Prescription Drug Prior Authorization: Costs to Pharmacies and Physicians

ABSTRACT

Objective: To determine the costs of prior authorization (PA) for prescription drugs in outpatient settings in the US and Canada.

Study Design: Systematic literature review.

Methods: We systematically searched PubMed, CINAHL, and Embase using keywords and controlled vocabulary to obtain English-language articles addressing PA and costs. Two authors reviewed each article to determine eligibility for the study.

Results: The search identified 14 articles that met inclusion criteria. Four examined the time and costs of PA to pharmacies. All were conducted at hospital clinics. Mean times to complete PA varied from 15 minutes to 24 minutes. Labor costs varied from $15 to $63. One study estimated the mean time to complete a PA appeal to be 74 minutes. Eleven studies measured the time and/or costs of PA in physician practices. Four understated and 1 may have overstated time spent on PA. Mean estimates from the remaining studies varied from 25 to 64 minutes. Six studies also measured labor-related costs. The studies that misstated time also misstated labor costs. The 3 remaining mean estimates varied from $21 to $49.

Conclusions: Providers incur substantial costs when completing PAs. However, not enough studies have been done to accurately estimate the magnitude of these costs or identify key sources of cost variation. Further, there is no research on the costs of PA to community, mail-order, or specialty pharmacies.

Am J Manag Care. 2026;32(6):In Press

Takeaway Points

A systematic review found 14 articles that measured costs of prior authorization (PA) for drugs in pharmacies and physician practices.

• Four studies measured the time and costs incurred by health system pharmacies when administering PAs for affiliated clinics and physicians. Time per PA varied from 15 to 24 minutes and labor costs from $15 to $63.
• No studies measured the costs of completing PAs in community, mail-order, or specialty pharmacies.
• Eleven studies measured the costs and/or time required to complete PAs in physician practices. The best studies indicated that time per PA varied from 25 to 64 minutes and costs from $21 to $49.

Prior authorization (PA) is “a health plan cost-control process that requires physicians and other health care professionals to obtain advance approval from a health plan before a specific service is delivered to the patient to qualify for payment coverage.”¹ PA is widely and increasingly used by managed care organizations and pharmacy benefit managers (PBMs). For example, Medicare Advantage plans completed nearly 50 million PAs in 2023, up from 37 million in 2021.² In 2022, the Medical Group Management Association reported that 79% of responding medical groups said that payer PA requirements had increased in the last 12 months, and this number had increased each year since 2016.³ A 2024 American Medical Association survey found physicians completed an average of 39 PAs per week,⁴ up from 29 in 2017.⁵

These statistics relate to all medical services, including drugs, imaging studies, and hospitalizations. The trends for drug-related PA are similar. A survey of commercial health plans found that 97% used PA in 2023.⁶ The Medicare Payment Advisory Commission found that 31% of drugs in Medicare Prescription Drug Programs (PDPs) and 28% of those in Medicare Advantage PDPs were subject to PA in 2022.⁷ Dieguez et al examined sole-source drugs for chronic myeloid leukemia, multiple sclerosis, autoimmune diseases, and depression and found that use of PA and step therapy for these products increased from 35% in January 2011 to 67% by January 2016.⁸

Providers and payers have different views about the costs of PA. From the payer’s perspective, PA generates savings by decreasing drug costs.^9,10 From the provider’s perspective, PA increases costs. Pharmacists experience increased labor costs from contacting physicians to have them initiate PAs, following up with physicians and patients to determine when insurers answer those requests, and then resubmitting prescriptions once the PA is approved. Physicians must contact payers, determine the information needed to complete the PA, compile and submit the information, and relay the PBM’s responses to patients and pharmacies. Further, there is great variation among the forms and processes used by different payers. This “lack of standardization is associated with significant administrative burdens for providers, who must identify and comply with each entity’s unique requirements.”¹¹ Physicians and pharmacists also incur nonfinancial costs. There is considerable stress involved in dealing with PA-related delays when getting a patient started on a medication, spending phone time on hold with the insurer, and dealing with patients who wonder why they cannot get medicines they need. A total of 89% of physicians report that PA increases physician burnout.⁴ This review will address only the financial costs of PA. We define cost as the financial expenses incurred by pharmacies and physician practices in completing PAs.

Larger practices may have specialized departments that administer PAs. In addition to salaries, these practices have overhead costs such as rent and utilities for the space occupied by employees who conduct PAs. All practices have opportunity costs. The opportunity cost of PA is the revenue that employees could generate, or the other useful work they could perform, if they were not conducting PAs.

Several studies have measured the time and costs of PA for drugs in physician practices and pharmacies. However, there have been no reviews of these studies. We focused on the costs of drug-related PA because we were interested in those specific costs, rather than costs for imaging or hospitalizations, and because the PA process is different for drugs. PAs for drugs are usually initiated after a prescription is written, the patient takes it to the pharmacy, it is submitted to the PBM, and the PBM denies payment. PAs for nondrug services are usually initiated by the physician, submitted to the health plan, and completed before the patient has received the service. Further, many drugs are subject to PA; each may have its own unique PA requirements, and drugs and requirements frequently change. As a result, it is reasonable to expect PA costs for prescription drugs to differ from those for other medical services. The objective of this study was to characterize the costs incurred by pharmacies and physicians for PA of prescription drugs in outpatient settings in the US and Canada.

METHODS

We systematically searched PubMed, CINAHL, and Embase using keywords and controlled vocabulary to obtain English-language articles addressing PA and costs. Complete search strategies are described in the eAppendix (available at ajmc.com).

We included research studies that provided quantitative information about costs or time associated with PA for prescription drugs in outpatient settings. We excluded literature reviews and commentaries, studies of nondrug services, studies conducted outside the US or Canada, and modeling studies. Prior research has examined changes in drug utilization and/or savings on drug or medical services resulting from PA.^10,12 Therefore, we did not include articles on these topics. We included studies in which respondents reported actually measuring time and/or costs of PA; surveys that asked respondents to estimate time or costs from memory were not included. We excluded studies that did not differentiate costs and time spent on PA for drugs vs nondrug services. We excluded published conference abstracts because they often lacked critical details on methods or results. We did not include studies of step therapy.

Two authors (N.V.C. and N.W.C.) reviewed the abstract of each article to determine whether it fit the inclusion or exclusion criteria. After reviewing 50 abstracts, the authors discussed any discrepancies in their selections. Discrepancies were resolved by consensus. After all articles had been reviewed by both authors, both met to resolve any remaining discrepancies.

Dollar values were converted to 2024 US$ using the Consumer Price Index for medical care services. Only 2024 costs are reported in the text. Costs included both salaries and benefits unless otherwise specified. Our study followed Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines.

RESULTS

Literature searches conducted on June 10, 2025, across PubMed, CINAHL, and Embase identified 2061 articles. After deduplication in EndNote 21 (Clarivate), 1552 were available for title/abstract screening. As shown in the Figure, 14 met the inclusion criteria. All were conducted in the US.

Pharmacy Time and Costs

Four articles examined the time and costs of PA to pharmacies (Table 1). Three were conducted at pharmacy-run PA clinics (PPACs) in health systems. PPACs consist of dedicated teams of pharmacists and pharmacy technicians who use standardized protocols to process PA requests. The process is centralized in that medical clinics within the health system send their PA requests to the PPAC for processing rather than processing the requests themselves.

Cutler et al examined the time to process a PA in a new PPAC. In addition to a standardized protocol, the PPAC had collaborative practice agreements with clinic prescribers. This prospective, observational study examined 47 PAs. Specialty drugs were excluded. PAs took an average of 15 minutes, and mean total labor costs were $15.¹³

Rim et al reported a case study of a PPAC initially implemented for specialty clinics but later expanded to include primary care clinics. During 2019, the clinic processed 13,110 PAs. The authors reported an average labor cost of $63 per PA. They noted that the costs of implementing the PA clinic, including the initial investment to build a new space, were substantial. They did not provide estimates of implementation or overhead costs.¹⁴

Jennings et al examined the time and costs of processing PA requests in a PPAC in a US Department of Veterans Affairs (VA) hospital. This PPAC managed PA requests for prescriptions written by non-VA physicians for outpatient veterans. These PAs did not involve submission to a third-party payer, but they were conducted for prescriptions written for non-VA National Formulary prescriptions. Essentially, the PPAC served the same role as the PBM in a commercial health plan. Pharmacists and technicians documented time spent on 100 nonapproved PA requests. Mean time spent was 24 minutes: 16 minutes for pharmacists and 8 minutes for technicians. Average cost was $22.¹⁵

Hvisdas et al studied the time clinical pharmacists spent managing 221 appealed PAs in outpatient clinics at a university health system. Pharmacists spent a mean of 74 minutes per appeal. This did not include time spent on the initial PA request or technician time spent following up with the payer.¹⁶

Physician Time and Costs

Eleven studies measured time or costs in physician practices (Table 2). Bukstein et al conducted a prospective, observational study in an outpatient allergy clinic in Wisconsin. A total of 117 PA requests were examined. Nurses spent a mean of 20.6 minutes, and physicians spent 4.1 minutes, for an overall mean of 24.7 minutes per PA completed. The mean wage cost was $36: $15 for nurses and $21 for physicians.¹⁷

In another prospective, observational study, Raper et al examined PA in an Alabama AIDS clinic. A nurse practitioner (NP) and clerk handled PAs. Over the 2006-2008 study period, the clinic processed 288 PAs. PA completions required a mean time of 26.8 minutes by the NP and 6.5 minutes by the clerk. Costs of NP and clerk time were $21 per PA. The authors calculated opportunity costs of $45 per PA completed based on the NP’s time.¹⁸

Epling et al examined PAs in 9 primary care clinics in central New York. All staff working on PA requests were asked to complete a standard “event form” for each PA request on which they worked during the 6-week study period. Participants completed 156 event forms for medications. These activities took an average of 16 minutes of prescriber and administrative staff time and had an average labor cost of $16 per event form.¹⁹

Cutler et al measured costs of processing PAs in 3 primary care clinics in a university health system. The clinics used medical assistants, nurses, and physicians to complete PAs. The study examined 77 PA requests. PAs took an average of 64 minutes and had a mean labor cost of $49.¹³

Dickens and Pollack examined PAs in a hematology/oncology clinic in a children’s hospital in Michigan. Over the 150-day study period, 142 PAs were processed. The median (IQR) clinical staff time per PA completed was 46 (25-80) minutes. The study did not specify time by type of clinical staff (eg, physician, nurse) or provide cost estimates.²⁰

Carlisle et al examined PAs at 11 sites of a university dermatology clinic in Utah. The study prospectively recorded the times required to complete 314 PA requests for drugs over a 30-day period. These took a median (IQR) of 13 (7-25) minutes and had a median (IQR) labor cost of $9 ($5-$18) per PA completed. The mean cost was $14. PA completions for biologics took longer and were more expensive; median (IQR) time was 30 (19-46) minutes, and the mean cost was $16 ($9-$23).²¹

Sundaram et al studied the costs of PA in a hospital-owned academic urology practice. In this prospective observational study, time spent completing PAs was recorded by licensed practical nurses, registered nurses, administrative support assistants, advanced practice providers, and physicians. Of 267 PA requests, 132 were for medications. Of these, PA was waived for 52 via a phone call to the insurer. Waivers required a mean of 13 minutes, at a cost of $7. Prescriptions requiring a full PA took a mean of 23 minutes and had a mean cost of $16. The article did not specify whether cost estimates included benefits. Also, it did not provide separate estimates for drug and imaging PAs, but it did indicate there was no statistically significant difference (P = .563) in time spent completing PAs between the two.²²

Habchi et al examined charts from 2014 to 2019 for patients with hepatitis C virus at 8 methadone maintenance clinics in Rhode Island. Time to complete a PA ranged from 45 to 120 minutes. The study suggested that only pharmacists were responsible for handling PA requests. No further information about time or costs was reported.²³

Dang et al examined the time required to complete PAs for anti-VEGF intravitreal injections at 9 privately owned retinal practices across the US. The median time to complete a PA was 100 minutes (range, 0-200 minutes). The mean time was 47 minutes. They did not report costs or the types of personnel who completed PAs.²⁴

Wilk et al conducted a national survey of psychiatrists during the first 4 months after the implementation of Medicare Part D. A total of 1183 psychiatrists met the inclusion criteria. Psychiatrists were asked to report the total time they or their staff spent on prescription drug administration for a dual-eligible patient and the plan features of the patient’s Part D plan. Patients in plans with PA required a mean of 57 minutes compared with 32 minutes for patients in plans without PA. This indicates that psychiatrists and their staff spent an additional 25 minutes on each prescription requiring PA. The study did not specify how much time was spent by psychiatrists vs their staff, nor did it estimate a cost.²⁵

Shah et al measured time and costs associated with telephone calls related to prescribing isotretinoin and nonisotretenoin acne prescriptions to 144 children. Prescriptions needing PA required an additional 0.74 calls to patients, pharmacies, and payers; call lengths ranged from 5 to 120 minutes. This resulted in additional costs of $3 to $73 per prescription.²⁶

DISCUSSION

We identified 4 PA studies conducted at pharmacies. Three were conducted at PPACs. Mean times to complete PAs varied from 15 minutes¹³ to 24 minutes.¹⁵ Also, Hvisdas et al reported that clinical pharmacists required 74 minutes to complete an appealed PA request.¹⁶ Labor costs varied from $15¹³ to $63.¹⁴ The lowest reported time and cost were from a dedicated PA clinic that did not administer PAs for specialty drugs.¹³ The pharmacies in 3 studies completed PAs for physicians’ practices.^13,14,16 The other pharmacy determined whether PA requests were approved.¹⁵ None were community, specialty, or mail-order pharmacies.

Eleven studies measured time and/or costs of PA in physician practices. (We use the term physician practices to refer to both private practices and outpatient clinics.) Four studies understated the time spent on PA. Epling et al reported 16 minutes per PA, which was measured per event form rather than per PA completed.¹⁹ Each PA could have comprised several event forms. Carlisle et al did not include time spent by physicians and was ended with 41% of PA requests pending.²¹ Sundaram et al did not include the time staff spent speaking with pharmacies, prescribing alternate medications, or talking with patients.²² The study by Shah et al did not include time spent on nonphone communications, documentation, Risk Evaluation and Mitigation Strategy portal activities, and electronic PA submissions.²⁶ Wilk et al may have overestimated time spent on PA.²⁵ This study was completed during the first 4 months after implementation of Part D, and there were substantial implementation issues during this time. Mean estimates from the remaining studies varied from 25 minutes¹⁷ to 64 minutes¹³ per PA. Also, Habchi et al reported a range of 45 to 120 minutes.²³

Six studies measured labor-related costs of PA in physician practices. Labor costs were based on time spent, so studies that understated time also understated costs.^19,21,22,26 The 3 remaining estimates varied from $21¹⁸ to $49.¹³

The studies we examined had several limitations. All were based on self-reports by the personnel completing the PAs. However, these self-reports were based on actual measurements of time spent, not on estimates based on memory or judgment. Most were conducted at hospital outpatient clinics. Only 3 included private practices. Few specified how much time each category of personnel spent on PA. This limited our ability to identify reasons for cost differences across practices. The studies predated advancements in informatics and artificial intelligence that could affect the costs and cost categories required to administer PAs.²⁷ Only 1 study considered costs other than labor.¹⁸ None included overhead costs.

Despite these limitations, our review indicates that pharmacies and physician practices bear substantial costs for administering PAs. The pharmacy studies estimated a cost of $22¹⁵ to $63¹⁴ per PA. As discussed earlier, these involved managing PA requests for physician practices. Physician studies estimated times of 25^17,25 to 64¹³ minutes and costs of $21¹⁸ to $49¹³ per PA. These costs would have a significant impact on physician practice profits. For example, Medicare reimbursement in the Richmond, Virginia, area for an office visit with a low level of decision-making (Healthcare Common Procedure Coding System code 99213) ranges from $64 to $90. A PA cost of $21 to $63 is a substantial proportion of this reimbursement. The impact on profitability is greater than our estimates suggest because they do not include the costs of facilities, equipment, and supplies needed for PA.

The studies provide conflicting evidence about the comparative costs of using pharmacists and pharmacy technicians to administer PAs vs physicians, nurses, and administrative staff. Cutler et al found that it was much cheaper to use pharmacists and technicians, at $15 vs $49.¹³ However, the estimate of $63 by Rim et al¹⁴ was higher than any of the estimates for PA conducted by physicians, nurses, and administrative staff.

Our review indicates that more research is needed to fully understand the costs of PA to pharmacies and physician practices. Only 11 studies have looked at the costs of PA to physician practices. Most had limited geographic scope. Only 2 used a national sample; the rest were conducted at a single hospital or in a single state. Only 1 looked at the cost of PA in community primary care practices.

We found no studies that measured time or costs incurred by community, mail-order, or specialty pharmacies for PA. This is a major gap in the literature because the vast majority of prescriptions are dispensed via these sites. PA-related activities differ substantially between these sites and PPACs and physicians’ practices.

CONCLUSIONS

Providers incur substantial costs when completing PAs. However, not enough studies have been done to accurately estimate the magnitude of these costs or to identify key sources of cost variation. Further, there is no research on the costs of PA to community, mail-order, or specialty pharmacies.

Author Affiliations: School of Pharmacy (NVC), College of Health Professions (NWC), and VCU Libraries (ERB), Virginia Commonwealth University, Richmond, VA.

Source of Funding: None.

Author Disclosures: The authors report no relationship or financial interest with any entity that would pose a conflict of interest with the subject matter of this article.

Authorship Information: Concept and design (NVC, NWC); acquisition of data (NVC, NWC, ERB); analysis and interpretation of data (NVC, NWC); drafting of the manuscript (NVC, ERB); critical revision of the manuscript for important intellectual content (NWC); and administrative, technical, or logistic support (NVC, ERB).

Address Correspondence to: Norman V. Carroll, PhD, School of Pharmacy, Virginia Commonwealth University, 1112 E Clay St, Box 980581, Richmond, VA 23298-0540. Email: nvcarroll@vcu.edu.

REFERENCES

1. What is prior authorization? American Medical Association. July 12, 2022. Accessed March 19, 2025. https://www.ama-assn.org/practice-management/prior-authorization/what-prior-authorization

2. Fuglesten Biniek J, Sroczynski N, Freed M, Neuman T. Medicare Advantage insurers made nearly 50 million prior authorization determinations in 2023. KFF. January 28, 2025. Accessed March 19, 2025. https://web.archive.org/web/20250324123217/https://www.kff.org/medicare/issue-brief/nearly-50-million-prior-authorization-requests-were-sent-to-medicare-advantage-insurers-in-2023/

3. Ernst C. Virtually all medical groups say payer prior authorization requirements aren’t improving. Medical Group Management Association. March 2, 2022. Accessed March 19, 2025. https://www.mgma.com/mgma-stats/virtually-all-medical-groups-say-payer-prior-authorization-requirements-aren-t-improving

4. 2024 AMA prior authorization physician survey. American Medical Association. 2024. Accessed April 16, 2025. https://www.ama-assn.org/system/files/prior-authorization-survey.pdf

5. 2017 AMA prior authorization physician survey. American Medical Association. 2018. Accessed March 19, 2025. https://www.ama-assn.org/sites/ama-assn.org/files/corp/media-browser/public/arc/prior-auth-2017.pdf

6. 2024 Trends in Specialty Drug Benefits Report. Pharmaceutical Strategies Group; 2024. Accessed March 19, 2025. https://www.psgconsults.com/industry-report/2024-trends-in-specialty-drug-benefits-report/

7. Health Care Spending and the Medicare Program: A Data Book. Medicare Payment and Advisory Commission; July 2023. Accessed April 16, 2025. https://www.medpac.gov/wp-content/uploads/2023/07/July2023_MedPAC_DataBook_SEC.pdf

8. Dieguez G, Sawhney TG, Mirchandani H. Evolution of the Use of Restrictions in Commercial Formularies. Milliman Inc; October 2016. Accessed March 19, 2025. https://edge.sitecorecloud.io/millimaninc5660-milliman6442-prod27d5-0001/media/Milliman/importedfiles/uploadedFiles/insight/2016/evolution-restrictions-commercial-formularies.pdf

9. Shoemaker S, Pozniak A, Subramanian R, Mauch D; Academy of Managed Care Pharmacy. Effect of 6 managed care pharmacy tools: a review of the literature. J Manag Care Pharm. 2010;16(6, suppl A):S3-S20. doi:10.18553/jmcp.2010.16.S6-A.1

10. Park Y, Raza S, George A, Agrawal R, Ko J. The effect of formulary restrictions on patient and payer outcomes: a systematic literature review. J Manag Care Spec Pharm. 2017;23(8):893-901. doi:10.18553/jmcp.2017.23.8.893

11. Prior authorization and utilization management reform principles. American Medical Association. Accessed April 8, 2025. https://www.ama-assn.org/system/files/principles-with-signatory-page-for-slsc.pdf

12. Happe LE, Clark D, Holliday E, Young T. A systematic literature review assessing the directional impact of managed care formulary restrictions on medication adherence, clinical outcomes, economic outcomes, and health care resource utilization. J Manag Care Spec Pharm. 2014;20(7):677-684. doi:10.18553/jmcp.2014.20.7.677

13. Cutler T, She Y, Barca J, et al. Impact of pharmacy intervention on prior authorization success and efficiency at a university medical center. J Manag Care Spec Pharm. 2016;22(10):1167-1171. doi:10.18553/jmcp.2016.22.10.1167

14. Rim MH, Thomas KC, Barrus SA, et al. Analyzing the costs of developing and operating an integrated health-system specialty pharmacy: the case of a centralized insurance navigation process for specialty clinic patients. Am J Health Syst Pharm. 2021;78(11):982-988. doi:10.1093/ajhp/zxab083

15. Jennings AJ, Brown JN, Britt RB, McNaughton LA, Durkee M, Hashem MG. Value of a pharmacy-adjudicated community care prior authorization drug request service. Fed Pract. 2022;39(8):356-360a. doi:10.12788/fp.0296

16. Hvisdas C, Goode N, Lee K, Kim D, Silvey M, Kasbekar N. Leveraging ambulatory care pharmacist expertise in medication appeals to overcome access barriers: a descriptive report. J Am Coll Clin Pharm. 2022;5(11):1128-1138. doi:10.1002/jac5.1712

17. Bukstein DA, Cherayil GA, Gepner AD, Luskin AT, Kooistra JB, Olson RM. The economic burden associated with prior authorizations in an allergist office. Allergy Asthma Proc. 2006;27(2):119-122.

18. Raper JL, Willig JH, Lin HY, et al. Uncompensated medical provider costs associated with prior authorization for prescription medications in an HIV clinic. Clin Infect Dis. 2010;51(6):718-724. doi:10.1086/655890

19. Epling JW, Mader EM, Morley CP. Practice characteristics and prior authorization costs: secondary analysis of data collected by SALT-Net in 9 central New York primary care practices. BMC Health Serv Res. 2014;14:109. doi:10.1186/1472-6963-14-109

20. Dickens DS, Pollock BH. Medication prior authorization in pediatric hematology and oncology. Pediatr Blood Cancer. 2017;64(6). doi:10.1002/pbc.26339

21. Carlisle RP, Flint ND, Hopkins ZH, Eliason MJ, Duffin KC, Secrest AM. Administrative burden and costs of prior authorizations in a dermatology department. JAMA Dermatol. 2020;156(10):1074-1078. doi:10.1001/jamadermatol.2020.1852

22. Sundaram P, Bhatt V, Feustel P, Mian B. Burden of prior authorization requirements on urology practice and patients. Urology. 2022;169:76-83. doi:10.1016/j.urology.2022.05.055

23. Habchi J, Thomas AM, Sprecht-Walsh S, et al. Optimizing hepatitis C virus (HCV) treatment in a US colocated HCV/opioid agonist therapy program. Open Forum Infect Dis. 2020;7(10):ofaa310. doi:10.1093/ofid/ofaa310

24. Dang S, Parke DW, Sodhi GS, et al. Anti-VEGF pharmaceutical prior authorization in retina practices. JAMA Ophthalmol. 2024;142(8):716-721. doi:10.1001/jamaophthalmol.2024.2217

25. Wilk JE, West JC, Rae DS, Rubio-Stipec M, Chen JJ, Regier DA. Medicare Part D prescription drug benefits and administrative burden in the care of dually eligible psychiatric patients. Psychiatr Serv. 2008;59(1):34-39. doi:10.1176/ps.2008.59.1.34

26. Shah N, Smith E, Czermak J, Ferris M, Kirkorian AY. The administrative burden of prescribing and treating with isotretinoin. J Am Acad Dermatol. 2022;86(5):1165-1167. doi:10.1016/j.jaad.2021.04.062

27. Lenert LA, Lane S, Wehbe R. Could an artificial intelligence approach to prior authorization be more human? J Am Med Inform Assoc. 2023;30(5):989-994. doi:10.1093/jamia/ocad016