DOJ Antitrust Division, Business Review Request Letter 10-3, Pacific Business Group on Health (2009)
Assistant Attorney General
Antitrust Division
United States Department of Justice
950 Pennsylvania Avenue, NW
Room 3109
Washington, DC 20530
Re: Hospital Value Initiative Request for a Business Review Letter
Dear Ms. Varney:
This letter constitutes a request for the issuance of a business review letter pursuant to Department of Justice Business Review Procedure, 28 C.F.R. § 50.6. On behalf of the Pacific Business Group on Health, the California Public Employees' Retirement System, and the California Health Care Coalition, we are requesting a statement of the Department's enforcement intentions regarding a proposed Hospital Value Initiative ("HVI" or the "Initiative") to collect, aggregate, and analyze hospital claims data and then to report the results, in the form of efficiency scores, to health care purchasers, providers, and third party payors. This letter addresses the required components of a request for a business review letter, as well as the relevant elements of the analysis of the permissibility of the initiative under federal antitrust laws.¹
We emphasize at the outset that the HVI program is initiated and directed by health care purchasers on behalf of health care consumers. The primary purpose of the Initiative is to improve transparency of health care cost-efficiency information in the California market so that purchasers and payors can make better-informed decisions based on the cost and quality of the health care items and services they are purchasing. A key secondary goal of the Initiative is to participate in and contribute to the current national policy dialogue aimed at facilitating the introduction of value-based purchasing into the health care arena. The Initiative proposes to employ novel and sophisticated analytic methods that the participants expect will improve the public policy
¹ This letter supersedes in its entirety, including all Appendices, the original request letter submitted by my colleague Thomas Susman on November 15, 2007 or any other draft request letters that may have been provided to the Department since that date. We have provided under separate cover a redlined document showing the changes we have made to revise and update the original request letter.
knowledge base regarding measurement of value in health care. Further information about the broader policy context of which this Initiative is a part is set forth in Part V of this letter.
This request letter is submitted by the Pacific Business Group on Health ("PBGH"), California Public Employees' Retirement System ("CalPERS"), and California Health Care Coalition ("CHCC"). Payors who will provide the Initiative with hospital claims information have also contributed to the development of this request.
The request letter solicits the opinion of the Department of Justice, Antitrust Division ("DOJ") whether, if the requesting parties implement the proposed Initiative, the DOJ would recommend, under an antitrust rule of reason standard, no enforcement action with respect to the activities of the Initiative described in this request.
The question arises under Section 1 of the Sherman Act.
Planning for the HVI commenced in 2005 among PBGH, CalPERS, and CHCC. The Initiative, conceived and designed as a complement to ongoing hospital quality measurement initiatives, seeks to improve transparency in the California health care marketplace, particularly with respect to the cost and quality of hospital services. Specifically, the HVI will collect and analyze hospital claims data reported by major payors in California to develop scores of hospital resource use and cost-efficiency, without reporting actual prices. In addition to obtaining commitments to participate from payors, the Initiative has solicited input from hospital stakeholders to refine methodological details.
The participants recognized that the proposed Initiative involves the collection and dissemination of information to participants that could implicate the antitrust laws and therefore engaged antitrust counsel to advise on appropriate methodologies, procedures, and safeguards to ensure that the Initiative achieves its procompetitive objectives without any material risk of anticompetitive effects. The purchaser and payor participants, in consultation with their respective antitrust counsel, strongly believe that the proposed Initiative is procompetitive and would not lessen competition in any relevant market. Nevertheless, certain reporting hospitals have expressed reservations about sharing the information contemplated by the Initiative without first obtaining an opinion from the antitrust authorities that there is no present intent to challenge the Initiative.
The HVI comprises a broad coalition of purchaser entities working closely with payors and providers. The collaborating entities include nonprofit, proprietary, and governmental organizations. By including representatives from each of these constituencies, the HVI will benefit from more robust and reliable methodologies, data inputs, and results and thus will be more useful and relevant to all participants in the California health care market.
The HVI is a prospective initiative. No actions have been taken to formalize the agreements of the parties with respect to the HVI. Although it is possible that limited data collection and preliminary data aggregation may be undertaken prior to formal commencement of Initiative activities, any such collection and aggregation activities would be undertaken solely so that the collaborators can continue to refine and clarify the proposed analytic and reporting methodology. Any such preliminary data collection and aggregation would be undertaken by an independent third party, Milliman, Inc., on behalf of the senior stakeholder advisors and technical advisors to the Initiative and would not be disseminated to participating payors or affected providers, except to report broad, concept-level findings impacting the development of the methodology of the Initiative.
We understand that the DOJ and the Federal Trade Commission ("FTC") work closely together in issuing opinion letters. Although we are submitting this request for a business review letter to the DOJ, because of the FTC's historical interest in health care information sharing initiatives, we have discussed this Initiative with FTC staff and are fully prepared to provide corresponding information to the FTC at the request of either agency. We welcome comments or questions from either agency and are prepared to engage in discussions with both agencies throughout the request process.
The Pacific Business Group on Health ("PBGH") is a tax-qualified charitable nonprofit association of many of the nation's largest purchasers of health care, based in California. PBGH represents both public and private health care purchasers who cover over 3 million Americans, seeking to improve the quality of health care and access to affordable care. Since 1989, PBGH has been a catalyst promoting performance measurement and public reporting at every level of the health care system to improve performance and to help consumers make better choices. PBGH currently oversees the analytics and publication of the state's HMO quality report card, www.opa.ca.gov, and is also overseeing a ground-breaking project, part of CMS' Better Quality Information initiative, to measure and report quality performance of individual physicians. For more information, see www.pbgh.org and see the attached member list at Appendix A.
The California Public Employees' Retirement System ("CalPERS") is responsible for purchasing health care for approximately 1.2 million public employees, retirees, and their families and more than 2,500 employers. CalPERS is one of the largest non-federal purchasers of health care in the United States. CalPERS is committed to engage and influence the health care marketplace to provide medical care that optimizes quality, access, and affordability. Specifically, CalPERS recognizes that to achieve affordable, higher quality and improved access to health care, it needs to improve access to meaningful information on the quality, cost, and performance of health care providers, including hospitals, pharmaceutical companies, and physicians. CalPERS remains committed to strategies that impact the drivers of health care cost and quality within the provider community. For more information see http://www.calpers.ca.gov.
The California Health Care Coalition ("CHCC") is a membership organization of forty-two employers, unions, and health and welfare funds (Taft-Hartley Trusts), public agencies including school districts, and local governments, currently representing approximately 3 million Californians. Members of the Coalition are committed to seeking ways to improve access to affordable care and enhance quality of care without reducing benefits. CHCC seeks working partnerships with accountable, high value providers and health plans. Its goal is to assist hospitals and physicians to enhance their performance and delivery of care so that California families and communities have timely access to appropriate medical care that is evidence-based, patient-centered, prevention-oriented, and efficiently and cost-effectively delivered. For more information see www.chccnet.org and see the attached member list at Appendix A.
Since it is critical to the Initiative to obtain a sufficient sample of comparable data points, the Initiative has sought to include all California health care payors with a statewide presence that can provide comparable data. Thus, every California payor with a statewide presence, other than Kaiser Foundation Health Plan ("Kaiser"), has been invited to participate in the first round of data reporting, and the following payors have agreed to participate:²
a) California Physicians Service, d/b/a Blue Shield of California, a California nonprofit ("Blue Shield")
b) Health Net of California, Inc. and Health Net Life Insurance Company (collectively, "Health Net")
c) United HealthCare Services, Inc. on behalf of itself and its affiliates, PacifiCare Life and Health Insurance Company and PacifiCare of California ("United")
d) Blue Cross of California, Inc. ("Blue Cross")
² Note that participating payor contracts have not been executed; the list of participating payors contained herein constitutes the list of payors who have agreed in principle to participate in the Initiative.
e) Aetna Health of California, Inc. and Aetna Life Insurance Company (collectively, "Aetna")
Payor participation entails both financial support and agreement to provide hospital claims data as described below.
Many other organizations offer health care coverage in the commercial health insurance marketplace in California. These entities are described in more detail in Part III of this submission.
Kaiser Foundation Health Plan ("Kaiser") is not participating in the HVI for practical reasons, as Kaiser is a closed system from which no comparable data may be derived. The "costs" incurred by Kaiser in its owned facilities do not constitute arms-length reimbursement data, but instead, are more accurately viewed as intracompany charges. Consequently, were Kaiser were to participate in the HVI as a "plan," it would not receive a meaningful plan-specific report regarding "allowed amounts" at participating hospitals because it does not generally contract with non-Kaiser hospitals for network coverage. The same methodological issue applies for Kaiser hospitals. Moreover, including Kaiser data with respect to out-of-network services at non-Kaiser facilities would skew the data, as such out-of-network payments are hospital charges that must be paid in full and are not analogous to Allowed Amounts data analyzed by the Initiative.
In 2006, the combined statewide market share of the health maintenance organization ("HMO") and preferred provider organization ("PPO") product lines offered by the five payors participating in the HVI was approximately 67%. In that year, Kaiser's market share was approximately 27%, and the remaining major payors constituted about 5% of the overall market share. Information pertaining to the enrollment of each participating payor in the HVI-designated regions is attached to this submission at Appendix B.
Hospital facilities that provide inpatient and outpatient acute care services on behalf of participating payors will be asked to cooperate with the HVI. A list of hospital facilities (numbering approximately 330) that provide services to enrollees of participating payors, and whose claims are expected to be submitted to the Initiative for analysis, is attached to this submission at Appendix B. Following a waiver or renegotiation of applicable confidentiality provisions in provider network contracts, participating payors will provide to the Initiative claims data submitted to them by the hospital facilities (in- and out-of-network) providing care to their enrolled beneficiaries on whose behalf the payors remit payment.
PBGH, CalPERS, and CHCC have joined together to undertake the Initiative in furtherance of the objective of increasing transparency of health care information to enhance quality, efficiency, and competition in the health care market. Payors and purchasers will be able to combine the efficiency information resulting from the Initiative with California hospital quality information that is already available. Chief among these sources is a web site, www.calhospitalcompare.org, which is hosted by the California HealthCare Foundation. The quality information is obtained through a collaboration of purchasers, plans, and providers known as the California Hospital Assessment and Reporting Taskforce ("CHART"). PBGH, CalPERS, and CHCC are all active participants in the CHART process. Quality measures include both clinical indicators, such as mortality and complication rates and compliance with evidence-based hospital practices, and patient experience information gathered through surveys. This information is collected and displayed by certain key clinical categories – for example, maternity, care for heart attacks, and pneumonia – rather than at the overall hospital level. One reason, therefore, that the Initiative proposes to obtain information on cost-efficiency at the same level of granularity is to allow for an analysis that combines cost-efficiency with quality information and produces a measure of value. The HVI collaborators believe that decisions by payors and consumers should be based on value, not on cost-efficiency or quality alone.
Two types of comparative measures will be calculated and ultimately reported: "cost-efficiency" and "resource use":
a) Cost-efficiency calculations, which are scores calculated by analyzing actual cost data, measure the relative total reimbursement paid by patients and payors for the treatment of specific conditions. Cost-efficiency allows a payor to compare the costs at one hospital facility relative to another facility and relative to the average of all facilities for the same or similar conditions.
b) Resource use calculations, which are developed using standardized and not actual costs, measure the quantity and mix of bed days by bed type used to treat a condition within the facility. Resource use identifies hospitals that are relatively inefficient in the use of resources, even when they appear to be relatively cost-efficient based on their reimbursement rates for given services. These measures will assist individual participants to help identify and improve the delivery of patient care on a cost-efficient basis that takes into account resource use.
Key principles identified and adhered to in the development of the HVI are:
a) The Initiative is led by purchasers who represent the interests of employers and consumers in the health care marketplace. Lack of transparency and information asymmetry in the health care marketplace has meant that purchasers and consumers are forced to make decisions about their health care services in the absence of adequate information about relative cost and quality of those services.
b) The purpose of the Initiative is to increase transparency in the health care marketplace to better inform purchasers and, ultimately, their constituent consumers when making health care purchasing decisions. Other mechanisms, such as health plan and provider "report cards," exist or are in development; all of these share common goals of improving transparency and information in the health care market, bolstering health care competition, and ultimately improving access to affordable, quality care.
c) Participation in the Initiative of health care payors that operate on a statewide basis confers at least three benefits: (1) it utilizes the largest number of comparable observations, which will enhance the robustness, accuracy, and usefulness of the results in any given geographic region; (2) it expands the geographic coverage of the Initiative to include more regions throughout the California marketplace; and (3) it facilitates the development of objective standards that are accepted on an industry-wide basis to measure hospital quality and efficiency.
(1) The more comparable data points that are provided as inputs, the more useful the Initiative outputs will be. This is because the relative performance scores that are planned to be assigned to hospital facilities will be more robust, accurate, and informative when they are based on underlying data from a comparable and sufficiently large sample of payors and providers. The participation of multiple statewide players proposed by the HVI therefore is necessary to meet the participants' goals of producing accurate, useful, and comparable information for purchasers and consumers to make informed health care decisions.
(2) Reporting from multiple payors and providers also will expand the geographic scope of the Initiative. The purchasers that are driving this Initiative represent employers and consumers that are located throughout California and would benefit from information on provider performance in regions throughout the state to better inform their health care coverage decisions.
(3) Hospital cooperation also is important to the success of HVI and other health information transparency initiatives as stakeholders continue to work to develop common standards to measure provider quality and efficiency. Provider involvement and input in developing appropriate quality and efficiency measures will not only help them to assess their current performance, but also will enable facilities to be responsive to consumer needs and preferences going forward.
d) The Initiative will provide information that can be used to evaluate hospital provider performance based on an analysis of historic, aggregated claims data. HVI participants will continue to make independent health care purchasing decisions, and they may use the data reports produced by the Initiative in any lawful way that they find useful.³ The Initiative is intended to allow employers and other health care purchasers and consumers to make informed decisions in selecting the most appropriate health care services to meet their needs.
In this section, we describe the nature of the information to be collected, analyzed, and disseminated by the HVI. Where available, illustrative hypothetical examples will be included or appended. A Milliman memorandum dated May 12, 2009 describing the Hospital Value Initiative Methodology (the "Milliman Methodology Memorandum") and a list of data fields to be reported to Milliman are attached to this submission at Appendix C. Although we are aware that agency guidance regarding the submission of an advisory opinion request recommends the inclusion of samples of the information to be exchanged, the sponsors of the HVI have not yet been able to obtain from hospitals or payors samples of the actual hospital claims data that would be assessed by the Initiative.⁴ Some minor methodological questions or issues will likely remain unresolved until actual claims data is aggregated and statistically analyzed. However, none of these minor issues will materially affect the content of the HVI reports discussed in Section II.D, below.
Data will be collected by the HVI from participating payors. Payors will provide inpatient and outpatient hospital services claims submitted by hospital facilities, whether contracted or not (in- or out-of-network). The claims data will be provided at the claim line level for units of service. Claims data will be provided by each payor to an independent third party, Milliman Inc., which will perform all data collection, aggregation, data analysis, and presentation of results at the direction of the HVI. No HVI participant will have access to any other participant's raw data or to any of the disaggregated data collected.
Data submitted to Milliman will include: (a) overall costs paid for specified services on a facility-specific basis, including both payor reimbursement and patient cost-sharing (such as coinsurance or copayments) ("Allowed Amount"); and (b) the dollar figure that a hospital charges for that specified set of services, which in many cases is greater than the Allowed Amount ("Charges" or "Billed Amount"). It is estimated that, for 2005, there are more than 1.1 million unique discharges that
³ As discussed below, participants will expressly agree in the HVI Participation Agreement that they will conduct all activities related to the Initiative in compliance with the antitrust laws, and the Initiative will implement an antitrust compliance program.
⁴ The proposed report format for payors and hospitals is discussed in Part II.D and at Appendix C. In developing and discussing the proposed analysis and reporting format, the collaborators have determined that it would be enormously complicated to prepare a "mock" database for the purpose of conducting a simulation of the proposed analysis because any such database would need to contain millions of fictional data points. The Milliman Methodology Memorandum outlines the proposed methodology using a small sample of claims records. An analysis and report based on large amounts of publicly available data is described in Part V and appended to this submission; however, as noted below, the utility of the analysis using public data is limited.
would be included in the Initiative database. The number of data points for each discharge is estimated to range from one to several dozen.
In its simplest terms, the cost efficiency values that will presented by the HVI will consist of a numerator that reflects the average Allowed Amount for a unit of service at a particular hospital or hospital peer group over a denominator that reflects the average Allowed Amount for that same service across all of the participating hospitals in California.⁵ Consequently, if the average allowed cost of a particular service at a particular hospital or hospitals is less than the statewide average Allowed Amount, the Buyer Cost Index (“BCI”) will be less than 1.0. Conversely, if the average allowed cost is greater than the statewide average Allowed Amount, the BCI will be greater than 1.0.
The sponsors and Milliman believe that Allowed Amounts provide the most accurate tools to measure efficiency in that they constitute, for a given unit of service, the overall costs paid by the payor plus the total cost-sharing payments made by the patient. To develop the BCI scores, hospital claims data will be collected by Milliman by unit of service provided, or by a grouping of units of service falling into a given category (e.g., laboratory services).
For the inpatient services, reported claims will be classified and grouped by diagnosis and then risk adjusted using 3M’s APR-DRG and Severity of Illness Grouper (“APR-DRG/SOI”) algorithms. There are approximately 350 APR-DRGs under 3M’s methodology,⁶ and the HVI anticipates that claims for all of these APR-DRGs will be reported and included in the analysis.
Because a single APR-DRG (e.g., cesarean delivery) can embrace everything ranging from a straight-forward c-section to a very complicated (and costly) delivery, using the Severity of Illness Grouper ensures the fairness of the process by grouping and averaging Allowed Amounts in both the numerators and denominator at the APR-DRG/SOI level. As a result, hospitals that face more than their fair share of complicated procedures because they have certain specialty practices will not be penalized for taking the more complicated cases.⁷ The methodology proposed by Milliman will
⁵ For the All-Payor Report, the denominator will reflect the statewide average Allowed Amount of all payors participating in the Initiative (“All-Payor Statewide Average”); for each Payor Specific Report, the denominator will reflect the statewide average Allowed Amount for that particular payor (“Payor Specific Statewide Average”). The All-Payor and Payor Specific Reports are defined and described in Section II.D.
⁶ The APR-DRG/SOI groups are developed by 3M and are not the same as Medicare’s Diagnosis Related Groups.
⁷ While results will be calculated at the APR-DRG/SOI levels, the current plan is to report results only down to the APR-DRG levels. Because the methodology accounts for more complicated and more costly procedures in this manner, the HVI does not plan, at this time, to make further adjustments based upon the particular profile of the hospital (e.g., teaching hospital, children’s hospital). Instead, as noted in greater detail below, hospitals will be assigned to one or also make certain other adjustments to claims data, such as excluding claims with third-party payments and linking related and newborn claims, to ensure consistent reporting across plans and facilities.
In calculating the denominator for inpatient services, the HVI presently plans to use the statewide average Allowed Amount (All-Payor and Payor Specific Statewide Averages, for the All-Payor and Payor Specific reports, respectively).
In addition to the Buyer Cost Index values derived using the processes described above, the HVI reports will also contain publicly available information relating to the relative weight of public and private reimbursement sources to each hospital's overall reimbursement profile. The California Office of Statewide Health Planning and Development ("OSHPD") makes public certain information about overall hospital payor mix (expressed as number of patient days attributable to certain public payors and private insurance options, including managed care, and uncompensated care). These data are reported at the hospital level; therefore, each report will include a payor mix value at the hospital level, reflecting a hospital's overall payor mix, and not payor mix information at the more granular APR-DRG or MDC levels. The hospitals have expressed the view that it is important for payors and purchasers to have this information readily at hand as they consider and compare costs across hospitals. It bears repeating, however, that the information relating to the payor mix that will be added to the reports is already freely available from the California OSHPD.
Outpatient services create more of a challenge in terms of establishing the average costs at the hospital or hospital peer group (the numerator) and the comparable statewide average benchmark (the denominator). First, the challenge of grouping like procedures and severity levels for outpatient services is much more complicated. As compared to inpatient services, where virtually all procedures can be captured with just 1,256 APR-DRG/SOIs, the HCPCS lists almost 14,000 different procedure codes for outpatient services. Second, outpatient codes tend to be consolidated and rolled up for purposes of billing and reimbursement, resulting in a lack of clarity regarding allocation of payments to specific outpatient procedures. Consequently, the relatively precise tools that are more readily available for inpatient services are simply not available to analyze and compare outpatient services.
To solve these issues, the HVI intends to aggregate results across five relatively broad categories (Emergency Room, Surgery, Radiology, Lab/Pathology and Other) to minimize the potential impact of such consolidation and utilize Milliman's RBRVS methodology to establish statewide benchmarks.
Milliman's RBRVS methodology uses cost and billed charge relationships from all payor sources to estimate the relative resources for each HCPCS. The time and resource requirements for each
more peer groupings (region, hospital specialty, size) and, by making the results sortable by peer group, the final reports will permit direct peer group comparisons.
HCPCS are specified by Relative Value Units ("RVUs"). To establish the denominator ("Statewide RBRVS-Based Benchmark"), Milliman will take the total billings for outpatient services across all reporting hospitals and divide that by the total number of statewide outpatient RVUs to arrive at a statewide outpatient conversion factor. That factor will then be multiplied by the RVU assigned to each of the HCPCS procedure codes to establish Statewide RBRVS-Based Benchmarks for each of the 14,000 codes. Numerators for each hospital or hospital peer group will be the average Allowed Amount for each of the 14,000 codes. Results for both the denominator and numerators will then be grouped and aggregated within one of the five general outpatient service categories, and BCIs for each hospital or hospital peer group will be calculated based on those results.
HVI reports of inpatient services – both All-Payor and Payor Specific (as defined and described in Section II.D, below) – will contain resource use efficiency scores. To calculate resource use efficiency scores by hospital, actual hospital claims values will be replaced with standardized cost amounts for specific bed day types (e.g., Maternity, ICU, CCU, NICU). These standard costs will be applied at the level of the revenue code for bed type. Each hospital's performance will be measured by comparing its use of resources for inpatient services (bed days by bed type) in connection with a given APR-DRG/SOI category against a resource use statewide average metric for that same APR-DRG/SOI. These standard scores also will be aggregated by MDC and globally within the hospital.
This resource use evaluation methodology, by replacing actual claims costs with standardized cost amounts, holds the element of cost constant so that hospitals are evaluated on their use of resources (by bed day type) for a given diagnosis at a given level of severity. This element of the methodology therefore identifies hospitals that are relatively inefficient in the area of resource use, even when they appear to be relatively efficient in the area of costs based on the reimbursement rates they receive for given services.
The Initiative participants recognize that under-utilization, as well as over-utilization, can lead to poor health outcomes, and that high value health care incorporates both efficiency and quality – in other words, they are committed to empowering consumers to access what consumers believe to be "the right care at the right time." The participants therefore expect that the resource use efficiency scores will be interpreted in conjunction with, and in light of, existing measures of facility quality.
The participants also understand that the methodology to calculate resource-use efficiency scores using standardized costs does not raise antitrust concerns because proxy resource use figures are used in the analysis. As described above, the resource-use evaluation methodology specifically eliminates claims cost data from the calculation so that hospital facilities may be evaluated on their use of resources while holding the element of cost constant. The use of standardized costs to produce resource-use performance scores and standardized averages does not allow payors and hospitals to identify rates negotiated by competitors. The project sponsors believe that, to obtain an accurate picture of a hospital's performance, it is necessary to have information about both cost efficiency and resource use, and to pair that information with existing public quality information.
Claims data will be collected from the payors based on dates of service within the prior calendar year. Claims data are finalized in the four- to six-month period after the end of the year; thus the Initiative will allow for a six-month "run-out" after the end of the year prior to commencing the analysis. The analysis and report production process is expected to take about four months to complete after the finalized claims data have been collected from each of the payors. Therefore, all of the data collected and reported by the Initiative will be between ten and twenty-two months old.
The HVI anticipates electronically disseminating three basic reports (the "All-Payor Report," the "Payor Specific Report" and the "Hospital Specific Report") to three different recipient groups (Purchasers, Payors and Hospitals). Each report will consist of one or more customized pivot table formats, which will present the results as described in greater detail below. Report recipients will access the reports they are entitled to see through a secure web-site created and maintained by Milliman or an equivalent independent third-party contractor. In each instance, site access will be user name and password protected and recipients will only be able to access those reports for which they have been approved. All underlying and unaggregated data and metadata will be unavailable to recipients either through the site or by other means.
It is anticipated that each report will consist of a "Hospital Comparison Summary" and/or a "Hospital Specific Summary." Exemplars of these tables are provided as Attachments A and B respectively to the Milliman Methodology Memorandum attached as Appendix C and are described in greater detail below. At the outset we would note that while these exemplars contain the information that will be reported, the precise formatting or presentation of the information contained in these forms is continuing to be refined and adjusted.
As noted below, both the All-Payor and Payor Specific reports will contain a Hospital Comparison Summary like the example in Attachment A to the Milliman Methodology Memorandum. The top portion of the form displays the pull-down options that are available to the user to expand or contract the scope of the reported results for inpatient and outpatient care. In this example, the user has asked for comparative information at the "all services" level for both inpatient and outpatient care for all hospitals located in the San Francisco-San Mateo-Marin HVI region, purchased through PPO plan products. By manipulating the pull-down screens, the user can expand the data by selecting "All State" in lieu of a particular region or, in the alternative, drill down by grouping hospitals by Specialty or Size or by selecting particular MDC and/or APR-DRG codes. The user can view results for either PPO or HMO plans.
The results are reported by all hospitals that fall within the criteria selected by the user. This example reports seven hospitals in the region and reports Discharges/Cases, the Buyer Cost Index ("BCI"), and Resource Use Efficiency ("RUE") (inpatient services only) for each, as well as group totals and averages. Each hospital specific result will also show the regional BCI and RUE values for that hospital to facilitate comparisons. Regional values will be calculated simply by averaging the scores of all of the hospitals in each region.
The Hospital Comparison Summary contained in the All-Payor report will be based upon aggregated data from all participating hospitals collected from all participating payors. For each Payor Specific report, the same table will be used, but the information contained in the report will be based upon only the claims data reported by that payor alone. As noted below, Payor Specific reports will only be made available as a feature to that particular payor. Other payors, purchasers, and hospitals will not have access to Payor Specific reports.
All of the reports will contain data in the form of the Hospital Specific Summary for which an exemplar is provided as Attachment B to the Milliman Methodology Memorandum. Like the Hospital Comparison Summary, the Hospital Specific Summary relies on pull-down menus and expandable and collapsible folders to present relevant information. In the example provided, the user has selected the PPO data for Hospital A (a 500+ bed teaching hospital in the San Francisco-San Mateo-Marin HVI region) and has asked for the information at the APR-DRG level under MDC 17. Like the Hospital Comparison Summary, results in the Hospital Specific Summary report will display Discharges/Cases, BCI, and RUE (inpatient services only) at each level, and will contain regional and statewide BCIs and RUEs for hospitals in a particular region or other grouping as well to facilitate comparisons. Regional values will be calculated simply by averaging the scores of all of the hospitals in each region. Statewide values are the average across the state.⁸
It is anticipated that each Payor and Purchaser will be given access to the "All-Payor" report, which will consist of the Hospital Comparison Summary described above as well as Hospital Specific Summaries for each of the hospitals reported under the HVI. User names and passwords will be required to access the All-Payor report. In its initial configuration, access to the All-Payor report will be conditioned on the recipient's agreement that the contents will not be made public and will not be shared with the hospitals.
⁸ Statewide values will deviate from 1.0 only when peer group-specific reports are selected; e.g., Teaching Hospitals statewide, where the average BCI value would represent the average BCI for all Teaching Hospitals statewide.
Each payor will also be given access to its own Payor Specific Report. While the Payor Specific Report will be presented in the same Hospital Comparison and Hospital Specific Summary form as the All-Payor Report, the cost efficiency and resource use efficiency indices presented therein will be drawn exclusively from the data reported by that specific payor for its claims from all participating hospitals. No other payor, and no purchaser or hospital, will have access to a Payor Specific Report. Again, the material will be user name and password protected.
The HVI sponsors deem the information in the Payor Specific Report to be much more competitively sensitive than that contained in the All-Payor Report and so, at present, do not anticipate making these reports available beyond each specific payor.
Each hospital facility for which claims are submitted to the HVI will be able to access the Hospital Specific Summary for its particular facility on-line using the same user name/password protections that protect other on-line information. The Hospital Specific Summary reports cost efficiency and resource use efficiency scores derived from the specific hospital facility to compare the facility to other facilities at the region and statewide level. Regional values are calculated simply by averaging the scores of all of the hospitals in each region. Statewide values are calculated simply by averaging the scores of all of the hospitals across the state. The Hospital Specific Summary report does not permit comparisons by hospital size or type (i.e., teaching or children's hospitals).
The HVI collaborators have engaged in extensive discussions about the nature of the data contained in the proposed reports to ensure the validity of the analysis and results while preventing any possibility that it could facilitate (or be perceived to facilitate) anticompetitive coordination among the participants on hospital reimbursement. To this end, the collaborators have investigated whether any particular reports may permit so-called "reverse engineering" of the reported data so that participants might be able to infer or calculate the reimbursement levels paid or received by their competitors.
Such reverse engineering might theoretically occur only if two conditions are satisfied: (1) a report contains sufficiently few hospital facility or payor data points, or payor market shares are distributed in such a way, as to permit a participant to calculate others' reimbursement levels based on the reported statistic for that participant, and (2) the data themselves are sufficiently granular, relevant, and related to the health care service level at which payors and providers actually negotiate contracts.
As noted above, data will be aggregated and reported to hospitals and payors at the APR-DRG and MDC level. The MDCs, or service line level categories, are mutually exclusive categories that aggregate related APR-DRGs, and they often correspond to particular medical specialties and/or organ systems. In general, however, participating payors pay hospital providers under one of the following broad types of reimbursement schemes: per diem (e.g., $1,000 per day), case rate (e.g., $10,000 for an admission), or capitation (e.g., a per member per month payment reflecting estimates of expected health care utilization and costs and independent of actual admissions).
Reports at the APR-DRG level are slightly more specific than the MDC level, but again are not the reimbursement methodologies typically adopted by payors and providers. Therefore, even if payor participants were able to derive reimbursement information from the relative hospital score reports, such reverse engineering would yield data only at the MDC or APR-DRG levels, which are not customary or standard reimbursement methodologies.
The collaborators therefore believe that the nature of the data proposed to be reported by the Initiative is not conducive to coordination on prices or price-related terms because it has little, if any, relevance to reimbursement levels for the units of service that are actually negotiated between payors and providers.
The collaborators have considered proposals to weight or "mask," or simply not to report, certain data points for legal, practical, or proprietary reasons. The collaborators recognize that the use of masking mechanisms or decisions not to report certain data points would yield data outputs that are less robust and less useful to the collaborators. Indeed, weighting or excluding certain data points belies the primary purpose of the Initiative, which is to improve transparency in the health care market and to equip market participants with accurate, valid, and useful data to make informed health care decisions. The collaborators therefore request that this submission be evaluated as if no weighting or exclusion methodologies will be adopted.
As described in this section, HVI reports are proposed to be provided to purchasers, participating payors and the hospitals at which their enrollees received services. Thus, the antitrust laws are implicated to the extent that, in the distribution of cost-efficiency reports, aggregated health care claims information may be shared – among payors, on one hand, and hospitals, on the other – through the operation of the Initiative.
The antitrust safety zone in Health Care Statement 6 published by the FTC and DOJ describes information sharing activities that pose little risk of anticompetitive effects and generally are not subject to challenge by the antitrust agencies. Based on guidance provided by the staff of both agencies in prior advisory opinion and business review letters, we understand that, to fall within the safety zone for information sharing initiatives, the Initiative must meet all elements of the safety zone with respect to data shared among both payors and hospitals.
The HVI participants believe that the Initiative may well operate well within the bounds of the antitrust safety zone with respect to payors, hospitals, or both, in many geographic areas in California. However, the collaborators also understand that in some geographic locations, the requirements of the safety zone may not be met in full. Nonetheless, as discussed throughout this submission, the collaborators have developed the Initiative taking into account the intent and objectives of the key principles and purposes of the safety zone. Specifically:
a) The survey is managed and data are collected by an independent third party (Milliman);
b) Information is aggregated before it is disseminated; no underlying data of any participant are provided to any other participant;
c) The aggregated information will be, at a minimum, almost one year old by the time it is reported; and
d) Participation in the Initiative is open to all market participants that meet objective eligibility criteria to ensure they can provide useful and comparable data.
As noted above, artificially restructuring the Initiative to conform to the safety zone (such as by excluding or weighting certain data points) is likely in some cases to severely compromise the validity and utility of the results and the geographic coverage of the Initiative. The collaborators remain fully committed to producing and reporting the most useful data possible while structuring the Initiative so that dissemination of these data is unlikely to have an anticompetitive effect, consistent with the rule of reason and the principles of the safety zone.
To ensure that the Initiative is able to produce the most reliable, accurate results (which will be critical to its marketplace acceptance and utility), in the initial stages of the Initiative claims information will be sought from payors that operate on a statewide basis and have comparable claims data. Although these participating payors do not represent all of the entities that offer health coverage in California, it is believed that the initial participating payors will provide a sufficient number of comparable observations that will be manageable to collect and analyze and thus will optimize the Initiative's ability to provide reliable results and reports in its critical first phase of operation.
As described in Part II above, wherever possible, the Initiative will include aggregated data from at least five participating payors and providers in each region surveyed. Furthermore, in many geographic markets, the numbers of payors and providers are far greater than five. These factors render the possibility of coordination on prices or output or other anticompetitive effects resulting from the dissemination of Initiative reports highly improbable.
In California, there are a multitude of health plans and other entities of all shapes and sizes that compete to offer provider networks and all types of health insurance. According to InterStudy data, there are dozens of entities offering health care coverage in the commercial health insurance marketplace in California today. These include:
1. National and multi-state carriers offering plans under California's Knox Keene Act, including Aetna, Blue Cross of California, United, CIGNA HealthCare of California, Health Net, and Kaiser;
2. Local Knox Keene plans operating in selected regions or cities in California;
3. Companies such as Great West Healthcare, Guardian Life Insurance Company of America, Metropolitan Life Insurance Company, Principal Life Insurance Company, and Trustmark Insurance Company, with insurance plans that are regulated in California by the Department of Insurance, and that may offer coverage in other states under those states' insurance regulations;⁹
4. Entities that act as third party administrators and/or provider networks, such as Beech Street Corporation, First Health, and MultiPlan.¹⁰
One of the principal findings from the economics literature on collusion and cooperative behavior is that coordinated behavior is likely to succeed only if there are a relatively small number of competitors in the marketplace.¹¹ This is true regardless of whether the coordination involves the downstream output market (in which payors compete to sell health care coverage to employers and individuals) or the input market (in which payors compete by contracting with providers to provide health care services to their members).¹²
The large number of entities purchasing health care services from providers in California renders coordinated interaction in the purchase of provider services or the sale of health insurance unlikely.
However, as described further below, even in regions where there are fewer than five participants or competitors, the nature of the data being disseminated, the safeguards adopted in the Initiative, and other characteristics of the California health care marketplace still make it highly improbable that the Initiative could facilitate coordination on prices or output among health care payors or otherwise have anticompetitive effects in any relevant market.
⁹ California Department of Insurance, "California Life & Annuity Insurance Industry, 2004 Market Share Report," August 2005, ("2004 Market Share Report"), Exhibit (4D), available at http://www.insurance.ca.gov/docs/FS-MarketShare.htm.
¹⁰ HealthLeaders-InterStudy Spring 2005 Managed Market County Surveyor.
¹¹ See, e.g., D. Carlton and J. Perloff, Modern Industrial Organization 3rd ed. (New York: Addison-Wesley, 2000), Chapter 5; A. Jacquemin and M. Slade, "Cartels, Collusion, and Horizontal Merger," in R. Schmalensee and R. Willig, Handbook of Industrial Organization Vol. 1 (Amsterdam: North Holland, 1989); G. Stigler, "A Theory of Oligopoly," Journal of Political Economy 72 (1964); L. Pepall, D. Richards, and G. Norman, Industrial Organization: Contemporary Theory and Practice 2nd ed. (Mason, OH: Southwestern, 2002), Chapter 7; and J. Tirole, The Theory of Industrial Organization (Cambridge, MA: MIT Press, 1988).
¹² See, e.g., R. Blair and J. Harrison, Monopsony: Antitrust Law and Economics (Princeton: Princeton University Press, 1993), pp. 42-44.
The California marketplace is characterized by a variety of competitors, offering differentiated products, with a range of different benefit designs. The diversity of product offerings in California ranges from Kaiser's staff model HMO products and other HMO models based on capitated financing or gatekeeper/non-gatekeeper variations to fee-for-service open access PPO products. Although all of these commercial payors and products compete to provide health care coverage in the marketplace, this diversity in product offerings makes reaching terms of coordination on price or other competitive terms difficult if not impossible.¹³
On the purchasing side, the story is no different.¹⁴ The wide range of contracts with substantial variations in payment terms, reimbursement rates, and other provisions makes coordination on provider pricing terms improbable. The variety of business models that health plans and providers use to organize and run their businesses and pay providers also strongly militates against any possibility of tacit or explicit coordination in input or output markets. These models differ in various ways, including the reimbursement method (fee-for-service, capitation, and salaries), degree of provider choice and network access (broad networks with no gatekeepers vs. narrow networks with gatekeepers), ownership/corporate structure (investor-owned vs. provider-owned, for-profit vs. not-for-profit), and product offerings (fully-insured vs. self-insured, HMO vs. PPO vs. Point-of-Service).
Moreover, health plans maintain their own internal claims processing practices and reimbursement policies. All of the health insurance companies use their own provider reimbursement policies. The Initiative does not provide additional information to payors that would allow them to learn what each of its competitors is doing with respect to those internal and confidential claims processing practices or physician reimbursement policies.
In sum, that these are differentiated products and that there is variation among payors and their products make it significantly more difficult for tacit coordination to occur than in industries involving commodity products and relatively few sellers.
¹³ See Department of Justice & Federal Trade Commission, HORIZONTAL MERGER GUIDELINES § 2.11, available at http://www.usdoj.gov/atr/public/guidelines/horiz_book/hmg1.html (hereinafter "MERGER GUIDELINES"); see also New York v. Kraft Gen. Foods, Inc., 926 F. Supp. 321, 342 (S.D.N.Y. 1995) (multi-dimensional competition makes coordinated conduct impractical). In addition, the variation in health plan costs, resulting from the many different relationships and negotiated rates that health plans have with often the same providers, complicates any attempts to coordinate.
¹⁴ See, e.g., D.Carlton and J.Perloff, Modern Industrial Organization 3rd ed. (New York: Addison-Wesley, 2000), Chapter 5; G. Stigler, "A Theory of Oligopoly," Journal of Political Economy 72 (1964).
The federal antitrust agencies and the courts have recognized the procompetitive benefits of information sharing initiatives, including that they can have significant benefits for health care consumers.15 Indeed, the objectives and intent of the Initiative are inherently procompetitive: to enhance health care marketplace transparency, efficiency, and competitiveness and ultimately to support purchasers in making more informed decisions when buying health care services.
A recent Congressional Research Service Report for Congress16 supported the hypotheses and goals of the HVI, concluding that it is reasonable to believe that greater price transparency would improve overall outcomes in the health care market, including decreasing price and increasing access to health care by the indigent. The report suggested that more health care price transparency efforts must be undertaken to further develop understanding of the operation of health care markets. The groups convening the Initiative seek to facilitate broad public policy goals by developing useful measures that can be reviewed, adopted, and adapted by other organizations – public and private – seeking to improve price and cost transparency in the health care marketplace.
The structure of the Initiative and characteristics of the reported data preclude tacit or explicit coordination. The HVI is a purchaser-driven initiative. Third party payors and hospital providers may agree to cooperate with the project, but these entities did not initiate and will not direct the activities of the HVI. Further, as the Horizontal Merger Guidelines recognize, there is little risk of anticompetitive coordination where firms “have substantially incomplete information about their rival’s business” and where products operating in the market are heterogeneous or differentiated.17 The Initiative would not give payors or providers any greater insight into the actual reimbursement rates or other price-related terms paid to or received by any individual participant. Furthermore, the ability to effectively coordinate on reimbursement rates would require not only that all participants have precise and current information on the myriad fee schedules and claims processing and reimbursement procedures that each utilizes, but would also require them to develop a common set of fee schedules and claims processing and reimbursement policies, notwithstanding significant differences in reimbursement methods (e.g., capitation vs. fee for service) and other differences in business models that currently exist.
As described in more detail in Part II above, the Initiative will not collect data or report any data on fee schedules or claims processing or reimbursement procedures. Thus, due to the nature and characteristics of the data reported by the Initiative, neither payors nor providers could derive data points that they would find useful for the purpose of setting reimbursement terms in their own contracts. In sum:
1. Relative scores, reported as a proportion, make "reverse engineering" of reimbursement infeasible and of little, if any, utility in assessing reimbursement rates
2. Reports at both MDC and APR-DRG levels are not useful for purposes of negotiation of contract reimbursement amounts
3. Reports are of past data on average more than one year old (and never less than ten months old)
4. Data are collected, aggregated, analyzed, and reported by an independent third party
5. Reports are for information only; participants will continue to make independent decisions on network contracting, price, quality, and other business decisions, and the HVI will implement an antitrust compliance policy
Thus, the nature and characteristics of the data collected ensure that any theoretical risk of tacit or explicit coordination on reimbursement rates will be de minimis.
It is theoretically possible that, after receiving Initiative reports, individual payors may be observed taking apparently simultaneous, but coincident, action to improve their networks by disfavoring poor-performing hospitals and favoring better performing ones. Any such actions are not evidence of collective action, much less of an unlawful collective boycott.¹⁸ Actions resulting from the Initiative are expected solely to be the result of a unilateral decision by an individual purchaser or
¹⁸ A per se analysis of potential boycott activity is reserved for situations in which there is a collective exercise of market power by firms to discourage suppliers or customers from doing business with a competitor. FTC v. Indiana Federation of Dentists, 476 U.S. 447, 458 (1986). The HVI participants have procompetitive purpose and intent; therefore, any potential for collective boycott resulting from the Initiative should be analyzed under the rule of reason.
payor to improve health care product offerings in light of the information presented by the Initiative and cannot reasonably be viewed as an inappropriate exercise of market power or a concerted refusal to deal. As discussed below and at Appendix D, efficient, high-quality providers are likely to benefit from the Initiative and less efficient providers will be incented to improve their performance.
As will be discussed below, options other than exclusion exist for addressing findings that a hospital may perform poorly in the areas of efficiency or quality. Indeed, as more information about provider performance becomes public each year, products have become more sophisticated, but collective boycotts have not resulted.
It is expected that third party payors or hospital providers will acquire, through Initiative reports, general information about how their reimbursement or price levels compare to those of their competitors. It is theoretically possible that one of the effects of the Initiative is that these entities could decide to offer prices or reimbursement that come closer to a presumed "average" price or reimbursement level.19 "Converging on the average" – a potential outcome for any price transparency initiative – is not inherently anticompetitive. The structure and performance of the health care market suggest that it is unlikely that the Initiative would lead to a uniform "averaging" of prices.
Public payor fee schedules – the most prominent of which are the Medicare provider fee schedules – have long influenced the commercial health care market. Notwithstanding the existence of these and other widely available health care price benchmarks, they do not appear to have driven commercial prices to inefficient levels. In fact, although health care payors often incorporate publicly available price and fee information into their reimbursement negotiations, they do not apparently follow them uniformly. Instead, payors have developed an enormous variety of reimbursement mechanisms, payment structures, and benefit designs; at the behest of consumers and purchasers, they employ these myriad business tools in a variety of ways to lower costs, improve quality, streamline delivery of services, and increase profits or surplus.
The HVI organizers and participants recognize that quality is an essential component of value-based purchasing. The primary objective of the Initiative is not necessarily to facilitate or contribute to an overall lowering of health care prices, but rather to improve overall information about the health care market so that purchasers and consumers can understand the relative value of care and can make informed decisions to pay for higher cost, better quality health care as they desire (discussed in Part IV.D., below). Even though public payor reimbursement information is already widely available, the U.S. health care market continues to be characterized by widespread opacity, leading to many potential operational inefficiencies because market participants do not have adequate
19 As discussed above, any determination of what that average price or reimbursement level is would have to be proactively derived by the individual payors and providers in the Initiative, as specific reimbursement information will not be reported by the Initiative.
information to make educated price, reimbursement, and purchasing decisions. Better alignment between prices on one side and quality and efficiency is fundamentally procompetitive. Indeed, the beneficial effects of cost and price transparency have long been recognized as helping to “avoid the waste which inevitably attends the unintelligent conduct of economic enterprise.”20
A variety of potential business activities may result from the Initiative. They include: (1) development or refinement of “narrow network” products or tiered-pricing solutions; (2) designation of hospital facilities as “Centers of Excellence” (by service line or at the facility level); (3) negotiations to reward hospitals for performance found to be more efficient/higher performing; and (4) collaborative programs to improve quality and value of services. Removal of a facility from a payor network is one, but certainly not the only (nor a primary) possible outcome of the initiative. A further description and explanation of potential uses of data reported by the HVI is attached as Appendix D. Potential uses of the data are expected to be consistent with the general trends that can be observed in the health care market, described briefly in Part V below.
As noted in the document attached as Appendix D, the Initiative will also encourage and enable hospital actions to improve efficiency and contain costs, since relevant and reliable comparative data will be provided to them as well. Boards of Trustees, parent corporations, and others can utilize the HVI reports to set performance goals and hold administrators more accountable.
As described above, the aggregated information that will be disseminated to participants is not of the type that poses significant antitrust risks. No information on any payor or hospital participant’s actual prices, costs, or strategic planning will be disclosed to any other participant. The conclusion that the Initiative will have procompetitive benefits without any material risk of anticompetitive effects is reinforced by the existence of competitive and relatively unconcentrated markets for health care coverage in California. In addition, competition in health care markets is driven by a range of variables beyond the cost of health care services, including the scope and quality of the provider network and terms of access, as well as a host of other benefit features.
In sum:
1. Procompetitive purposes and effects
2. No material likelihood of anticompetitive effects
3. No material likelihood that information exchange would facilitate tacit or explicit collusion or any other anticompetitive conduct/effects, including collusion on price/reimbursement and collective boycott
20 Maple Flooring Mfrs Assn v. U.S., 268 U.S. 563, 584 (1925).
The collaborators that have conceived of and are working to implement the HVI are leaders in the value-based purchasing ("VBP") movement, which seeks to improve the operation of the health care market by introducing methods of measuring and rewarding efficiency and quality in health care delivery. Two cornerstones – indeed, preconditions – of VBP are transparency of market price and quality information. The prior Administration recognized and reinforced the need for price and quality transparency when it issued Executive Order 13410 in August, 2006. E.O. 13410 defined the following four cornerstones for employers purchasing health insurance: (1) interoperable health care information technology; (2) reporting of quality-of-care measures; (3) reporting of health care price information; and (4) incentives for high-quality, cost-effective care. The overarching goal of the initiative was to help consumers make informed health care choices. E.O. 13410 directed federal agencies to develop health care quality measurement programs and to make available the prices that federal health care programs and their health insurance issuers or plans pay to providers for procedures.
The elements of E.O. 13410 are consistent with ongoing efforts by the Centers for Medicare and Medicaid Services ("CMS") to introduce quality measures and pay-for-performance metrics in its payment system. CMS has developed the "Hospital Compare" website providing comparative information on hospital quality. CMS has also launched the Physician Quality Reporting Initiative, which it views as an important element in the move toward VBP – "the key mechanism for transforming Medicare from a passive payer to an active purchaser." More recently, CMS announced that it is developing the Performance Measurement and Reporting System. Under this system of records, CMS will release physician identifiable health care data so that stakeholders can calculate and publicize physician quality and efficiency measures. CMS continues actively to work on implementing a VBP system for hospitals, first explored in an April, 2007 VBP Options Paper, and those efforts have bipartisan support in Congress.
These federal efforts, however, are limited to data arising from federal health care programs. State-based initiatives, public-private coalitions, and other efforts by private purchasers are essential to reinforce and supplement federal efforts. Through collaborative efforts such as HVI, private market data can be used to reinforce federal efforts to increase transparency and enhance publicly available information about health care providers.
In 2007, the Commonwealth Fund issued a report evaluating VBP efforts in four states.21 The report found that, if VBP initiatives are successful at overcoming identified political, technological, and logistical challenges, they have the potential to "influence providers to enhance quality and efficiency of care" and, ultimately, "to raise all boats ... for all users of the health care system, not just the current participants of VBP initiatives."22 The report identified transparency and public
21 SHARON SILOW-CARROLL & TANYA ALTERAS, VALUE-DRIVEN HEALTH CARE PURCHASING: FOUR STATES THAT ARE AHEAD OF THE CURVE (Commonwealth Fund 2007), available at http://www.commonwealthfund.org/usr_doc/1052_Silow-Carroll_value-driven_purchasing.pdf?section=4039.
22 Id. at x.
reporting of health care cost information as a “critical component” of VBP, and it suggested that the efforts of “mixed coalitions” of health care purchasers and suppliers have the potential to make a significant positive impact on the health care market.
In California, broad, coalition-based efforts are underway to improve the availability of health care cost and quality information. Many of the same payors participating in the HVI have agreed to provide quality-of-care information for their PPO product lines to the California Department of Insurance to develop a Health Insurance Report Card for California consumers. Report cards indicating quality of care and patient satisfaction are already produced for HMO products by the California Department of Managed Health Care.
Beginning in 2005, California hospital facilities were required, by an act of the state legislature, to begin submitting their chargemaster data to OSHPD. Unfortunately, the Congressional Research Service report cited above notes that this “price transparency” initiative has thus far had a limited effect on the level and dispersion of medical costs.23 The organizers of the HVI recognize that the chargemaster data now available in OSHPD is severely limited in its ability to improve price transparency in the market because chargemasters do not reflect actual prices paid on a service line basis.24 The HVI is thus designed to bring significant improvements to information that can be derived from the chargemaster data that is already publicly available by incorporating into the analysis actual reimbursement information at the service line level. In this way, the HVI is entirely consistent with California legislative intent to increase price transparency; the Initiative is designed – and expected – to improve upon the results of current efforts underway in California.
The Initiative participants and collaborators constitute a “mixed coalition,” and their efforts are consistent with the expressed policy goals of federal and state legislators and regulators. The HVI is therefore an integral component of the VBP policy movement in California. It is consistent with ongoing efforts to achieve the price transparency that purchasers need to make informed health care purchasing decisions and is likely to achieve generalizable results on which other private health care purchasers and policy decision makers can build.
At least three other health care improvement collaborations either are planning or are already engaged in efficiency measurement projects:
A. Care Focused Purchasing, Inc. (“CFP”) is a nationwide nonprofit employer- and carrier-led initiative involving over 50 employers and 9 carriers spanning the country. CFP members believe that a more transparent, rational market for health care will reduce cost pressures, improve quality of care, and reverse declines in consumer confidence. CFP is working to enable transparency through adoption of national standard measures of provider performance and aggregation of data from multiple sources to yield the most credible
23 AUSTIN & GRAVELLE, supra note 19, at CRS-26–CRS-30.
24 PBGH commissioned a cost-efficiency analysis that is based on this chargemaster data, which confirmed that the chargemaster data has limited usefulness. The published report of that analysis is attached to this submission as Appendix E.
measurement results. Current measurement efforts undertaken by CFP analyze hospital resource use efficiency using standardized, as opposed to real, costs.
B. The Integrated Healthcare Association ("IHA") (www.iha.org) is a not-for-profit statewide collaborative leadership group of California health plans, physician groups, and health care systems – plus academic, consumer, purchaser, pharmaceutical, and technology representatives – that promotes quality improvement, accountability, and affordability for the benefit of all California consumers through special projects, policy innovation, and education. IHA is developing, as part of its Pay for Performance ("P4P") program, an Efficiency Domain with the goal and desired outcome of lowering the cost of care without compromising quality of care. To achieve this goal, IHA is working toward: (1) developing a reliable, transparent, and valid set of efficiency measures; (2) implementing a trusted process of data collection and aggregation, yielding information to physician groups that improves the efficiency of care delivery; and (3) collaborating with health plans to implement meaningful incentives for physician groups to promote more efficient health care delivery. IHA anticipates measuring efficiency using the physician group as the unit of analysis, and including all aspects of health care services while maintaining confidentiality of financial and contractual arrangements between health plans, physician groups, and hospitals. IHA intends to implement these measures by incorporating them into its current P4P program.
C. The Leapfrog Group (www.leapfroggroup.org). On behalf of the millions of Americans for whom many of the nation's largest corporations and public agencies buy health benefits, The Leapfrog Group aims to use its members' collective leverage to initiate breakthrough improvements in the safety, quality, and affordability of health care. Founded in November 2000 by the Business Roundtable, Leapfrog promotes the use of nationally standardized measures of health care provider performance, public reporting on those measures, and the use of incentives and rewards to drive improvements in care. Using measures of risk-adjusted length of stay and readmission rates for particular procedures and conditions in hospitals, Leapfrog works to assess resource use and provide motivation for improved efficiency tied directly to improving quality of care. These measures had been collected through the Leapfrog Hospital Insights program and were integrated into Leapfrog's Hospital Quality and Safety Survey, a national, voluntary online survey of hospitals, in 2008.
Thus, just as HVI itself constitutes an important extension of federal transparency and accountability initiatives into the private sector, other organizations around the United States, including the three described above, also are considering how to appropriately bring to the health care marketplace information about health care value that is useful for both payors and patients.
The most recent drafts of documents reflecting or representing the agreement(s) among the parties to exchange information are attached to this submission at Appendix F.
One document discussing or relating to the legality or illegality under the antitrust laws of the information exchange or the impact of the information exchange is attached to this submission at Appendix G.
The parties to this request appreciate your consideration and a timely response. The HVI participants are confident that the Initiative is procompetitive and will yield significant benefits to the health care market in California and beyond. Indeed, the HVI is a methodologically robust, coalition-based approach to improve health care cost-efficiency and the overall operation of the health care market. Its efforts and goals are aligned with federal and state initiatives to bring value-based purchasing to health care and are ultimately of great consequence to the public fisc and the public's health.
Please do not hesitate to contact me if you have questions or seek further information about this request or the Initiative. We look forward to your comments and response.
Sincerely,
Mit Spears
Anne B. Claiborne
Enclosures
cc: Joshua Soven, Antitrust Division, United States Department of Justice