What is the national hospital price transparency study?
The study is an ongoing employer-led initiative to measure and report publicly the prices paid for hospital care at the hospital- and service-line level. These studies are the first of its kind in that it is an employer-led initiative that uses claims data to compare hospital prices publicly.
The core goals of the study are:
- to enable employers to be better-informed shoppers for health plans and provider networks;
- to hold hospitals, hospital systems, and health plans accountable for the prices they have negotiated;
- to report hospital prices relative to a Medicare benchmark.
- conduct all study analyses
- prepare study final reports and supplemental material
- co-develop study design
- co-recruit nationally for study participation
Employers’ Forum of Indiana (EFI) Role:
- commission and partner with RAND Corp to conduct Round 1.0, Round 2.0, and Round 3.0 analyses per MOU
- co-develop study design
- co-recruit nationally for study participation
Why is it important for employers, and coalitions of employers, to participate?
Our health care system consumes vast economic resources, without producing commensurate health benefits. Employers, in their role as purchasers of health benefits, have been too passive and have not aligned around a strategy that is capable of materially increasing value from the health care system. Becoming an active, informed purchaser starts with “turning on the lights” and recognizing, as Zack Cooper and colleagues put it, “the price ain’t right.”1
How will Round 3.0 be different from Round 1.0 and Round 2.0?
The study has expanded over time.
- Round 1.0, aka RAND 1.0 (the pilot, Indiana hospital price transparency)
- included 120 hospitals in Indiana
- included claims data from mid-2013 through mid-2016
- included claims data from around one dozen self-funded employer participants
- fully funded by the Robert Wood Johnson Foundation (RWJF)
- public report released in September, 2017
- Round 2.0, aka RAND 2.0, (the current round, national hospital price transparency)
- includes 1598 hospitals in 25 states
- includes claims data from 2015 through 2017
- includes claims data from dozens of self-funded employers, two state-based all payer claims databases (APCDs), and several health plans
- funded jointly by participating self-funded employers, RWJF, the National Institute for Health Care Reform, and The Health Foundation of Greater Indianapolis
- public report released May 9, 2019
- participating self-funded employers who helped fund the study received employer-specific private reports, based just on their enrollees’ claims data
- Round 3.0, aka RAND 3.0, (the next round)
- will include additional states and hospitals
- will include more claims data for hospitals in the Round 2.0 study
- will be updated to include claims data through 2018
- will be publicly released first quarter or early second quarter of 2020
How are the Studies Funded?
Through a combination of foundation grants and contributions from participating employers.
- Round 1.0, fully funded by The Robert Wood Johnson Foundation (RWJF)
- Round 2.0, contributing funders include RWJF, the National Institute for Health Care Reform, The Health Foundation of Greater Indianapolis, and self-funded employers across the country. No funding accepted by hospitals or health plans.
- Round 3.0, will be a combination of funding from self-funded employers, employer coalitions, and foundations.
Can I see reports from Rounds 1.0 and 2.0?
Reports and Supplemental material are freely and publicly available.
- Round 1.0
- Final report, including summary findings, methodology and detailed findings: https://www.rand.org/pubs/research_reports/RR2106.html/ 2
- Summary slide deck: http://employersforumindiana.org/media/2017/09/Hospital-Prices-in-Indiana-Findings-Chapin-White-9-20-17-updated.pdf 3
- Interactive online map that allows users to pinpoint hospital locations and view their prices: https://www.rand.org/health/projects/indiana-hospital-prices.html/ 4
- Round 2.0
- RAND website: www.rand.org/pubs/research_reports/RR3033.html
- Home Page: We created this public website to host all information related to Round 1.0, Round 2.0 and Rand 3.0 studies. Included on this website are the following:
- Round 1.0 PDF report
- Round 2.0 PDF report
- Round 2.0 Supplemental Excel database which provides information on all 1598 hospitals and tabs per State. We encourage stakeholders to utilize this information to conduct their own analyses of interest.
- Round 2.0 Interactive Map of US notes a circle for each of the 1598 hospitals. This user-friendly Tableau tool permits one to easily view hospital price variation by region, state, or heath-system.
- Bios about the researchers
- News media about the Round 2.0 study
- Sign up form for Round 3.0 study
How do I sign up for the Round 3.0 study?
Enrollment for Round 3.0 is now closed. Please contact us if you have further questions.
How much does it cost for employers to participate?
For Round 3.0 there were 2 options for participating in the study:
- No charge (free) – for employers who wish to contribute claims data to RAND for the PUBLIC report (we have grant funding to support this work).
- $0.20 per member (min of $1,000, up to a max of $15,000) – for employers who wish to have a PRIVATE employer-level report showing the prices they paid to each hospital relative to average prices paid. For example, an employer with 1000 covered lives would contribute $1000, an employer with 10,000 covered lives would contribute $2,000, and an employer with 75,000 lives or more would contribute the maximum of $15,000. These employers’ claims data will also be included in the public report.
Why are only self-funded employers asked to contribute?
Self-funded employers are asked to contribute because they are the ultimate purchasers and have the most to gain. They will receive private individual employer-level reports in addition to the published aggregate reports. Their data comes in messy from the health plans, thus it takes considerable effort to scrub their data clean for analyses. Also, often multiple data files are sent per plan which require data streamlining.
All-payer claims databases (APCDs) can contribute their claims data without charge, because the organizations that maintain APCDs are either not-for-profit entities or governmental agencies. Including claims data from APCDs strengthens the study and provides public benefit to the APCDs and other participants, but those APCDs do not generally have revenue streams or funding available to support analytics by external research organizations. In addition, their data typically comes in clean.
For the fully insured health plans, there is no charge for this group because we decline to accept funding directly from health plans. As opposed to self-funded employers, employers purchasing these products do not own their own claims data and generally have relatively few enrollees, which means that securing study funding from them would be a considerable challenge. At the same time, including claims data from fully insured health plans strengthens the study and provides a public benefit.
What information will the employer-specific reports provide?
The employer-specific reports will include:
- total allowed amounts paid to all hospitals for inpatient and outpatient care as a percent of what Medicare would have paid for the same services from the same hospitals;
- total amounts paid per service for hospital care relative to Medicare, and relative to average commercial rates; and
- allowed amounts paid to individual hospitals and hospital systems identified by name, relative to Medicare reimbursement for that same set of services.
- Partial Example of employer report:
Please contact Chapin White (firstname.lastname@example.org) the lead study researcher, if you would like to receive an example of the tables and figures in an employer-specific price report.
What agreements need to be in place for a self-funded employer to participate?
A self-funded employer who participates in the study will enter into a contract with RAND (1. in the figure below) that describes the services RAND will perform and the contribution the employer will provide. The employer will send an authorization (2.) to their third-party administrator (TPA) instructing them to supply RAND with a copy of their claims data from at least 2015 through 2018. (3.) RAND and the TPA will enter into a data use agreement (DUA) that specifies the data being transferred and the privacy safeguards that will be in place. RAND and EFI will have a memorandum of understanding (MOU) (4.) that describes the purposes of the study and the roles of the two organizations—that MOU will be referenced by the other 3 agreements.
To view the four agreements, please contact Chapin White (email@example.com).
How will RAND ensure data security and privacy of protected health information (PHI)?
As defined under the Health Insurance Portability and Accountability Act (HIPAA) Privacy Rule, protected health information (PHI) refers to information that identifies an individual and that relates to the individual’s medical conditions or health care services.5 Health insurers, health care providers, and employers that self-fund their health plans are all considered “covered entities,” and are, therefore, subject to the HIPAA Privacy Rule, meaning that they must take safeguards to maintain the privacy of PHI.
RAND will enter into DUAs with the TPAs and any other suppliers of claims data, and those DUAs will obligate RAND to adhere to HIPAA privacy standards. The DUAs will specify a data safeguarding plan for protecting privacy of PHI, including physical access controls, network security, and a process for securely deleting PHI once it is no longer needed for the study.
In general, RAND will avoid receiving any data elements that are unnecessary for the study or that could be used to directly identify patients, and RAND will erase data containing PHI as soon as those data have been processed and are no longer necessary. RAND will also limit publication of results based on the number of data points available. For example, hospital-specific prices will only be reported if a minimum number of claims are available. (For the first round, hospital-specific prices were reported only if a minimum of 11 claims were available. That minimum will be reduced to 5 for the current study in order to include more hospitals, while still maintaining statistical reliability and ensuring patient privacy.)
To illustrate the data safeguarding procedure, the key steps in the data processing are summarized in the figure below.
TPAs and other data suppliers will create extracts of their claims data (1. in the figure) that will contain the minimum fields necessary for the study. Those raw claims data will not include direct identifiers (e.g. patient names or medical record numbers) but they will identify the employer and will include detailed information (including dates of service) on health care services. Employer identifiers, when combined with the health and medical records of their employees, are considered PHI because they could, in small firms, be linked to individual employees. Because service dates and employer identifiers are included in the raw claims data, those data must be considered PHI even though direct identifiers are not included.
The TPAs and other data suppliers will transmit the PHI either by secure file transfer protocol (SFTP) or by encrypted drive. Once RAND receives the raw claims data (2. in the figure), it will be loaded onto an “airgapped” workstation (i.e., a computer that it is permanently disconnected from the RAND network and from the internet) in a “cold room” (a locked, high-security workspace that requires a passkey for entry), and the SFTP files and encrypted drive will be securely erased.
RAND analysts will then create two derivative files in the cold room. The first derivative file will be a deidentified claims dataset (3. in the figure) that excludes service dates (except for year) and excludes employer identifiers and thereby satisfies the HIPAA safe harbor standard for deidentification. The second file will be an employer-claim link file (4. in the figure) that only includes two fields: a unique identifier for each claim (this unique identifier will also be included in 3.) and a unique identifier for each employer.
Once RAND has created a deidentified claims dataset in the cold room, it will be transferred using an encrypted drive to a limited-access folder on RAND’s secure server (5. in the figure). RAND analysts will then go through the process of repricing the claims using Medicare’s payment formulas, resulting in a deidentified dataset (6. in the figure) containing actual allowed amounts from the raw claims data in addition to simulated Medicare payment amounts. RAND will then produce summarized price data for the public report (7. in the figure).
To produce employer-specific reports, RAND will transfer the deidentified claims data with Medicare prices (6.) back to the cold room using an encrypted drive. Those claims data will then be merged in the cold room with the employer-claim link file to produce a dataset (8. in the figure) containing claims data with actual allowed amounts, simulated Medicare payment amounts, and employer identifiers. That claims dataset will then be processed in the cold room to create employer-specific summarized price data (9. in the figure)—that summary data will include employer identifiers but will not include any individual-level health records and will not, therefore, include PHI.
Which employers and coalitions have been invited to participate?
We have reached out to a wide range of employer coalitions and individual employers, including:
- The Alliance (Wisconsin-based not-for-profit cooperative)
- Colorado Business Group on Health (participated in round 2.0)
- Economic Alliance for Michigan (participated in round 2.0)
- Employers Health
- Florida Health Care Coalition (participated in round 2.0)
- Houston Business Coalition on Health (participated in round 2.0)
- Kentuckiana Health Collaborative
- Memphis Business Group on Health
- Midwest Business Group on Health
- Minnesota Health Action Group
- Montana Association of Health Care Purchasers (participated in round 2.0)
- National Alliance of Healthcare Purchaser Coalitions
- New Mexico Coalition for Healthcare Value (participated in round 2.0)
- Northeast Business Group on Health
- Pacific Business Group on Health
- Rhode Island Business Group on Health
- South Carolina Business Coalition on Health
- St. Louis Area Business Health Coalition (participated in round 2.0)
- Washington Health Alliance
- Wyoming Business Coalition on Health (participated in round 2.0)
- But, all others are also invited to participate
What is RAND? What is EFI?
From RAND’s website (https://www.rand.org/about.html), “The RAND Corporation is a research organization that develops solutions to public policy challenges to help make communities throughout the world safer and more secure, healthier and more prosperous.” RAND is a nonprofit 501(c)(3) headquartered in Santa Monica, California with offices in Washington, D.C., Pittsburgh, and Boston.
From EFI’s website (https://employersforumindiana.org/), “The Forum is an employer-led health care coalition of employers, physicians, hospitals, health plans, public health officials and other interested parties. Our goal is to improve the value payers and patients receive for their health care expenditures.”
Who can I contact for more information?
- Emily Hoch (firstname.lastname@example.org);
- Chapin White (email@example.com, 703-413-1100 x5684);
- Chris Whaley (firstname.lastname@example.org, 310-393-0411, x7969);
- Gloria Sachdev (email@example.com, 317-847-1969).
Does this study fall in the antitrust “safety zone”?
The Federal Trade Commission (FTC) and the U.S. Department of Justice (DOJ) share responsibility for monitoring mergers and anti-competitive behavior, and protecting consumer interests through enforcement of antitrust law. The FTC and DOJ in 1996 released guidance describing their general approach to antitrust enforcement in the health care industries,6 and the FTC and DOJ have issued more-recent guidance relating specifically to Accountable Care Organizations7 and to the public disclosure of contracts between health plans and providers.8
Hospitals and health systems would put themselves in legal jeopardy with the FTC and DOJ if they engaged in private exchanges of information regarding prices and costs for anticompetitive purposes (“price fixing”). The FTC and DOJ recognize, however, the potential benefits of public exchanges of health care price and cost information, and they have defined a “safety zone” for such exchanges. Those exchanges will not be challenged if “(1) the survey is managed by a third-party, … (2) the information provided by survey recipients is based on data more than 3 months old; and (3) there are at least five providers reported data upon which each disseminated statistic is based, no individual provider’s data represents more than 25 percent on a weighted basis of that statistics, and any information disseminated is sufficiently aggregated such that it would not allow recipients to identify the prices charged or compensation paid by any particular provider.”9
This study satisfies conditions (1) and (2) for the safety zone, but not condition (3)—the reporting of hospital-specific prices falls outside the safety zone. But, as the FTC and DOJ make clear, “public, non-provider initiated surveys may not raise competitive concerns,” as long as they are “for procompetitive purposes.” The current study, given that it is initiated and supported by employers in their role as purchasers of health care, is clearly procompetitive in its intent, execution, and impact.
The Center for Improving Value in Health Care (CIVHC), the not-for-profit organization that administers Colorado’s all payer claims database, analyzes and publicly reports provider-specific price and cost data similar to the public price reports that will result from this study. CIVHC has shared a legal opinion supporting those exchanges, with the key takeaway being that public reports, even if they fall outside the safety zone, are generally permissible “unless competitor recipients of the reports used the information to enter into price-fixing agreements.”10
What is the timeline for round 3.0?
Updated March 17, 2020. At this time, the release of results from RAND 3.0 is unknown due to the changing situation with COVID-19. We will provide more information as it becomes available.
Initial timeline: Every spring, most self-funded employers begin their design their health plans and benefits in preparation for open enrollment for the following plan year. The timeline for the project has been set with the goal of making the reports available in time to be relevant and useful to that process.
|March, 2019||Begin recruitment of self-funded employers, APCDs, and health plans|
|July, 2019||Have agreements in place between RAND and employers, DUAs in place between RAND and health plans/APCDs, and authorizations sent by self-funded employers to their TPAs|
|October, 2019||Data delivery complete|
|November, 2019||Data testing and analysis, drafting of public report|
|December, 2019||Draft public report submitted to RAND quality assurance process (QA), private employer-level reports generated|
|Early 2nd Quarter, 2020||Public report finalized and made public online, private employer-level reports distributed|
The release of the public report from round 2.0 was delayed due to: 1) ironing out DUAs between RAND and health plans, and 2) confirming that data deliveries include the necessary fields and re-extracting the data when necessary. Given our experience with round 2.0, we have overcome these challenges, but we also know how difficult this undertaking is. Achieving the milestones listed above for round 3.0 will require steady progress by RAND and all participating employers, health plans and APCDs. Please let Chapin White (firstname.lastname@example.org) know if you have questions or concerns about the timeline.
1 Cooper, Zack, Stuart V. Craig, Martin Gaynor, and John Van Reenen, The Price Ain’t Right? Hospital Prices and Health Spending on the Privately Insured, May 7, 2018. http://www.healthcarepricingproject.org/sites/default/files/20180507_variationmanuscript_0.pdf
2 White, Chapin, Hospital Prices in Indiana: Findings from an Employer-Led Transparency Initiative, RR-2106-RWJ, October, 2017. https://www.rand.org/pubs/research_reports/RR2106.html
3 White, Chapin, Hospital Prices in Indiana: Findings from an Employer-Led Transparency Initiative, September 20, 2017. http://employersforumindiana.org/media/2017/09/Hospital-Prices-in-Indiana-Findings-Chapin-White-9-20-17-updated.pdf
4 RAND Corporation, Hospital Price Comparisons in Indiana, 2017. https://www.rand.org/health/projects/indiana-hospital-prices.html
5 Department of Health and Human Services, Guidance Regarding Methods for De-identification of Protected Health Information in Accordance with the Health Insurance Portability and Accountability Act (HIPAA) Privacy Rule November 26, 2012. https://www.hhs.gov/sites/default/files/ocr/privacy/hipaa/understanding/coveredentities/De-identification/hhs_deid_guidance.pdf
6 U.S. Department of Justice, and Federal Trade Commission, Statements of Antitrust Enforcement Policy in Health Care, August, 1996. http://www.ftc.gov/sites/default/files/attachments/competition-policy-guidance/statements_of_antitrust_enforcement_policy_in_health_care_august_1996.pdf
7 Federal Trade Commission, and Department of Justice, “Statement of Antitrust Enforcement Policy Regarding Accountable Care Organizations Participating in the Medicare Shared Savings Program; Notice,” Federal Register, Vol. 76, No. 209, October 28, 2011. http://www.gpo.gov/fdsys/pkg/FR-2011-10-28/pdf/2011-27944.pdf
8 Federal Trade Commission, Amendments to the Minnesota Government Data Practices Act Regarding Health Care Contract Data, June 29, 2015. https://www.ftc.gov/system/files/documents/advocacy_documents/ftc-staff-comment-regarding-amendments-minnesota-government-data-practices-act-regarding-health-care/150702minnhealthcare.pdf
9 U.S. DOJ and FTC, 1996, p. 50.
10 Center for Improving Value in Health Care, Antitrust Legality of Reports and Analytic Data Sets Generated based on All Payer Claims Data, 2014. https://www.apcdcouncil.org/sites/apcdcouncil.org/files/media/state/final_anti_trust_summary_05-02-14.pdf