This is a pragmatic, prospective, propensity-matched, cohort study using routine data, with an embedded patient survey, economic evaluation, and qualitative and preference studies. Individual sites will be classified according to one of the four follow up strategies to compare outcomes in a natural experiment. Individual participants will be clustered within their treating hospital.
The main cohort for the study will be identified from the National Disease Registration Service Cancer Registry. Data on treatments, diagnoses, clinic attendances and other events that occur in secondary care from 3 years prior to diagnosis will be extracted from databases linked to the Cancer Registry including Cancer Pathway, Radiotherapy Dataset (RTDS), Systemic Anti-Cancer dataset (SACT) and Hospital Episode Statistics (HES). These datasets are referred to collectively as 'HES-linked data'. HES-linked data will be collected from 3 years prior to diagnosis, to enable analyses to be adjusted for patient medical history and pattern of health service use prior to diagnosis with prostate cancer, to the latest available date (expected to be 31 December 2026).
To identify the most clinically and cost-effective follow up strategy for patients, five interrelated work packages (WPs) will address specific objectives:
\- WP1: Determine safety and clinical effectiveness of different follow up strategies
Stage 1: The utility of HES-linked data to study treatment for prostate cancer recurrence will be validated. It will be confirmed if databases contain the parameters required and are complete enough to robustly assess outcomes in secondary care including time to treatment for cancer recurrence and treatment for side-effects from primary treatment. Detailed statistical methods and computer code will be developed for determining whether follow up strategies are equivalent for treatment of recurrence, including methods for dealing with missing patients and treatment variables. A preliminary investigation of hospital adherence to stated follow up strategies will be conducted.
Stage 2: All primary and secondary outcomes will be measured using both HES-linked datasets and information collected through the patient survey in WP2. The methods developed in stage 1 will be used to assess the equivalence of the four follow up strategies for the primary outcome of treatment for cancer recurrence at 3 years, the key secondary outcome of time to hospital treatment for cancer recurrence, and remaining secondary outcomes.
-WP2: Measure side effects of prostate cancer treatments, patient quality of life and recurrence outcomes not captured in routine data
A survey will be designed and administered to capture data from a subgroup of study participants on adverse effects of initial cancer treatment, how those adverse effects were managed, health service utilisation and HRQoL. The survey will ask patients whether and when they received hormone therapy for prostate cancer recurrence, providing data for WP1 on recurrence treatments not captured in HES-linked databases.
-WP3: Identify key patient perspectives on the important features of follow-up
A qualitative interview study will be conducted with patients with lived experience of prostate cancer and follow up, purposively sampling geographic spread, socioeconomic class and ethnicity (WP3 is sponsored by University of Aberdeen and is described in a separate protocol and has separate ethics approvals). Early findings will inform the design of the Discrete Choice Experiment (DCE) to be deployed in WP5.
-WP4: Evaluate the cost-effectiveness of each follow up strategy
The use of primary and secondary health care services during follow up to estimate the costs of the four follow up strategies will be assessed using HES-linked and patient survey data. An economic evaluation model will be developed to compare the patient lifetime costs and outcomes of the four strategies, and identify the approach that provides the best value for money for the NHS.
-WP5: Define patient preferences for follow-up
A DCE will be conducted to understand how patients might respond to follow up strategies with different attributes and elicit patient preferences. Questions for the DCE will be designed using early qualitative findings (WP3) and will be included as a module in the patient survey (WP2) for a subgroup of survey recipients.
