Industry payments associated with oncologists’ use of preferred cancer treatments

Researchers used fee-for-service Medicare claims to identify patients with a new cancer diagnosis from 2014 to 2020. For 14 cancer scenarios involving variation in clinical benefit among National Comprehensive Cancer Network-recommended options, they determined the preferred treatment at each patient’s diagnosis date. The analysis assessed whether patients received that treatment in relation to drug-industry payments their medical oncologist had received during the preceding year.

Among the 15,607 patients, 41.6% whose oncologist received no payments received the preferred treatment, compared with 51.5% when the oncologist had received payments tied only to the preferred treatment, 34.7% when payments were tied only to nonpreferred treatment, and 42.9% when payments were tied to both. After adjustment, payments tied to preferred treatment were associated with greater use of that treatment (risk ratio [RR], 1.06; 95% CI, 1.01–1.13); payments tied to nonpreferred treatment were associated with lower use (RR, 0.93; 95% CI, 0.89–0.98). Associations were stronger at higher payment amounts.

The findings indicate an association, not proof that payments caused prescribing choices. The study was observational and limited to Medicare patients and the specified cancer scenarios.