Revealed Opportunism: How Physicians "Game" Prior Authorization Protocols Until They Are Rescinded

Prior authorization (PA) is a management technique that has been implemented to manage the utilization of expensive drugs and to improve the precision of drug prescribing. PA requirements may incentivize physicians to document adverse effects, sometimes falsely, to meet the eligibility requirements.

To identify documentation patterns that may facilitate the quantitative analysis of physician gaming and underreporting behaviors associated with the prescribing of angiotensin-converting enzyme (ACE) inhibitors in a primary care setting, and to evaluate the effect of a PA requirement on the documentation of adverse events as a way to receive approval for more expensive drugs.

We conducted a retrospective analysis of physician electronic reporting of adverse effects associated with ACE inhibitors before and after the revocation of a PA requirement for angiotensin receptor blockers (ARBs) between 2004 and 2013 in an Israeli HMO. The data were stratified into 2 groups-patients who were newly prescribed an ACE inhibitor or those who had been receiving an ACE inhibitor for at least 1 year. The annual rate of adverse events related to ACE inhibitors (ie, the number of reported cases of adverse events per 1000 patients receiving an ACE inhibitor) was calculated from data captured on the date the events were first reported for the 5 years before and 5 years after the revocation of the PA constraint.

A total of 151,845 patients treated with ACE inhibitors were identified during the 10-year study period. The reported adverse events among patients newly treated with an ACE inhibitor peaked in 2007 to 10 cases per 1000 patients, and gradually decreased to 4.6 cases in 2012, which was the year after the PA requirements for the ARBs valsartan and candesartan were rescinded by the HMO. Among previously treated patients, adverse events rates decreased from a high of 5.4 per 1000 patients in 2008 to 1.9 in 1000 patients in 2012, the year after the PA restraints for the last 2 ARBs with a requirement were revoked.

The PA requirement influenced physician propensity for reporting drug side effects, possibly encouraging reporting inaccuracies. The decline in the incidence of reported side effects, in both subpopulations in the study, with the revocation of the PA requirement confirms our hypothesis that physicians were incentivized to document the side effects related to ACE inhibitors to meet the eligibility requirements for the approval of an ARB by the HMO.