Our work programme consists of three key work packages.
Our work programme consists of three key work packages.
We will utilise an observational study design to ensure relevance of our results to real-world practice. Data will be compared between an “intervention” period, where sites had HELP Flag active, to a prior “control” period before the innovation was in place. The intervention period will last a minimum of 12 months and the control period will be the 12 months prior to activation of HELP Flag. Patients aged 60+ with tests meeting the age-sex thresholds over these periods will be followed-up for a minimum of 12 months from the test date to capture outcomes. Data will be supplied by the NHS Southwest Secure Data Environment (Devon) and the NHS North West Secure Data Environment.
Data will be extracted at three timepoints: an initial extract to capture data on the control period (November 2025), an interim extract in May 2026, and a final extract in February 2027. The primary outcome is a record of an incident cancer diagnosis within one year of index test date. Secondary outcomes will include referral to secondary care services to a speciality that may result in a cancer diagnosis during one year and the proportion of cancers diagnosed at Stage I or II and the route to diagnosis (e.g. emergency or urgent suspected cancer referral). Multilevel logistic regression, clustering by site, will explore how odds of outcomes differ between the control and intervention periods, and by age, sex, and ethnicity/deprivation.
Further to the requirements for Work Package 1 we will also extract information where available on treatments received, progression-free survival and mortality for patients who have been diagnosed with cancer for use in Work Package 2. We will use stage at diagnosis to estimate prognosis where necessary. We will assess the feasibility of use of Open Safely (www.opensafely.org) to track referrals in close to real-time via an interactive dashboard. Open Safely is an open-source software platform for analysis of electronic health records data and facilitates access to CPRD, HES and other datasets.
We will calculate whether or not HELP Flag produces enough benefit to patients in improved quality of life and reduced mortality and to the NHS in cost reductions from longer term savings to justify the upfront costs associated with increased referrals.
We will use data from earlier work (Work Package 1), published studies, and standard cost sources to fill in missing information, such as long-term outcomes and quality of life.
The analysis will examine how shifting cancer diagnoses to earlier stages and changing treatments affect overall costs—considering whether savings from fewer emergency cases and less expensive late-stage care outweigh the costs of new tests and pathway changes.
If some inputs remain uncertain, the value of collecting more data on them will be assessed. The study will look not only at whether the intervention is good value for money overall, but also at how its costs and benefits are shared across different groups of people. In other words, it will explore whether some groups – such as people from poorer areas or certain ethnic backgrounds – gain more or less benefit than others, and how that affects the overall conclusions about cost-effectiveness. This helps decision makers see if an intervention is both efficient and fair in reducing health inequalities.
The final model will be open source so that decision makers can explore the findings.
An important part of the evaluation is for us to understand how acceptable HELP Flag is to those who are using it. In this respect we want to know how easy or difficult it is to put HELP Flag into practice and what it is like to manage patients through the pathway.
To do this, we will talk with the group of individuals who had a key role in setting up HELP Flag within the participating hospital trusts. This will include members of the cancer alliance and local clinical and managerial staff from haematology, pathology, the non-site-specific pathway and general practice. We will also talk with individual GPs and hospital staff who had experience of managing patients through the pathway.
We will use a use a popular research framework, the Theoretical Framework of Acceptability, to structure our discussions with these groups. This framework will allow us to tap into a range of issues related to perceptions of acceptability and will provide insight into individual’s attitudes toward HELP Flag, their confidence implementing or using it, and its impact on their workload, along with a range of other factors.
We expect our discussions to provide us with some practical insight into the challenges involved in implementing and using HELP Flag as well highlighting what is working well. This information will help us to create a blueprint for implementation, that could be used for the wider rollout of HELP Flag across the NHS and help us improve the HELP Flag innovation itself.
An updated version of HELP Flag will be shared with our patient and public advisor group who will support the production of resources for patients affected by HELP Flag. This is an important part of the update and will allow the public to feed into the wider consideration of HELP Flag’s acceptability.