PCBaSe Xtend; Pathfinder project for enrichment of a research database with individual-level healthcare data

Population-based screening with prostate-specific antigen (PSA) testing for prostate cancer, the most common cancer in Sweden, is not recommended since benefits have not been shown to outweigh adverse effects from overdiagnosis and overtreatment.

This balance may now be shifting, by use of magnetic resonance imaging (MRI) of the prostate a new diagnostic tool, and of active surveillance as treatment strategy for low-risk cancer.

However, the benefit-risk balance is difficult to evaluate due to the long latency phase before diagnosis and the several decade-long disease course.

A very long follow-up with longitudinal data is required to assess all relevant effects of screening and new treatments.

Conventional study methods cannot generate timely results to guide public health policy decisions.In Sweden there is a unique possibility to link national health registers, but longitudinal data on laboratory and imaging results (PSA, MRI, biopsy) and hospital medications (hormonal therapy and chemotherapy) are not available in any of these national registers.

Such data must be extracted directly from healthcare.The purpose of this project is to extract individual-level patient data from all 21 regions in Sweden on these critical variables.

The project is in line with the Government’s efforts to facilitate secondary use of healthcare data and EU initiatives such as the European health data space (EHDS).

In our pathfinder project PCBaSe Xtend, individual-level patient data from routine healthcare from all 21 Swedish regions will be combined with data from The National Prostate Cancer Register and other national registers.Project feasibility has been verified in a subset of regions, where we have established a detailed administrative strategy, extracted  targeted data from all IT systems in use, and found good data coverage and quality.

Several hurdles, but also their remedies, have already been identified.

The project will deliver generalizable strategies for secondary use of healthcare data as well as internationally unique statistical models of prostate cancer screening and long-term outcomes.PCBaSe Xtend will be a nationwide, population-based cohort with comprehensive longitudinal data on 250 000 prostate cancer cases and more than 4 million men not diagnosed with prostate cancer.

We will use PCBaSe Xtend for advanced statistical modelling of the effects of various prostate cancer screening and treatment strategies on  long-term disease outcomes.