Pharmacist-Led Education to Discontinue Inappropriate Prescribing—Reply

Letters

Figure. Estimates of the Effect of Low vs Intermediate Tidal Volume Ventilation Strategies in Subgroups Defined Post Hoc in the PReVENT Trial

Low Tidal Volume Intermediate Tidal Volume Ventilator-Free Ventilator-Free Days,a Mean Days,a Mean Mean Difference Favors Intermediate Favors Low P Value Subgroups (SD) No. (SD) No. (95% CI) Tidal Volume Tidal Volume Cardiac arrest Yes15.2 (12.4) 110 15.1 (12.4) 120 –0.00 (–3.17 to 3.29) .80 No15.2 (11.3) 367 15.6 (11.1) 364 –0.38 (–2.02 to 1.25) Sepsis Yes12.0 (11.7) 50 14.2 (10.8) 46 –2.20 (–5.46 to 1.06) .17 No16.0 (11.4) 427 15.7 (11.6) 438 –0.25 (–1.38 to 1.89) Postoperative ventilation Yes17.7 (11.1) 82 17.8 (10.8) 79 –0.08 (–3.26 to 3.09) .87 No14.6 (11.6) 395 15.0 (11.5) 405 –0.38 (–2.01 to 1.25) Risk of ARDSb Higher14.0 (11.4) 292 14.2 (11.4) 290 –0.19 (–2.06 to 1.66) .96 Lower17.0 (11.6) 185 17.3 (11.3) 194 –0.27 (–2.61 to 2.07) Pneumonia Yes13.3 (11.2) 77 11.7 (10.6) 77 1.56 (–1.94 to 5.08) .29 No15.6 (11.6) 400 16.2 (11.5) 407 –0.58 (–2.18 to 1.01) Airway protection Yes17.4 (11.2) 39 16.1 (12.2) 39 1.30 (–4.02 to 6.64) .53 No15.0 (11.6) 438 15.4 (11.4) 445 –0.41 (–1.93 to 1.11) Cardiac failure Yes15.4 (11.2) 28 18.4 (12.3) 17 –2.93 (–10.39 to 4.53) .45 No15.2 (11.6) 449 15.4 (11.4) 467 –0.18 (–1.68 to 1.31) Head trauma or brain surgery Yes14.3 (12.0) 25 14.5 (10.5) 31 –0.18 (–6.32 to 5.95) .97 No15.3 (11.6) 452 15.6 (11.5) 453 –0.29 (–1.80 to 1.21) Aspiration Yes14.1 (11.6) 20 14.9 (11.7) 24 –0.80 (–5.41 to 3.79) .81 No15.3 (11.6) 457 15.5 (11.4) 460 –0.23 (–1.78 to 1.32) All patients –0.27 (–1.74 to 1.19)

–6 –5 –4 –3 –2 –1 0 1 2 3 4 5 6 Mean Difference (95% CI)

a Ventilator-free days and alive at day 28. b Higher risk of acute respiratory distress syndrome (ARDS) is defined as a Lung Injury Prediction Score of at least 4.

Corresponding Author: Ary Serpa Neto, MD, MSc, PhD, Department of Critical pharmacist-led education directed at older patients who were Care Medicine, Hospital Israelita Albert Einstein, Albert Einstein Ave 700, prescribed Beers Criteria medications led to discontinuation of São Paulo, Brazil ([email protected]). inappropriate prescriptions.1 The study demonstrated validity Conflict of Interest Disclosures: None reported. because there were few differences between treatment and con- 1. Simonis FD, Serpa Neto A, Binnekade JM, et al; Writing Group for the trol groups, randomization was concealed, participants were PReVENT Investigators. Effect of a low vs intermediate tidal volume strategy on ventilator-free days in intensive care unit patients without ARDS: a randomized blinded, and few participants were lost to follow-up. However, clinical trial. JAMA. 2018;320(18):1872-1880. doi:10.1001/jama.2018.14280 there are 2 aspects of the trial that raise concerns. 2. Simonis FD, Binnekade JM, Braber A, et al. PReVENT—protective ventilation First, a significant portion of the eligible pharmacies and in patients without ARDS at start of ventilation: study protocol for a randomized patients were not included in the study. Half of all eligible phar- controlled trial. Trials. 2015;16:226. doi:10.1186/s13063-015-0759-1 macies declined to participate either because of competing pri- 3. Serpa Neto A, Cardoso SO, Manetta JA, et al. Association between use of orities or lack of interest in research. Of the participating phar- lung-protective ventilation with lower tidal volumes and clinical outcomes among patients without acute respiratory distress syndrome: a meta-analysis. macies, more than half of the eligible patients (1805 of 2815) JAMA. 2012;308(16):1651-1659. doi:10.1001/jama.2012.13730 did not provide consent to be contacted by the research team. 4. Determann RM, Royakkers A, Wolthuis EK, et al. Ventilation with lower tidal Analysis of the pharmacies and patients excluded from the volumes as compared with conventional tidal volumes for patients without study should be done to assess for selection bias. We would acute lung injury: a preventive randomized controlled trial. Crit Care. 2010;14(1): be especially interested to see if there were any defining char- R1. doi:10.1186/cc8230 acteristics or geographic locations of the pharmacies that re- fused to participate, which consequently excluded their pa- Pharmacist-Led Education to Discontinue tients as well. Inappropriate Prescribing Second, reproducibility in the United States may be lim- Tothe Editor In the D-PRESCRIBE cluster randomized clinical trial ited by differences in pharmacist reimbursement for cognitive conducted in Canada, Dr Martin and colleagues showed that services and the predominance of large for-profit chain

jama.com (Reprinted) JAMA April 2, 2019 Volume 321, Number 13 1313

Downloaded From: https://jamanetwork.com/ by a McGill University Libraries User on 12/11/2020 Letters

pharmacies. Most pharmacies in the United States are large chain [range, 1-28]), frailty rates of 27%, and almost 40% of pa- retail stores that focus primarily on profit generation and met- tients taking more than 10 medications per day. The 2014 rics, such as the number of prescriptions per day, over-the- EMPOWER trial recruited a comparable sample of pharma- counter medications sold, and the number of immunizations cists and patients to an educational deprescribing interven- administered. Reimbursement for patient education services un- tion and achieved discontinuation rates of 27% for chronic ben- der Medicare Part D is variable, which may make it a much lower zodiazepine users.4 Taken together, the EMPOWER and priority for these businesses. Pharmacists play an integral role D-PRESCRIBE trials provide strong evidence that the depre- in patient care and are trained to provide counseling to pa- scribing intervention will work effectively if applied by high- tients. In the inpatient and outpatient clinic settings, pharma- volume, for-profit pharmacy chains in the United States. cists are frequently embedded and often provide direct dis- Whether reimbursement strategies for US pharmacists will ease and medication management services for patients. Patient preclude uptake is a distinct issue. The pay-for-prescription, empowerment in the community setting can lead to a reduc- fee-for-service model of pharmacy care in the United States tion in inappropriately prescribed medications, but reimburse- is similar to the model in Canada. While pharmacists in Quebec ment and incentives for pharmacies and pharmacists to pro- had access to an additional small, government-sponsored ser- vide these services need to be improved. vice fee of Can$19 for sending evidence-based pharmaceuti- cal opinions to physicians, this fee is not available in many US Madeline McCrary, MD jurisdictions. According to the American Pharmacists Asso- Jennifer Goldstein, MD, MSc ciation, new payment models that are less volume based and more value based are becoming increasingly common, de- Author Affiliations: Christiana Care Health System, Newark, Delaware. pending on the organization pharmacists work within.5 How- Corresponding Author: Madeline McCrary, MD, 4755 Ogletown Stanton Rd, ever, financial incentives and performance measures alone do Ammon Education Bldg, Ste 2E70, Newark, DE 19899 (madeline.mccrary@ not lead to successful deprescribing.6 The uptake of im- christianacare.org). proved quality-of-care practices depends on the motivation of Conflict of Interest Disclosures: None reported. individual practitioners to apply evidence-based interven- 1. Martin P, Tamblyn R, Benedetti A, Ahmed S, Tannenbaum C. Effect of a pharmacist-led educational intervention on inappropriate medication tions. Insurers may also be motivated by approaches that re- prescriptions in older adults: the D-PRESCRIBE randomized clinical trial. JAMA. duce waste, diminish costs, and increase value. The evolu- 2018;320(18):1889-1898. doi:10.1001/jama.2018.16131 tion of health care systems toward a self-driven, internally reflective microcosm that embraces evidence-based interven- In Reply We deliberately designed the D-PRESCRIBE study as tions and limits low-value care remains a work in progress. a pragmatic clinical trial to mirror real-life conditions and test the effectiveness, rather than the efficacy, of the pharmacist- Cara Tannenbaum, MD, MSc led educational intervention for patients and primary care Robyn Tamblyn, PhD clinicians.1 The more similar the participants and conditions in a trial are to real life, the more reproducible the findings will Author Affiliations: Faculty of Pharmacy, Université de Montréal, Montreal, be in practice. For this reason, the trial was rolled out in the Quebec, Canada (Tannenbaum); Department of Epidemiology, Biostatistics and Occupational Health, McGill University, Montreal, Quebec, Canada (Tamblyn). community, outside of the academic setting, without finan- Corresponding Author: Cara Tannenbaum, MD, MSc, Centre de Recherche de cial incentives offered to pharmacists or patients. We re- l’Institut universitaire de gériatrie de Montréal, 4545 Queen Mary Rd, Montreal, cruited pharmacies through partnership with 3 large for- QC H3W 1W5, Canada ([email protected]). profit pharmacy chains serving rural and urban areas in Quebec. Conflict of Interest Disclosures: Dr Tannenbaum reports receipt of funding Although only half of the eligible pharmacies agreed to par- from the Fonds de Recherche en Santé de Quebec. No other disclosures were ticipate, and we could not record the profiles of pharmacies reported. that declined, the characteristics of the 69 pharmacies that en- 1. Martin P, Tamblyn R, Benedetti A, Ahmed S, Tannenbaum C. Effect of a pharmacist-led educational intervention on inappropriate medication rolled in the trial were similar to the characteristics of a popu- prescriptions in older adults: the D-PRESCRIBE randomized clinical trial. JAMA. lation-based sample of 1742 pharmacies in the province of 2018;320(18):1889-1898. doi:10.1001/jama.2018.16131 2 Quebec. Pharmacies in the intervention group filled a mean 2. Winslade N, Tamblyn R. Determinants of community pharmacists’ quality of number of 411 prescriptions per day (range, 140-1000), with care: a population-based cohort study using pharmacy administrative claims more than 50% of their clientele composed of adults aged 65 data. BMJ Open. 2017;7(9):e015877. doi:10.1136/bmjopen-2017-015877 years and older. This large-chain, high-volume prescribing pro- 3. Westrick SC, Patterson BJ, Kader MS, Rashid S, Buck PO, Rothholz MC. National survey of pharmacy-based immunization services. Vaccine. 2018;36 file of consenting pharmacies compares favorably—and even (37):5657-5664. doi:10.1016/j.vaccine.2018.07.027 exceeds—the mean prescription volume reported in 2017 by a 4. Tannenbaum C, Martin P, Tamblyn R, Benedetti A, Ahmed S. Reduction of random sample of 292 nationally representative community inappropriate benzodiazepine prescriptions among older adults through direct pharmacies in the United States.3 The latter sample of US phar- patient education: the EMPOWER cluster randomized trial. JAMA Intern Med. macies reported a mean prescription volume of 228 prescrip- 2014;174(6):890-898. doi:10.1001/jamainternmed.2014.949 tions per day (range, 0-900). Furthermore, the following char- 5. American Pharmacists Association. Billing Primer: A Pharmacist’s Guide to Outpatient Fee-for-Service Billing. Washington, DC: American Pharmacists acteristics of participants in the trial reflected the type of patient Association; 2018. that physicians and pharmacists are likely to see in practice: 6. Rat C, Penhouet G, Gaultier A, et al. Did the new French pay-for-performance mean age of 75 years (range, 66-96 years) with significant poly- system modify benzodiazepine prescribing practices? BMC Health Serv Res. pharmacy (a mean number of 9 different medications per day 2014;14(1):301. doi:10.1186/1472-6963-14-301

1314 JAMA April 2, 2019 Volume 321, Number 13 (Reprinted) jama.com

Downloaded From: https://jamanetwork.com/ by a McGill University Libraries User on 12/11/2020