
Although there were already two software developers already on the project, their software didn’t talk to each other and didn’t perform all the functions that were needed. The project team needed someone with software development skills to remedy this and initially approached the Health eResearch Centre (HeRC) who in turn asked Research IT.
Andrew designed and developed the applications that provided an interface to their software and recorded the data being transmitted from them. The software he wrote was a suite of applications including a web application and several small batch service applications. This required a lot of technical liaison to make sure the correct information was being sent between the 3 systems at the right time and make pragmatic decisions (with the researchers) about when to spend time automating a process or just making it a manual process. Andrew wrote a specification that all parties could agree to and wrote the software himself.
Andrew also led the testing phase, liaising with the other companies involved to ensure the software was fully integrated and worked together. He trained the researchers and administration staff in the use of the software, also writing a user guide and other FAQs, and providing ‘translation’ between the technical staff at the other software companies and the research team.
Since the initial development phase has been completed he has been available to make small changes (as research is never static!) and provide support such as access management and user queries.
Sarah Sampson explained “As the Programme Coordinator on this project, it has been a pleasure working alongside Andrew. The software that he developed is user-friendly and successfully meets the ever changing demands of the research. I have consistently found Andrew to be approachable and diligent, always making time to effectively communicate technical information in an easy-to-understand form in order to resolve any issues that arise. Thank you Andrew for all your hard work!”.
If you want to read more about the development of the system and the associated study please see:
What are the benefits and harms of risk stratified screening as part of the NHS breast screening Programme? Study protocol for a multi-site non-randomised comparison of BC-predict versus usual screening (NCT04359420).
French et al. BMC Cancer (2020) 20:570 https://doi.org/10.1186/s12885-020-07054-2