Life Sciences Blog

Measuring the effectiveness of risk minimisation: A growing driver of post-authorisation studies in Europe




Just recently, the EMA published its first summary of a risk-management plan for a centrally authorised product, Neuraceq for PET imaging in Alzheimer’s disease. The EMA will be piloting publication of these plans throughout 2014 as part of its drive for transparency and as a result of the 2010 pharmacovigilance reforms. Last month, the Agency also completed its review of cardiovascular risk pertaining to osteoporosis medicine Protelos/Osseor. The product will remain available on the European market, subject to further restrictions on use and implementation of additional risk minimisation measures.

The latter will include an evaluation of these risk minimisation measures themselves – an evaluation that also figures in the recently published risk-management plan for Neuraceq. This evaluation is something that is very much part and parcel of the pharmacovigilance reforms. In particular, that legislation saw a renewed emphasis on the role of post-authorisation safety studies as a means to further characterise the effectiveness of risk minimisation in the EU. Given the time that has elapsed since implementation of the reforms, I thought it timely to consider the use of post-authorisation studies for this purpose, both in terms of some of the questions they raise and what they may look like in practice.

Measuring across geographies Through its register of post-authorisation safety studies requested by European regulators, EnCePP - the European network of Centres for Pharmacoepidemiology and Pharmacovigilance – offers some insights into how this landscape is evolving. A growing number of the additions to this register are studies which do in fact centre on evaluation of risk minimisation measure effectiveness.

Most of these studies take a “representative” approach when it comes to geographical inclusion, such that results in a certain number of markets (often including France, Germany and the UK, for example) will be extrapolated across the EU. Existing data infrastructure and linkages in a number of markets – notably the UK and some Scandinavian countries – have meanwhile lent themselves to pharmacoepidemiological studies looking to measure the incidence of adverse events, or other events of interest such as whether patients are sent for required laboratory testing.

This is interesting as one of the first things to come to mind regarding the conduct of these studies is the perennial question over geography: as a collection of 28 member states, the EU represents a grouping of diverse healthcare systems, with variations that are potentially relevant for risk minimisation planning. This may be most pertinent in specific settings, such as pregnancy prevention programmes, for example, where differences in service design may affect how risk minimisation measures apply. However, in a more general sense, fundamental differences, such as whether GPs tend to be based in independent practices or in group medical centres, may have tangible impact on how measures are conceived and their effectiveness interpreted.

The question is, of course, whether results will necessarily be generalisable, or indeed whether regulators may wish to see specific geographies targeted in certain cases. Past research conducted by the IHS Life Sciences team has revealed that it can work both ways: industry stakeholders point to some cases where a representative sample of countries will suffice, and to others cases where regulators have asked for programmes to be conducted across a large number of markets.

Making measuring commensurable With regards to the effectiveness of risk minimisation measures, some members of the European regulatory community have proposed a “dual-evidence” approach for its evaluation. The idea is that a risk minimisation measure may prove ineffective for two reasons – either because the measure itself has not been conceived properly, and/or because it has not been effectively implemented. As such, evaluation of effectiveness may need to track two different types of indicators: both the level of specific adverse events of interest, and the knowledge and behaviours of healthcare professionals and/or patients. Understanding how these different audiences perceive messages, and how to drive the desired behaviours, is a challenge that should probably not be underestimated.

In thinking about the actual design of these measures, it is interesting to consider an idea that we have heard voiced by industry: that the tools used to evaluate the effectiveness of a risk minimisation measure ought to be commensurate both with the measure itself and with the nature of the underlying risk in question. That is to say, the labour intensiveness of the evaluation process ought to correlate in some way with the labour intensiveness of the design and implementation of the corresponding measure, as well as with the severity of the risk posed by the medicine.

This harks to the experience that some industry stakeholders have had under the Risk Evaluation and Mitigation Strategies (REMS) programme implemented by the US FDA. As REMS has included an evaluation of risk-minimisation measure effectiveness component, it has the potential to serve as something of a reference point in the European context. Here, industry has pointed to pushback against REMS by US physicians as evidence that these programmes can overburden the system; there is some concern that Europe may be going full throttle to embrace these sorts of schemes even as the FDA has cooled off a bit in response to this reality.

One important factor in determining the scale of the task facing industry will be whether these evaluations are requested solely in the context of additional risk minimisation measures – for example, when providing comprehensive educational material to physicians such as in the case of Protelos/Osseor – or whether more routine risk minimisation measures will also qualify.

Measuring what is meaningful It appears that measuring the effectiveness of risk minimisation measures will be a key driver of post-authorisation safety studies going forward. Clearly the design, implementation and interpretation of these studies present a number of key questions that must be answered on a case-by-case basis. Ultimately, this environment offers industry and European regulators the opportunity to explore new ways of generating meaningful data.

About The Author

Cameron Lockwood manages the EMEA consulting and multi-client study team. He has a background in the life sciences and specialises in market access and pricing and reimbursement issues.