Skip to content
Back to catalogue

Verified intelligence · Expert-reviewed from public records, platform data, primary sources, and private research beyond AI search.

AGENTSPharmacy and benefitsGLOBAL

AI agents for pharmacy, PBMs and medication workflows

Pharmacy and PBM AI should prove a drug-class access ledger: benefit rule, formulary evidence, prior-auth result, channel, patient cost, paid fill, continuation and safety escalation.

The colophon

Landscape reviewed
2026-06-01
Latest revision
2026-06-01
Kinetics
Kinetics: activeImpact 0.87 · momentum 0.89 · vector 0.778 balls · 66 px/s

Medication AI depends on drug-class access proof: paid fills, exceptions, therapy continuation and safety escalation by benefit rule. ePA standards, PBM scrutiny and drug prior-authorization policy work make benefit-to-fill evidence more specific.

To watch
  • 15 June 2026: CMS drug prior-authorization proposal comment period closes; U.S. drug prior-authorization lanes may need revised API, response-clock and reporting assumptions before ePA rollout 1.
Known unknowns
Drug-class and channel-level medication-workflow telemetry linking benefit rules, prior-authorization outcomes, paid-fill yield, abandonment, patient cost, pharmacist intervention, specialty continuation and safety escalation.
  1. Executive summary
  2. What changed
  3. Decision map
  4. Signature proof
  5. Provider lanes
  6. Workflow selection
  7. Evidence fit
  8. Private eval design
  9. Risks and controls
  10. Operating model
  11. Economics
  12. Public claim tests
  13. Diligence directory
  14. Recommended next steps
  15. Known unknowns
  16. Source ledger
  17. Glossary

Medication AI should be judged by therapy access, not transaction completion: paid fill, exception clarity, continuation and safety escalation by drug class. Judge electronic prior authorization and formulary logic by medication access. The proof object is a drug-class ledger that follows the request through benefit rule, patient cost and channel. Fill evidence belongs in the paid-fill field 21. Continuation belongs in specialty-support evidence 10. Safety escalation belongs in pharmacovigilance evidence 7.

CMS and NCPDP can make drug authorization more structured 13, while FTC specialty-generic evidence raises the cost of opaque steering or affiliated-channel optimization 11.

The trap is automation that clears a transaction but loses the therapy outcome. A medication can be approved and still fail through abandonment, high out-of-pocket cost, channel redirect, refill break or safety handoff. CMS and ONC guidance make the governance test more explicit: individual evidence, public criteria, bias controls, reviewer accountability and DSI risk management 1213.

Start with high-cost, high-friction drug classes and require a benefit-to-fill scorecard before scaling formulary, PA, dispensing, specialty support or adherence agents. The scorecard should distinguish requirement discovery and documentation capture 1819. It should also separate PA exchange and attachment evidence 1720. Real-time benefit records need their own checks 14. Formulary and claim-response records need separate checks 1516.

Make paid fill, denial or appeal reason, patient-cost exposure, channel redirect, continuation and pharmacist or safety action the required denominator.

Keep formulary and network steering inside human review until the buyer can see who benefited, who abandoned therapy and why.

The buying decision

QuestionAnswerAction
What is the object of approval?A drug-class access ledger from benefit rule and PA result to paid fill, continuation and safety escalation.Approve only lanes that return patient-cost, appeal, fill, continuation and safety fields before expanding beyond transaction automation.
Which lane goes first?High-cost or high-friction therapies where authorization, patient cost, pharmacy channel and continuation are measurable.Baseline clean ePA, denial yield, appeal yield, paid fill, abandonment, refill continuation and pharmacist intervention.
What remains blocked?Opaque formulary steering, channel redirect or adherence automation without patient-cost, continuation and safety evidence.Separate benefit automation from pharmacy-channel incentives and safety reporting.

Public evidence is now structured enough to define private approval tests. The trigger is the gap between faster medication transactions and provable therapy access: ePA structure, PBM scrutiny, patient-cost standards and algorithm-governance evidence now have to connect to fill outcomes, continuation and safety escalation. Fill outcome evidence belongs in the private eval 21. Continuation and safety evidence need specialty and FAERS controls 107. CMS and NCPDP can make drug ePA transactions faster and more structured 13. The proof still comes from therapy access: formulary version, step therapy, patient cost and pharmacy channel. Fill outcomes need ePA-to-fill evidence 21. Refill continuation needs specialty workflow support 10. Pharmacist or safety action needs safety-reporting evidence 7. FTC PBM evidence makes hidden steering, affiliated-pharmacy routing and acquisition-cost opacity a commercial risk 11.

Start with high-cost, high-friction drug classes and contract for a benefit-to-fill scorecard: PA status, denial or appeal reason, exception rationale, patient out-of-pocket exposure, pharmacy-channel redirect, paid-fill date, refill continuation, pharmacist intervention and safety escalation. Scale formulary or network optimization only when those fields are returned by drug class and channel, with individual-patient evidence and bias controls for coverage workflows 12.

For high-risk workflows, the approval record has to connect source evidence, accepted output, reviewer action, exception path, outcome signal and retained audit record before volume or staffing claims matter.

Triggers that change approval timing

TriggerWhy it mattersDecision use
CMS drug prior-auth proposalMedication authorization may move onto API-based requirements, response clocks and public usage evidence 1.Demand transaction-to-fill instrumentation rather than ePA completion metrics alone.
NCPDP ePA standardMedication authorization has pharmacy-specific request and response structure 3.Test by drug class, formulary version, real-time benefit, payer channel and fallback path 1415.
FTC PBM scrutinyBenefit design, specialty economics and pharmacy steering are under public policy pressure, with official specialty-generic markup evidence now available 11.Require policy, acquisition-cost, conflict, channel, patient-cost and appeal evidence for optimization.

Workflow approval changes by authority, source boundary and proof burden. The drug-class access ledger is the proof object: each medication request must resolve into a paid fill, transparent exception, continued therapy or safety escalation. The decision map should start with a drug-class access ledger with benefit rule, formulary version, PA result, pharmacy channel, patient cost, paid-fill status, continuation and safety escalation: what source record is read, what output is drafted, who approves it, where it is stored and what event forces reapproval.

A system can pass a low-risk support test and still fail production work: source evidence, accepted output, integration state, exception handling, reviewer burden and downstream rework have to hold together.

The useful vendor question is whether the system can make the named task inspectable, reversible and measurable inside the operating record.

Workflow selection

WorkflowAccepted workApproval gateApproval rule
Drug-class access ledgerRequest-to-fill record with PA result, exception, pharmacy channel, patient cost, paid fill, refill continuation and escalation outcome 21.NDC or therapy class, benefit type, formulary version, policy source, reviewer action, evidence attachment and outcome timestamp 1417.Fund the ledger before funding broad autonomy.
Electronic drug prior authorizationMedication PA request, response, denial reason, appeal trigger or approved exception tied to benefit and drug class 320.NCPDP or payer channel, source policy and required evidence 3, plus response-clock and fallback evidence where CMS drug-authorization rules apply 1.Require ePA to connect to fill and abandonment outcomes.
Formulary, exception and PBM steeringRecommendation or exception rationale with policy source, patient-cost effect, pharmacy-channel impact and conflict evidence.Formulary version, step-therapy rule, acquisition-cost/NADAC comparison, rebate or network conflict check, human review and appeal path 1115.Block optimization that hides steering, access or affiliated-pharmacy risk.
Specialty therapy start and continuationBenefits investigation, enrollment, first fill, refill continuation, patient-support task or abandonment reason 10.Patient consent, pharmacist review, specialty requirements, adverse-event escalation and support-contact evidence.Measure therapy continuation and abandonment, not only staff minutes.
Dispensing, adherence and safetyRefill intervention, counselling prompt, interaction alert, corrected claim or safety escalation.Pharmacist action, medication history, patient communication, safety check and escalation log 710.Keep dispensing and safety authority with licensed pharmacy staff.

Write authority changes the proof burden

Source path
Reviewer gate
System action
Retest trigger
Drug-class access ledgerRequest-to-fill record with PA result, exception, pharmacy channel, patient cost, paid fill, refill continuation and escalation outcome 21.
Source namedNDC or therapy class, benefit type, formulary version, policy source, reviewer action, evidence attachment and outcome timestamp 1417.
Human gateFund the ledger before funding broad autonomy.
Accepted workRequest-to-fill record with PA result, exception, pharmacy channel, patient cost, paid fill, refill continuation and escalation outcome 21.
ReapprovalRetest when policy, workflow, model, data access or system write rights change.
Electronic drug prior authorizationMedication PA request, response, denial reason, appeal trigger or approved exception tied to benefit and drug class 320.
Source namedNCPDP or payer channel, source policy and required evidence 3, plus response-clock and fallback evidence where CMS drug-authorization rules apply 1.
Human gateRequire ePA to connect to fill and abandonment outcomes.
Accepted workMedication PA request, response, denial reason, appeal trigger or approved exception tied to benefit and drug class 320.
ReapprovalRetest when policy, workflow, model, data access or system write rights change.
Formulary, exception and PBM steeringRecommendation or exception rationale with policy source, patient-cost effect, pharmacy-channel impact and conflict evidence.
Source namedFormulary version, step-therapy rule, acquisition-cost/NADAC comparison, rebate or network conflict check, human review and appeal path 1115.
Human gateBlock optimization that hides steering, access or affiliated-pharmacy risk.
Accepted workRecommendation or exception rationale with policy source, patient-cost effect, pharmacy-channel impact and conflict evidence.
ReapprovalRetest when policy, workflow, model, data access or system write rights change.
Specialty therapy start and continuationBenefits investigation, enrollment, first fill, refill continuation, patient-support task or abandonment reason 10.
Source namedPatient consent, pharmacist review, specialty requirements, adverse-event escalation and support-contact evidence.
Human gateMeasure therapy continuation and abandonment, not only staff minutes.
Accepted workBenefits investigation, enrollment, first fill, refill continuation, patient-support task or abandonment reason 10.
ReapprovalRetest when policy, workflow, model, data access or system write rights change.
Dispensing, adherence and safetyRefill intervention, counselling prompt, interaction alert, corrected claim or safety escalation.
Source namedPharmacist action, medication history, patient communication, safety check and escalation log 710.
Human gateKeep dispensing and safety authority with licensed pharmacy staff.
Accepted workRefill intervention, counselling prompt, interaction alert, corrected claim or safety escalation.
ReapprovalRetest when policy, workflow, model, data access or system write rights change.
Private eval can start hereRequires named owner or reviewerHard approval gate before scale

Use the matrix to decide what proof is required before each workflow becomes accepted work.

Terms used in this report

TermPlain meaning
RouteThe approved path from work to accepted output: source set, data boundary, model or vendor channel, tools, reviewer, evidence record and reapproval trigger.
APIApplication programming interface: a governed software connection used by an app or workflow to call a model or service.

Definitions are reader aids, not separate evidence claims.

MCP PREVIEW INVENTORY

What the locked report contains

Pharmacy and PBM AI should prove a drug-class access ledger: benefit rule, formulary evidence, prior-auth result, channel, patient cost, paid fill, continuation and safety escalation.

Schema endpoints

  • Sample object: /api/v1/reports/RPT_PHARMACY_BENEFITS_AI_AGENTS_MEDICATION_WORKFLOWS/sample
  • Inventory object: /api/v1/reports/RPT_PHARMACY_BENEFITS_AI_AGENTS_MEDICATION_WORKFLOWS/inventory
  • Schema contract: /api/v1/reports/RPT_PHARMACY_BENEFITS_AI_AGENTS_MEDICATION_WORKFLOWS/schema

Coverage fields

  • Category: AGENTS
  • Sector: Pharmacy and benefits
  • Region: GLOBAL
  • Topics: pharmacy AI, PBM, medication prior authorization, formulary, specialty pharmacy, claims, adherence, dispensing, NCPDP, pharmacy benefit

Schema contract

  • Contract: colophon.report_document v2026-05-28
  • Payload schemas: document, summary, sample, inventory, full_report
  • Required top-level fields: title, subtitle, abstract, category, sector, region, topics, sources, evidence, claims, colophon, body
  • Integrity rules: evidence.sourceId must reference sources[].id claim.evidenceIds[] must reference evidence[].id evidence.claimIds[] must reference claims[].id section and block evidence references must resolve colophon.kinetics.speed must match the momentum band for the business impact x momentum model

Data-point inventory

  • vendor_or_actor_ledger: entity names; product categories; source references; confidence notes
  • buyer_decision_fields: pricing signals; deployment constraints; risk flags; monitoring fields
  • evidence_index: source class; access date; claim linkage; verification status

Section inventory

  1. 1. Executive summaryPreview

    decision-ready conclusions; major findings; confidence caveats

  2. 2. What changedPreview

    structured findings; supporting evidence; buyer-use fields

  3. 3. Decision mapPreview

    selection criteria; tradeoff map; recommended buyer actions

  4. 4. Signature proofLocked

    structured findings; supporting evidence; buyer-use fields

  5. 5. Provider lanesLocked

    structured findings; supporting evidence; buyer-use fields

  6. 6. Workflow selectionLocked

    structured findings; supporting evidence; buyer-use fields

  7. 7. Evidence fitLocked

    claim support; known gaps; verification notes; 29 evidence objects

  8. 8. Private eval designLocked

    structured findings; supporting evidence; buyer-use fields

  9. 9. Risks and controlsLocked

    structured findings; supporting evidence; buyer-use fields

  10. 10. Operating modelLocked

    structured findings; supporting evidence; buyer-use fields

  11. 11. EconomicsLocked

    structured findings; supporting evidence; buyer-use fields

  12. 12. Public claim testsLocked

    structured findings; supporting evidence; buyer-use fields

  13. 13. Diligence directoryLocked

    structured findings; supporting evidence; buyer-use fields

  14. 14. Recommended next stepsLocked

    structured findings; supporting evidence; buyer-use fields

  15. 15. Known unknownsLocked

    structured findings; supporting evidence; buyer-use fields

  16. 16. Source ledgerLocked

    normalized entity rows; source ledger; evidence-to-claim mapping

  17. 17. GlossaryPreview

    structured findings; supporting evidence; buyer-use fields

Relevance signals

  • Confidence score: 0.84
  • Source classes: proposed_rule, agency_program, pharmacy_standard, regulatory_report, regulation, safety_reporting, network_docs, industry_report, pharmacy_provider_docs, agency_guidance, health_it_rule, draft_operating_rule, implementation_guide, peer_reviewed_study, industry_survey, vendor_directory
  • Known unknowns: Drug-class and channel-level medication-workflow telemetry linking benefit rules, prior-authorization outcomes, paid-fill yield, abandonment, patient cost, pharmacist intervention, specialty continuation and safety escalation.
  • Watch signals: 15 June 2026: CMS drug prior-authorization proposal comment period closes; U.S. drug prior-authorization lanes may need revised API, response-clock and reporting assumptions before ePA rollout [[E1]].

Locked continuation

Purchase or subscribe to unlock the full findings, evidence ledger, and buyer-ready recommendations.

HumanAgent