PH: I wanted to comment on the paper by Clauson et al. that we read for a quiz in our PDA class. The paper was published in the March 8, 2007 edition of BMC Medical Informatics and Decision Making. I invite anyone else who has read the paper to post comments.
The paper rates 7 products, including: ePocrates, Clinical Pharmacology and Lexi-Comp among others. It rated Clinical Pharmacology (CP) the highest but had three other products ranked closely to CP on its top tier of products. They rated the products on the basis of ease of use, thoroughness and scope.
I found the study had a lot of potential for bias, primarily because of the funding of the study by the company that sells CP. The authors don't describe why they chose to weight the topics that were analyzed the way they did, or how they developed the list of drug related questions used to compare the products. It would certainly be conceivable that questions would be chosen in a way that they knew would favor the CP product, or that topics such as pregnancy and lactation were weighted to a lesser degree so that it would favor CP which performed less favorably in this category than some of the other products.
I will not go so far as to say that the authors did anything that blatant to favor one product, but the association of the report with the company that markets the leading product certainly has to raise some suspicions. I would also like to point out that even if the study was done with all intentions of being unbiased that the method of weighting different topics and how the questions are developed might have a huge impact on the outcome. I would not be surprised if two completely independent lists of questions would be chosen that there would be a difference in the outcome. I think it is very hard to do this comparison in an extremely meaningful way because the authors choices in what is important in a product are subjective and can vary depending on who is performing the analysis.
Subscribe to:
Post Comments (Atom)
2 comments:
TL: I agree with PH about the potential for bias. While it was good that there were two authors evaluating the databases, it might have been better to have more.
SE PH was right on target about the bias in the Clauson article.
The writing of Kevin Clauson’s article, Clinical Decision Support Tools: analysis of online drug information databases” – BMC Med Inform Decis Mak. 2007,7:7. was a good idea yet lacked validity based on obvious bias by the author. Mr. Clauson’s financial involvement with the producer of the leading ranked drug information database, Clinical Pharmacology, was too suspect for a complete endorsement of the research results. Particular areas of concern were the research study design, draft manuscript and data collection/evaluation methods.
In my experience I am inclined to believe that Clinical Pharmacology, Micromedex Clinical Xpert, Lexi-Comp Online and Epocrates are very worthwhile legitimate databases. While these resources are all high quality I would prefer to see a study of these databases by an independent research source. Incorporated in this unbiased study should also be an evaluation of the calculators and diagnostic tools found on these programs. Only then can one compare apples to apples and rate the programs for their full intended use. – End- SE
Post a Comment