A. SYNOPSIS This is a request for information. It is not a solicitation for proposals, proposal abstracts, or quotations. Release of this information should not be construed as a commitment to procure items discussed or to incur costs for which reimbursement would be required or sought. The purpose of this notice is to acquire information from the health care and public health surveillance industries providing large, customized healthcare related data sets or data access platforms to researchers, evaluators, and public health program specialists. The ability to provide individual and customized data products that include: Electronic Health Records (inpatient, outpatient and/or emergency) representing visits to U.S. physicians that include details on prescriptions written, and dispensed, diagnoses, procedures, treatment, immunizations, medical imaging, testing (laboratory, microbial, and other types of testing to detect a condition, determine a diagnosis, plan treatment, or monitor treatment or condition over time) and prescriptions;. Aa Software-As-A-Service (SaaS) based platform to manage major Real-World Evidence (RWE) datasets in one place, thereby allowing users to access and analyze the data for research, surveillance or other programmatic or public health use-cases. CDC requires industry to provide written responses on available data sets and resources for potential augmentation of their portfolio of information available to all Centers, Institutes and Offices Agency-wide in support of collaborative work with institutions of higher learning and other Agencies within the Department of Health and Human Services. Market resources are expected to meet the following core requirements: 1. Electronic Health Records The data should ideally preferably be longitudinally linked and updated regularly and include information on patient demographics like race/ethnicity, age, sex, and geographic location (zip3–5) and metropolitan statistical area status, health status; as well as information on date/time of service, date/time of discharge, payor, provider elements like NPI, provider type and specialty, facility or facility type and location, for patients within the United States or territories. The data can be used to link key clinical variables such as vital measures (e.g. temperature, heart rate, blood pressure), diagnoses, procedures, testing orders and results for laboratory, microbiology and other testing measures used to detect a condition, determine a diagnosis, plan treatment, or monitor treatment or condition, medical imaging, findings, underlying medical conditions, reasons for the visit/chief complaints, current diagnoses, immunizations and biologics, drug or treatment, procedures, administered medications, prescriptions, treatment (written and dispensed), and therapeutic outcomes. Data should include patient condition or other transformation elements that are linkable with a common data model, such as OMOP. The data should also be able to connect patient demographics, biometrics (e.g. weight, height), health behaviors (e.g. smoking, alcohol use, substance use, alcohol, sexual health), and health insurance to patient’s clinical information. Availability of other salient visit related information such as provider type, providers specialty, location of provider, date and time of visit, patient location will be important in the overall considerations about the data procedures. All data must be HIPAA compliant to protect patient privacy. The data should be formatted and relational to easily allow linkages of the patient record over time, over record type, and record location within a data source. 2. Medical Claims Data Medical claims data can be adjudicated or pre-adjudicated and should include patient-level data that can be longitudinal and updated regularly and should include inpatient, outpatient and emergency or urgent care payor-based claims. The medical claims data source should include information on patient demographic information (race/ethnicity, age, sex), vital status(date of death), eligibility status and beneficiary relationship and coverage, patient and beneficiary geographic location (3-5zip), and metropolitan statistical area status, encounter and discharge dates, as well as information on payor, and payor elements of service charge, and/or provider elements like NPI, provider type and specialty, and facility type.-based claims. These may be for outpatient services, inpatient admissions, nursing home services, laboratory and imaging services, and medications (i.e. pharmacy claims). Medical claims should include (when applicable) patient diagnosis, and/or reason for testing (in case of lab related claims) and demographics. Depending on the basis of the claim, the data may also include reasons for visit/chief complaints, procedures, laboratory testing, medical devices, medication and treatment prescriptions (written and dispensed), and tests done, medications administered, and other clinical and visit related information during visits to United States (or territories) office-based medical providers, health care systems, ambulatory and general healthcare sites or clinics, emergency or urgent care sites, inpatient hospitals or medical service facilities, or specialized treatment care or nursing homes. Information on date and time of service, location of the healthcare related encounter, and patient location (preferably) should be available. Preferably, the data should enable linking of all data related to and individual healthcare visit, an individual (anonymized) patient. Such data should potentially enable identification and tracking of disease occurrence, medical service utilization, and the utilization of lab services, medications, immunizations, vaccines, drug or treatment prescriptions. 3. Hospital or Facility or HealthCare Encounters-Related System Data Provide regularly updated data sources should provide data on hospital or treatment facilities, healthcare related systems that occur in the United States and territories, including population coverage, medical or specialty service focus, ICU, bed and operational capacity, staffing elements including providers and affiliations, provider networks information, care-quality metrics, technology capacity, geographic location (3-5zip) and metropolitan statistical area status, if applicable, are preferred. Additionally, information on elements that include hospital or health system financial stability or profile, patient costs and billing are encouraged. Data sources with supporting information on patient visits, patient-level diagnosis, procedures, treatments, and discharge information including vital status, payor, and charges. These data may be generated from hospitals’ electronic administrative systems and/or information systems of the hospital. The data should include patients with any pay types and provide details about healthcare related events that take place during a patient’s hospital visit in both inpatient and outpatient settings. Healthcare related events include drugs administered, immunization provided, patient diagnosis, surgeries, procedures, and tests performed. Additionally, information of patient demographics, visit/admission date and time, discharge date and time, discharge status and destination, source of admission (when applicable), visit type (e.g. emergency department visit, out-patient clinic visit, inpatient visit), patient vitals-biometrics-underlying medical conditions, immunization status, health related behaviors would be preferable. Availability of information about the hospital such as type of hospital, bed-size, annual patient population (by type), location of the facility, academic affiliation, would also be important in the overall considerations. Preferably, patient information should be longitudinally linkable across all patient settings and all information pertaining to each individual visit should also be identifiable. 4. Lab Data (See attached spreadsheet) Data sources should provide patient-level records (or other SAS usable formats) representing lab tests (orders and as well as test results) for general maintained health checks, or specific diseases/health conditions, microbiology and cultures testing, and resistance testing for patients within the United States or territories. The lab data may include general lab tests, lab tests related to(COVID-19 and ), and microbiology lab tests (including) Laboratory test data should include all tests and reflex tests ordered and the results for those tests, regardless of test result (both positive and negative results) ( being or), and other details associated with the test such as specimen type, collection date, test date, ordering provider, reason for test/diagnosis .In addition collection type and site, test collected, test performed , isolate/culture id, a non-identifiable unique patient ID should enable linking of all lab data specific to each individual, supporting longitudinal analyses. Patient demographic and clinical data elements shall be included when available. Demographic and clinical data elements should include patient month and year of birth, gender, race, ethnicity (if available), patient residence (preferably 5-digit or 3-digit Zip Code). Provider data elements should include provider specialty, provider NPI code, and 3–provider location (preferably 5-digit or 3-digit provider Zip Code). Data values shall include standard codes; local codes may be additionally included where available or provided if otherwise necessary. 5. Pharmacy Prescriptions Data (See spreadsheet) Data sources should provide regional, or large area prescription activity for all pharmaceutical products. Data should allow for enable measuring of demand for and dispensing of prescription drugs, including ordered and dispensed pharmaceuticals. The data may include pharmaceuticals dispensed to patients across retail and mail service channels, inpatient, treatment, long term or other facilities channels. Data should include elements on prescription including NDC or patient age and gender, date dispensed, standard drug code, generic codes, amount dispensed, type of pharmacy (e.g. retail vs mail-order), NPI number (if available), and commercial drug names, drug class or hierarchical description of use, quantity dispense, therapeutic days, refill or location of pharmacy (when applicable). Additional information can include ordering provider type (if available), whether it a new prescription starter refill, number of refills, controlled substance schedule classification, dispensing pharmacy geographic location where applicable, pharmacy type (retail, hospital, mail). National, region and state drug dispensing data should include aggregated information on patient demographic information (race/ethnicity, age, sex), drug cost, duration (if applicable), patient residence location (e.g. 3-digit zip code - when available), charges and provider elements like provider type and specialty payments, and payor elements, and date elements by week or month. Patient level data should include information that provides a longitudinal record and patient demographic information (race/ethnicity, age, sex), order and dispense dates, as well as information on payor, and payor elements of service charge, provider elements like NPI, provider type and specialty. Supporting clinical data elements should include diagnosis codes, procedure codes when applicable and reason for medication (e.g. insurance type) information. 6. Privacy-Preserving Record Linkage PPRL can include different data sources around laboratory, claims, imaging, prescription data, EHR, clinical trials, chargemaster data, social determinants of health, social mobility/media, medical devices, and other epidemiologic data from multiple sources. Vendors that are able to provide PPLR products should include regularly updated line level data by data source with tokens that link the data between data sources. Ideally data that can provide a longitudinal record is preferred. Vendors should indicate if HIPPA certification and CCPA compliance is included as part of the Vendor’s work. Vendors should indicate methods for matching and provide data on number of sources and vendor names (if they are able), for each of the described source data (lab, prescription, claims…). A description of how the Vendor combines data from internal data sources, external data sources, partner data and other sources, and include details on how the data is de-identified and matched/tokenized and developed into a consolidated data product. Information on how mortality is included and updated in the data is useful. Information on number of transactions/records per data source, estimated number of unique individuals (per data source) and average longitudinal record time period is requested, as well as the geographic detail available for the data record, 3–5-digit Zip Code is preferred. Additionally, information on data platforms and tools associated with the platform should be included when applicable. B. REQUESTED INFORMATION To help inform the types and quality of health data available for acquisition by the CDC, please respond with a whitepaper that provides answers to all of the following questions. In addition, please complete the attached Excel spreadsheet titled, CDC Data Acquisition Requirements related to the core requirements by providing additional detail on your capability within that area. Describe your background and experience working with federal, state or local government. Describe the availability of a comprehensive Data Dictionary (standard/proprietary) and Data Description Documents. CDC receives appropriations for public health collaborations with higher education institutions, federal, state, and local governments. Please describe your restrictions and exceptions as it relates to the CDC’s treatment of data in collaborative analysis and post-market surveillance. Please provide the most recent date for which the data is available; frequency of data updates, and average lag time between the patient’s heath related encounter and the data being available to CDC. Describe the types of technical assistance that are available. Describe the types of training that are available. Describe how data are made available, e.g. files electronically delivered, made available for download, or accessed only through a cloud-based platform. Describe the general size of data sets (in bytes). Describe system requirements and software(s) needed to access, manipulate, and analyze the data. Describe how Data Security is ensured. Describe how Quality control is maintained. Describe the programming languages needed for accessing, manipulating, and analyzing the data. Describe the seat or user-group capacity for data users. Describe deidentification and certification provided. Describe your policy on 3rd party clinical data cleaning. QUESTIONS DUE: Questions about the RFI must be sent via email to Liz Cole-Greenblatt at
[email protected] by Monday, March 15, 2021 by 12:00PM EST . Questions must be sent in the Questions & Answers Template (See attached Excel Spreadsheet). RESPONSES DUE: Interested parties are requested to respond to this RFI by completing the attached Excel spreadsheet titled "CDC Data Acquisition Requirements," and submitting a white paper to answer the fifteen (15) questions above. White papers should follow the flow of this RFI’s Section C - questions 1 through 15, should not repeat information addressed in the spreadsheet, and are limited to fifteen (15) single-spaced pages in length with a minimum 12-point Times Roman font. Responses must be sent via email to Liz Cole-Greenblatt at
[email protected] by Thursday, April 1, 2021 at 12:00PM EST . The attached Excel spreadsheet titled, "CDC Data Acquisition Requirements," contains three (3) tabs. Please refer to ALL three tabs to provide substantive responses to questions in Section B above. Tab 1 (Instructions) shall provide instructions to completing the excel spreadsheet. Tab 2 (Core Requirements) shall provide core requirements to respond to questions in Section C. Please fill in the right-hand columns (E, F, G) with details related to what components and functionalities are offered with your solution. Tab 3 (Additional Comments) - please provide any additional comments you may have here. Edited 3/23/2021 - responses to questions posted