Attachment_1-SOW.pdf

Order for Implementation of SHIELD-Harmonized Laboratory Data Standards in 6 Active Healthcare Institutions

(To support CDRH priorities to build a National Evaluation System for Healthcare Technology (NEST) and the HHS Strategy for Patient-Centered

Outcome Research (PCOR))

United States Food and Drug Administration (FDA) Center for Devices and Radiological Health (CDRH)

Office of In Vitro Diagnostics and Radiological Health (OIR)

A. Statement of Work

Background Summary:

Although laboratory in vitro diagnostic (IVD) testing consumes approximately 5% of a provider institution’s budget, IVD data makes up a significant portion of EHRs and is leveraged in approximately 60-70% of decisions made influencing patient care1,2. The loss or compromise of IVD test data due to interoperability failures have had deadly consequences3.

IVDs are used by healthcare professionals to ‘ask’ a question of a clinical specimen to gain insight into patient’s physiologic status (e.g., glucose levels, disease presence/absence, etc.), which helps guide clinical decisions and subsequent therapeutic actions. Semantic interoperability is a lynchpin to ensuring that the questions that IVD tests ask and results from those tests are accurately and consistently described within electronic healthcare records (EHRs) – essential for ensuring that the life-saving IVD data benefit individual and public healthcare.

This project focuses on implementing infrastructure that directly addresses laboratory data interoperability failures at their root by harmonizing how laboratory data informatics standards are practically applied within healthcare institutions through a sole authoritative source4,5,6. The infrastructure intended for deployment within this project was developed/adopted by the multi-agency/stakeholder SHIELD (Systemic Harmonization and Interoperability Enhancement for Laboratory Data) initiative. SHIELD is a public-private partnership that was assembled with a singular focus on improving the interoperability and utility of diagnostic test data. SHIELD works in a collaborative consensus framework to adopt/develop informatics standards and procedures to improve the interoperability

1 Forsman RW. Why is the laboratory an afterthought for managed care organizations? Clin Chem. 1996 May;42(5):813-6. https://www.ncbi.nlm.nih.gov/pubmed/8653920 2 Hallworth, M. The (True) Value of Laboratory Medicine. https://thepathologist.com/issues/1015/the-true-value-of-laboratory-medicine/ 3 “Death by a Thousand Clicks: Where Electronic Health Records Went Wrong.”

http://fortune.com/longform/medical-records/ 4 2015. “FDA/CDC/NLM Workshop on Promoting Semantic Interoperability of Laboratory Data.”

https://wayback.archive-it.org/7993/20170111193824/http:/www.fda.gov/MedicalDevices/NewsEvents/WorkshopsConferences/ucm453 897.htm 545 CFR 170.207(c)(2): https://www.gpo.gov/fdsys/pkg/CFR-2014-title45-vol1/pdf/CFR-2014-title45-vol1-sec170-207.pdf 6 Federal Register, List of Recognized Standards, Recognition List Number 038 (under FDAMA Section 514(c)): https://www.gpo.gov/fdsys/pkg/FR-2015-01-27/pdf/2015-01420.pdf https://www.ncbi.nlm.nih.gov/pubmed/?term=Why+is+the+laboratory+an+afterthought+for+managed+care+organizations%3F https://www.ncbi.nlm.nih.gov/pubmed/8653920 https://thepathologist.com/issues/1015/the-true-value-of-laboratory-medicine/ https://thepathologist.com/issues/1015/the-true-value-of-laboratory-medicine/ http://fortune.com/longform/medical-records/ https://wayback.archive-it.org/7993/20170111193824/http:/www.fda.gov/MedicalDevices/NewsEvents/WorkshopsConferences/ucm453897.htm https://wayback.archive-it.org/7993/20170111193824/http:/www.fda.gov/MedicalDevices/NewsEvents/WorkshopsConferences/ucm453897.htm https://wayback.archive-it.org/7993/20170111193824/http:/www.fda.gov/MedicalDevices/NewsEvents/WorkshopsConferences/ucm453897.htm https://www.gpo.gov/fdsys/pkg/CFR-2014-title45-vol1/pdf/CFR-2014-title45-vol1-sec170-207.pdf https://www.gpo.gov/fdsys/pkg/CFR-2014-title45-vol1/pdf/CFR-2014-title45-vol1-sec170-207.pdf https://www.gpo.gov/fdsys/pkg/FR-2015-01-27/pdf/2015-01420.pdf

(To support CDRH priorities to build a National Evaluation System for Healthcare Technology (NEST) and the HHS Strategy for Patient-Centered laboratory data. These efforts to harmonize the application of standardized nomenclatures and streamline high-quality health information exchange directly address the priorities within the 21st Century Cures Act7 to improve electronic data interoperability and the CDRH priorities to build a National Evaluation System for Healthcare Technology (NEST)8.

Fundamentally, all IVDs ask a question of a specimen to get an answer. For instance, when a patient comes into a physician’s office with the signs and symptoms of influenza, the physician may collect a specimen from the patient and order the lab to conduct a test to query if the specimen harbors antigens or nucleic acids from a specific influenza strain (question).

The lab will then process the specimen, conduct a test and provide the findings (answer) observed from that test back to the physician (see Figure 1 below). The physician will then take those results into consideration and diagnose the patient, which often influences medications prescribed, e.g. a neuraminidase inhibitor for influenza A.

Figure 1. IVD descriptive coding.

In the scenario described above, the question being asked of the specimen can be described by a LOINC code (Logical Observations Identifiers Names and Codes) and the answer can be described with a SNOMED-CT code (Systematized Nomenclature of Medicine – Clinical Terms) and the specific device can be described using the Device Identification component of the Unique Device Identification (UDI) system.

This project will help ensure that the IVD descriptive information that feeds into all of these processes is consistent and unambiguous. These semantic standards are key to this project by facilitating the storage, exchange, and pooling of results for clinical care, outcomes

7 “H.R.34 - 21st Century Cures Act.” https://www.congress.gov/114/bills/hr6/BILLS-114hr6rfs.pdf 8 2016-2017 CDRH Strategic Priorities:

https://www.fda.gov/downloads/aboutfda/centersoffices/officeofmedicalproductsandtobacco/cdrh/cdrhvisionan dmission/ucm481588.pdf https://www.congress.gov/114/bills/hr6/BILLS-114hr6rfs.pdf https://www.fda.gov/downloads/aboutfda/centersoffices/officeofmedicalproductsandtobacco/cdrh/cdrhvisionandmission/ucm481588.pdf https://www.fda.gov/downloads/aboutfda/centersoffices/officeofmedicalproductsandtobacco/cdrh/cdrhvisionandmission/ucm481588.pdf

(To support CDRH priorities to build a National Evaluation System for Healthcare Technology (NEST) and the HHS Strategy for Patient-Centered management, and research by providing a set of standard names and identifiers that can be used across heterogeneous computing environments.

Based on feedback from public workshops and the SHIELD network of experts:

1. FDA recognized LOINC in the context of IVDs9,

2. FDA recognized SNOMED-CT in the context IVDs10,

3. SHIELD identified UCUM (Unified Codes for Units of Measure) as the optimal standard for representing units for quantitative tests,

4. FDA published a Guidance11 outlining regulatory considerations when associating

LOINC codes to IVDs,

5. FDA procured a “Guide for Using LOINC Microbiology Terms for in vitro diagnostic

(IVD) tests” from the Regenstrief Institute with a list of examples and helpdesk support accounting for approximately 12,000 codes representing almost all infectious disease tests12,

6. the IVD Industry Connectivity Consortium (IICC) developed LIVD13 - a structured format for IVD vendors to share LOINC codes with laboratories incorporating the Unique Device Identification (UDI) system14,

7. Health-Level 7 (HL7) and the IICC developed a Fast Healthcare Interchange Resource (FHIR) profile for LIVD15,

8. the IICC developed the Laboratory Analytical Workflow (LAW) standard16 - a plug-and-play mechanism for an IVD analyzer to communicate directly with a Laboratory Information System (LIS).

This work has been lauded by FDA Commissioner Dr. Gottlieb as a landmark effort in the combat of antimicrobial resistance17 and is paving the way towards laboratory data interoperability. By empowering the entire healthcare ecosystem to use the same terminology to describe the same diagnostic observation the same way every time, critical public health information can be instantly transmitted across the planet and analyzed. By leveraging deidentified, semantically interoperable diagnostic information from patients, real-time data

9 FDA Recognized Consensus Standards Database: LOINC https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfstandards/detail.cfm?id=32889 10 FDA Recognized Consensus Standards Database: SNOMED-CT https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfstandards/detail.cfm?id=33224 11 2018. FDA Guidance “Logical Observation Identifiers Names and Codes for In Vitro Diagnostic Tests.”

https://www.fda.gov/downloads/MedicalDevices/DeviceRegulationandGuidance/GuidanceDocuments/UCM61 0636.pdf 12 2018. “Guide for Using LOINC Microbiology Terms” https://loinc.org/guides/micro/ 13 2017. “LIVD – Digital Format for Publication of LOINC to Vendor IVD Test Results” http://ivdconnectivity.org/livd/ 14 FDA Unique Device Identification System https://www.fda.gov/medicaldevices/deviceregulationandguidance/uniquedeviceidentification/ 15 2018. “LIVD – FHIR Mapping Project” http://hl7.org/fhir/uv/livd/2018Sep/index.html 16 2018. “LAW – Laboratory Analytical Workflow Profile” https://ivdconnectivity.org/law-profile/ 17 FDA Commissioner call to improve utility of RWE: https://www.healthcare-informatics.com/blogs/david-raths/interoperability/why-fda-s-scott-gottlieb-wants-create-national-utility-real-world https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfstandards/detail.cfm?id=32889 https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfstandards/detail.cfm?id=33224 https://www.fda.gov/downloads/MedicalDevices/DeviceRegulationandGuidance/GuidanceDocuments/UCM610636.pdf https://www.fda.gov/downloads/MedicalDevices/DeviceRegulationandGuidance/GuidanceDocuments/UCM610636.pdf https://loinc.org/guides/micro/ http://ivdconnectivity.org/livd/ https://www.fda.gov/medicaldevices/deviceregulationandguidance/uniquedeviceidentification/ http://hl7.org/fhir/uv/livd/2018Sep/index.html https://ivdconnectivity.org/law-profile/ https://www.healthcare-informatics.com/blogs/david-raths/interoperability/why-fda-s-scott-gottlieb-wants-create-national-utility-real-world https://www.healthcare-informatics.com/blogs/david-raths/interoperability/why-fda-s-scott-gottlieb-wants-create-national-utility-real-world

(To support CDRH priorities to build a National Evaluation System for Healthcare Technology (NEST) and the HHS Strategy for Patient-Centered can be used to drive innovative solutions and 21st Century Cures to significant public health challenges.

B. Statement of Need:

The purpose of this procurement is to sponsor an interoperability upgrade to a diverse18 range of existing laboratory information systems (LISs) by incorporating high-quality, industry-supported and 3rd party vetted, standardized descriptive coding for IVDs through LIVD, FHIR and LAW (as appropriate). Although largely digitized, IVD data is often represented differently between institutions (or even within an institution) leading to ambiguity and incapacitating its utility in research or other purposes. Semantic interoperability failures for IVDs has been frequently cited as a significant impediment to patient care and public healthcare overall. Maintaining electronic health record systems that lack harmonized, semantically interoperable coding is costly and encumbers healthcare professionals whose time could be put to much better use19.

This project will ‘pressure test’ the implementation of SHIELD-approved infrastructure prior to roll-out on a national scale. Electronic healthcare system interoperability will be assessed pre- and post-implementation, to include comparative associations with ICD (International Statistical Classification of Diseases and Health Problems) codes and CPT (Current Procedural Terminology) codes, used for diagnosis and reimbursement claims.

This effort is expected to also have a significant impact on the ability to use Real World Evidence (RWE) in regulatory decisions, realize real-time epidemiology, enhance clinical decision support and enable research related to key diseases, including (but not limited to) opioid overdose, antimicrobial resistance, sepsis, cancer, renal failure and much more.

C. Requirements:

Note: Every healthcare institution is different; some activities will not be possible or appropriate for some healthcare institutions, e.g., institutions not leveraging FHIR could not implement the LIVD FHIR profile. Demonstrating semantic interoperability can be practically achieved irrespective of underlying infrastructure differences is essential for establishing a platform for national health data interoperability. The implementing healthcare institution (IHI) will work with SHIELD at the time of contract to determine which requirements are reasonable and appropriate in consideration of the existing IHI infrastructure.

18 In order to be optimally impactful, demonstrating implementation in diverse IHI environments within the United States is key; for the this project, SHEILD will determine if the IHI is appropriately diverse.

19 American Clinical Laboratory Association (ACLA): “Benefits for the adoption of eDOS” https://www.acla.com/wp-content/uploads/2016/07/2016-Jul-8-SIFramework_eDOSIG_BenefitsPosition- Statement-Final.pdf https://www.acla.com/wp-content/uploads/2016/07/2016-Jul-8-SIFramework_eDOSIG_BenefitsPosition-Statement-Final.pdf https://www.acla.com/wp-content/uploads/2016/07/2016-Jul-8-SIFramework_eDOSIG_BenefitsPosition-Statement-Final.pdf

(To support CDRH priorities to build a National Evaluation System for Healthcare Technology (NEST) and the HHS Strategy for Patient-Centered

This project will implement processes and standards adopted/developed and approved by the SHIELD collaborative to improve IVD data interoperability (as possible/appropriate) into an active healthcare institution. The overall goal of this work is to improve the interoperability of laboratory data within and between a diverse20 range of healthcare institutions, beginning by describing the same type of test the same way.

The following informatics infrastructure will be intended for implementation:

1. Semantic Standards:

i) LOINC21

ii) SNOMED-CT22

iii) UCUM23

iv) UDI (Device Identifier component only)24

2. Transmission/Mapping:

i) LAW25

ii) LIVD26

iii) LIVD FHIR profile27

To help ensure that the same type of test is described the same way, ensuring that descriptive coding has been vetted through a sole authoritative source is key. The IHI will work together with SHIELD to select 100 tests across the following domains:

3. Microbiology, e.g.,

i) Human Papilloma Virus (HPV)

ii) Hepatitis C Virus (HCV)

iii) Hepatitis B Virus (HBV)

iv) Human Immunodeficiency Virus (HIV)

v) Influenza Type A

vi) Antimicrobial Susceptibility Tests (ASTs)

vii) Mycobacterium tuberculosis

viii) Streptococcus pyogenes

ix) Clostridium difficile

4. Chemistry, e.g., 20 SHEILD will determine if the IHI is appropriately diverse with respect to other implementation pilots associated with this funding.

21 Logical Observations Identifiers Names and Codes (LOINC) https://loinc.org/get-started/loinc-term-basics/ 22 Systematized Nomenclature of Medicine – Clinical Terms (SNOMED-CT) https://www.snomed.org/snomed-ct/five-step-briefing 23 Unified Codes for Units of Measure (UCUM) https://unitsofmeasure.org/trac 24 Unique Device Identification (UDI) https://www.fda.gov/medicaldevices/deviceregulationandguidance/uniquedeviceidentification/udibasics/default.

htm 25 Laboratory Analytical Workflow (LAW; CLSI AUTO 16) https://ivdconnectivity.org/law-profile/;

https://clsi.org/standards/products/automation-and-informatics/documents/auto16/ 26 Digital Format for LOINC to Vendor IVD Test Results (LIVD) https://ivdconnectivity.org/livd/ 27 LIVD FHIR Profile: http://hl7.org/fhir/uv/livd/2018Sep/index.html https://loinc.org/get-started/loinc-term-basics/ https://www.snomed.org/snomed-ct/five-step-briefing https://www.snomed.org/snomed-ct/five-step-briefing https://unitsofmeasure.org/trac https://www.fda.gov/medicaldevices/deviceregulationandguidance/uniquedeviceidentification/udibasics/default.htm https://www.fda.gov/medicaldevices/deviceregulationandguidance/uniquedeviceidentification/udibasics/default.htm https://ivdconnectivity.org/law-profile/ https://clsi.org/standards/products/automation-and-informatics/documents/auto16/ https://ivdconnectivity.org/livd/ http://hl7.org/fhir/uv/livd/2018Sep/index.html

(To support CDRH priorities to build a National Evaluation System for Healthcare Technology (NEST) and the HHS Strategy for Patient-Centered

i) Troponin

ii) Blood glucose

iii) Hemoglobin A1c

iv) Creatinine

5. Drug/Toxicology, e.g.,

i) Opioids

ii) Lead

6. Allergy, e.g.,

i) Antibiotics

ii) Food

7. Serology/Hematology, e.g.,

i) Prothrombin time/International Normalized Ratio (PT/INR)

ii) White blood cell count

8. Molecular Pathology/Cell Markers, e.g.,

i) Human Epidermal Growth Factor Receptor 2 (HER2)

ii) Breast Cancer Gene 1,2 (BRCA 1,2)

iii) Epithelial Growth Factor Receptor (EGFR)

iv) Prostate-Specific Antigen (PSA)

To understand the benefits and challenges of implementing the new interoperable infrastructure, a thorough evaluation of the current existing infrastructure within the IHI will be necessary. This evaluation should include how each of the 100 IVD tests are currently represented within the LIS and the EHR. For instance, some LISs leverage semantic standards, e.g., LOINC, SNOMED-CT, UCUM, etc., and some use internally-developed terminology to electronically represent the question and answer associated with each IVD test in the LIS. This may be similar or different to how the same test is represented in the EHR. With an understanding of how much change is necessary to achieve semantic interoperability for these tests, we will begin to gain an understanding of the type of support that will be needed for other healthcare institutions as we begin to approach addressing interoperability on a national scale.

During the test selection period for the 100 tests, each IHI will begin preparation to receive SHIELD pre-vetted IVD descriptive coding in LIVD format, including a plan to test the sending and reception of the codes in a test environment. For tests that run on instrument platforms that can interface with an LIS, the IHI will work with SHIELD to partner with IVD manufacturers (i.e., through the IVD Industry Connectivity Consortium – IICC) to implement the next-generation IVD instrument interface Laboratory Analytical Workflow (LAW – CLSI AUTO16), which will enable plug and play functionality between the analyzer and the LIS. Note, unlike LIVD, LAW implementation only requires IVD manufacturer and EHR vendor interaction; no work is required from the IHI. For systems with FHIR, the FHIR LIVD profile should be implemented to help guide mapping of IVD test transmission codes.

(To support CDRH priorities to build a National Evaluation System for Healthcare Technology (NEST) and the HHS Strategy for Patient-Centered

Once the SHIELD-vetted LOINC mappings in LIVD and associated answer sets are collated for the tests, they will be provided to the IHI. The IHI will then compare their data dictionary mappings to those from the authoritative source and evaluate the suitability of these codes. If the IHI determines that a code provided is inappropriate, the IHI engage in collaborative conversation with the IVD manufacturer and SHIELD (as appropriate) to resolve any challenges that may arise. Differences/inconsistencies should be carefully documented between:

i) Codes/mappings provided by the authoritative source (e.g., LOINC codes in LIVD format) and

ii) Representation of laboratory data for each of the tests within each IHI.

SHIELD will work together with the IHI (as appropriate/needed) to ensure that this data is stored accurately in the LIS and reliably transferred, such that the data can feed into the EHR and be associated with ICD28 and CPT29 codes. Assessments will be made to demonstrate that data was accurately and reliably transferred by, showing that the appropriate IVD descriptive coding is in the IHI electronic system. SHIELD will also communicate with the IHI to learn how the process of interoperable infrastructure upgrade could be optimized.

Once a successful transmission of IVD descriptive data has been completed in a test environment, the appropriate coding and informatics infrastructure should be moved in to a production environment.

Successful completion of this project relies on collaborative interaction between IVD manufacturers, EHR vendors and the IHI. The resources provided for this project include the funds an IHI may need to enlist the assistance of EHR vendors, IVD manufacturers and an appropriate third party (as needed). Throughout this process, the IHI should work with EHR vendors and IVD manufacturers (as appropriate for each entity) to aid in the process for implementation. As needed, the IHI should work with EHR vendors to develop an interface with the IHI. As appropriate, the EHR vendor can aid in the creation of aliases in the alias tables for the mapping of LOINC codes transmitted to the EHR for both retrieval and storage of the LOINC codes transmitted to it from the lab.

For the implementation of LAW, the IHI should enlist the services of the EHR vendor and the appropriate IVD manufactures to build an interface between the device and the LIS and test the transmission from the device to the LIS. SHIELD can provide assistance in this process (as needed/appropriate).

28 International Statistical Classification of Diseases (ICD) Codes:

https://www.cms.gov/Medicare/Coding/ICD10/2019-ICD-10-PCS.html 29 Current Procedural Terminology (CPT) Codes: https://www.ama-assn.org/amaone/cpt-current-procedural-terminology https://www.cms.gov/Medicare/Coding/ICD10/2019-ICD-10-PCS.html https://www.ama-assn.org/amaone/cpt-current-procedural-terminology https://www.ama-assn.org/amaone/cpt-current-procedural-terminology

(To support CDRH priorities to build a National Evaluation System for Healthcare Technology (NEST) and the HHS Strategy for Patient-Centered

In order to assess the return on investment (ROI) from this project, it will be important to be able to evaluate the status of a system before and after implementation of the updated informatics infrastructure. To do this, an existing mechanism should be used or a new application should be created that will operate in the system (e.g., leveraging FHIR and/or HL7 V2) to be able to perform queries to attain the IVD descriptive coding associated with the range of different tests in this project. At a minimum, the IHI will need to query which test codes are associated with laboratory test orders and results along with associated ICD and CPT codes, in general. This should be done for all of the tests of interest in this study.

Although this study does not intend to include direct patient information, in order to assess the potential for impact of this project, the IHI should count the number of times each test was ordered and the number of times a result for that test was returned during the evaluation period. The IHI will work with SHIELD to determine the appropriate length of time for the evaluation period. The IHI should also provide an estimate of the number of patients they currently service. Any relevant demographic information would be welcome, if available.

Quarterly meetings between healthcare institutions and SHIELD will be held to discuss progress and findings, in addition to updates described in the Tasks described in the outline below. Findings from this project will be compiled and prepared for publication/dissemination by the IHI and SHIELD with the direct intent to publish content useful for the betterment of public health.

This project will increase the semantic interoperability of each implementing institution with other healthcare systems as well as be an excellent semantic interoperability project to share with the healthcare informatics industry.

If the Offeror chooses to submit alternatives technical information substantiating their equivalence shall also be submitted. The FDA will make final determinations based on this technical information as to the acceptability and equivalence of substitute items.

A. Statement of Work