Ethics Committees and IRBs: Boon, or Bane, or More Research Needed?

Cambridge: summary of remarks in keynote talk at WECSR 2012 Institutional review boards in the USA, and ethics committees in the UK, have their roots in medical research. In the US Tuskegee scandal, black patients with syphilis were left untreated even after an effective treatment became available in the form of penicillin; in the UK Alder Hey scandal, pathologists retained body parts from deceased children without informing their parents. Yet simply having a committee of doctors review other doctors' research proposals isn't foolproof, as it disregards the differing perspectives and cultural assumptions between doctors and patients. For example, ethics committees were already well established in Britain by the time of Alder Hey, and it's not entirely obvious that a committee of half a dozen randomly-chosen white doctors in the deep south in the 1940s would have acted any differently from the Tuskegee team. The current tussle in the UK is between a medical research establishment that wants access without consent to medical records that have been " pseudo anonymised " in that the patients' names and addresses have been removed, and a privacy community which points out that most such records can be re-identified easily. Computer scientists know that anonymity is hard, thanks to the work of Denning, Sweeney, Dwork and others; this knowledge is slowly percolating through to the policy community via Ohm's work. Yet we have already had an incident were over eight million " pseudo anonymised " records were lost when a researcher's laptop was stolen; should such a haul end up on wikileaks or pastebin, we might have a scandal like Alder Hey that could damage public confidence in medical research. Could such a dilemma be fixed by ethics committee? Here is a second example. One UK university has data on the movements of millions of vehicles taken from automatic number-plate recognition cameras. This has been " pseudo-anonymised " by hashing the license plate numbers, yet someone who knew that a target drove on road X at time t could search for all other sightings of that vehicle. Yet the Department for Transport asserts this is no longer personal data. It follows that anyone should be able to obtain a copy using the Freedom of Information Act – and by that I mean anyone, not just any researcher working within the framework of an ethics committee. The comfort that the committee's existence gave to civil servants …