Surescripts ePA impact figures need separate cohorts

Freeze two populations before using a percentage

The direct answer is that a network growth percentage is not the denominator for the narrower outcome comparison. Surescripts' official 2025 impact report describes 19.9% growth in electronically processed prior authorizations and 26.1% growth for specialty medications, while its automation comparison concerns requests from Fairview and Cleveland Clinic to Optum Rx. These are provider-reported measures with different eligible populations. A buyer cannot apply the cohort's 22%-to-4% prescriber-abandonment figures to the full Surescripts network, a medical-benefit authorization program, another pharmacy benefit manager, or a particular plan without matching definitions and independent evidence.

Build one metric card per measure: reporting entity, transaction type, medication class, health system, prescriber population, plan or PBM, NCPDP pathway, observation window, baseline method, numerator, denominator, excluded requests, definition of abandonment, comparison method, result, source URL, extraction date, and responsible reviewer. If the report does not disclose a count, definition, confidence interval, or adjustment method, label it not reported rather than reconstructing it from another percentage. An 18-percentage-point difference between the displayed abandonment percentages is arithmetic on reported values, not a measured effect for an unobserved buyer population.

Do not collapse route, request, response, and access

The subject is pharmacy electronic prior authorization, not the CMS-0057-F medical-service API program. A pharmacist or prescriber can encounter a requirement, open an electronic question set, transmit the prescriber's evidence, receive network acknowledgment, obtain a PBM or plan decision, revise a prescription, appeal, and later see a claim and dispense. The report's electronically processed percentage counts a route or process category as Surescripts defines it; it does not establish that every request was approved, fulfilled, paid, or clinically appropriate. Prescriber abandonment is likewise not a direct measure of medication adherence or patient outcome.

Surescripts says its network migrated toward NCPDP SCRIPT v2023011 in 2025 and describes clearer communications and more precise medication specification as expected benefits. A purchasing test should nevertheless name actual message versions, connected prescribers and EHRs, pharmacy benefit managers, drug identifiers, question-set provenance, payer receipt, response correlation, duplicate handling, resubmission, renewal, and exceptions. Published standards work and provider positioning do not establish that a particular customer's channels or plan rules were deployed, tested, or available throughout 2025.

Compare like cases before attributing improvement

A buyer evaluating automation should stratify prospective pharmacy-ePA cases by plan, PBM, medication and specialty status, prescriber, EHR, requirement source, reason for authorization, baseline intake channel, available evidence, urgency, period, and missing-data state. Compare the same definition of submitted, incomplete, denied for missing information, appealed, abandoned, dispensed, and paid in pre- and post-implementation periods. Preserve late corrections and a non-automated path. Otherwise a change in drug mix, participating clinics, processor policy, enrollment, or reporting can masquerade as a product effect.

The two-health-system Optum Rx comparison is a useful prompt for a buyer-led measurement design, not an independently reproduced causal study here. The annual report does not give this publication raw records, absolute transaction counts for every displayed figure, randomized assignment, a matched control group, every exclusion, or a cross-PBM comparison. Ask Surescripts and the buyer's own participants for the underlying measure specifications and de-identified cohort audit before selecting a target, estimating savings, or making a clinical-access promise. Plans and PBMs remain responsible for benefit policies and determinations; prescribers retain clinical judgment.

Test a specialty request across the measured states

Use synthetic cases for one common medication and one specialty medication. For each, record a requirement check, precise drug and member-plan context, initial question set, attachment, incomplete-data denial, corrected resubmission, appeal, alternate channel, abandoned prescriber attempt, final determination, later pharmacy claim, and dispense or non-dispense. Identify which event belongs to the network-growth population and which, if any, would have qualified for the named automation cohort under its published definitions. Do not infer a real patient's treatment or a buyer's performance from this scenario.

Surescripts' report supports the attributed 2025 figures, the named Fairview/Cleveland Clinic-to-Optum Rx comparison, and its description of network migration toward NCPDP SCRIPT v2023011. It does not establish absolute denominators for every highlighted figure, generalizability, causation, full network migration, plan-specific payer reach, deployed version, question-set accuracy, authorization appropriateness, coverage, payment, dispensing, adherence, patient outcome, or a customer's return on investment. The controlling evidence for those judgments belongs to the accountable provider, plan, PBM, pharmacy, clinical, finance, data, privacy, and compliance owners.

Enterprise buyer test

Translate this change into the exact population, record type, workflow stage, decision owner, effective date, and evidence that could be affected. Ask current or prospective providers to demonstrate the named workflow with representative data and an exception—not a polished feature tour. Record what official documentation establishes, what a provider states, what the team observes, and what remains unresolved.

A defensible review also identifies the dependency outside the product. Authority interpretation, policy configuration, data quality, integrations, human judgment, approval rights, release governance, training, and retained evidence may remain customer or service responsibilities. The evaluation should preserve those boundaries instead of treating a technology claim as the complete operating model.

What we will watch next

Prior Auth Monitor will watch the named source and affected market records for later evidence that changes status, scope, availability, implementation timing, workflow consequence, or the limits of the initial report. A later announcement does not silently overwrite this dated account; the change ledger preserves the sequence.