Do observational studies have a place in clinical research?

Guest blog post on Observational Studies by Alison Coletta, Principal Medical Writer

Randomised controlled trials (RCTs) are the acknowledged gold standard in clinical research, they are used to evaluate the use of novel interventions in a carefully selected groups of subjects under strictly controlled conditions. In contrast, observational studies can provide an important source of information about the use of licenced medicines in a heterogeneous population of subjects in real life clinical practice.

observational studies

The great advantage of observational studies is that they reflect normal clinical practice more accurately than RCTs.

In RCTs the intervention is specified in the protocol, subjects are randomly allocated to treatment groups and there is usually an extensive list of inclusion and exclusion criteria in order to ensure the similarity of the treatment groups and to minimise bias. Conversely, in observational studies, treatment is allocated to a diverse population of subjects in accordance with the normal clinical practice of the treating physician, with data being collected either prospectively or retrospectively. Observational studies are therefore an invaluable source of information about the use of treatments within the terms of the marketing authorisation.

The importance of data generated in observational trials in evidence based medicine is increasing. It is anticipated that the use of data from observational studies is likely to increase as research and development costs continue to rise whilst financial constraints continue to squeeze research budgets.

Historically observational studies have had a poor reputation, often this is due to the inadequate interpretation and reporting of results. Recent developments including the introduction of a number of new guidance documents should help to improve the status of observational studies in the future. The Strengthening the Reporting of Observational studies in Epidemiology (STROBE) guidelines, aim to provide guidance in the publication of the results of observational studies. It has been suggested that journal editors should invoke the STROBE guidelines to help ensure appropriate reporting of observational studies in the literature.

Unlike randomised controlled trials, there is no mandatory requirement for registering observational studies, however it is both ethically and scientifically sound to do so. Observational studies can be registered on the website. In Europe, the European Network of Centres for Pharmacology and Pharmacovigilance (ENCePP) administers the European Union electronic register of post-authorisation studies (EU PAS Register) which aims to increase transparency, reduce publication bias, promote the exchange of information and facilitate collaboration among stakeholders and to ensure compliance with EU pharmacovigilance legislation requirements.

Ethics committee approval of observational studies can be a minefield of confusion. Some countries waive the requirement for ethics approval since there is no experimental intervention, however other countries will not accept studies unless they have been approved by an appropriately constituted ethics committee. The most prudent approach is probably to obtain approval or if that is not feasible to request a supporting letter from the ethics committee in countries where approval is not a prerequisite.

In summary, the results of RCTs may not always be directly applicable to the heterogeneous and diverse patients encountered in daily clinical practice. Observational studies provide the opportunity to assess “real world” prescribing of an intervention in a large number of subjects in routine clinical practice. In addition, observational studies provide the opportunity to assess safety and efficacy across multiple subgroups which may be difficult to achieve in a RCT.