Doses/treatments delivered ÷ planning denominator, from tallies. Disaggregable by sex, age band, delivery strategy and geography.

Implementation-unit coverage: settlements/areas treated ÷ total targeted (coverage-unit = implementation-units). The ESPEN supervision-form 'villages treated / total' figure, with non-treatment reasons as a stratifier.

PC-NTD treatment coverage: persons treated ÷ at-risk/eligible population, disaggregated by sex, age band and treatment disposition (treated vs not-treated reason). The Measure behind the ESPEN treatment-form tally.

Independently-measured coverage (cluster survey / LQAS / RCM). A separately-sourced measure of the same quantity as administrative coverage; the two routinely diverge and must never be merged.

Zero-dose children reached in the round ÷ children reached, disaggregated by prior-dose status (dose-history). Turns the polio SIA tally's never/previously/no-recall split into a first-class measure of how many zero-dose children a campaign reaches — the equity signal behind zero-dose reduction. Placeholder CQL pending executable logic (forms-v1 / jul3-form-analysis §Aggregate #1).

Structured pre-campaign readiness/preparedness validation at operational level (UNICEF Preparedness Validation form), grouped: microplan, cold chain & logistics, social mobilization, trainings. Representative items, not the full instrument; countries extend (forms-v1).

Structured supervision checklist for community-directed MDA (ESPEN Forms 5 & 6), grouped: supplies, CDD observation, stock & wastage, social mobilization.

Administrative coverage: doses/treatments delivered ÷ planning denominator, computed from tallies. Carries denominator provenance and is permanently marked source=administrative.

An adverse event following a campaign intervention — vaccine (AEFI) or MDA drug (pharmacovigilance) alike. The suspected product/event is referenced via suspectEntity; causality uses the WHO/CIOMS classification; seriousness/severity carry the minor-vs-serious distinction the field forms collect. Carries the required campaign-vs-routine record origin. Intervention-neutral so it serves both the immunization and MDA arms (working doc §17.2 C1).

A specific campaign execution — the keystone resource. Begins life as a microplan (intent=plan) and evolves into the execution record as Tasks complete and coverage accumulates. Rounds are sibling ICRCampaigns under an umbrella campaign via partOf (working doc 7.2, 7.2,

A discrete work type within a campaign — 'administer albendazole to children 5–14', 'distribute ITNs to households' — instantiated as ICRCampaignTask resources (working doc §7.3).

The reusable, version-controlled template for a campaign type — what a measles SIA is (products, age bands, activity sequence, coverage goals), instantiated by every execution in every country (working doc §7.1).

The assignable, trackable operational unit of work — one Task per site-session (Type A, focus = the site Location) or per household (Type B, focus = the household Group). Tasks may be pre-planned from the microplan or field-registered on discovery (the required task-origin code records which). Whether Tasks are assigned at village or household level is a configuration choice (working doc §7.4).

The campaign delivery team — the vaccinators/CDDs who do the work and the supervisor who oversees them and typically files the report. Referenced from ICRCampaign.careTeam (the roster) and Task.owner/Task.performer (the team that worked a Task); the supervisor surfaces again as the MeasureReport.reporter on rolled-up coverage and often owns the supervisory-area Location via the oversees-area extension. With workload-target it carries the microplan workload assigned to the team (working doc 5.5, 5.5,

Records the permission governing collection, storage, and cross-border sharing of a registered individual's (or a household's) campaign data. A v1 starting point: ICR holds named people (ICRPatient, §6.4), so a privacy/sharing permission and its scope are first-class. The grantor is typically the head of household or caregiver; the subject is the ICRPatient the data is about.

The actual Group of people a campaign Task acts on — a household (Type B house-to-house), a community (Type C MDA), or a school cohort (school-based delivery), distinguished by the required group-kind code. The validated Group + Location pattern, generalized: the Group is who, the Location (group-location extension) is where it lives or is based — the dwelling for a household, the settlement or community point for a community, the school for a school cohort — with the Location carrying the Overture GERS ID for stable cross-campaign identity. Type A's delivery unit is a site, which is a Location, not a Group (working doc 3.2, 3.2,

The accountable facility entity — the mCSD-style pairing partner of a facility ICRLocation. A health facility is two things: an Organization (the conceptual/legal entity that owns registry codes, classification, and accountability) and a Location (the physical place where care happens), linked Location.managingOrganization → Organization. Organization.type is the source of truth for the national facility classification (ICRFacilityTypeVS) and ownership (ICROwnershipVS) alongside the generic 'prov' provider coding; Location.type carries the generic 'facility' functional code and MAY additionally carry copies of the classification codings — the mCSD-sanctioned duplication for consumers that only query Locations (geospatial exports, map layers). On any disagreement the Organization is authoritative. Organization.partOf carries the administrative reporting hierarchy (facility → LGA/district health office → state agency), which is deliberately distinct from — and need not mirror — the geographic hierarchy on Location.partOf: a facility can report to one authority while sitting in territory that authority does not govern. Per the georegistry rule this profile still holds only identify/classify/contact data (working doc 5.3, 5.3,

A vaccination delivery event: CVX-coded, with lot accountability and the required campaign-vs-routine record origin.

The most-customized ICR resource — the ICR's georegistry layer, covering the WHO IDHC administrative boundary, health facility and school master lists: nested administrative hierarchy (6+ levels in campaign countries), operational geography linkable-but-distinct from admin units, GeoJSON boundaries, and multi-system geospatial identity — Overture Maps GERS IDs (building / place / division) as the preferred cross-campaign join key, with P-codes and national codes as coequal aliases. Per the IDHC georegistry rule, this layer holds only identify/classify/locate/contact data; programmatic data references it but never lives in it (working doc 7.7, 7.7,

An MDA treatment event: ATC-coded preventive chemotherapy with directly-observed consumption, dosage derived from a dose-pole height Observation, and the required record origin.

A beneficiary — an individual registered in a campaign, a household or community member, and the subject of every person-level delivery event. This is the WHO IDHC 'beneficiary list' master registry entry; the resource is FHIR Patient by necessity, but human-facing text (forms, reports, training material) should say 'beneficiary' or 'individual', never 'patient'. Person registration is a mainline capture mode in community-and-household campaigns, not an exception. Base R4 Patient aligned to WHO IMMZ.Patient (required gender/birthDate; MS name/phone/address), with a sliced cross-campaign identifier (national ID preferred, registry-assigned ID as fallback) so the same individual is rejoinable across rounds. A beneficiary need not belong to any Group: they may be a household member, a community member, or simply the subject of a standalone event. The caregiver is a RelatedPerson (WHO IMMZ.Caregiver), not an ICRPatient (working doc §6.4).

A STRUCTURED supervision-visit / QA record (ESPEN Forms 5 health-facility and 6 CDD-observation): a QuestionnaireResponse against the icr-mda-supervision-checklist Questionnaire, so each checklist answer (supplies present, DOC observed, height-chart used correctly, stock concordance, population informed, channels used…) links to a defined, coded question. Subject is the supervised area/community; author is the supervising ICRCareTeam. Replaces the v0.19 lightweight text-component Observation (working doc §17.3).

A commodity distribution event — ITNs, IRS consumables, vitamin A capsules — GS1 GTIN-coded where applicable, with the required record origin.

Independently-measured coverage — post-campaign cluster survey, LQAS, or RCM — with method, sample design, and date. A separately-sourced, first-class measure of the same quantity as administrative coverage: the two routinely diverge (Cuamba, Mozambique: ~99% admin vs ~76% survey) and must never be collapsed.

A target-population denominator: a conceptual cohort (actual=false) with a count, eligibility characteristics, and — critically — source and date provenance. Multiple competing estimates per geography are retained; exactly one is flagged as the planning denominator (working doc 7.6, 7.6,

Round number of this campaign execution. Each round is its own CarePlan instantiating the same protocol, linked to an umbrella campaign via CarePlan.partOf (working doc §6.3).

The measurement lineage of this coverage report: administrative vs survey/LQAS/RCM. The two lineages routinely diverge (Cuamba: ~99% admin vs ~76% survey) and must never be merged (working doc §4.1).

What the coverage figure counts: people, or implementation units (villages/areas = geographic coverage). Absent ⇒ people (espen-v3 / §17.2 B1).

How this activity/site/task reaches its target population. First-class and coded because a single campaign routinely mixes strategies, and the strategy governs which data elements exist (working doc §3).

The method/source behind this population estimate (census, microcensus, WorldPop, GRID3…). Reuse is only safe with provenance (working doc 2.3, 2.3,

Whether the denominator is the TOTAL population or the AT-RISK/eligible subset — programme vs epidemiological coverage in NTD MDA. On a coverage MeasureReport it qualifies the figure; on an ICRTargetPopulation it labels the estimate (espen-v3 / §17.2 B1).

Whether the community drug distributor physically observed the individual swallow the medication (MDA DOC protocol).

The measured dose-pole height band that determined the tablet count for a PC-NTD treatment — makes the height-band → dose logic machine-readable rather than buried in dosage.text. Bands are drug-specific; coded extensibly (espen-v3 minor-issue).

Number of eligible persons absent at the visit(s) — feeds same-day mop-up lists. Program-neutral counterpart of eligible-present.

Number of eligible persons — per the campaign protocol's target definition — present at the visit(s). Program-neutral by design: children under 5 for polio, household members for ITN registration, the eligible age band for MDA or vitamin A.

Free-text or coded confidence qualifier on a population estimate.

When this population estimate was made. Denominators decay fast (1–3 years); a stale denominator silently reused produces confident, wrong coverage.

Why a present, age-eligible person was clinically excluded from the intervention this round (under dose-pole height/age, pregnant, breastfeeding, acutely ill). Distinct from missed (not reached) and refusal (declined): the person was there and age-eligible but contraindicated. MDA treatment forms tally these per drug (espen.md rec 2 / §17.4).

Whether the child was finger-marked — the in-field 'already covered' flag of house-to-house campaigns.

The physical Location of a delivery-unit Group — the validated Ona pattern: Group (who) + Location (where). For a household the Location is the dwelling (carrying the GERS building ID); for a community it is the settlement or community point; for a school cohort it is the school. This is the group's RESIDENCE/BASE, not its service point: where service actually happened is Task.location and the delivery event's own location — a household served at a village distribution center keeps its dwelling here unchanged (working doc 7.5, 7.5,

Number of houses visited — aggregate output of a house-to-house task.

True when this estimate is the one flagged as the campaign's planning denominator among competing estimates for the same geography.

Boundary geometry for a Location — district polygons, settlement areas, catchment zones — as a GeoJSON attachment. R4 extension mirroring the R5 standard boundary extension; alignment path is working doc §10 question 6.

Why eligible person(s) were missed at this visit. House-to-house campaigns produce this natively (working doc §3.1).

Links an operational-geography Location (supervisory area, operational area) to the administrative unit(s) it overlays. Operational ≠ administrative geography: partOf can only express one hierarchy, so this extension is what makes operational areas linkable-but-distinct rather than just distinct (working doc §9, identity principle 3).

The supervisory/operational-area Location(s) a CareTeam's supervisor covers — ties the team to operational geography (working doc 5.5, 5.5,

The target-population Group flagged as this campaign's planning denominator. Multiple competing estimates may exist per geography; exactly one is the planning denominator (working doc §4.2).

The person's prior-dose status for this antigen at the time of the campaign contact: zero-dose (never received before), previously-received, or no-recall. The per-event carrier of the polio SIA tally's never/previously/no-recall split; aggregates to the dose-history coverage stratifier and feeds zero-dose reach / dropout measures. Distinct from Immunization.protocolApplied.doseNumber, which counts this series' doses, not prior status of the antigen (forms-v1 / jul3-form-analysis §Aggregate #1).

Data lineage of this record: the real-time operational stream or the post-campaign reconciled stream. One structure serves both; consumers filter by lineage — dashboards read realtime, JAP exports read reconciled (working doc §4.3). Default semantics: ABSENT MEANS REALTIME — live-stream records may omit the flag, but reconciled records must carry it, and coverage MeasureReports always carry it (1..1 on both coverage profiles).

Whether this delivery event was recorded during a campaign or a routine facility visit. REQUIRED on all ICR delivery-event profiles — without it, SIA doses contaminate routine coverage analytics (working doc §4.4). Also valid on AdverseEvent so a campaign-context AEFI/side-effect stays separable from routine pharmacovigilance (v0.19.0).

Why the household/caregiver refused the intervention (WHO IDHC: 'intervention refusal') — drives social mobilization and mop-up targeting.

Sample design / method detail of an independently-measured coverage estimate — e.g. 'WHO 30×10 cluster survey, district-representative', LQAS lot definition, RCM site-selection note. Survey coverage without its design is uninterpretable (working doc §4.1).

Why an adverse event is serious — the WHO/CIOMS criterion/criteria behind AdverseEvent.seriousness = serious (death, life-threatening, hospitalization, disability, congenital anomaly, other medically important) (espen-v4 / §17.2 C1).

The settlement or special-population classification of a place (urban / rural / slum / refugee-IDP / nomad-pastoralist / security-compromised / hard-to-reach / cross-border / immigrant) — the recurring 'type of settlement' axis on the polio SIA monitoring/tally forms. A vulnerability/equity attribute of a Location that drives hard-to-reach-area targeting and equity disaggregation (forms-v1 / jul3-form-analysis §Aggregate #5).

Demand-generation for a campaign/round: whether the population was informed beforehand and the channels used (ESPEN supervision Form 5). The demand axis of §17.3 (espen-v4).

Vial/commodity accountability and wastage on a supply event — received / used / remaining / not-usable (expired/damaged) / returned, plus physical-vs-theoretical concordance and (vaccines) the VVM stage. Reusable for vaccines, drugs and ITNs; the ESPEN supervision Form 5 stock block (espen-v4 / §17.2 C2).

The geography (admin unit or operational area) this campaign targets.

Whether this Task was generated in advance from the microplan or registered in the field on discovery. REQUIRED on ICRCampaignTask — field-registered counts per area measure how incomplete the microplan's enumeration was, feeding the next round's denominators (working doc §10 q1).

The microplan workload assigned to a CareTeam — the area(s) it covers and the expected volume (population / households / days). With oversees-area, makes the CareTeam the typed team-area-workload unit of the microplan (the ICRCampaign with intent=plan is the microplan itself) (espen-v4 / §17.3).

WHO/CIOMS causality categories. Binding: extensible on ICRAdverseEvent causality assessment.

Serious vs non-serious. Binding: extensible on ICRAdverseEvent.seriousness — uses the HL7 adverse-event-seriousness code system.

Campaign types by delivery model. Binding: required on ICRCampaignProtocol.type and ICRCampaign.category.

Social-mobilization channels. Binding: extensible on the social-mobilization extension's channel.

All coverage measurement lineages.

Standard coverage disaggregation axes. Binding: extensible on the Measure/MeasureReport stratifier code (espen-v3 / §17.2 B1).

People vs implementation-units (geographic coverage). Binding: required on the coverage-unit extension.

Real-time vs reconciled lineage. Binding: required on the realtime-vs-reconciled extension.

Delivery strategies. Binding: required on the delivery-strategy extension.

Sources of population denominators. Binding: extensible.

Total-population vs at-risk/eligible denominator. Binding: required on the denominator-type extension.

Prior-dose status (zero-dose / previously-received / no-recall). Binding: required on the prior-dose-status extension; also the value space of the dose-history coverage stratifier (forms-v1).

Reasons a present, age-eligible person was clinically excluded from treatment (under-height/age, pregnant, breastfeeding, acutely ill). Binding: extensible — countries may add local codes.

Facility ownership / managing authority, on the facility Organization's type. Binding: extensible — countries add local ownership categories.

National facility-classification tiers, on the facility Organization's type. Binding: extensible — countries add their national facility kinds.

Delivery-unit Group kinds (household / community). Binding: required on ICRDeliveryUnit.code.

Independently-measured coverage lineages only (survey / LQAS / RCM) — the ICRSurveyCoverage binding.

Campaign-relevant location types, including operational geography. Binding: extensible on ICRLocation.type — countries may add local types.

WHO ATC-coded preventive-chemotherapy medications. Binding: extensible — local formulary codes map back via ConceptMap. Includes the full ATC system; typical PC-NTD codes are albendazole P02CA03, ivermectin P02CA01, praziquantel P02BA01, azithromycin J01FA10, diethylcarbamazine P02CB02. Restricting to a PC-NTD subtree is deferred until country formularies are reviewed.

MDA medicine packages, single and combined (espen-forms).

Reasons an eligible person/household was missed. Binding: extensible — countries may add local codes.

Diseases covered by an MDA campaign (espen-forms).

Campaign vs routine record origin. Binding: required on the record-origin extension.

Reasons for intervention refusal (WHO IDHC term; ESPEN forms say noncompliance). Binding: extensible — countries may add local codes.

Outcome of a follow-up revisit (already-vaccinated / vaccinated-on-revisit / still-missing). Binding: extensible on the revisit-outcome extension (forms-v1).

WHO/CIOMS serious-event criteria. Binding: extensible on the serious-criteria extension.

Settlement / special-population classification of a Location. Binding: extensible on the settlement-type extension — countries add local codes (forms-v1).

Coded products distributed via ICRSupplyDelivery. Binding: extensible. Drug commodities (MDA receipts/distribution) use WHO ATC; physical commodities (ITNs, IRS consumables, vitamin A) use GS1 GTIN where a local binding exists, otherwise text. Added v0.18.0 (espen.md rec 3) so a drug receipt → administration → reconciliation share one ATC code.

Pre-planned vs field-registered task origin. Binding: required on the task-origin extension.

Campaign CareTeam member roles. Binding: extensible on ICRCareTeam.participant.role.

WHO/CIOMS causality classification of an adverse event following an intervention — intervention-neutral: covers AEFI (immunization) and MDA drug pharmacovigilance alike. Top-level WHO categories; immunization implementations may use WHO IMMZ.AdverseEvent's finer A1–A4 subtypes (espen-v3 / §17.2 C1).

The type of public health campaign, grouped by delivery model rather than disease (working doc §3).

Social-mobilization / demand-generation channels used to inform the population ahead of and during a campaign (ESPEN supervision Form 5). Bound extensible (espen-v4 / §17.3).

The measurement lineage of a coverage figure. Administrative and independently-measured coverage are separately-sourced, first-class measures of the same conceptual quantity and must never be collapsed (working doc §4.1).

The standard disaggregation axes a coverage MeasureReport stratifies by (and a Measure declares). Formalises the v0.18.0 stratified-tally pattern into a named contract (espen-v3 / §17.2 B1).

What is being counted in a coverage figure: people, or implementation units. Implementation-unit coverage is geographic coverage — villages/areas treated ÷ total (ESPEN supervision Form 5; espen-v3 / §17.2 B1).

Whether a record belongs to the real-time operational stream or the post-campaign reconciled stream. One structure serves both; consumers filter by lineage (working doc §4.3).

How a campaign activity reaches its target population. A first-class, coded attribute of every campaign activity, site, and task — campaigns routinely mix strategies, and the available data elements change with the strategy (working doc §3).

The method/source behind a target-population estimate. Every estimate carries its source and date — the denominator is the dominant source of error in campaign analytics (working doc §4.2).

Whether a coverage figure (or a target-population estimate) is measured against the TOTAL population or the AT-RISK/eligible subset — the difference between programme coverage and epidemiological coverage in NTD MDA (espen-v3 / §17.2 B1).

A person's prior-dose status for the campaign antigen, at the time of a campaign contact — the polio SIA tally's core split (never received / previously received / no recall) and the zero-dose vs not-zero-dose axis. Carried per event by the prior-dose-status extension and aggregated as the dose-history coverage stratifier; feeds zero-dose reach and the proposed dropout/fully-immunized measures (forms-v1 / jul3-form-analysis §Aggregate #1).

Why a PRESENT, age-eligible person was not treated because they are clinically ineligible/contraindicated for the intervention this round — as opposed to being MISSED (not reached, ICRMissedReasonCS) or DECLINING (ICRNoncomplianceReasonCS). The MDA case (ESPEN treatment Form 3): under the dose-pole minimum, pregnant, breastfeeding, acutely ill. Added v0.18.0 (espen.md rec 2 / §17.4 NTD specifics).

Ownership / managing-authority classification for the facility Organization — a second Organization.type axis (base FHIR Organization has no ownership element; carrying it as a type coding is the mCSD/OpenHIE convention). The governmental tier detail (local/state/federal) travels as display/text or a country localization.

National facility-classification tiers for the mCSD-style facility Organization (working doc §5.3 facility pairing). The authoritative classification axis lives on Organization.type — not Location.type — because facility level is an administrative designation (funding, staffing norms, reporting), which is an Organization concern. Codes are the broad tiers shared across national MFLs; the country-specific kind (e.g. Nigeria NHFR 'Primary Health Center' vs 'Health Post') travels as the coding display/text or a country localization system.

Characteristic codes for ICR Group profiles — the geographic-scope characteristic that links a target-population estimate to its Location, making estimates computably joinable to the location hierarchy (working doc §7.6), plus the age-band characteristic used to scope age-specific denominator Groups (espen-forms).

The kind of delivery-unit Group a campaign Task acts on. Households (Type B house-to-house) and communities (Type C MDA) share the same Group + Location pattern; this code distinguishes them (working doc 3.2, 3.2,

Campaign-relevant location types, including operational geography (supervisory and operational areas) as linkable-but-distinct from the administrative hierarchy (working doc §9).

The medicine package distributed in an MDA round — single drugs and the standard co-administration combinations (ESPEN medicine list, espen-forms).

Why an eligible person or household was not reached during a campaign visit.

The preventive-chemotherapy NTDs an MDA campaign addresses — the disease-scope axis of the ESPEN MDA instruments (espen-forms).

Project/programme provenance tags applied as meta.tag on example instances, so the example gallery can be filtered by the source project or instrument set (espen-forms).

Whether a delivery event originated in a campaign or a routine facility visit. Required on every ICR delivery event so SIA doses never contaminate routine coverage analytics, and routine history observed during campaigns stays analyzable (working doc §4.4).

Why a household or caregiver refused the intervention (WHO IDHC: 'intervention refusal'; ESPEN forms say noncompliance) — drives social-mobilization targeting and mop-up planning.

The outcome of a follow-up visit to a household/person previously missed — the 'outcome of the revisit' block of the missed-children recording forms. Carried on the follow-up Task via the revisit-outcome extension (forms-v1 / jul3-form-analysis §Aggregate #4).

WHO/CIOMS criteria that make an adverse event SERIOUS — the reason behind AdverseEvent.seriousness = serious. Intervention-neutral (AEFI and MDA pharmacovigilance) (espen-v4 / §17.2 C1).

The settlement or special-population classification of a place — the recurring 'type of settlement' axis on the polio SIA monitoring/tally forms (urban / rural / slum / refugee-IDP / nomad-pastoralist / security-compromised / hard-to-reach / cross-border / immigrant). A vulnerability/equity attribute of a Location, carried by the settlement-type extension; drives hard-to-reach-area (HTRA) targeting and equity disaggregation (forms-v1 / jul3-form-analysis §Aggregate #5). Bound extensible — countries add local codes.

Whether a campaign Task was generated in advance from the microplan or registered in the field on discovery. Field-registered counts per area measure how incomplete the microplan's enumeration was — input to the next round's denominators (working doc §10 q1).

Role of a member within a campaign CareTeam — the front-line delivery and supervision roles (working doc §5.5). Bound extensible: countries add local roles. Supervisor level (national/regional/district/partner/health-facility, ESPEN Forms 5/6) is carried by managingOrganization or the overseen area, not multiplied into roles.

Maps ICRAdverseEventCausalityCS (WHO/CIOMS A/B/C/D) to the WHO SMART Immunizations AEFI causality categories. Provisional — target codes to be confirmed against the published IMMZ IG (§18.3).

ESPEN MDA demo Form 1 (village/location registration and census): admin cascade, population by age band, GPS. Template-based extraction: one ICRLocation for the village and ICRTargetPopulation Groups for the total, eligible, and age-band denominators (espen-forms).

ESPEN MDA demo Form 2 (medicine receipt at health facility): disease and medicine-package scope, per-medicine received totals. Template-based extraction: one ICRSupplyDelivery per answered medicine total (espen-forms).

ESPEN MDA demo Form 3 (the treatment tally): village census plus per-drug treated counts by sex and age band with reasons-not-treated. Template-based extraction: one ICRDeliveryUnit community Group for the village, one Group-subject ICRMedicationAdministration per answered drug block (the treatment event — tablets into people are treatment, not a supply transfer), and one ICRAdministrativeCoverage MeasureReport per answered drug block — measure icr-mda-treatment-coverage, stratified by sex, age-band, and disposition, the same cube as example-mda-treatment-tally (espen-forms, remapped espen-remap).

ESPEN MDA demo Form 4 (medicine use and case management): per-drug distributed totals, side-effect counts, other-NTD case counts. No template-based extraction (espen-remap): a distributed total is not a custody transfer, so it does not mint a SupplyDelivery — the treatment events are Form 3's Group-subject ICRMedicationAdministrations, and the distributed counts are folded into the stock-accountability extension on the Form 2 receipt ICRSupplyDelivery by the ingestion pipeline (a cross-form merge extraction cannot express). Side-effect and other-NTD counts remain on the QuestionnaireResponse: person-level ICRAdverseEvent records cannot be minted from aggregate counts.

ESPEN MDA demo Form 5 (supervision at health facility): geographic coverage of villages treated, per-drug stock triplets (remaining/expired/concordance), distributor training, social mobilization, supervision area, pharmacovigilance, and free-text challenges/solutions/recommendations. No extraction templates: per ICRSupervisionReport (working doc §4.6) the QuestionnaireResponse itself is the supervision record.

ESPEN MDA demo Form 6 (CDD observation supervision): direct observation of a community drug distributor's technique and attitude during treatment, MDA-supplies availability, CDD training coverage, and free-text challenges/solutions/recommendations. No extraction templates: per ICRSupervisionReport (working doc §4.6) the QuestionnaireResponse itself is the supervision record.