Intended for healthcare professionals

Views And Reviews

The European Medicines Agency is still too close to industry

BMJ 2016; 353 doi: (Published 06 May 2016) Cite this as: BMJ 2016;353:i2412
  1. Silvio Garattini, director
  1. IRCCS Istituto di Ricerche Farmacologiche Mario Negri, Via Giuseppe La Masa 19, 20156 Milan, Italy
  1. silvio.garattini{at}

Two decades after its inception, the agency still fails to put patients’ interests first

Fifteen years ago I raised criticisms about the policy and attribution of the then young European Medicines Agency (EMA) and its Committee for Medicinal Products for Human Use (CHMP).1

With the same aim of encouraging this institution to put the interests of patients and public health services first, I now offer an updated picture (see table).

The current supervising directorate general of the European Commission is certainly more appropriate than before: first it was DG Enterprise, which also represents drug companies—a considerable conflict of interest. Today it is DG Sanco, which is responsible for health and consumers.

Withdrawal by a company of an application for marketing authorisation no longer precludes the publication of a negative opinion by the CHMP with its reasons. And now, when the CHMP does not unanimously approve a medicinal product, it publishes the minority opinion.

Access to data

Each dossier supporting an application for a drug’s approval used to be completely confidential; now the public has some access to the data, for instance in redacted clinical trial reports.2 Making dossiers and assessment processes public would improve openness further.3 4

Some of the EMA’s problems have not been solved, and the situation is essentially unchanged. For example, the criteria for evaluating drugs are still quality, efficacy, and safety, but a previous suggestion was to include “added therapeutic value” to require comparisons with the best treatments available for the same indication. Today it’s still possible to obtain approval without studies of comparative efficacy: trials of superiority versus placebo and of non-inferiority versus active comparators are still welcome, as is the clinical evidence supporting marketing authorisations that relies too often on surrogate outcome measures.

The fact that independent pivotal trials are not required favours an incredible conflict of interest: even today, only studies promoted by industry are accepted for evaluation. The documentation accompanying the marketing authorisation—that is, the summary of product characteristics and the leaflet—is still prepared by the industry and omits important information for doctors and patients, such as comparison with other drugs of the same therapeutic class.

Unfortunately, some aspects of the EMA are getting worse. For instance, the contribution from industry was once about €39m (£30.2m; $44.1m), equal to 71% of its whole budget; today it is about €250m (83%). This dependence is incompatible with the EMA being seen to be independent. Nobody receiving 80% of a salary from industry would be admitted to any committee that deals with drug evaluation.

Too much power

Today the CHMP concentrates even more functions and too much power. For substantial fees it gives scientific advice to industry figures to help them prepare studies that are then judged by the committee itself. It considers appeals against its own decisions. And it is responsible for pharmacovigilance and the withdrawal of drugs that it has authorised. Scientific advice, appeals, and withdrawals should be dealt with by committees independent of the CHMP.

It used to be that drugs could be approved under exceptional circumstances—that is, despite a lack of comprehensive evidence of efficacy and safety. Today we also see “conditional approvals,” which are often not followed by the studies that the EMA requires from the manufacturer.5 The agency is also starting an adaptive licensing process described as “. . . a prospectively planned, flexible approach to regulation of drugs,”6 which, in essence, aims to allow medicines to the market more quickly and is based on lower evidence requirements than under a conventional marketing authorisation. This may be dangerous for patients, in that it further shifts the burden of evidence from pre-marketing to post-marketing.7

More than 20 years have elapsed since the decision to have a single agency consider drug approvals, with a view to facilitating the preparation of dossiers by the industry and harmonising the availability of drugs in Europe.8 Unfortunately, however, the economic interests of member states and the industrial lobby have so far hampered the adoption of legislation and rules that favour the interests of patients.

The European Medicines Agency in 2000 and 2015

View this table:


  • I thank Vittorio Bertele’ for contributing to this article, and I thank Judith Baggott for editing.

  • Competing interests: I have read and understood BMJ policy on declaration of interests and have no relevant interests to declare.

  • Provenance and peer review: Not commissioned; not externally peer reviewed.


View Abstract