Recently, Rafael Fonseca, MD and John A. Tucker, MBA, PhD, issued a critical analysis of a recent study that drew correlations between meals provided by prescription drug companies and physician prescribing behavior.
The initial study, authored by Hadland SE, Cerda M, Krieger MS, and Marshall BDL, explored data which related opioid prescribing to opioid manufacturer payments and impliedly concluded that there is a cause-and-effect relationship. However, as Fonseca and Tucker note, as is typical with such studies, the authors did not make any direct claims of a cause-and-effect relationship and that “their observations are based on correlation alone.” In the conclusion of the Hadland study, however, the authors recommend “Amidst national efforts to curb the overprescribing of opioids, our findings suggest that manufacturers should consider a voluntary decrease or complete cessation of marketing to physicians. Federal and state governments should also consider legal limits on the number and amount of payments.”
In performing their study, Hadland et al. linked physician-level data from the 2014 CMS Open Payments database to 2015 opioid prescribing behavior as described in the Medicare Opioid Prescribing Database. They hypothesized that meals and other payments made by pharmaceutical manufacturers increase physician opioid prescribing. In their findings, they noted the following,
1. A nearly linear relationship between the number of opioid manufacturer-provided meals accepted by a prescriber and the number of opioid prescriptions written. Prescribers who received nine meals from opioid manufacturers in 2014 prescribed opioid analgesics at slightly more than 3x the rate of those who accepted only one meal.
2. When broken down by physician specialty, those who accepted any payment from opioid manufacturers wrote between 1.2% more and 11% more opioid prescriptions as those who did not accept any such payments.
Fonseca and Tucker looked at the data and took it a step further, which helped to disprove the Hadland et al. theory. Fonseca and Tucker performed an independent analysis of the same data, repeated the same data extraction as cited in the Hadland paper, associated payments with prescribing behavior in the same fashion as Hadland described, and were able to closely reproduce the number of opioid prescribers, opioid prescribers accepting payments, and the total number of payments as described in the original Hadland paper.
What Fonseca and Tucker found was that doctors who accepted no free meals from opioid manufacturers wrote anywhere between 0 and 1000 prescriptions in 2015 – the same as those who accepted 50 or more free meals from opioid manufacturers.
As you can see from the above graph created by Fonseca and Tucker, the correlation between opioid manufacturer meals accepted and opioid prescriptions written is not as drastic as one may think after reading the Hadland paper.
In their article rebutting the Hadland paper, Fonseca and Tucker outline how the math actually works and the way Hadland et al. almost molded the numbers in an attempt to create their desired outcome, instead of using the straight numbers as found in reality.
In sum, Fonseca and Tucker conclude that there are many shortcomings that articles which correlate prescribing with industry payments and sponsored meals have, and that such articles should always be taken with a grain of salt as the math is not necessarily what it seems on the surface. They conclude by noting,
“In the case of opioid prescribing, we believe that we have presented a strong case that 1) the relationship between industry payments and prescribing is much weaker than has been presented in the literature, and 2) that prescribing and attendance at manufacturer-sponsored informational lunches are both driven by practice characteristics, rather than the meals themselves driving prescriptions”