Intended for healthcare professionals

Letters News story on Tamiflu study

Authors’ reply to Mark Jones’s critique of the study by Muthuri and colleagues reported in The BMJ

BMJ 2014; 348 doi: (Published 30 April 2014) Cite this as: BMJ 2014;348:g2990
  1. Puja R Myles, public health specialist/epidemiologist1,
  2. Jo Leonardi-Bee, associate professor in medical statistics1
  3. On behalf of the PRIDE research consortium investigators
  1. 1University of Nottingham, City Hospital, Nottingham NG5 1PB, UK
  1. puja.myles{at}

This is a point by point response to Mark Jones’s critique by the authors of Muthuri and colleagues’ paper1 2:

“‘A crude analysis of the data shows an increased risk of mortality associated with neuraminidase inhibitor treatment,” suggesting that the finding of a reduced risk of death was incorrect.”

It is true that, based on simple number counts, more people in the group treated with the antiviral agent died compared with those who were not treated. However, this could be explained by people in the treated group having a higher baseline risk of dying than those in the untreated group. In this kind of work we encounter the problem of non-equivalent comparison groups.

Ideally, the data that we used would have come from randomised controlled trials (experimental studies; RCTs), where equivalent patients are randomly assigned to treatment or placebo. We would then have been more confident that any differences in patient outcome were attributable to treatment status alone. In a pandemic situation, however, it would have been unethical to randomly deny antiviral treatment to patients, and we found no RCT data during our extensive search for data pertaining to the pandemic period. Thus, our only option was to study actual treatment practice and resultant patient outcomes during the 2009-10 pandemic.

To overcome the problem of comparing non-equivalent patient groups, we used statistical methods to adjust for any patient differences and allow us to disentangle treatment effects from outcomes arising from fundamental differences between patients. This is why the adjusted results are paramount, not the crude (unadjusted) results, which Jones chose to highlight. Nonetheless, we thought it important to present the crude results in the interest of transparent scientific reporting.

Of rather more concern is the fact that Jones has ignored the clustering of effects by study centre in …

View Full Text

Log in

Log in through your institution


* For online subscription