DiGA Fast-Track

Healthcare Regulatory DiGAV BfArM Germany

Overview

The DiGA Fast-Track skill supports the preparation and validation of Fast-Track applications for listing Digital Health Applications (DiGA) in the BfArM directory under the Digitale-Gesundheitsanwendungen-Verordnung (DiGAV). It covers the full application lifecycle: structuring the submission with all mandatory sections, mapping claimed Positive Versorgungseffekte to clinical evidence, auditing technical documentation against ISiK interoperability standards, and validating data protection compliance. The skill provides two validation tools, application templates, evidence mapping frameworks, and reference documents covering DiGAV requirements and ISiK/FHIR specifications.

Getting Started

Upload your clinical study protocols, technical documentation, and CE conformity materials to the project workspace and activate the DiGA Fast-Track skill. The agent will guide you through the process: capture your DiGA context (name, therapeutic area, risk class, CE marking), select the appropriate workflow (new application, evidence review, or interoperability audit), populate the BfArM-required sections, and run automated validations. The result is a structured application draft with validation reports and actionable findings.

Example Conversation

Below is an example of a typical session preparing a DiGA Fast-Track application. The agent captures the DiGA context, structures the submission, validates the evidence against claimed effects, and audits the technical documentation.

You: We’re preparing a Fast-Track application for our mental health app “MindBalance” — a CE-marked Class I medical device for treating mild to moderate depression. We want a permanent listing in the BfArM directory. Our clinical study (RCT, N=240) is complete and we have the technical documentation ready.
Agent: I’ll help you prepare the Fast-Track application. Let me gather some details:
What Positive Versorgungseffekte do you claim? For example: improvement of health status (symptom reduction), therapy adherence, health literacy?
What validated instrument did you use to measure the primary endpoint (e.g. PHQ-9, BDI-II)?
Is the clinical study published or registered (DRKS/ClinicalTrials.gov)?
Which ISiK Stufe does your app support, and which FHIR resources do you exchange?
You: We claim improvement of health status (PHQ-9 reduction) and improvement of therapy adherence. The RCT used PHQ-9 as primary endpoint and MARS (Medication Adherence Rating Scale) as secondary. Study is registered at DRKS00012345. We support ISiK Stufe 2 with FHIR R4 — Patient, Observation, and Condition resources. Technical docs are in docs/technical_documentation.md and the study protocol in evidence/study_protocol.md.
Agent: I’ve loaded the DiGAV requirements and created the application from the template. Here’s what I’ve done:
Populated all 8 mandatory sections with your information
Mapped both Versorgungseffekte to DiGAV §14 categories
Included the ICD-10-GM codes for F32 (depressive episode) and F33 (recurrent depressive disorder)
Now running the evidence cross-check…
diga_evidence_orchestrator completed:
Claimed effects found: Verbesserung des Gesundheitszustands, Therapieadhaerenz — both detected
Study design: RCT with control group, validated instrument (PHQ-9), sample size N=240, ethics approval — all present
DiGA class I: Prospective comparative study confirmed — adequate for permanent listing
1 WARNING: No explicit mention of Minimally Important Difference (MID) for PHQ-9. Consider documenting that the observed reduction exceeds the established MID of 5 points.
Running the interoperability audit…
interoperability_standard_audit completed:
ISiK Stufe 2, FHIR R4 — correctly cited
FHIR profiles: Patient, Observation, Condition — referenced
Data export: FHIR Bundle export documented
DSGVO: Data processing location (EU), TOMs, consent mechanism — all present
1 WARNING: No LOINC codes referenced for Observation resources. Consider adding standardized codes for PHQ-9 scores.
1 INFO: No gematik/TI reference. Consider documenting whether ePA integration is planned.
The application draft is saved. Two minor findings to address before submission.

Document Excerpt

Below is a representative excerpt from a DiGA Fast-Track application prepared with the skill.

DiGA Fast-Track Antrag — MindBalance (Auszug)

Hersteller: HealthTech Solutions GmbH DiGA-Risikoklasse: I Aufnahme beantragt: Dauerhaft

3. Positive Versorgungseffekte

3.1 Uebersicht

Nr.	Versorgungseffekt	Kategorie	Evidenz
1	Reduktion depressiver Symptome (PHQ-9)	Medizinischer Nutzen — Verbesserung des Gesundheitszustands	RCT (DRKS00012345), N=240
2	Steigerung der Therapieadhaerenz (MARS)	pSVV — Therapieadhaerenz	RCT (sekundaerer Endpunkt)

3.2 Versorgungseffekt 1: Reduktion depressiver Symptome

Kategorie: Medizinischer Nutzen — Verbesserung des Gesundheitszustands

Beschreibung: MindBalance reduziert depressive Symptome durch ein evidenzbasiertes kognitiv-verhaltenstherapeutisches Programm (iCBT) ueber 12 Wochen, bestehend aus 8 Modulen mit psychoedukativen Inhalten, Verhaltensaktivierung und kognitiver Umstrukturierung.

Messung: PHQ-9 (Patient Health Questionnaire-9), validiert fuer die deutsche Population.

Ergebnis: Mittlere PHQ-9-Reduktion von 6.8 Punkten (95%-KI: 5.2–8.4) gegenueber Kontrollgruppe (Warteliste). p < 0.001. Effektgroesse d = 0.72 (mittlerer Effekt). Klinisch relevante Response-Rate (≥50% PHQ-9-Reduktion): 62% vs. 28% in der Kontrollgruppe.

6. Interoperabilitaet

6.1 ISiK-Konformitaet

Angabe	Wert
ISiK-Stufe	Stufe 2
FHIR-Version	R4 (4.0.1)
Unterstuetzte Profile	ISiKPatient, ISiKDiagnose, ISiKObservation

6.2 FHIR-Ressourcen

Ressource	Profil	Richtung	Beschreibung
Patient	ISiKPatient	Import/Export	Patientenstammdaten (Name, Geburtsdatum, Versichertennummer)
Condition	ISiKDiagnose	Export	Diagnose (ICD-10-GM: F32.1, F33.0)
Observation	ISiKObservation	Export	PHQ-9-Scores, Therapiefortschritt, Modulabschluss

6.3 Datenexport

Maschinenlesbar: FHIR Bundle (type: document), alle patientenbezogenen Observations, Conditions und Verlaufsdaten
Menschenlesbar: PDF-Report mit Symptomverlauf, Modulabschluss und Therapieempfehlung
Zugang: Einstellungen → Meine Daten → Daten exportieren

This excerpt is illustrative. Final content must reflect the specific DiGA’s data and current BfArM requirements.

Validations by Extension Tools

The skill includes two extension tools that automatically validate the evidence alignment and interoperability compliance of DiGA Fast-Track application documents.

diga_evidence_orchestrator

Analyzes a clinical study protocol or evidence document (Markdown) and cross-checks it against the claimed Positive Versorgungseffekte per DiGAV §14. Detects missing effect categories, study design gaps, and evidence adequacy for the DiGA risk class.

Check	Description
Versorgungseffekte categories	Scans for all DiGAV §14 categories: medical benefit (health status, symptom reduction, survival) and patient-relevant structural/process improvements (adherence, health literacy, patient sovereignty, care coordination, access to care, patient safety, caregiver support, reduced burden). Missing categories are flagged.
Study design elements	Checks for study type, primary endpoint, sample size, control group, validated outcome instruments (PROMs), study duration, statistical analysis, ethics approval, and DiGA risk class. Missing elements are reported as ERROR or WARNING.
DiGA class adequacy	Evaluates whether the evidence level matches the DiGA class: Class I requires a prospective comparative study (ideally RCT); Class II allows systematic data evaluation for provisional listing. Mismatches are flagged.
Placeholders	Detects unfilled placeholders such as `[DESCRIBE]`, `[VALUE]`, `[TODO]`. Reports total count.

diga_evidence_orchestrator path="evidence/study_protocol.md"

interoperability_standard_audit

Validates DiGA technical documentation against the required ISiK interoperability standards, FHIR profiles, data export capabilities, and data protection measures per DiGAV §4–§5.

Check	Description
ISiK standards	Verifies ISiK reference, ISiK Stufe (version), FHIR version, supported FHIR profiles (Patient, Observation, Condition, Medication), data export capability, interface documentation, and vesta reference. Missing citations are reported as ERROR or WARNING.
Terminology	Checks for required coding systems: ICD-10-GM, LOINC, SNOMED CT. Missing terminology references are flagged.
Data protection (DiGAV §4)	Verifies DSGVO reference, data processing location (EU/EWR), TOMs (Art. 32), consent mechanism, special category health data handling (Art. 9). Missing elements are reported.
gematik / TI	Checks for Telematikinfrastruktur, ePA, and KBV MIO references. Reported as INFO-level recommendations.
Placeholders	Detects unfilled placeholders. Reports total count.

interoperability_standard_audit path="docs/technical_documentation.md"

Interpreting Results

RESULT: PASS — Documentation appears aligned with requirements. Ready for final review.
RESULT: REVIEW NEEDED — Warnings indicate potential gaps that should be checked before submission.
RESULT: INCOMPLETE — Errors indicate missing required elements that will block BfArM review. Address all errors and re-validate.

These tools serve as structural quality assurance — they do not replace regulatory review. Final DiGAV compliance must be confirmed by the manufacturer and, where applicable, regulatory consultants.