> For clean Markdown of any page, append .md to the page URL. > For a complete documentation index, see https://docs.emtelligent.com/llms.txt. > For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.emtelligent.com/_mcp/server. # Result format `GET /v1/jobs/{id}/result` returns one of two shapes, told apart by `type`. The [API reference](/formfill/api) has the full JSON Schema of each; this page explains what the fields mean. ## Questionnaire result ```json { "id": "job_8f2c1d0e9b7a6c5d4e3f2a1b", "type": "questionnaire", "questionnaire_response": { "resourceType": "QuestionnaireResponse", "...": "..." }, "stats": { "...": "..." }, "pipeline_version": "2026.10.12-1" } ``` | Field | Meaning | | ------------------------ | ----------------------------------------------------------------------- | | `id` | The job id | | `type` | `questionnaire` | | `questionnaire_response` | A FHIR R4 QuestionnaireResponse answering your Questionnaire; see below | | `stats` | Counts for the fill; see [Stats](#stats) | | `pipeline_version` | The API release that produced the result | ### The QuestionnaireResponse | Element | Content | | --------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | | `resourceType` | `QuestionnaireResponse` | | `status` | `completed`, or `in-progress` when a required item is unanswered (`stats.required_missing` lists them) | | `authored` | When the response was produced (UTC) | | `questionnaire` | Your Questionnaire's `url`, when it has one | | `item` | One node per answered item, nested as in your Questionnaire. Each node has the item's `linkId` and `text`, its `answer` list, and its answered children in `item`. Items without an answer, and groups with no answered item inside, are left out | A repeating item holds one `answer` per value. A repeating group occurs once per entry: several sibling nodes that share the group's `linkId`. Each answer's `value[x]` follows the item's `type`: | Item type | Answer | | ------------------------------------------------------------------------------------ | -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | | `choice`, `coding`, `open-choice`, or any item with `answerOption`s | The chosen option as given in your Questionnaire: `valueCoding` with its `system`, `code` and `display`, or the option's own `valueString`, `valueInteger`, `valueDate` or `valueTime` | | Open items (`open-choice`, `answerConstraint` `optionsOrString`) when no option fits | `valueString`: the value as the document states it | | `answerConstraint` `optionsOrType` when no option fits | A value of the item's type; for a coded item, a `valueCoding` with a `display` only, never an invented `code` | | `quantity` | `valueQuantity`: `value`, plus `unit`, `system` and `code` from the item's `questionnaire-unit` extension | | `string`, `text` | `valueString` | | `decimal`, `integer` | `valueDecimal`, `valueInteger` | | `boolean` | `valueBoolean` | | `date`, `dateTime`, `time` | `valueDate`, `valueDateTime`, `valueTime` | | `url` | `valueUri` | ### Extensions Answers carry emtelligent extensions alongside their value: ```json { "valueCoding": { "system": "http://loinc.org", "code": "LA6560-2", "display": "Oriented" }, "extension": [ { "url": "https://emtelligent.com/fhir/StructureDefinition/answer-confidence", "valueDecimal": 0.9 }, { "url": "https://emtelligent.com/fhir/StructureDefinition/answer-evidence", "valueString": "A/O x3" } ] } ``` | Extension `url` | Value | Present on | Meaning | | -------------------------------------------------------------------- | -------------------------------------- | --------------------------------------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ | | `https://emtelligent.com/fhir/StructureDefinition/answer-confidence` | `valueDecimal`, 0 to 1 | Answers taken from the document, and calculated answers | The service's estimate of how well the document supports the answer; see [Confidence](/formfill/guides/results-and-evidence#confidence). On a calculated answer, the lowest confidence of the answers it was calculated from | | `https://emtelligent.com/fhir/StructureDefinition/answer-evidence` | `valueString` | Every answer taken from the document | The verbatim quote the answer was taken from, without offsets; see [Evidence](/formfill/guides/results-and-evidence#evidence) | | `https://emtelligent.com/fhir/StructureDefinition/answer-derivation` | `valueCode`: `calculated` or `implied` | Answers the service derived, instead of `answer-evidence` | `calculated`: a score your form defines as a sum of option weights, calculated from the answers chosen. `implied`: a yes/no control implied by an answered item it gates. See [Answers the service derives](/formfill/guides/results-and-evidence#answers-the-service-derives) | Each extension appears at most once on an answer, and only on `QuestionnaireResponse.item.answer`. Software that does not know them can ignore them: the answers are complete without them. ## Questions result ```json { "id": "job_8f2c1d0e9b7a6c5d4e3f2a1b", "type": "questions", "answers": [ { "...": "..." } ], "stats": { "...": "..." }, "pipeline_version": "2026.10.12-1" } ``` `answers` holds one object per question, in the order you asked them: | Field | Meaning | | ------------ | ----------------------------------------------------------------------------------------------------------------------------------------------------------- | | `id` | Your question's `id`; where you gave none, `q1`, `q2`, … by its position | | `question` | The question text | | `type` | `text`, `number`, `boolean`, `date` or `choice` | | `status` | `answered`, `not_found` or `unmatched_option`; see [Questions](/formfill/guides/questions#the-result) | | `answer` | The typed answer; `null` when not found. For a `multiple` question, a list of `{answer, confidence, evidence}` objects, one per value (`[]` when not found) | | `confidence` | 0 to 1, when answered | | `evidence` | `{quote, spans}` when answered: the verbatim quote and where it occurs in `document.text`; see [Evidence](/formfill/guides/results-and-evidence#evidence) | ## Stats Both result types carry the same `stats`: | Field | Meaning | | ------------------ | ----------------------------------------------------------------------------------------------------------------------- | | `answerable` | Items or questions that can hold an answer (not groups or display items) | | `answered` | Items or questions that received at least one answer | | `answer_values` | Total answers; a repeating item or `multiple` question may hold several | | `unmatched_option` | Coded answers dropped because they matched none of your options | | `suppressed` | Answers dropped because your form's `enableWhen` logic disables the item | | `evidence_located` | Answers whose evidence quote was found in the document | | `required_missing` | linkIds of required items your form logic enables that the document does not answer. Always `[]` for a questions result | | `ungrounded` | Free-text answers dropped because the document does not support them |