Attach._01_SOW_Capstone.pdf

Statement of Work

Background As of April 2018, the AHRQ Registry of Patient Registries (RoPR) (see https://patientregistry.ahrq.gov/) contains records for almost 4000 registries for many different reasons including research, quality improvement, public reporting, and post market surveillance. These registries represent a huge investment in data infrastructure and a potential wealth of data that could be reused for patient centered outcomes research. However, the lack of standardization of definitions in the registries makes analysis across registries impossible.

To address this issue, AHRQ has developed a core set of harmonized definitions of patient centered outcome measures in five different clinical areas (atrial fibrillation, asthma, depression, lung cancer and lumbar spondylolisthesis) in an existing Patient Centered Outcomes Research Trust Fund (PCORTF) funded project (see https://effectivehealthcare.ahrq.gov/topics/registry-of-patient-registries/outcome-measures-framework) . Despite the recognized value of these harmonized core outcome measures, stakeholders are reluctant to put resources into implementing them because of three main perceived barriers:

1. Burden on patients and clinicians to collect patient reported outcomes from patients;

2. Disruptions to clinical care when collecting structured data from clinicians or challenges in extracting high quality information from unstructured data already in patient records; and

3. Difficulty working with electronic health records.

This project will address the barriers cited by registries and demonstrate feasibility and value by incorporating data collection in a way that fits into the workflow of busy practices and facilitates communication between specialists and primary care. This project will demonstrate how incorporation of harmonized outcome measures to clinical systems could improve clinical care, and improve compliance and sustainability for registry and clinical system developers. For example, this project will allow streamlined reporting of practices to CMS Qualified Clinical Data Registries (QCDRs) which will lead to less burden and higher payments for the practices under the Medicare Incentive Payment System (MIPS) and thus be more sustainable. Further, by collecting higher quality data into registries, the registries may be able to become platforms for pre and post market studies of medical products (for example, by enabling nesting of clinical trials in registry infrastructure) leading to revenue for the clinical sites while simultaneously lowering the cost of clinical research (see https://www.ctti-clinicaltrials.org/projects/registry-trials).

This project will specifically address data infrastructure for treatment of depression.

Depression is a common condition managed by both primary care and specialty care providers. Antidepressants are one of the most frequently prescribed drug classes https://effectivehealthcare.ahrq.gov/topics/registry-of-patient-registries/outcome-measures-framework https://effectivehealthcare.ahrq.gov/topics/registry-of-patient-registries/outcome-measures-framework https://www.ctti-clinicaltrials.org/projects/registry-trials https://www.ctti-clinicaltrials.org/projects/registry-trials during physician office visits (https://www.cdc.gov/nchs/fastats/drug-use-therapeutic.htm) yet depression remains poorly recognized and inadequately treated in clinical care in the United States (https://jamanetwork.com/journals/jamainternalmedicine/fullarticle/2546155). Second line therapies include antipsychotics and implantable medical devices, but there is little evidence on the effectiveness of these treatments and which patients would benefit from these therapies. Clinical care has important uncertainties that could be addressed by data in registries and patient care encounters.

The final list of harmonized outcome measures for depression are:

OMF

CATEGORY

OUTCOME

MEASURE

DEFINITION

MEASUREMENT

NOTES

Survival All-cause mortality

Death from any cause, reported in 12-month intervals.

Survival Death from suicide

Patient with a diagnosis of major depression or dysthymia who died from suicide, reported in 12-month intervals.

This should be captured where feasible;

however, it should be noted that this information may not be recorded accurately or available to all providers.

https://www.cdc.gov/nchs/fastats/drug-use-therapeutic.htm https://www.cdc.gov/nchs/fastats/drug-use-therapeutic.htm https://jamanetwork.com/journals/jamainternalmedicine/fullarticle/2546155

CATEGORY

OUTCOME

MEASURE

DEFINITION

MEASUREMENT

NOTES

Clinical Response

Improvement in Depressive Symptoms – Remission

Patient age 18 or older with a diagnosis of major depression or dysthymia and an initial PHQ-9* score > 9 who demonstrates remission defined as a PHQ-9 score less than 5.

*The PHQ-9 or another brief, publicly available, validated patient-reported instrument with empirically derived cutpoints equivalent to the PHQ-9 cutpoints for remission and response and for which an evidence-based crosswalk to the PHQ-9 exists should be used to measure clinical response.

Other measures may be used in addition for research or other purposes.

Timeframe for measurement:

6 months (+/- 60 days)

12 months (+/- 60 days)

In some implementations, it would beneficial to capture earlier responses and remissions and to obtain higher degrees of follow-up. Additional measurements outside of the windows listed above are recommended as supplemental measures.

CATEGORY

OUTCOME

MEASURE

DEFINITION

MEASUREMENT

NOTES

Clinical Response

Improvement in Depressive

Response dysthymia and an initial PHQ-9* score > 9 who demonstrates a response to treatment defined as a PHQ-9 score that is reduced by 50% or greater from the initial PHQ-9 score.

available, validated patient-reported instrument with empirically derived cutpoints equivalent to the PHQ-9 cutpoints for remission and response and for which an evidence-based crosswalk to the PHQ-9 exists should be used to measure clinical response.

Other measures may be used in addition for research or other and remissions and to obtain higher degrees of follow-up. Additional measurements outside of the windows

CATEGORY

OUTCOME

MEASURE

DEFINITION

MEASUREMENT

NOTES

Clinical Response

Worsening in Depressive

Recurrence dysthymia and an initial PHQ-9* > 9 who demonstrates remission (defined as a PHQ-9 score < 5) of at least two months’ duration and subsequently experiences a recurrence of a depressive episode, defined as a 50% increase in PHQ-9 score or defined as a PHQ-9 score > 9 OR hospitalization for depression or suicidality.** available, validated patient-reported instrument with empirically derived cutpoints equivalent to the PHQ-9 cutpoints for remission and response and for which an evidence-based crosswalk to the PHQ-9 exists should be used to measure clinical response. Other measures may be used in addition for research or other

**This definition was proposed by the workgroup. Data accruing from ongoing registries are needed to assess the feasibility of using this definition to capture recurrence.

and remissions and to obtain higher degrees of follow-up. Additional measurements outside of the windows

Events of Interest

Adverse Events Depression treatment-related adverse events. Use of a brief, publicly available, validated measurement tool to capture adverse events is recommended. Reported in 12-month intervals.

CATEGORY

OUTCOME

MEASURE

DEFINITION

MEASUREMENT

NOTES

Events of Interest

Suicide Ideation & Behavior

Selection of ‘several days’, ‘more than half the days’ or ‘nearly every day’ option on PHQ-9 item 9 (“Thoughts that you would be better off dead or of hurting yourself in some way”).

Supplemental assessments of suicide ideation and behavior should be completed for patients who screen positive for suicide ideation on the PHQ-9 or when a clinician has concerns about suicidality.

Supplemental assessments should be completed using an appropriate, brief, validated instrument, such as the Concise Health Risk Tracking (CHRT) scale. Includes nonfatal suicide attempts/suicide attempt behaviors, planning/preparatory acts, active suicidal ideation.

Reported in 12-month intervals (in conjunction with the PHQ-9 suicide item).

Patient Reported

Depression-specific Quality of Life

Depression-specific quality of life should be measured using a brief, validated, publicly available instrument that is appropriate for the population of interest, such as the Quality of Life Enjoyment and Satisfaction Questionnaire (Q-LES-Q). Reported in 12-month intervals.

Resource Utilization

Depression-related Resource Utilization

All resource utilization (as measured by cost) related to treatment or management of depression, including medications, psychotherapy, office visits, urgent care center visits, ED visits, and hospitalizations. Reported in 12-month intervals.

Resource Utilization

Work Productivity Work productivity loss (overall work impairment/absenteeism plus presenteeism), as measured by the Work Productivity and Activity Impairment Questionnaire (WPAI), reported in 12-month intervals.

The AHRQ Confidentiality Statute, 42 U.S.C. 299c-3(c) and (d), applies to the activities conducted under the contract. The statute protects both identifiable individuals and identifiable establishments and provides, in part, that data may not be used for any purpose other than the purpose for which it was supplied absent the consent of the individual or establishment supplying the data. Neither compliance with HIPAA nor IRB approval necessarily ensures that the requirements of the AHRQ Confidentiality Statute are met.

References:

AHRQ RoPR system: https://patientregistry.ahrq.gov/

Information about the RoPR system and Outcome Measure Repository including the library of harmonized outcome measures, design documents:

https://effectivehealthcare.ahrq.gov/topics/registry-of-patient-registries/abstract/

AHRQ Applications and Systems Development Requirements. Appendix 2A in

AHRQ Publishing and Communications Guidelines. AHRQ Publication No. 08(-09)-

M049, April 2009. Agency for Healthcare Research and Quality, Rockville, MD.

https://www.ahrq.gov/research/publications/pubcomguide/index.html

AHRQ Web Site Guidelines in AHRQ Publishing and Communications Guidelines.

AHRQ Publication No. 08(-09)-M049, April 2009. Agency for Healthcare Research and

Quality, Rockville, MD.

https://www.ahrq.gov/research/publications/pubcomguide/pcguide2.htmlAHRQ

Information Security and Privacy Policy. Agency for Healthcare Research and Quality, Rockville, MD. https://www.ahrq.gov/policy/electronic/privacy/infosecurity.html

Overall Objectives/Goals of the Contract

The overall objectives of this contract are to facilitate and advance the use of registries and registry research.

Specifically, the objectives of this contract are:

(1) Collect electronic health data from existing structured and unstructured information in Electronic Health Records (EHRs) and directly form patients at multiple clinical sites at fixed intervals for patient outcomes associated with treatment for depression

(2) Transfer the collected data to electronic patient registries

(3) Develop governance and data use agreement templates with the registries so that data may be made available to external researchers.

(4) Develop infrastructure and tools to allow replication for other clinical registries and clinical systems. This project will examine whether this approach to data collection and transfer enhances the ability to use registries for research on patient outcomes.

The anticipated outcomes of this project are:

https://patientregistry.ahrq.gov/ https://effectivehealthcare.ahrq.gov/topics/registry-of-patient-registries/abstract/ https://www.ahrq.gov/research/publications/pubcomguide/index.html https://www.ahrq.gov/research/publications/pubcomguide/pcguide2.htmlAHRQ https://www.ahrq.gov/policy/electronic/privacy/infosecurity.html

Phase I:

1. Risk analysis of number and ability of EHRs to send/exchange data to the registries

Phase II:

1. Tools for clinicians and patients to facilitate integration of the harmonized depression outcome measures into electronic health records (EHRs) and registries so that these data will be available for clinical research, patient-centered outcomes research, quality improvement and implementation research.

2. Proof of concept for a standards-based approach (an approach that uses technical standards for data extraction/collection/exchange) for collecting and reporting patient outcomes information to clinicians within their workflow and simultaneously transmitting the data to registries to make it available for research.

3. Tools to make it easier for researchers and registry developers to integrate registries with clinical systems.

4. Governance and data use agreement templates that can be used by registries to make data available to external researchers.

AHRQ is only funding the feasibility of creating a connection between the registries and the EHRs. AHRQ will neither own the connection nor will we fund the ongoing operations and maintenance cost for the connection

Phase I: Risk Analysis (May 15, 2019 – Sept 23, 2019)

Task 1: Project management and Reports

Subtask 1.1: Kick-Off Call

At the beginning of the project, the contractor shall participate in an extended initial conference call with the COR and appropriate AHRQ staff to discuss the overall objectives and specific elements of work to be performed. During this call, any changes that are needed to the protocol and timeline that was submitted with the proposal shall be discussed.

Contact information for participants and signed conflict of interest statement for any new project team members who have been added since the original proposal was submitted to AHRQ will be provided from the contractor to the COR. The contractor shall schedule the call within 3 days of the Effective Date of the Contract (EDOC). Within 3 days of the call, the contractor shall send by email to the COR a summary of the conference call.

Deliverable: Call summary and a signed conflict of interest statement for any new members of the project team members who have been added since the original proposal was submitted to AHRQ.

Subtask 1.2: Conference Calls / Status Reports

Beginning one month after the Kick-Off Call, the contractor shall participate in monthly conference calls with the COR and other appropriate AHRQ staff to discuss the report and progress to completing the work.

Within 3 days of the call, the contractor shall send by email to the COR a summary of the conference call along with a concise progress report. The progress report shall document progress in completing tasks and any problems encountered and how they were addressed, and any foreseeable problems.

Deliverables: Call summaries and progress reports

Task 2: Convene a Stakeholder Panel

The contractor shall in consultation with AHRQ and ASPE, organize a stakeholder panel of no more than 20 members representing consumers of outcomes information, such as patient organizations, provider organizations, payers, quality measurement groups, life sciences organizations, research funders, and federal agencies. The contractor shall prepare the stakeholder panel roster and meetings agendas and other meeting materials at least one week before the meetings and submit to the COR. The panel will provide initial feedback on the protocol through a web-based meeting. This initial feedback will include how the project can contribute to overcoming barriers for clinical sites (especially primary care) to participate in data collection infrastructure that can be used for multiple purposes, including quality improvement and research. The feedback will also include how to incorporate elements into the project that will enhance sustainability and adoption of lessons learned at additional sites after the end of this project to build data capacity. The contractor shall incorporate suggestions from project partners, AHRQ and the stakeholder panel review into the modified protocol (task 3).

The contractor shall prepare meeting notes and submit to the COR no later than one week after the stakeholder meetings.

Deliverables:

Stakeholder panel roster, meeting agendas and other meeting materials, meeting notes.

Task 3: Recruitment and Configuration

The contractor shall recruit at least 10 sites from each of 1) a primary care focused registry (e.g., PRIME Registry), 2) a specialty-focused registry (e.g., APA registry)

(registries will have been identified during the proposal), and 3) an integrated healthcare delivery system. The contractor shall prepare all documents necessary for Institutional Review Boards (IRB) approval and submit to the COR and the IRBs as required for each site. The contractor shall also begin to prepare information needed for documents for OMB Paperwork Reduction Act clearance.

The contractor shall evaluate the registries against the requirements for registry-readiness to receive/exchange data that was submitted with the proposal and modify the protocol to meet these requirements. The contractor shall also evaluate the specific EHRs at each clinical site for readiness to send/exchange data to the registries. The contractor shall submit the modified protocol to the COR together with a risk analysis of the whether the objectives of the contract can be met in Phase II.

Deliverables: IRB documentation, modified protocol and risk analysis.

Note: Dependent on findings of the risk analysis, AHRQ and ASPE will make determination on whether to proceed to Phase II. AHRQ and ASPE may consult experts at the Office of the National Coordinator of Health Information Technology (ONCHIT) to review the risk analysis to determine whether technical feasibility is appropriately assessed.

Phase II: Tool development. (Sept 24, 2019 to May 15, 2021)

Task 1: Project Management and Dissemination

Subtask 1.1: Project Management and Reports: The contractor shall participate in monthly conference calls with the COR and other appropriate AHRQ staff to discuss the report and progress to completing the work.

Within 3 days of the call, the contractor shall send by email to the COR, a summary of the conference call along with a concise progress report. The progress report shall document progress in completing tasks and any problems encountered and how they were addressed, any foreseeable problems.

Deliverables: Call summaries and progress reports

Subtask 1.2: Dissemination Plan

The contractor shall finalize the preliminary dissemination plan that was submitted with the contractor proposal and submit to the COR. The dissemination plan shall address the objectives, benefits, and activities of this project, and how the tools may be used by other registries and clinical sites. The plan should also include a list of professional meetings (for example, ISPOR, DIA, Health Datapalooza) where information about this project may be presented.

The contractor shall submit abstracts and present the project at at least five professional meetings over the period of performance.

Deliverables: Final dissemination plan

Subtask 1.3: The contractor shall work with AHRQ to obtain OMB approval for data collection under the Paperwork Reduction Act (PRA). The contractor shall submit all required documents and responses to OMB questions to the COR and other AHRQ personnel as needed.

Deliverables: OMB PRA documents.

Task 2: Convene Stakeholder Panel

The contractor shall update the stakeholder panel on project progress four times per year through web based meetings. The contractor shall prepare the stakeholder panel roster and meetings agendas and other meeting materials at least one week before the meetings and submit to the COR. The contractor shall prepare meeting notes and submit to the COR no later than one week after the stakeholder meetings. In the final quarter of the contract, the stakeholder panel will review the final reports and outputs of the project, prioritize research questions on depression care that can be answered using data from harmonized depression outcome measures, and provide feedback on the deliverables.

Deliverables:

Stakeholder panel roster, meeting agendas and other meeting materials, meeting notes.

Task 3: Configuring registries and Final IRB approval

The contractor shall work with the two registries to assure their clinical registries are configured to support the receipt, exchange and delivery of clinical and patient reported outcomes specific to AHRQ’s harmonized depression measures (receipt from EHRs and a patient mobile application (task 5.2); exchange between the registries and delivery back to the clinical environment). The contractor shall test whether the two registries are ready for the next steps to using test data. The contractor shall develop and submit to the COR a report on registry configuration and challenge using test data.

The contractor shall obtain final IRB approval for data collection and exchange and submit documentation of IRB approval at all sites to the COR.

Deliverables: Report on registry configuration and challenge using test data Documentation of final IRB approval

Task 4: Develop products to facilitate the adoption of the AHRQ core set of depression outcome measures in EHRs

In order to facilitate the adoption of the AHRQ core set of depression outcome measures in EHRs, the contractor shall develop standardized implementation models, Fast Healthcare Interoperability Resources (FHIR, a standard for exchanging healthcare information electronically) profiles as appropriate, and an implementation guide following all of the procedures specified for the HL7 collaboration (https://www.hl7.org/fhir/implementationguide.html). The contractor shall deliver these documents to COR and work with Health Level 7 (HL7) to make the documents freely available on the HL7 website and provide information needed to implement the outcome measure definitions in EHRs.

Deliverables: Standardized implementation models, FHIR and/or similar profiles and implementation guide.

Task 5: Data extraction and exchange

The contractor shall maintain engagement with at least 10 clinical sites from each of two registries and an integrated health system and work with the registries on ongoing issues that may arise with registry configuration. The contractor shall pilot collection of the harmonized outcome measures as described in the modified protocol (task 3).

Subtask 5.1: Structured data collection

The contractor shall pilot collection of outcome measures linked to existing codes and value sets and extract the information from EHRs.

Subtask 5.2: Patient Reported Outcome Measures

The collection of the Patient Reported Outcomes (PROs) at fixed intervals will be particularly important to enable comparisons of outcome measures across provider types and care settings; currently, patients are seen and PROs collected at a wide variety of time points depending on their provider type, access to care/insurance status, and depression severity. Some of the sites have some capacity to collect PRO data at their sites when the patient is seen in an office visit. The contractor shall augment this capacity through mobile/web technologies to achieve the collection at fixed intervals independent of visits (i.e. mobile, web).

If possible, the contractor shall leverage the FY17 OS-PCORTF PRO project ‘Advancing the Collection and Use of Patient-Reported Outcomes (PROs) through Health Information Technology,’ outputs (in particular the specifications and standards for integration into the EHR and allow collection at fixed intervals independent of patient visit to the provider). In addition, the contractor shall develop a function for automated reminders to patients to fill out the forms at fixed time intervals. The contractor shall conduct electronic data collection of patient-reported outcomes (including the Patient Health Questionnaire (PHQ)-9) https://www.hl7.org/fhir/implementationguide.html at baseline and every six months for one year, and generate outcome measurements that are presented back to the clinicians through standards-based integration with EHRs. The contractor shall combine the electronic data collection from patients with the clinical data extracted from the EHR, and provide comparison reporting based on the OMH measures.

Subtask 5.3: Use of Natural Language Processing (NLP) for Unstructured Text Data

Much of the information in EHRs is collected as unstructured text data in “notes” sections. Natural language processing is increasingly used to extract data from EHRs and several commercial and open source options for implementing NLP exist. Where direct collection of structured data from EHRs, clinicians and patients is not feasible, the contractor shall investigate and, where feasible, collect relevant unstructured data from the EHRs. The contractor shall process the data with existing tools using natural language processing and machine learning to identify specific data elements from the text information for specific characteristics or outcomes, such as adverse events and death by suicide, that could not be collected using the OMH standards directly. If such natural language processing and machine learning models are used for processing textual information, the contractor shall validate the models for performance by sampling patient records and having physician expert review compared to the output of the algorithms.

Subtask 5.4: Data Quality

A premise of this work is that high quality data can be captured in the EHR and sent to registries for use in research. To show that the predecessor steps yield data of high quality, the contractor shall validate data quality using an AHRQ-approved method. Examples of techniques include source system loop back verification, ongoing source-to-source verification, data-issue tracking, and sampling data from EHRs and data sent to registries comparing results of outcome measure computation.

Subtask 5.5: Reporting to registries and back to clinicians

The contractor shall collect and report outcomes information from and to clinicians within their workflow through their EHR. These outcomes data will be submitted to registries that use the same data elements. Providing outcomes information within the clinician workflow through the EHR will enable clinicians to use that information in clinical decision-making. The selection of the standards-based approach will be determined during the contract period, with consideration of the EHRs used at the participating sites and whether those EHRs have fully implemented FHIR resources.

Task 6: Pilot Data Analysis and Prioritizing Topics for Patient-Centered Outcome Research

There are many patient-centered outcomes research questions that could be answered using real-world, robust data (i.e., EHR data sent to registries). The contractor together with the stakeholder panel shall design a pilot where the baseline registry data from the 2 selected registries, including those extracted from clinical records, can be obtained and used in analysis to answer a sample research question. This pilot analysis will serve as proof of concept that clinical data harmonized on depression outcome measures sent to registries can be robust enough to answer research questions.

The contractor shall work with the registries to produce governance and data use agreement templates. The contractor together with the stakeholder panel will create and prioritize a list of research topics related to depression care that can be answered through the use of depression outcome measures harmonized across data sources.

The contractor shall prepare and submit to the COR a report detailing pilot data analysis and high priority topics for patient-centered outcome research on depression care using real world, robust data and the approach used to determine prioritization.

Deliverable: report detailing pilot data analysis and high priority topics for patient-centered outcome research on depression care using real world, robust data and the approach used to determine prioritization

Task 7: Final report and artifacts

The contractor shall develop a draft and final report including methods used for data extraction, submitting data to the registries and back to clinicians including barriers encountered (both technical and institutional) and solutions to those barriers and assessment of lessons learned applicable to implementation of harmonized outcome measures in other clinical areas. Specifically, the final report and artifacts shall include:

Tools for clinicians and patients to facilitate integration of the harmonized depression outcome measures into electronic health records (EHRs) and registries so that these data will be available for clinical research, patient-centered outcomes research, quality improvement and implementation

Proof of concept for a standards-based approach (an approach that uses technical standards for data extraction/collection/exchange) for collecting and reporting patient outcomes information to clinicians within their workflow and simultaneously transmitting the data to registries to make it available for

Tools to make it easier for researchers and registry developers to integrate registries with clinical systems.

Governance and data use agreement templates that can be used by registries to make data available to external researchers.

The draft report will be posted on the AHRQ website by AHRQ for public review and also reviewed by invited peer reviewers, project partners AHRQ and the stakeholder panel. The contractor shall prepare a final report and written disposition of comments received and submit to the COR.

The contractor shall also prepare all artifacts from the project that may be reused in other clinical sites and registries to AHRQ in a format that can be made accessible to all interested stakeholders.

All reports and artifacts shall be consistent with AHRQ guidelines (see references in background section) as applicable, including 508 compliance.

In addition, the contractor shall develop drafts of at least two manuscripts suitable for publishing in peer reviewed journals based on the project final report and deliver to the COR.

Deliverables: Draft report, Final report and disposition of all comments, Drafts of two manuscripts, Project artifacts