Organised cancer screening programme with

Dear all,

I am Aapeli Nevala from Finnish Cancer Registry, and I am working in European Comission’s Joint Action (EUCanScreen) to define a common data model for cancer screening research and monitoring.

We are scoping possibilities for monitoring the European cancer screening programmes using OMOP, and it seems that the Episodes-concepts could be really something that could be used to define cancer screening programme data.

A brief background:

Cancer screening consists of different medical observations done in sequence. Typically, these sequences vary by screening programme and country, but they could be (roughly) separated into five phases (with example):

Primary phase (HPV-testing in cervical cancer screening)

Triage phase (Pap-smear for HPV-positives)

Further assessment/confirmation phase (colposcopy)

Treatment phase (first initial treatment after the further assessment)

Interval phase (cancer-related events happening after four actual phases but before the start of the next screening round).

Typically, for each phase of the sequence there exists some events of interest, that may be procedures (testing, endoscopies) or observations (findings such as cancer). In themselves, there is nothing special about the events, but together they form the “screening chain”, that is of interest for reporting.

For this, we have been planning to use the novel Episode-table. Each phase of screening would need an Episode-concept, and for each individual the events related to that particular phase of screening would be mapped to proper Episodes using Episode events. Additionally, the Episodes would be defined to form a sequence such that triage follows the primary, further assessment follows triage, etc.

Additionally, there could be a top-level concept “screening round”, that is followed up by another “screening round” to form up a sequence of screening rounds.

Currently, there exists concepts for individual screening observations, f.ex. 36685900 – “Cervical cancer screening offered”, but often in a screening programme it is desired to link all possible observations that are part of the organised screening process into the “chain” of screening observations.

The current version of the CDM has only a certain set of Episode definitions — some of them are directly applicable to the treatment phase, but primary phase of screening, triage phase of screening, further assessment/confirmation phase, interval phase and the screening round are something that should be added to make this possible. What would you think about this approach? Would it be feasible to add new Episodes to the vocabularies?

Additionally, if there are existing OMOP CDM implementations of cancer screening data, I would be very interested on getting into contact!

Br and thanks in advance,
Aapeli

episode_concepts.drawio

Hi Aapeli:

Sounds like a good idea to get into, but what is the actual evidence you want to generate? Effectiveness of screening programs? Risks? Cost?

Usually, everybody agrees with screening programs, but in some cases it turns out that the screening will reveal cases which, if left alone, would have never bothered the patient. Instead, a whole lot of “chain” activities happen, i.e. heavy treatment with nasty side effects. Are you after those situations?

Hi, and thanks for the reply!

Indeed, overdiagnosis is a problem in cancer screening programmes. We are after the “whole chain” in the sense that we want to enable both studying effectiveness of screening, but also to monitor all the different steps and procedures that are initiated by screening, so we can indeed follow-up for all the harms (in addition to the benefits).