global_trial_search
ChatGPTSearch for global clinical trials from clinicaltrials.gov, ChiCTR, ISRCTN, ANZCTR, and other sources. It searches title, description, conditions, interventions, eligibility, outcomes, etc. Args: query (str): The search term. Uses simple_query_string syntax with AND as default operator. offset (int, optional): Number of results to skip for pagination. Defaults to 0. size (int, optional): Maximum number of results to return. Defaults to 25. sort_by: Field to sort by. Defaults to '_score'. Acceptable values: "_score", "start_date". sort_order: Sort order. Defaults to 'desc'. Acceptable values: "asc", "desc". study_phase (optional): Filter study phase. Acceptable values: 'Early Phase 1', 'Phase 1', 'Phase 2', 'Phase 3', 'Phase 4', 'Not Applicable'. intervention_type (optional): Filter by intervention type. Acceptable values: 'Behavioral', 'Biological', 'Combination Product', 'Device', 'Diagnostic Test', 'Dietary Supplement', 'Drug', 'Genetic', 'Procedure', 'Radiation', 'Other'. overall_status (optional): Filter by overall status. Acceptable values: 'Not yet recruiting', 'Pending', 'Recruiting', 'Enrolling by invitation', 'Active, not recruiting', 'Suspended', 'Terminated', 'Completed', 'Temporary halt', 'Withdrawn', 'Unknown', 'Available', 'Ongoing', 'Approved for marketing', 'Stopped early'. facility_country (optional): Filter by facility country. Acceptable ISO 2-letter country codes, like 'US', 'GB', 'CA', 'MX', etc. start_date_from (datetime, optional): Filter by start date from. start_date_to (datetime, optional): Filter by start date to. Returns: dict: A dictionary containing: - total_hits (int): Total number of matching trials found - trials (list): List of clinical trials, each containing: - id: the unique trial identifier of a clinical trial - official_title - brief_title - overall_status - intervention_type - enrollment_count - trial_phase - interventions - start_date - official_url: the url to clinical trial official page. You can retrieve the webapge and find more details. - facility_country: ISO 2-letter country codes for facilities, like 'US', 'GB', 'CA', 'MX', etc.
usfda_company_search
ChatGPTSearch for applicant companies with FDA-approved devices by name. You can get the company id before searching for devices for a specific company. Args: query (str): The search term to match against company names and aliases. Uses simple_query_string syntax with AND as default operator. Try to search parent companies for better results, for example, "siemens" instead of "siemens healthineers". offset (int, optional): Number of results to skip for pagination. Defaults to 0. size (int, optional): Maximum number of results to return. Defaults to 10. Returns: dict: A dictionary containing: - total_hits (int): Total number of matching companies found - companies (list): List of company objects, each containing: - id (int): Unique company identifier - name (str): Company name - avg_review_days (int): Average review days for the company - device_cnt (int): Num of approved devices
usfda_device_search
ChatGPTSearch for FDA-approved devices. It supports pagination and returns a list of matching devices along with the total count. Args: query (str): The search term to match against device id, device name, indications, and device classification name. Uses simple_query_string syntax with AND as default operator. When you want to fetch a list of devices from a specific company/product code, you can use the empty string "" as the query and filter by application_company_id/product_code. offset (int, optional): Number of results to skip for pagination. Defaults to 0. size (int, optional): Maximum number of results to return. Defaults to 25. sort_by: Field to sort by. Defaults to '_score'. Acceptable values: "_score", "decision_date", "device_name", "fda_review_days". sort_order: Sort order. Defaults to 'desc'. Acceptable values: "asc", "desc". device_class (optional): Filter by device class. Acceptable values: "1", "2", "3", 'Unclassified', 'Not classified', 'HDE'. submission_type (optional): Filter by submission type. Acceptable values: "510(K)", "PMA", "HDE", "EUA", "De Novo". application_company_id (int, optional): Filter by application company id. product_code (str, optional): Filter by product code. Devices with the same product code are considered similar. decision_date_from (datetime, optional): Filter by decision date from. decision_date_to (datetime, optional): Filter by decision date to. clinical_trial_flag (bool, optional): Filter by if devices have associated clinical trials. combination_product_flag (bool, optional): Filter by if devices are combination products. Returns: dict: A dictionary containing: - total_hits (int): Total number of matching companies found - devices (list): List of device objects, each containing: - id (int): Unique device ID - device_name (str): Name of the device - device_class (str): Device class (e.g., "1", "2", "3", 'Unclassified', 'Not classified', 'HDE') - application_company_id (int): Unique ID for the applicant company - application_company_name (str): Name of the applicant company - date_receive (datetime): Date when the device was received by the FDA - decision_date (datetime): Date when the device was approved by the FDA - submission_type (str): Type of submission (e.g., "510(K)", "PMA", "HDE", "EUA", "De Novo") - medical_specialty_name (Optional[str]): The medical specialty associated with the device - product_code (Optional[str]): Product code assigned to the device. Devices with the same product code are considered similar. - clinical_trial_flag (bool): Indicates if the device has associated clinical trials - combination_product_flag (bool): Indicates if the device is a combination product - summary_pdf_link (list): List of URLs to summary PDF documents for the device. You can read the PDFs to get more detailed information about the device, including bench testing, clinical data, and more. - aggregations (dict): Aggregation results for device class, medical specialty, submission type, product code, decision date by year, and average review days.
usfda_guidance_search
ChatGPTSearch for FDA latest guidance. It searches guidance title, topic, docket number, and pdf content. Args: query (str): The search term. Uses simple_query_string syntax with AND as default operator. offset (int, optional): Number of results to skip for pagination. Defaults to 0. size (int, optional): Maximum number of results to return. Defaults to 25. sort_by: Field to sort by. Defaults to '_score'. Acceptable values: "_score", "issue_date". sort_order: Sort order. Defaults to 'desc'. Acceptable values: "asc", "desc". issue_date_from (datetime, optional): Filter by issue date from. issue_date_to (datetime, optional): Filter by issue date to. Returns: dict: A dictionary containing: - total_hits (int): Total number of matching guidances found - guidances (list): List of guidances, each containing: - title - docket_number (optional) - issue_date (optional) - fda_url: the url to fda guidance page. You can retrieve the webapge and find more details. - pdf_link (optional): the direct link to the guidance pdf document. You can read the PDF to get more detailed information about the guidance.
usfda_inspection_citation_search
ChatGPTSearch FDA inspection citations Args: query (str): The search term to match against FDA inspection citation description Uses simple_query_string syntax with AND as default operator. offset (int, optional): Number of results to skip for pagination. Defaults to 0. size (int, optional): Maximum number of results to return. Defaults to 10. sort_by: Field to sort by. Defaults to '_score'. Acceptable values: "_score", "end_date". sort_order: Sort order. Defaults to 'desc'. Acceptable values: "asc", "desc". fei_num (int, optional): Filter by facility's FEI number. Useful to check a specific facility's inspection records. cfr_num (str, optional): Filter by cited CFR number, like '820.100'. end_date_from (datetime, optional): Filter by inspection end date from. end_date_to (datetime, optional): Filter by inspection end date to. Returns: dict: A dictionary containing: - total_hits (int): Total number of matching inspections found - inspections (list): List of regulation object, each containing: - inspection_id (int): inspection id - fei_num (int): FEI number of the facility being inspected - end_date (datetime): end date of the inspection - program_area (str): program area of the inspection - cfr_number (str): CFR number of the citation - iso_number (str): ISO number of the inspection - legal_name (str): legal name of the facility being inspected - citation_description (str): description of the inspection citation - aggregations (dict): Aggregation results for fei number, CFR number, ISO Number, program area, end date by year.
usfda_maude_search
ChatGPTSearch for medical device adverse events in FDA MAUDE database. Args: query (str): The search term to match against adverse event report. Uses simple_query_string syntax with AND as default operator. When you want to fetch adverse events from a specific deivce/product code, you can use the empty string "" as the query and filter by device_id/product_code. offset (int, optional): Number of results to skip for pagination. Defaults to 0. size (int, optional): Maximum number of results to return. Defaults to 25. sort_by: Field to sort by. Defaults to '_score'. Acceptable values: "_score", "date_received". sort_order: Sort order. Defaults to 'desc'. Acceptable values: "asc", "desc". product_code (str, optional): Filter by product code. device_id (str, optional): Filter by device id. date_received_from (datetime, optional): Filter by date received from. date_received_to (datetime, optional): Filter by date received to. Returns: dict: A dictionary containing: - total_hits (int): Total number of matching companies found - maude (list): List of maude objects, each containing: - report_number (str) - report_title (str) - date_received (datetime) - device_id (str, optional): device id associated with the report - device_name (str, optional): device name associated with the report - product_code (list, optional): list of product codes associated with the report - event_type (str) - device_problem (list, optional) - patient_problem (list, optional) - brand_name (list, optional) - generic_name (list, optional) - manufacturer_name (list, optional) - aggregations (dict): Aggregation results for device problem, patient problem, event type, product code, date received by year.
usfda_recall_search
ChatGPTSearch for FDA medical device recalls. It supports pagination and returns a list of matching recalls along with the total count. Args: query (str): The search term to match against recall title, recall number, reason for recall, product description, action, recalling firm, and application_company_name. Uses simple_query_string syntax with AND as default operator. When you want to fetch a list of recalls from a specific company/product code/device, you can use the empty string "" as the query and filter by application_company_id/product_code/device_id. offset (int, optional): Number of results to skip for pagination. Defaults to 0. size (int, optional): Maximum number of results to return. Defaults to 25. sort_by: Field to sort by. Defaults to '_score'. Acceptable values: "_score", "date_initiated". sort_order: Sort order. Defaults to 'desc'. Acceptable values: "asc", "desc". recall_status (optional): Filter by recall status. Acceptable values: "Completed", "Open, Classified", "Terminated" application_company_id (int, optional): Filter by application company id. product_code: (str, optional): Filter by product code. Devices with the same product code are considered similar. device_id (str, optional): Filter by device id. recall_class (optional): Filter by recall class. Acceptable values: "1", "2", "3". date_initiated_from (datetime, optional): Filter by date initiated from. date_initiated_to (datetime, optional): Filter by date initiated to. Returns: dict: A dictionary containing: - total_hits (int): Total number of matching companies found - recalls (list): List of recall objects, each containing: - recall_title (str): Title of the recall - recall_number (str): Unique recall number - recall_class (int): Class of the recall (1, 2, or 3) - date_initiated (datetime): Date when the recall was initiated - date_posted (datetime, optional): Date when the recall was posted - date_created (datetime, optional): Date when the recall was created - date_terminated (datetime, optional): Date when the recall was terminated - recall_status (str): Status of the recall - reason_for_recall (str, optional): Reason for the recall - root_cause_description (str): Root cause of the recall - recalling_firm (str): Name of the recalling firm - product_code (List[str], optional): List of 3-letter FDA product codes associated with the recall. Devices with the same product code are considered similar. - device_id (List[str], optional): List of device IDs associated with the recall - application_company_id (List[int], optional): List of application company IDs associated with the recall - application_company_name (List[str], optional): List of application company names associated with the recall - official_url (str): the url to fda recall official page. You can retrieve the webapge and find more details. - aggregations (dict): Aggregation results for application company name, device id, root cause description, product code, and initiated by year.
usfda_registration_search
ChatGPTSearch FDA establishment registration Args: query (str): The search term to match against FDA establishment registration records. It searches the registration name, business scope (contractor, importer, etc), and registration listings (include device names and IDs). This is useful to search Class I device manufacturers. Uses simple_query_string syntax with AND as default operator. offset (int, optional): Number of results to skip for pagination. Defaults to 0. size (int, optional): Maximum number of results to return. Defaults to 10. Returns: dict: A dictionary containing: - total_hits (int): Total number of matching registrations found - registrations (list): List of regulation object, each containing: - reg_key (int): registration key - reg_num (int): registration number - fei_num (int): FEI number of the facility being registered - reg_name (str): name of the registration - reg_status (str): if the registration is active or not - business_scope (str): business scope of the facility - reg_listing (str): list of device listings, including device name, ID, and proprietary names. - official_url (str): url to the official FDA registration page. You can visit the page to get more details.
usfda_regulation_search
ChatGPTSearch CFR title 21 by part, subpart, section, and content Args: query (str): The search term to match against part, subpart, section, and content. Uses simple_query_string syntax with AND as default operator. offset (int, optional): Number of results to skip for pagination. Defaults to 0. size (int, optional): Maximum number of results to return. Defaults to 10. Returns: dict: A dictionary containing: - total_hits (int): Total number of matching regulations found - regulations (list): List of regulation object, each containing: - id (str): regulaton id, like '870.5225' - part (str): part number, like 'PART 870—CARDIOVASCULAR DEVICES' - subpart (str): subpart number, like 'SUBPART F—CARDIAC VALVE AND GRAFT DEVICES' - section (int): section number, like 'External counter-pulsating device' - content (str): content of the regulation
usfda_warning_letter_search
ChatGPTSearch FDA warning letters Args: query (str): The search term to match against FDA warning letters. It searches the company name, issuing office, subject, and warning letter text. Uses simple_query_string syntax with AND as default operator. offset (int, optional): Number of results to skip for pagination. Defaults to 0. size (int, optional): Maximum number of results to return. Defaults to 10. sort_by: Field to sort by. Defaults to '_score'. Acceptable values: "_score", "letter_issue_date". sort_order: Sort order. Defaults to 'desc'. Acceptable values: "asc", "desc". letter_issue_date_from (datetime, optional): Filter by letter issue date from. letter_issue_date_to (datetime, optional): Filter by letter issue date to. Returns: dict: A dictionary containing: - total_hits (int): Total number of matching registrations found - warning_letters (list): List of regulation object, each containing: - fda_url (str): url to the official FDA warning letter page. You can visit the page to get more details. - company_name (str): company involved in the warning letter - issuing_office (str): FDA office who issued the warning letter, for example, CDRH, CBER, CDER, etc. - subject (str): subject of the warning letter - title (str): title of the warning letter - letter_issue_date (datetime): issue date of the warning letter, e.g., 2025-01-01. - main_content (str): main content of the warning letter
