SoW.docx

Scope of Work (SOW):

Trends in Nursing Home-Hospice Contracting and Common Ownership between Hospice Agencies and Nursing Homes

REMIS: DA-17-007

Purpose of Study Recent research suggests that there may be financial incentives for hospices and nursing homes with joint ownership to increase hospice referral to their patients in nursing homes. The purpose of this study is to examine the dynamics of nursing home-hospice contractual relationship and how they have evolved over time, and to explore implications of joint ownership in payment, quality and appropriateness of patient care, and oversight. Findings from this study will also lay the ground work for future research on the appropriateness of care delivery to beneficiaries receiving care in nursing homes who are owned by hospices.

Background The Medicare hospice benefit covers palliative and support services for beneficiaries who are terminally ill with a life expectancy of six months or less as documented by a physician. Once the beneficiary elects hospice, they must forgo Medicare coverage for curative care for the underlying terminal condition. Hospice services typically include physical care, counseling, drugs, equipment, and supplies for the terminal illness and related conditions. Care is generally provided at home, although it can also be provided in other places of residence, such as nursing homes and assisted living facilities. In cases where hospice care is provided in a nursing home, the hospice manages the contracted services and provides core hospice care, while other non-core services (i.e., those which the hospice is not required by law to provide) may be provided by the nursing home, in accordance with their contract. For nursing home patients who are dually eligible for Medicare and Medicaid, Medicaid pays the hospice for the patient’s room and board, and Medicare pays the hospice for the hospice benefit. The hospice then pays the nursing home for the beneficiary’s room and board, for daily care, and, depending on the arrangement, for other services.

Although the hospice benefit represents a small portion of total Medicare Part A expenditures, (about 2 percent), it has grown considerably over the past several years (CMS 2016), totaling $15.9 billion in 2015 (MedPAC 2017). In 2015, more than 1.38M Medicare beneficiaries received hospice services from about 4,200 providers, an increase in the number of hospice providers of about 2.6 percent, with most of the growth coming from for-profit hospices (MedPAC 2017). Most notably, however, hospice use among Medicare beneficiaries has substantially increased over the years, more than quadrupling in the last decade (Stevenson, 2017). This increase in service use has raised concerns about the financial incentive created by the hospice payment system to aggressively target more profitable patients, such as those residing in nursing homes (Stevenson 2012).

Hospice services may be appropriate and beneficial to terminally ill residents in nursing homes; however, based on prior studies, hospice care in nursing homes may also present vulnerabilities to Medicare beneficiaries that are worth exploring. Exclusive arrangements and common ownership between the hospices and nursing homes may result in monetary gains, influencing medical decisions that may affect the appropriateness of care to Medicare beneficiaries. Nursing home patients often have longer and less complex care, which contribute to higher profits for hospices. Efficiencies in mileage and staff travel cost, care coordination, and care efforts in a centralized location, as well as facilities serving as referral sources for new patients, may also contribute to financial efficiencies and gains. A recent ASPE study (Stevenson 2017) found that for-profit hospice agencies had considerably higher proportions of patients living in nursing homes, compared to those operated by non-profit and government agencies. In a 2011 study, the Office of the Inspector General (OIG) found that the number of hospice beneficiaries in nursing homes increased by 40 percent, primarily in the for-profit sector. The study also found that Medicare spending on hospice care for nursing home residents grew by almost 70 percent ($2.6 to $4.3 billion) from 2005 to 2009. Because of potential financial advantages, some hospices may be seeking out or depend heavily on beneficiaries in nursing homes, resulting in incentives for hospice agencies to contract with or own nursing homes to generate higher revenue. Despite these concerns, however, to date, there have been few studies focusing on the contracting and common ownership structures between hospice agencies and nursing homes.

Recent research suggests that there may be financial incentives for hospices and nursing homes with joint ownership to increase hospice referrals to patients in their nursing homes. Although it is unclear how the quality and appropriateness of hospice care may be affected, these financial incentives may represent a conflict of interest that influences admissions to and utilization of hospice care. In addition to focusing broadly on nursing home-hospice contracting, this project will produce descriptive statistics about joint ownership and explore the implications of such ownership structures on hospice use and referral patterns for patients in their nursing homes. Outcomes of interest will include any use of hospice before death (e.g., among nursing home decedents and residents with limited prognosis), length of hospice use, live discharge from hospice, and frequency of skilled hospice visits overall and at the very end of life. The research will also examine in detail how facility level hospice use is associated with such ownership arrangements. Findings from the project will inform policy discussions concerning hospice payment and quality of care in the Medicare hospice benefit and will lay the ground work for future research on the appropriateness of care delivery to beneficiaries receiving care in nursing homes who are owned by hospices.

Justification for Sole-source contract:

This project builds on a recently completed hospice project, Tracking the Impact of Ownership Changes in Hospice Care Provided to Medicare Beneficiaries. Vanderbilt (the contractor) has already coded the chain and ownership status of all hospice agencies nationally. This process involved coding Medicare costs reports from 2000-2013 for all hospice agencies that were free-standing or affiliated with hospitals, home health agencies, and skilled nursing facilities. To ensure the quality of this coding, the project also cross-checked this information with other available data about hospice agencies, including Provider of Service files and Provider Enrollment, Chain, and Ownership System (PECOS) data.

Although their use for the prior project was exploratory, PECOS data will be essential in this follow-on project. This project will integrate the initial hospice ownership coding that was achieved using cost reports and supplementary data sources with the rich ownership data contained in PECOS. In particular, PECOS data can be used to identify instances of common ownership between hospice agencies and nursing homes, as individual and organizational owners with ownership stakes of 5% or higher are listed in the PECOS data and can be cross-referenced.

Vanderbilt has already performed preliminary analyses of the PECOS data to define and explore joint ownership between hospice agencies and nursing homes and has extensive experience and the necessary background knowledge from the prior research to move forward with this research immediately. These data have a very different structure from the longitudinal data structures that are typical of cost reports and Provider of Service files. In particular, PECOS data are current to the date at which they are extracted, and investigators must populate observations back in time based on records of changes observed in the data. Based on previous work by Vanderbilt comparing PECOS data to other provider datasets (e.g., including cost reports, OSCAR/CASPER, and provider of service files), PECOS data are somewhat reliable back to 2005, with even better reliability from 2010 forward.

Finally, Vanderbilt plans to integrate all ownership details with information from other Medicare administrative datasets, including the Minimum Data Set and Medicare hospice claims, both of which were completed in the context of the prior project (through 2013). Reflecting the likely availability of Medicare claims and MDS data, the proposed project will seek to integrate these datasets through 2015.

Research Questions Completion of this research will result in a final report that will address the following research questions:

1. Determine whether joint ownership relationships have increased over the years, by examining nursing home-hospice contracting relationships over time (from 2005 to most current year data is available), including the number of hospice agencies that nursing homes work with and the percent of facility hospice users in each agency (e.g., the percent being served by the dominant hospice or a more detailed measure of concentration, such as the Herfindahl Hirschman Index (HHI)).

2. Determine if joint ownership practices over the years lead to changes in behavior, specifically how care is provided to nursing home residents served by nursing homes under these arrangements.

3. Analyze nursing home-hospice common ownership and how it has evolved over time, including its prevalence among hospice agencies and nursing homes and whether it is happening primarily in certain parts of the country. What are the major companies that seem to be strategically focusing across the hospice and nursing facility sectors?

4. Determine whether there are any differences in the role of commonly owned hospice agencies for nursing homes hospice users relative to other agencies. For instance, do commonly owned agencies serve all or almost all of a nursing home’s hospice users? Where new instances of common ownership are identified (e.g., where a company makes a strategic acquisition), how does the role of the commonly owned agency evolve over time within the facility (from 2005 to most current year data is available)?

5. Do referral patterns, characteristics of hospice enrollees (e.g., age, hospice diagnosis, and cognitive status), timing of hospice use, and the quantity and type of hospice visits provided differ for residents living in nursing homes with: i) highly concentrated referrals (e.g., based on HHI) compared to facilities with a lower concentration of referrals; and ii) facilities with joint ownership of a hospice compared to facilities without such joint ownership?

6. Using new acquisitions of hospice agencies by nursing homes that are observable in the data, identify any before and after trends in hospice referral and use patterns before and after the acquisitions (if possible, given the data)?

Tasks The contractor shall provide the necessary personnel, materials, equipment, support and supplies (unless explicitly indicated in this statement of work to be provided by the government) to complete the tasks listed below within the timeframe specified in Appendix A. All work done under this task order shall be performed under the general guidance of the ASPE Contracting Officer’s Representative (COR) and is subject to the COR’s approval.

Task 1: Post Award Meeting Within 2 weeks following the effective date of the task order, the project lead and other relevant staff shall have a conference call with the COR to discuss the objectives of the task order and any related project issues. A written draft agenda for the meeting shall be prepared by the contractor and sent to the COR three days prior to the meeting. Specific topics to be discussed include, at a minimum: the purpose and objectives of the project, the scope of work, work plan, analyses, sources of data, project timetable (see Appendix A for the Schedule of Tasks), and tasks.

Within 3 weeks following the effective date of the task order, the contractor shall prepare a brief draft memorandum summarizing the discussions and any decisions reached during the post-award meeting for the review and approval of the COR. The revised memo is due one (1) week after comments from the COR are received by the contractor, and will be considered final upon approval by the COR.

Task 2: Work Plan The contractor shall develop a work plan to guide the activities of the project in the Data Analysis phase of this project (Task 5). The work plan shall reflect preliminary discussions and decisions reached during the Post Award Meeting (Task 1) and will serve as a blueprint for the contractor’s approach to carrying out project activities and timelines.

At a minimum, the work plan should also include the following activities:

· A statement of the objectives of the research;

· A detailed explanation of the proposed approach for completing the data analysis in Task 5, identifying potential problems and how to mitigate them;

· A description of the data to be used for the analyses;

· An updated and more refined list of the Research Questions to be addressed; and,

· A timeline for project completion.

The work plan shall be submitted to the COR for review four (4) weeks after contract award. The revised work plan is due two (2) weeks after receiving comments from the COR and will be considered final upon approval by the COR. Upon review and approval of the COR, the work plan will be considered final and will be used to guide the project activities and timelines.

Task 3: Monthly Progress Reports and Interim Conference Calls The contractor shall submit monthly administrative progress reports (not to exceed two pages) outlining all work accomplished during the previous month. At a minimum, such reports shall cover the following items:

· Progress made during the month (e.g., meetings, data analysis, or Tasks);

· Difficulties encountered and remedial action(s) taken;

· Personnel and other management changes;

· Discussion of the progress of the data analysis during the data analysis phase along with a brief discussion of substantive findings to date, if any; and,

· Budget status and any concerns.

Monthly reports are due within the first week of every month following the approval of the work plan and shall be sent directly to the COR via electronic mail. The contractor and the COR shall also have interim conference calls or meetings as deemed necessary (but not to exceed a total of five) to discuss issues that need the COR’s input or approval. Within one (1) week of such meetings, the contractor shall submit a short summary of the meeting, highlighting the important steps and action items decided during that meeting.

Task 4: Quantitative Analysis Plan Subtask 4.1 Prepare Draft Analysis Plan The contractor shall develop a quantitative analysis plan based on available CMS administrative data and informed by the revised and finalized Research Questions and research objective. The contractor shall submit a draft Analysis Plan for review by the COR 8 weeks following contract award. A revised plan incorporating comments from the COR is due 2 weeks after review from the COR and will be considered final upon approval by the COR.

The analysis plan should include key components, at a minimum:

· The available data resources (e.g., Medicare claims data, Provider Enrollment Chain Ownership System-PECOS). This may include discussions about the particular use of the data, year of data that will be used for this project, quality of the data, comparability, timeliness of data needed, and any other component that may be of relevance to the analyses;

· For the PECOS data files, the contractor should include a description of how these files would be useful in refining the methodology for identifying providers with joint ownership.

· Data and methods used in this research;

· A description of the methodology that identifies joint ownership and contractual relationships between hospices and nursing homes; and

· Limitations of the analysis.

Subtask 4.2 Negotiate Data Use Agreement and Obtain Data

In Subtask 4.1, the contractor shall identify and purchase the data sources needed for the project and shall prepare a CMS Data Use Agreement (DUA) to acquire the necessary data directly through RESDAC. As the project completion is dependent on the analysis of CMS data, timely request for the data is essential. The DUA documents shall be submitted to the COR for review within 12 weeks after the contract award.

Task 5: Quantitative Data Analysis

The contractor shall conduct analyses of the data according to the Analysis Plan developed in Task 4. The analyses under Task 5 shall be completed no later than 36 weeks (assuming DUA approval within timeframe specified in Subtask 4.2) after contract award to begin work on the draft final report (Task 6.1). The contractor shall notify the COR if any significant problems arise in analyzing the data that could affect their ability to address the purposes of this contract or significantly impact the timeline for completion of the project.

The analysis will be a retrospective analysis of hospice and nursing home enrollment and care practices used by Medicare beneficiaries (age 65 and older) at the end of life using data from years specified in the Analysis Plan, as determined by the analysis plan. The study will consider all Medicare certified hospices, including freestanding hospice agencies and those associated with nursing homes.

Task 6: Draft and Final Report Subtask 6.1 Draft of Final Report The contractor shall prepare a draft of the final report explaining the analyses and addressing the questions outlined in the finalized Research Questions and Work Plan sections above.

The report shall include an executive summary that provides:

· An executive summary

· An overview of the objective of the study, including the research questions;

· Data and methods;

· Main outcomes and measures;

· Results and conclusions; and,

· Policy implications of the findings.

In the body of the report, the contractor shall include, at a minimum:

· An introduction with background information;

· A description of the research questions addressed;

· The study data and methods used in this research;

· Study results;

· Potential future research;

· Limitations of the research, significant data gaps, and areas that warrant further study;

· A final discussion session about policy implications; and,

· Appendices for technical and other documentation relevant to the study. A technical appendix describing the methodology used for classifying “chain” vs. “non-chain” providers shall also be included.

The draft of the final report shall be submitted to the COR for review no later than 42 weeks after contract award.

Subtask 6.2 Final Report The contractor shall prepare a final report, incorporating comments from the COR, within two (2) weeks after receiving comments on the draft report from the COR. Since the audience for this report is broad, the contractor shall ensure that the report is written in plain language and includes appropriate examples, figures, and graphs. The final report should be no longer than 30 pages in length (not inclusive of executive summary and appendices). The final report shall also conform to ASPE guidelines for the delivery of digital copies of reports (see Appendix B). The report will be considered final upon review and approval of the COR. The final version of this document is intended for publication on the ASPE website and is thus subject to 508 compliance requirements.

Contractor shall allow up to four (4) weeks of review for the ASPE clearance process.

Task 7: Briefing for DHHS Officials

The contractor shall conduct a briefing for HHS and other federal government officials and other interested parties. The briefing shall take place at HHS headquarters in Washington, D.C., and present an overview of the study, summary of the data used in the analyses and methodology, major findings, and the implications of the findings for public policy. The contractor shall prepare all necessary handouts and overheads for the briefing, and submit them to the COR for review and approval one week before the date of the briefing. The briefing should occur after the draft of the Task 6.1 report has been prepared and reviewed by the COR, but before the final version is submitted to incorporate any last changes recommended from this briefing (not to exceed 52 weeks).

Appendix A: Schedule of Deliverables

Summary Memo of Post-Award Conference Final Summary Memo of Post-Award Conference Week 2 Week 3 1 weeks after COR comments

Final Work Plan Week 4 2 weeks after COR comments

Interim Meetings Within the 1st week of every month As deemed necessary (but not to exceed five); notes within 1 week of meeting.

4.1 Draft of Analysis Plan

Final Analysis Plan

4.2 Negotiate DUA and Obtain Data

Week 8 2 weeks after COR comments Week 12

6.2 Final Report

Week 42 2 weeks after COR comments

* Deadlines assume DUA approval within timeframe specified in Subtask 4.2. As there is some variability in DUA process time, contract will be 6 months longer in case DUA approval takes longer than anticipated.

Appendix B: Specifications for the Delivery of Digital Copies of Reports 508 Compliance Section 508 of the Rehabilitation Act of 1973 (29 U.S.C. 794d) requires Federal agencies to purchase electronic and information technologies (EIT) that meet specific accessibility standards. This law helps to ensure that federal employees with disabilities have access to, and use of, the information and data they need to do their jobs. Furthermore, this law ensures that members of the public with disabilities have the ability to access government information and services.

Three regulations address the requirements detailed in Section 508. The Section 508 technical and functional standards are codified at 36 CFR Part 1194 and may be accessed through the Access Board’s Web site at http://www.access-board.gov. The second regulation issued to implement Section 508 is the Federal Acquisition Regulation (FAR). FAR Part 39.2 requires that agency acquisitions of Electronic and Information Technology (EIT) comply with the Access Board’s standards. The entire FAR is found at Chapter 1 of the Code of Federal Register (CFR) Title 48, located at http://www.acquisition.gov. The FAR rule implementing Section 508 can be found at http://www.section508.gov. The third applicable regulation is the HHS Acquisition Regulation (HHSAR). Finally, HHS has created a web site with applicable standards, rules, and checklists, http://www.hhs.gov/web/, that must be met.

Regardless of format, all Web content or communications materials produced for publication on or delivery via HHS Web sites - including text, audio, or video - must conform to applicable Section 508 standards to allow federal employees and members of the public with disabilities to access information that is comparable to information provided to persons without disabilities. All contractors (including subcontractors1) or consultants responsible for preparing or posting content intended for use on an HHS-funded or HHS-managed Web site must comply with applicable Section 508 accessibility standards, and where applicable, those set forth in the referenced policy or standards documents below. Remediation of any materials that do not comply with the applicable provisions of 36 CFR Part 1194 as set forth in the SOW or PWS, shall be the responsibility of the contractor or consultant retained to produce the Web-suitable content or communications material.

The following Section 508 provisions apply to the content or communications material identified in this SOW: Access Board Final Rule 36 CFR Part 1194.22(a)-(p).

The following specifications are required for the electronic submission of ASPE reports and published research.

1. All submissions must be in Microsoft Word (or Word Perfect) format — the material will be converted into PDF by ASPE staff. Contractors may include an unlocked PDF if they wish, but it is not required.

2. If the product is longer than 50 pages, please segment the document into sections.

3. File names should correspond to the material with the following conventions:

a. All sections should be titled with the same root name (e.g. ChildWelfare, HealthPolicy, RuralAccess).

b. Each subsection should be numbered sequentially (e.g. ChildWelfare1, ChildWelfare2; HealthPolicy1, HealthPolicy2; RuralAccess1, RuralAccess2 or ch1, ch2, ch3, etc.).

4. Tables, graphs, and charts should remain in their original format (e.g. Word, Word Perfect, or Excel) and use the following conventions:

a. Tables should remain in text format — do not convert to jpg or gif.

b. Graphs, bar, and line charts should be in jpg or gif format. All such charts must have an associated spreadsheet, either in the Word document or as a separate spreadsheet included with the set of files sent to us.

5. Any presentation materials such as PowerPoint should be sent as separate files.

6. Datasets must have complete documentation.

Any questions regarding the submission of electronic material should be directed to ASPE.

References The Centers for Medicare and Medicaid Services (2016) Medicare Hospice Transparency Data (CY2014), Available at: https://www.cms.gov/Newsroom/MediaReleaseDatabase/Fact-sheets/2016-Fact-sheets-items/2016-10-06.html (Accessed: April 2017).

Medicare Payment Advisory Commission (MEDPAC). March 2017. Report to Congress: Medicare Payment Policy, Washington, DC: MedPAC.

Stevenson, Krone1, Gambrel, Meneades, and Huskamp (2017) Tracking the Impact of Ownership Changes in Hospice Care Provided to Medicare Beneficiaries. ASPE Stevenson DG. Growing Pains for the Medicare Hospice Benefit. N Engl J Med. 2012;367(18):1683-1685. doi:10.1056/NEJMp1208465.

DHHS Office of Inspector General (2011) Medicare Hospices That Focus on Nursing Facility Residents, Inspector General: HHS.