Linked datasets offer comprehensive population-level evidence on health outcomes. This abstract highlights the value of linked data in informing prostate cancer outcomes, based on our experiences from the South Australia Prostate Cancer Data Linkage Project (SA-PCDLP).

The SA-PCDLP combined data from various sources, including cancer registries, health service utilisation records including hospital admissions, prescription medications data, and death registries (Fig1). The dataset includes >30,000 men diagnosed 2001-2021.

Using the linked data, we answered a range of research questions, including evaluating the impact of comorbidities on prostate cancer, exploring mental wellbeing of men with prostate cancer, comparing prostate cancer treatment outcomes, analysing patterns of care in men with advanced prostate cancer, and measuring cancer health disparities. The findings have shown that: • comorbidities influenced treatment choice and survival outcomes of patients (although to a lesser extent than age), emphasising the complex interplay between existing health conditions and prostate cancer management and prognosis. • significant impact of prostate cancer diagnosis on mental health, with increased psychotropic medication and mental health service use around the time of diagnosis, highlighting the need for early mental health screening and intervention. • radiotherapy was associated with an increased risk of genitourinary and lung cancers compared with radical prostatectomy (RP). • RP patients experienced more post-treatment urinary procedure rates, while radiotherapy patients had higher colorectal procedure rates in the longer term. • temporal shifts in treatment modalities, with increased RP and active surveillance (AS), and decreased hormone therapy alone as primary treatment. • Socioeconomic disparities in primary treatment, with men from higher socioeconomic areas more likely to undergo RP. The issues encountered during SA-PCDLP included linkage process took over two years, a lack of complete coverage of hospital admissions from private practice, and incomplete data for key variables (tumour characteristics). Despite these challenges, our studies demonstrate the value of linked data for providing comprehensive, population-based, real-world evidence on prostate cancer outcomes.

The linked data analyses revealed critical insights into prostate cancer management and prognosis which has the potential to inform quality of care and patient outcomes.