Randomized trial needed, says hematologist
, of the departments of medicine, hematology, and medical oncology, at the Icahn School of Medicine at Mount Sinai, New York, had a more guarded view of the findings than Dr. Wakefield and Dr. Houghton.
“[The investigators] certainly seem to be doing a lot of statistical gymnastics in this paper,” Dr. Edwards said in an interview. “They used all kinds of surrogates in place of real data that you would get from a randomized trial.”
For example, Dr. Edwards noted the use of prescription refills as a surrogate for medication adherence, and emphasized that inpatient observational data may not reflect outpatient therapy.
“Inpatients are constantly missing their medicines all the time,” she said. “They’re holding it for procedures, they’re NPO, they’re off the floor, so they missed their medicine. So it’s just a very different patient population than the outpatient population, which is where venous thromboembolism is treated now, by and large.”
Although Dr. Edwards suggested that the findings might guide treatment selection “a little bit,” she noted that insurance constraints and costs play a greater role, and ultimately concluded that a randomized trial is needed to materially alter clinical decision-making.
“I think we really have to wait for randomized trial before we abandon our other choices,” she said.
The investigators disclosed relationships with Merck, Celgene, UCB, and others. Dr. Wakefield reported awaiting disclosures. Dr. Houghton and Dr. Edwards reported no relevant conflicts of interest.