We often get confused when we use the terms ‘retrospective’ or ‘prospective’ in the context of a non-interventional study.

By definition [ref 1], a retrospective study collects data that was generated before the study was approved and a prospective study collects data that was generated after the study was approved.  Seems simple enough! 

In Europe, countries such as Germany [ref 2], Italy [ref 3], and Spain [ref 1] use the terms ‘retrospective’ and/or ‘prospective’.  Whereas, Austria has revised their non-interventional study guidance [ref 4] to state that:

“Retrospective data evaluations are studies that are carried out exclusively on data already collected for other reasons (in particular on the occasion of a curative treatment). Since retrospective data collection does not  involve a human study – which is required for the ‘clinical study’ under Article 2(2)(1) of the EU CTR – the  term  ‘clinical study’ (and thus also the legal definition of ‘non-interventional study’) is not fulfilled.”

Similarly, the research system in France focuses on research directly involving humans (RIPH e.g., clinical trials and non-interventional studies) [ref 5] and research which doesn’t (Non-RIPH e.g., secondary use of existing data).

Meaning?  In Austria and France, retrospective non-interventional studies aren’t actually…non-interventional studies. No wonder we get confused!

It could be argued that France is a leader in terms of the pivot from ‘retrospective and prospective’ to more contextual descriptors of ‘primary data and secondary data’.

This is very topical in the context of the European Health Data Space (EHDS) [ref 6], where ‘primary use of electronic health data’ means the processing of personal electronic health data for the provision of health services etc, and ‘secondary use of electronic health data’ means the processing of electronic health data for purposes such as scientific research.

Clearly, there is much work to do at the local level before we see the terms ‘retrospective’ and ‘prospective’ phased out and replaced with unified (and less confusing?) data descriptors, such as  ‘primary data’ and ‘secondary data’.  Until then…let the confusion abound!

References:

[1] Article 2(1)(b) of Royal Decree 957/2020 of November 3: https://www.boe.es/eli/es/rd/2020/11/03/957/con

[2] BfArM and PEI Joint Recommendations (Dec 2022): https://www.pei.de/SharedDocs/Downloads/DE/regulation/klinische-pruefung/awb-nis-pass-empfehlungen.pdf?__blob=publicationFile&v=4

[3] AIFA Determination March 20, 2008 – Guidelines for the classification and conduct of observational drug studies: https://www.medicoeleggi.com/argomenti000/italia2008/400078-a1.htm

[4] Federal Office for Safety in Health Care (BASG) – Guideline for the Differentiation of Clinical Studies: https://www.basg.gv.at/fileadmin/redakteure/06_Gesundheitsberufe/Klinische_Studien/NIS/Leitfaden_NIS_BMSGPK_BASG_2022-10-11_v01.pdf

[5] Article L1121-1 of the Public Health Code (CSP): https://www.legifrance.gouv.fr/codes/article_lc/LEGIARTI000032722870/

[6] European Commission – European Health Data Space: https://health.ec.europa.eu/ehealth-digital-health-and-care/european-health-data-space_en