Impact of clinical pharmacists' recommendations on a proton pump inhibitor taper protocol in an ambulatory care practice.

BACKGROUND Previous studies have demonstrated an association between chronic proton pump inhibitor (PPI) utilization and adverse events such as fractures, infections, hypomagnesemia, and vitamin B12 deficiency. Because patients taking PPIs for an extended period of time are more susceptible to these adverse events, an approach to tapering patients off PPIs is clinically warranted. OBJECTIVE To evaluate the impact of clinical pharmacists' recommendations to clinicians to decrease PPI use in patients when chronic therapy is not indicated. METHODS Clinical pharmacists electronically sent PPI taper recommendations for qualifying patients to primary care providers the day before each patient's appointment. Using insurance claims data, an average pills per month (PPM) count was calculated for the 5-month period prior to initiating the PPI taper as well as for the 5-month period after the date of taper initiation. The PPM count was calculated by dividing the total number of pills a patient received by the total number of days in that period, multiplied by 30. The primary outcome for the study was the change in average PPM count from baseline (pretaper period) to follow-up (posttaper period) and was assessed using a paired t-test. Secondary outcomes included change in total annualized PPI costs to the organization, proportion of patients who began the taper protocol after primary care provider recommendation, and whether baseline characteristics were predictors of successful response. Change in annualized PPI costs to the organization was calculated by multiplying the average unit cost per pill (determined using a weighted average of the average wholesale price of the individual drugs) by the PPM change seen with the primary outcome and by the number of patients included in the study and expressed over the period of a full year. Logistic regression analysis was used to determine whether baseline variables including alcohol and tobacco use, diagnosis related to PPI use, PPI dose, dosing frequency, gender, and length of prior PPI use significantly impacted successful tapering. RESULTS Average PPM count decreased by 8.7 pills (95% CI: 6.4, 11.1), from 25.6 at baseline (95% CI: 23.1, 28.1) to 16.9 at follow-up (95% CI: 14.3, 19.5; P less than 0.001). For the 117 evaluable patients in the study, there was an annualized PPI cost reduction of $18,151. 37.6% (44/117) of pharmacist-recommended tapers were enacted upon by primary care providers at the patient visit. Baseline patient characteristics were not found to be predictors of a successful taper response. CONCLUSION Clinical pharmacist intervention may decrease overutilization of PPIs and associated costs in the primary care setting. While a decrease in PPI use was observed in this study, these findings do not imply improvement in clinically meaningful patient outcomes.