Use StartFHIRExportJob with an AWS SDK or CLI

To start a FHIR export job

The following start-fhir-export-job example shows how to start a FHIR export job using AWS HealthLake.

aws healthlake start-fhir-export-job \
    --output-data-config '{"S3Configuration": {"S3Uri":"s3://(Bucket Name)/(Prefix Name)/","KmsKeyId":"arn:aws:kms:us-east-1:012345678910:key/d330e7fc-b56c-4216-a250-f4c43ef46e83"}}' \
    --datastore-id (Data store ID) \
    --data-access-role-arn arn:aws:iam::(AWS Account ID):role/(Role Name)

Output:

{
    "DatastoreId": "(Data store ID)",
    "JobStatus": "SUBMITTED",
    "JobId": "9b9a51943afaedd0a8c0c26c49135a31"
}

    @classmethod
    def from_client(cls) -> "HealthLakeWrapper":
        """
        Creates a HealthLakeWrapper instance with a default AWS HealthLake client.

        :return: An instance of HealthLakeWrapper initialized with the default HealthLake client.
        """
        health_lake_client = boto3.client("healthlake")
        return cls(health_lake_client)


    def start_fhir_export_job(
        self,
        job_name: str,
        datastore_id: str,
        output_s3_uri: str,
        kms_key_id: str,
        data_access_role_arn: str,
    ) -> dict[str, str]:
        """
        Starts a HealthLake export job.
        :param job_name: The export job name.
        :param datastore_id: The data store ID.
        :param output_s3_uri: The output S3 URI.
        :param kms_key_id: The KMS key ID associated with the output S3 bucket.
        :param data_access_role_arn: The data access role ARN.
        :return: The export job.
        """
        try:
            response = self.health_lake_client.start_fhir_export_job(
                OutputDataConfig={
                    "S3Configuration": {"S3Uri": output_s3_uri, "KmsKeyId": kms_key_id}
                },
                DataAccessRoleArn=data_access_role_arn,
                DatastoreId=datastore_id,
                JobName=job_name,
            )

            return response
        except ClientError as err:
            logger.exception(
                "Couldn't start export job. Here's why %s",
                err.response["Error"]["Message"],
            )
            raise