This paper correctly reports a 6% increase in odds ratio, and not a 6% increase in mortality, as reported in most media reports of this paper. Based on the paper's data, there is, in fact, a 0.07% increase in mortality.

An example of one such error is at the BBC's web site: http://news.bbc.co.uk/1/h...

The details of this calculation are given below.

The following is a simple analysis of the data presented in the paper based on the total numbers given in the following text from the paper:

In the two cohorts derived from the two dates of admission, between 2000 and 2008, there were a total of 299,741 admissions (151,844 admitted on the last Wednesday in July and 147,897 admitted on the first Wednesday in August) and 4,409 deaths in hospital within a week of admission (2,182 for the former week and
2,227 for the latter week).

Define the week before junior doctors begin work as week1, and the next week as week2.

The mortality in week1 was therefore

2182/151844 = 0.0144 (ie 1.44%),

and in week2 was

2227/147897 = 0.0151 (ie 1.51%).


This represents an increase in mortality of

0.0151 - 0.0144 = 0.00070 (about 0.07%).

--

Additionally, there is an error in the paper:

Adjustment for year, gender, age group, socio-economic deprivation and co-morbidity led to a 6% higher odds ratio of death for the second week (OR 1.06 ...

This should say:

... led to 6% higher odds of death,

not

... led to a 6% higher odds ratio of death.

--

Dr JV Stone, Sheffield University, January 2010.