Skip to content

    From care gaps to EHR-embedded clinical impact

    We work across five disciplines that usually sit in separate teams: medical education, clinical workflow redesign, EHR optimization, SMART on FHIR development, and clinical-impact evaluation. This is the public reference for how we do it, grounded in open standards.

    Four layers, one continuum

    A tool changes care only when clinicians adopt it, so we cover all four layers.

    01

    Clinical workflow design

    Map how care is delivered, role by role, before changing anything.

    02

    EHR optimization

    The native vendor toolkit. Usually the lowest-friction way to change behavior at the point of care.

    03

    SMART on FHIR applications

    Vendor-neutral apps that do what native tools cannot: live calculation, multi-source synthesis, custom interfaces.

    04

    Implementation science

    Methods for planning adoption and measuring whether it lasts.

    Layer 2 · EHR optimization

    Start with the lowest-friction tool that solves the problem

    Many clinical-workflow problems can be solved inside the EHR's native toolkit before anyone writes code. In Epic, that toolkit looks like this.

    SmartPhrases (.phrases)

    Reusable text blocks expanded by a dot-prefixed token. They standardize documentation.

    SmartText

    Pre-configured template content that pre-populates a note.

    SmartLists / SmartLinks / SmartForms

    Pick-lists, dynamic chart data pulls, and structured input inside the note.

    SmartSets

    Modular bundles of orders plus documentation that standardize a care pathway.

    Order Sets

    Grouped, pre-built orders aligned to a clinical scenario.

    Best Practice Advisories (BPAs)

    Point-of-care alerts and reminders driven by patient data. A native form of decision support.

    Physician Builder

    Epic's certification track that grants clinicians build access (CLN150 basic, CLN155 orders and CDS).

    Epic, SmartPhrase, SmartSet, BPA, and Physician Builder are proprietary to Epic Systems. HMHCP references them descriptively and is not affiliated with or certified by Epic. Oracle Health (Cerner) and MEDITECH expose analogous capabilities through their own tooling and FHIR APIs.

    Layer 3 · SMART on FHIR

    When native tools are not enough, extend the EHR

    A SMART on FHIR app handles what native tools cannot: live calculation, multi-resource synthesis, a custom interface, a registry. It runs vendor-neutral on open standards.

    How a SMART app launches

    • Built on OAuth 2.0. The current SMART App Launch IG (v2.2.0) requires PKCE for every app, public or confidential.
    • EHR launch: the EHR passes context from an active session. Standalone launch: the app requests context during authorization.
    • Scopes are resource-prefixed (patient/, user/, system/) with launch context and identity scopes layered on top.

    A worked example

    A .ASTHMAFOLLOWUP SmartPhrase produces a standard note. A SMART app goes further: it pulls FeNO, eosinophils, and exacerbation history, applies guideline-driven biologic-eligibility logic, and proposes a note recommendation for the clinician to confirm. The app supports the decision; the clinician makes it.

    For registries, quality measurement, and population health, SMART Backend Services run server-to-server with no user in the loop.

    Layer 3 · Decision support

    Decision support that meets the clinician in the workflow

    CDS Hooks (v2.0.1) is a vendor-neutral way to surface decision support as Cards at defined moments in the workflow, built on FHIR R4 and maintained by HL7.

    A Card carries a summary, an indicator (info, warning, or critical), a source, and optional suggestions and links. A discovery endpoint at {baseURL}/cds-services advertises the available services and their prefetch templates. A Card can launch a SMART app, so the two standards can be used together.

    Illustrative use cases: suggest a biologic referral, prompt CRSwNP screening, flag an EoE evaluation, or surface an asthma action plan, each at a defined point in the encounter.

    Standard hooks

    • patient-view
    • order-select
    • order-sign
    • order-dispatch
    • medication-refill
    • appointment-book
    • encounter-start
    • encounter-discharge

    FHIR R4 and US Core

    SMART on FHIR and CDS Hooks both build on HL7 FHIR R4. US Core is the US baseline profile set (Patient, Observation, Condition, MedicationRequest, Encounter, and more) that underpins ONC-certified API access. It advances annually with USCDI, so build to the current published release rather than a frozen version.

    Layer 4 · Implementation science

    Why tools succeed or fail

    We use the frameworks below to design for adoption and to measure whether it happened.

    CFIR (2022)

    Consolidated Framework for Implementation Research. Five domains: Innovation, Outer Setting, Inner Setting, Individuals, and Implementation Process. Explains why a tool is or is not adopted in a given setting.

    RE-AIM

    Reach, Effectiveness, Adoption, Implementation, Maintenance. Evaluates whether a tool reaches its population and sustains its effect.

    Proctor outcomes

    Eight measurable implementation outcomes: acceptability, adoption, appropriateness, feasibility, fidelity, implementation cost, penetration, and sustainability.

    Audit and feedback

    An evidence-based behavior-change mechanism for sustaining guideline adherence over time.

    The tool ladder and the pipeline

    Choose the lowest rung that solves the problem. Then follow a pipeline that continues past deployment to measurement and publication.

    The tool ladder

    1. 1SmartPhrases / dot phrases: documentation standardization
    2. 2SmartText / SmartLinks: pull structured data into the note
    3. 3SmartSets / Order Sets: standardize care pathways
    4. 4BPAs / alerts: point-of-care reminders
    5. 5CDS Hooks: vendor-neutral decision-support cards
    6. 6SMART on FHIR apps: embedded tools with UI, analytics, registries
    7. 7Backend FHIR services: registries, quality measurement, population health

    The pipeline

    1. Clinical gap and needs assessment
    2. Workflow mapping across physicians, nurses, APPs, pharmacists, and care coordinators
    3. EHR build-option selection
    4. Vendor and site engagement
    5. Prototype design
    6. Clinical governance review
    7. Privacy and security review
    8. Pilot
    9. Adoption measurement
    10. Clinical-impact evaluation
    11. Publication and dissemination package

    Where these disciplines connect

    Education, EHR build, and app development are usually run as separate tracks. We cover all three and finish with measurement and publication. The standards are public; we connect them and measure the result.

    The playbook, as a PDF

    A version of this reference with citations, formatted for print and sharing. We email a secure link.

    EHR Workflow + SMART on FHIR Playbook

    Eleven pages: the four build layers, the native EHR toolkit, SMART on FHIR, CDS Hooks, FHIR R4 and US Core, implementation science, and the HMHCP delivery pipeline.

    Contact details

    Provide contact details only. Do not enter patient information or health details. We use your details to handle this request. Privacy policy.

    Talk with us about your program

    Tell us what you are planning. We can discuss the study design or the implementation plan.