Electronic letters to:

Research:
Colin R. Dormuth, Malcolm Maclure, Kenneth Bassett, Ciprian Jauca, Carl Whiteside, and James M. Wright
Effect of periodic letters on evidence-based drug therapy on prescribing behaviour: a randomized trial
CMAJ 2004; 171: 1057-1061 [Abstract] [Full text] [PDF]
*eLetters: Submit a response to this article

Electronic letters published:

[Read eLetter] Credibility of Printed Letters
Colin R. Dormuth   (1 March 2005)
[Read eLetter] Credibility of periodic letters
David A Cabral   (27 January 2005)

Credibility of Printed Letters 1 March 2005
Previous eLetter Top
Colin R. Dormuth
Therapeutics Initiative UBC, Brigham and Womens' Hospital, Harvard School of Public Health

Send letter to journal:
Re: Credibility of Printed Letters

cdormuth{at}hsph.harvard.edu Colin R. Dormuth

Two criticisms raised by Dr

Dr. Cabral makes three comments with respect to our study [1,2] that warrant a response. The first is that we did not support our statement that the Therapeutics Initiative (TI) is a credible and trusted source for drug therapy evidence. That statement is supported by a survey undertaken in 1996 during the 1994 to 1997 period in which the Letters in the study were distributed. The survey was sent to a 10 percent randomized sample of physicians and pharmacists who received the Therapeutics Letter. The survey achieved a 71 percent response rate from the physicians. Eighty percent of the physicians and 91percent of the pharmacists who responded answered yes to the following question: Do you trust that the information in the Therapeutics Letter is accurate and unbiased?  In addition, 70 percent of physicians indicated that the letters had influenced their prescribing. These survey data were included in our original submission to the CMAJ but were excluded from the shortened published version.

 

The second is that the Therapeutics Letters are not peer-reviewed. A draft of each Therapeutics Letter is sent out for review to at least 25 external specialists and family physicians prior to publication. Feedback from these individuals is incorporated into the final version of the Letter by the Editorial team and the Letters are frequently changed substantially as a result of this feedback. 

 

The third is that there is no avenue for reader response to the Letters. In fact contact information is provided on both the printed copy and the website version of each Letter. We frequently receive both positive and negative feedback.  We respond to the feedback as much as is possible given our limited staff. 

 

Colin R. Dormuth

James M. Wright

 

Authors:

From the Therapeutics Initiative (Dormuth, Wright), Vancouver, BC; the Division of Pharmacoepidemiology and Pharmacoeconomics (Dormuth), Brigham and Womens’ Hospital, Boston, Mass; the Department of Epidemiology (Dormuth), Harvard School of Public Health, Boston, Mass; the Department of Pharmacology and Therapeutics (Wright), University of British Columbia, Vancouver, BC.

 

Reference:

1.   Cabral DA. Credibility of Periodic Letters. CMAJ 27 January 2005.

2.   Dormuth CR, Maclure M, Bassett K, Jauca C, Whiteside C, Wright JM. Effect of periodic letters on evidence-based drug therapy on prescribing behaviour: a randomized trial. CMAJ 2004; 171: 1057-1061.

 

Conflict of Interest:

None declared
Credibility of periodic letters 27 January 2005
Next eLetter Top
David A Cabral
UBC

Send letter to journal:
Re: Credibility of periodic letters

dcabral{at}cw.bc.ca David A Cabral

Although the authors may have demonstrated a clinically significant effect in promoting new therapies (versus discontinuing therapies), this article does not demonstrate that they are a "credible and trusted source" other than that they are affiliated with a university. Indeed the authors of individual reports in the Therapeutics Letter are anonymous and therefore personally unaccountable for their publication, the reports are not peer reviewed as with most conventional credible journal publications, and there is no avenue for response to their reports (such as letters to the editor).

Arguably simply advertising a drug product may have just as much impact as their letter, especially given that the impact of the 'letter' in stopping prescribing was minimal.

Conflict of Interest:

None declared