Two recent articles on the alleged negative impact of pharma providing meals to physician practices are first rate examples of how the editors of JAMA and the authors ignore their own questionable data to arrive at a pre-ordained conclusion. And worse, JAMA peddled these articles to the media as proof positive of this claim.
The articles reinforce the assumption that doctors are influenced to prescribe more expensive brand drugs. There is has never been any causal evidence of any sort to support this claim, just anecdotes and the torturing of data that is conducted to fit the narrative.
But let’s say for the sake of argument that the more freebies and lunches doctors received is directly associated with more brand prescribing or specific prescribing of brands. Indeed, that is the hypothesis these two articles seek to test. Except that now, as opposed to even 10 years ago, the amount spent on pharma freebies like lunches and trips worth about $150 has declined. And the number of doctors who get them has fallen too for a number of reasons. At the same time brand prescribing has declined. So the reduction in payments has led to less brand prescribing right?
In Association of Industry Payments to Physicians With the Prescribing of Brand-name Statins in Massachusetts James S. Yeh, MD, MPH; Jessica M. Franklin, PhD; Jerry Avorn, MD; Joan Landon, MPH; Aaron S. Kesselheim, MD, JD, MPH (which I will refer to as Avorn and Co.) claim:
“Industry payments to physicians are associated with higher rates of prescribing brand-name statins. As the United States seeks to reign in the costs of prescription drugs and make them less expensive for patients, our findings are concerning.”
In Pharmaceutical Industry–Sponsored Meals and Physician Prescribing Patterns for Medicare Beneficiaries Colette DeJong, BA concluded:
“Receipt of industry-sponsored meals was associated with an increased rate of prescribing the brand-name medication that was being promoted. The findings represent an association, not a cause-and-effect relationship.”
Dejong and company looked at four specific drugs in different therapeutic class. They note that Crestor was 8.8% statin prescriptions; Benicar 3.3% beta-blocker prescriptions; Benicar was 1.6% of ACE inhibitor and ARB prescriptions; and Prestiq was 0.6% of SSRI and SNRI prescriptions and like Avorn and Co. conclude that prescribing rate was influenced by drug reps passing out donuts and Chipotle.
Except that in both ‘studies’ the brand prescribing rates were BELOW national averages for Medicare part D
The Medicare Payment Advisory Commission reported that
“Generic drugs accounted for 81 percent of all prescriptions filled in 2012 compared with 77 percent and 61 percent in 2011 and 2007, respectively. In 2015, generic fill rate increased again but as Express Scripts Drug Trend Report notes, the fill rates differed by plan type, with Medicare Advantage and stand-alone Part D plans with similar generic fill rates (87.5% and 87.2%, respectively), and Employer Group Waiver Plans with the lowest generic fill rate (82.4%).”
The Avorn group estimated that doctors who got lunch prescribed brand name statins 23 percent of the time vs 18 percent that were deprived of a free lunch in 2011. But the 23 percent is the same as prescribing of all brand drugs in Part D, a percentage that began and continued to decline as more medicines went off patent. We don't know what Avorn and Co.'s data would show in 2012 or 2013 as Lipitor went off patent... We will never know because taking that into account might undermine the conclusion they want to make.
Meanwhile DeJong shows that the undue influence of meals leads to much lower brand utilization in part D than the national market share of each drug.
And here is the trend in brand vs generic over the past decade. And neither study took the time to control for this critical variable?
Finally, neither study tested the reverse assumption: that the biggest prescribers of brand drugs were more likely to have drug reps visit their office, provide samples and schmooze than those that prescribe generic. If they had done that, both groups of authors could have controlled for patent expirations, co-pay effects, etc. that are more highly correlated with prescribing and generic uptake than snacks.
But that wouldn’t fit JAMA's distorted narrative of unscrupulous drug companies seducing dumb doctors with free lunches.
The articles reinforce the assumption that doctors are influenced to prescribe more expensive brand drugs. There is has never been any causal evidence of any sort to support this claim, just anecdotes and the torturing of data that is conducted to fit the narrative.
But let’s say for the sake of argument that the more freebies and lunches doctors received is directly associated with more brand prescribing or specific prescribing of brands. Indeed, that is the hypothesis these two articles seek to test. Except that now, as opposed to even 10 years ago, the amount spent on pharma freebies like lunches and trips worth about $150 has declined. And the number of doctors who get them has fallen too for a number of reasons. At the same time brand prescribing has declined. So the reduction in payments has led to less brand prescribing right?
In Association of Industry Payments to Physicians With the Prescribing of Brand-name Statins in Massachusetts James S. Yeh, MD, MPH; Jessica M. Franklin, PhD; Jerry Avorn, MD; Joan Landon, MPH; Aaron S. Kesselheim, MD, JD, MPH (which I will refer to as Avorn and Co.) claim:
“Industry payments to physicians are associated with higher rates of prescribing brand-name statins. As the United States seeks to reign in the costs of prescription drugs and make them less expensive for patients, our findings are concerning.”
In Pharmaceutical Industry–Sponsored Meals and Physician Prescribing Patterns for Medicare Beneficiaries Colette DeJong, BA concluded:
“Receipt of industry-sponsored meals was associated with an increased rate of prescribing the brand-name medication that was being promoted. The findings represent an association, not a cause-and-effect relationship.”
Dejong and company looked at four specific drugs in different therapeutic class. They note that Crestor was 8.8% statin prescriptions; Benicar 3.3% beta-blocker prescriptions; Benicar was 1.6% of ACE inhibitor and ARB prescriptions; and Prestiq was 0.6% of SSRI and SNRI prescriptions and like Avorn and Co. conclude that prescribing rate was influenced by drug reps passing out donuts and Chipotle.
Except that in both ‘studies’ the brand prescribing rates were BELOW national averages for Medicare part D
The Medicare Payment Advisory Commission reported that
“Generic drugs accounted for 81 percent of all prescriptions filled in 2012 compared with 77 percent and 61 percent in 2011 and 2007, respectively. In 2015, generic fill rate increased again but as Express Scripts Drug Trend Report notes, the fill rates differed by plan type, with Medicare Advantage and stand-alone Part D plans with similar generic fill rates (87.5% and 87.2%, respectively), and Employer Group Waiver Plans with the lowest generic fill rate (82.4%).”
The Avorn group estimated that doctors who got lunch prescribed brand name statins 23 percent of the time vs 18 percent that were deprived of a free lunch in 2011. But the 23 percent is the same as prescribing of all brand drugs in Part D, a percentage that began and continued to decline as more medicines went off patent. We don't know what Avorn and Co.'s data would show in 2012 or 2013 as Lipitor went off patent... We will never know because taking that into account might undermine the conclusion they want to make.
Meanwhile DeJong shows that the undue influence of meals leads to much lower brand utilization in part D than the national market share of each drug.
And here is the trend in brand vs generic over the past decade. And neither study took the time to control for this critical variable?
Finally, neither study tested the reverse assumption: that the biggest prescribers of brand drugs were more likely to have drug reps visit their office, provide samples and schmooze than those that prescribe generic. If they had done that, both groups of authors could have controlled for patent expirations, co-pay effects, etc. that are more highly correlated with prescribing and generic uptake than snacks.
But that wouldn’t fit JAMA's distorted narrative of unscrupulous drug companies seducing dumb doctors with free lunches.
