Email updates

Keep up to date with the latest news and content from BMC Family Practice and BioMed Central.

Open Access Highly Accessed Research article

The external validity of published randomized controlled trials in primary care

Ritu Jones, Robert O Jones, Colin McCowan, Alan A Montgomery and Tom Fahey*

BMC Family Practice 2009, 10:5  doi:10.1186/1471-2296-10-5

PubMed Commons is an experimental system of commenting on PubMed abstracts, introduced in October 2013. Comments are displayed on the abstract page, but during the initial closed pilot, only registered users can read or post comments. Any researcher who is listed as an author of an article indexed by PubMed is entitled to participate in the pilot. If you would like to participate and need an invitation, please email info@biomedcentral.com, giving the PubMed ID of an article on which you are an author. For more information, see the PubMed Commons FAQ.

Industry sponsorship is associated with a lower probability of publishing data required to calculate trial characteristics

Ronan Conroy   (2009-07-20 11:58)  Royal College of Surgeons in Ireland email

In table 5, Jones and colleagues present odds ratios for the impact of various types of trial funding on the probability that the trial will report the data required to determine the number of patients needed to be screened in order to randomize one participant, a key indicator of the external validity of any study.

However, the associated statistical analysis contrasts each of five funding sources with funding by charities. This has two disadvantages: by disaggregating their data into small groups, the authors' power to detect significant effects is diminished. But, more important, there is no a priori reason to compare each source of funding with charitable funding - why should charities be different?

On the other hand, there are compelling reasons to explore the effect of industry sponsorship on disclosure of the data required to assess external validity. Fifteen trials were sponsored by the pharmaceutical industry, of which eight (53%) provided data. Of the remaining trials with identifiable sources of funding (charities, governing body, government) 68 of 86 (79%) provided such data.

The relative risk of data disclosure associated with industry sponsorship is therefore 0.67 (P=0.0.033).

This is perhaps a lesson that for hypothesis tests to be meaningful there has to be a meaningful hypothesis being tested.

Competing interests

None.

top

Post a comment