Emma Davies at Chemical Watch has put together a nice piece summarising the thoughts of Lisa Bero, professor at the University of California San Francisco’s Department of Clinical Pharmacy Institute for Health Policy Studies (Chemical Watch, 27 June 2013).
Bero recently published a paper (Krauth et al. 2013) examining various instruments used for assessing the quality of animal studies, finding that only one of the instruments had been validated. This means the instruments are being used in the hope they measure study quality rather than being used on the basis of proof they measure study quality.
This casts something of a shadow on the credibility of risk assessments and expert opinion – if there aren’t any validated methods for assessing study quality, how can we tell if current evaluations of study are sufficiently objective? And if two committees disagree on the quality of a study, how do we decide which committee was using the most credible methodology for assessing study quality if neither has been validated?
As Bero says, the solution is to assess animal toxicology studies in the same way as is routinely done in human trials. It’s a fair job developing the quality criteria, but with 30 instruments already available we can see that validation should be prioritised in on-going research.
(My quote? Thrillingly, I said: “Developing a validated method for evaluating study quality would be a big step forward for improving the consistency and transparency of risk assessment. It would also help a lot with improving how studies are reported and conducted, as it would form a blueprint for best practice in the conduct of research – everyone would know what standard they should meet and could report their studies accordingly.”)