The Evidence

The evidence

What the published research actually says.

We built ConfirmMyMeds around findings we didn’t generate. Here they are, with sources, so you can check them. Everything below describes the problem — none of it is a claim about our results, because we don’t have any yet.

On care fragmentation and the ownership gap

  • Seeing multiple providers without proper coordination risks suboptimal outcomes, particularly for chronic disease medication management, where ongoing monitoring benefits from a continuous, longitudinal relationship. Risk of admission due to medication noncompliance increased with the number of physicians regularly seen, and the number of prescribers was the strongest predictor of potentially inappropriate drug combinations. — Pham et al., J Gen Intern Med, PMC3965732
  • Fewer than half of US primary care doctors reported getting information from specialists about changes to their patients’ medications or care plans, compared to 70% or more in Norway, France, and New Zealand. — Mathematica / Annals of Family Medicine and Health Services Research
  • Multiple prescribers who may not be in communication with each other lead to unintended polypharmacy; patients receiving specialized care encounter multiple prescribers, and formulary substitutions, prior authorizations, and fragmentation compound nonadherence risk. — Marcum et al., Clin Geriatr Med, PMC3335752
  • An increasing number of care providers involved in chronically ill patients’ care could lead to discontinuity and fragmentation, which in turn may reduce trust and result in poorer medication adherence. — Uijen et al., BMC Health Serv Res, PMC3515359
  • Prescriber barriers to minimizing inappropriate medications include poor communication with multiple prescribers and specialists, inadequate transfer of information at care interfaces, fragmented and difficult-to-access patient medical records, and failure of patients to know or disclose their full medication lists. — Anderson et al., BMJ Open, PMC4265124
  • In primary care consultations, discrepancies arose when specialists changed a dose without communicating it to the GP; the GP was unaware until the patient raised it. — Grønning et al., BMC Prim Care, PMC9639081
  • In fragmented care, the patient is left to reconcile conflicting perspectives — deciding which explanation should guide action, which concerns remain unresolved, which interpretation to trust. These are clinical judgments being made outside the clinical setting. — Bell, KevinMD 2026 (opinion piece, not peer-reviewed)

On patient confidence across providers

  • Patient confidence differs by strength of agreement, not just presence. For their primary provider, 62.1% of patients strongly agreed the provider knew all their medications (95.8% agreed or strongly agreed). For providers other than their primary doctor, strong agreement fell to 45.6% (92.6% agreed or strongly agreed). The near-identical total-agreement figures mask a 16-point gap in strong confidence. This is a community-wide study, not nationally representative data. — PeaceHealth shared medication list study, AHRQ/NCBI NBK43762

On shared medication lists as a coordination tool

  • In a controlled pre-post study, an electronically Shared Medication List (eSML) was implemented across GP, home care, and pharmacy in Oslo; the aim was to reduce discrepancies by giving all providers the same list. Multiple countries are developing such systems, and the scientific evidence on their effects is still limited. — Rangberg et al., BMC Health Serv Res, PMC8670071
  • Denmark’s Shared Medication Record study found that all medication information from the primary sector that flows through the SMR during admission requires a pharmacist to obtain the actual patient medication list to avoid errors during hospitalization — the shared record alone does not suffice without patient confirmation. — Jørgensen et al., Pharmacy, PMC10037631

On medication list accuracy

  • EMR-reconciled medication lists were accurate in only 21.9% of cases in an emergency department study of 502 subjects. Prescription lists included medications the patient did not report taking (78.9%) and failed to capture non-prescriptions (76.1%). — Monte et al., J Emerg Med 2015
  • More than 40% of medication errors trace to inadequate reconciliation. — Monte et al. 2015
  • In a VA ambulatory population with EHR–pharmacy linkage, discrepancies occurred in 60% of patients. The authors concluded that relying on EHRs alone will not ensure an accurate medication list. — Linsky & Simon, BMJ Qual Saf 2013
  • In an indigent primary care clinic, pharmacist-led reconciliation found frequent chart inaccuracies, often related to OTC use and to communication/documentation gaps. — Stewart & Lynch, J Am Pharm Assoc 2012
  • Lack of updated and uniform medication lists poses a problem for continuity of care; only 46% of currently-used medications were registered in hospital discharge letters. — Glintborg et al., Qual Saf Health Care 2007

On time

  • Median medication review in ambulatory visits: 2.1 minutes, ~10% of visit time, addressing 3 of 7 medications, with multitasking EHR use during 47% and focused clinician–patient talk during only 24%. — Matta et al., JMIR Med Inform 2018
  • Thorough reconciliation takes 30–60 minutes per hospital admission; three disciplines are involved with little agreement on roles, and data gathering is often duplicated. — AHRQ, Patient Safety and Quality, Ch. 38
  • At a VA skilled nursing facility, comprehensive reconciliation required substantial pharmacist time across multiple information sources. — J Am Med Dir Assoc 2021

On harm

  • Adverse drug events are the most common cause of iatrogenic harm in health care; ambulatory patients may experience ADEs at even higher rates than inpatients. — AHRQ PSNet
  • Roughly 25% of outpatients in adult primary care experienced an ADE. — Gandhi et al., NEJM 2003
  • 99,628 emergency hospitalizations annually for ADEs among US adults 65+. — Budnitz et al., NEJM 2011

On patients as error-finders

Selection-bias note: this study surveyed patients who already read their notes online — present as “among patients who read their records,” not “1 in 5 of all patients.”

  • Among 29,656 patients who read their visit notes, 1 in 5 reported a mistake; 40% considered it serious. Medications were among the most common categories. — Bell et al., JAMA Netw Open 2020
  • Older and sicker patients were twice as likely to report a serious error. — Bell et al. 2020, PMC7284300
  • The authors note that systems for checking the accuracy of notes are almost nonexistent, and overburdened practitioners may import inaccurate medication lists or propagate errors by copying forward.

On whether patients will complete a pre-visit tool

  • OurDX was completed in 65% of eligible adult primary care visits and 39% of pediatric subspecialty visits. Of the concerns patients raised, 63% were verified by physicians as probable safety opportunities. — Bell et al., JAMIA 2023
  • We cite this as the most relevant available benchmark. It is not our data, and it measured a different tool for a different purpose — its patients were portal-enrolled, which is precisely the population constraint we’re built to avoid. Do not present 65% as our expected completion rate.

On the portal constraint — and the honest prior result

  • In a cluster-randomized trial of MedTrue across 6 primary care clinics, pre-visit data was available only for patients who completed a portal questionnaire — approximately one third of patients have portal accessibility. — Gionfriddo et al., JMIR Form Res 2022
  • We should be straight about the rest of that trial: MedTrue did not improve medication list accuracy over usual care (7.5% vs 7.6% with zero discrepancies, OR 1.01, P=.98). The most common discrepancy in both arms was patients reporting they were no longer taking a listed medication. That result is why we built differently — outside the portal. Read the trial →

On what patients contribute

On why the pharmacy can’t close the gap

  • Nearly half of US patients use multiple pharmacies. — Open Forum Infect Dis 2023
  • Dispensing-based adherence scores are pharmacy-siloed. — Int J Clin Pharm
  • Adherence estimated from pharmacy dispensing data alone was 42% versus 56% from claims; for multiple-pharmacy users, 35% versus 67%. — PMC8990834
  • A missing dispensation is ambiguous between nonadherence, error, and outside fill. — Int J Med Inform 2020
  • Community pharmacists’ transitions-of-care role is constrained by lack of EMR access, inconsistent communication, and inadequate reimbursement. — J Pharm Pract Res 2026
  • Community pharmacy count declined 2023–2025; average 217 Rx/day. — NCPA / Cardinal Health (vendor-sponsored)
  • Workload and staffing are pharmacists’ top two stressors. — ASHP
  • Admission medication reconciliation requires ~15–30 minutes per patient. — Chen et al., BMC Health Serv Res 2026
  • MTM barriers include workflow integration and lack of patient engagement. — Pharmacy Times
  • Post-discharge reconciliation barriers center on staffing and volume peaks. — Kennelty et al., PMC4409924

On documentation requirements

CMS MIPS Measure #130 asks for best-effort documentation of a current, complete, accurate medication list during each encounter, including OTCs, herbals, vitamins, minerals, supplements, and cannabis/CBD. — CMS QPP Measure #130 (PDF)

Full reference list

Every source cited on this site

  1. Pham HH, et al. J Gen Intern Med. — free full text
  2. Mathematica / Ann Fam Med & Health Serv Res. — link
  3. Marcum ZA, et al. Clin Geriatr Med. — free full text
  4. Uijen AA, et al. BMC Health Serv Res. — free full text
  5. Anderson K, et al. BMJ Open. — free full text
  6. Grønning K, et al. BMC Prim Care. — free full text
  7. Bell SK. KevinMD, 2026 (opinion, not peer-reviewed). — link
  8. PeaceHealth shared medication list study, AHRQ/NCBI (book chapter, not a standalone trial). — free
  9. Rangberg A, et al. BMC Health Serv Res. — free full text
  10. Jørgensen SK, et al. Pharmacy. — free full text
  11. Monte AA, et al. J Emerg Med. 2015;49(1):78–84. — free full text
  12. Linsky A, Simon SR. BMJ Qual Saf. 2013;22(2):103–9. — abstract
  13. Stewart AL, Lynch KJ. J Am Pharm Assoc. 2012;52(1):59–66. — abstract
  14. Glintborg B, et al. Qual Saf Health Care. 2007. — abstract
  15. Matta GY, et al. JMIR Med Inform. 2018;6(2):e10167. — open access
  16. AHRQ. Patient Safety and Quality, Ch. 38. — free
  17. VA skilled nursing facility reconciliation. J Am Med Dir Assoc. 2021. — link
  18. AHRQ PSNet. Medication Errors and Adverse Drug Events. — link
  19. Gandhi TK, et al. NEJM. 2003;348:1556–64. — link
  20. Budnitz DS, et al. NEJM. 2011;365:2002–12. — link
  21. Bell SK, et al. JAMA Netw Open. 2020;3(6):e205867. — open access
  22. Bell SK, et al. JAMIA. 2023;30(4):692–702 (OurDX). — free full text
  23. Bell SK, et al. BMJ Qual Saf. 2024;33(9):597–608. — free PMC
  24. Bourgeois FC, et al. Appl Clin Inform. 2023;14(5):903–12. — free full text
  25. Patients’ ability to review EHR to identify errors. JMIR cross-sectional. — free full text
  26. Gionfriddo MR, et al. JMIR Form Res. 2022;6(3):e33488. — open access
  27. Nearly half of US patients use multiple pharmacies. Open Forum Infect Dis. 2023. — free
  28. Dispensing adherence pharmacy-siloed. Int J Clin Pharm. — free
  29. Adherence 42% vs 56%; 35% vs 67% multi-pharmacy. PMC8990834. — free
  30. Missing dispensation ambiguous. Int J Med Inform. 2020. — link
  31. Community pharmacist TOC constrained. J Pharm Pract Res. 2026. — link
  32. Pharmacy count declined; 217 Rx/day. NCPA / Cardinal Health (vendor-sponsored). — link
  33. Workload/staffing top stressors. ASHP. — link
  34. Chen et al. Admission MedRec ~15–30 min/patient. BMC Health Serv Res. 2026;26:81. — link
  35. MTM barriers incl. patient engagement. Pharmacy Times. — link
  36. Kennelty et al., pharmacist med rec barriers: staffing, volume. PMC4409924. — free full text
  37. CMS. Quality ID #130. — PDF
  38. Nephrology reconciliation. PMID 29218971. — link
  39. Predictors of medication discrepancies. PMC8453145. — link
  40. Federal requirement for standardized, read-only EHR data access. — link
  41. The Joint Commission. National Patient Safety Goal NPSG.03.06.01 (medication reconciliation). — link
  42. Portal reconciliation vs. pharmacy technician, RCT. — link
  43. PGHD clinician concerns: reliability, workflow. npj Digital Medicine 2025. — open access
  44. Physicians recognize PGHD value, cite concerns. JMIR 2026;28:e86368. — open access
  45. Lavallee DC, et al. Dismissive attitudes cause patients to stop sharing. mHealth. — open access
Scroll to Top