Kladno – A research group at the Faculty of Biomedical Engineering of the Czech Technical University in Prague (CTU) is using analysis of real patient pathways and artificial intelligence methods to find a more effective division of oncology care between comprehensive cancer centres and regional facilities.
Centralizing specialized oncology procedures in experienced centres is of major importance. This is also shown by the results of a yet unpublished nationwide Czech study that links data from the National Cancer Registry with health insurance data.
The analysis included 1,414 adult patients with early-stage (I–II) pancreatic cancer, diagnosed between 2017 and 2023, who underwent surgery to remove the tumour. In hospitals performing fewer than 17 pancreatic resections per year, 90-day postoperative mortality was 12%, compared with 4.9% at hospitals with a higher volume of these operations.
The difference did not disappear after accounting for patients’ age and sex, comorbidities, tumour stage, or treatment given before surgery. Statistical analysis showed that the odds of death within 90 days after surgery were approximately 2.5 times higher in hospitals with lower volumes of these procedures.
The results show that for technically demanding surgical procedures, concentrating care in experienced centres can be crucial. However, this does not mean that the entire further course of a cancer patient’s care must automatically take place in a specialized centre.
According to CTU experts, it is therefore equally important to determine which parts of care truly must take place in the centre and which can be safely provided closer to the patient’s home.
“We should not centralize the patient, but the part of their care where centralization demonstrably brings higher quality,” says Associate Professor Aleš Tichopád, head of the research group at FBMI.
The centre should play a key role as soon as there is reasonable suspicion of a serious oncological disease. Well-coordinated diagnostics and rapid involvement of the necessary specialists can shorten the path to diagnosis and the start of treatment. It is at this stage that a specialized centre can help the patient gain valuable time.
Conversely, once treatment has started and stabilized, or during remission, part of the care for suitable patients can be moved to the region, for example through regular check-ups, long-term follow-up, delegated prescribing where possible, or other activities that do not need to be performed in a highly specialized centre.
Decisions must take into account not only the diagnosis itself, but also the patient’s age and overall condition, prognosis, treatment options, preferences, and willingness to undergo demanding therapy.
The patient’s own perspective is also important. A person whom the centre helped during a critical phase of illness may understandably have exceptional trust in it and want to remain in its care. From the perspective of the whole system, however, it may not always be optimal for every stabilized patient to stay long-term in a highly specialized centre.
Communication and navigation of the patient are therefore key, and trained care coordinators should help with them. Patients must understand why part of their care is being moved closer to home, and that this does not mean lower quality or loss of contact with the specialized centre.
Overcrowded centres mean a heavy workload for doctors and nurses and less time for patients who truly need their specialized expertise. Centralization that is too broad can therefore lead to inefficient use of the centres’ limited capacity.
CTU therefore uses data on the actual course of patient care and artificial intelligence methods to look for better ways of organizing care. Large volumes of anonymized data make it possible to track the pathways patients take through the healthcare system, what care they received, and what the outcomes were. These methods can be used to compare different model approaches to organizing care and to identify at which stages a patient should be treated in a specialized centre and when part of their care can be safely provided closer to home.
Success in centralization is not about bringing as many patients as possible into the centre. It is about ensuring the patient receives the right expertise at the right moment, and preserving the centre’s capacity for the times when it is truly needed.
Media contact: Ing. Anna Erfányuk Kamenská, Ph.D., erfanann@fbmi.cvut.cz