Department of Health & Human Services
Health Care Financing Administration
Medicare Managed Care
Operational Policy Letter #47 [revised]
OPL96.047
Date: April 14, 1997
Subject: New Reporting Requirements for Medicare Health Plans in 1997: HEDIS 3.0 Measures and the Medicare Beneficiary Satisfaction Survey
Note: This Operational Policy Letter replaces OPL96.047, dated December 23, 1996. There are several important changes which are identified here, along with reference to the page on which each appears. In addition, there are clarifications and updating related to implementation issues. This OPL does NOT reflect a change in HCFA's policy.
1. Plans are to follow the General Guidelines for Data Collection and Reporting in the final HEDIS 3.0 Volume 2, Technical Specifications, for Effectiveness of Care Measures, Access/Availability of Care Measures and Use of Services Measures.
See Page 4, No Minimum Size Requirement.
2. The specific reporting elements, including beneficiary member month contribution, for the measures requiring patient identifier are provided in Attachment II.
3. The HEDIS measure, Language Translation Services, under the Informed Health Care Choices domain, shall be reported directly to HCFA, since it will not be included in the software package developed for HEDIS reporting.
See page 7, Where Data Should be Sent.
4. Failure to comply with the requirements of this OPL may be considered as a factor for the imposition of possible sanctions as provided for in Section 1876(i) of the Social Security Act, as amended by Section 215 of the Health Insurance Portability and Accountability Act of 1996 (P.L. 104-191).
See page 4, Implications for Failure to Comply.
In addition, we are providing an addendum of Questions and Answers based on questions we have received since the release of OPL#47, dated December 23, 1997.
INTRODUCTION:
Effective January 1, 1997, HCFA requires Medicare managed care plans to report on performance measures from the Health Plan Employer Data and Information Set 3.0 (HEDIS 3.0) relevant to the Medicare managed care population, and participate in an independently-administered Medicare beneficiary satisfaction survey, the Medicare version of the Consumer Assessments of Health Plans Study (Medicare CAHPS).
These reporting requirements apply to all Section 1876 Risk and Cost managed care plans. These requirements are consistent with our regulatory and statutory authority and terms of our contracts with health plans to obtain the information necessary for the proper oversight of the program. It is critical to HCFA's mission that we collect and disseminate information that will help beneficiaries choose among health plans, contribute to improved quality of care through identification of quality improvement opportunities, and assist HCFA in carrying out its responsibilities.
HCFA has worked closely with the American Association of Health Plans (AAHP), the National Committee for Quality Assurance (NCQA), as well as the Agency for Health Care Policy and Research (AHCPR) and the Centers for Disease Control and Prevention (CDC), both of the Department of Health and Human Services, on developing initiatives that will help us achieve these objectives. HCFA conducted a meeting on September 5, 1996 with health plans and other interested parties to advise them of our intentions and to hear their views. We received a number of helpful suggestions related to implementation of these initiatives. We also received widespread acknowledgment of the importance and value of these requirements.
HCFA was particularly cognizant of the plans' concerns that the new requirements would impose an additional burden upon them, and that the time frame for reporting the HEDIS 3.0 measures was short. As an active member of the HEDIS 3.0 Committee on Performance Measurement (CPM), HCFA worked with the CPM to minimize, to the extent possible, the burden associated with the measures and time frame. Thus, HCFA has taken steps to limit the burden, while at the same time working to ensure that it receives valuable information about the performance of managed care plans in a timely manner, and that plans have valid information to use for quality improvement.
This Operational Policy Letter addresses specific HCFA requirements regarding how health plans must implement HEDIS 3.0 and Medicare CAHPS and provides guidance plans need to prepare to implement the two new initiatives. It also addresses questions that have been raised by plans concerning HCFA's new requirements.
Important: The final HEDIS 3.0 Volume 2, Technical Specifications is available from NCQA. If you do not have a copy, please call NCQA Publications at 1-800-839-6487.
We expect the final Medicare CAHPS survey instrument will be available in late spring. The final questionnaire will have 80 - 90 items. Plans may obtain a copy of the final Medicare CAHPS survey instrument, publication number AHCPR 97-R051, by calling 1-800-358-9295 after May 1.
Please note that HCFA's requirements are different in some ways from the HEDIS 3.0 document. Several major areas are: 1) this OPL applies to Section 1876 cost-based plans, as well as Section 1876 risk-based plans; 2) we are not requiring reporting on all the HEDIS measures identified as applicable to Medicare in the specifications of each measure and in Appendix 3; and 3) plans must submit separate reports for each contract with HCFA (see explanation below at A.2, Sampling and Reporting Unit).
Further, by late spring, HCFA will send additional information to health plans regarding the reporting and collection of HEDIS 3.0 data, and the implementation of the Health of Seniors measure (one of the Effectiveness of Care measures) and Medicare CAHPS survey.
IMPLEMENTING HEDIS 3.0 MEASURES AND MEDICARE CAHPS
A. Specifics Applicable to Both HEDIS 3.0 and Medicare CAHPS
1. Plans Required to Participate
Regarding HEDIS 3.0: All Section 1876 Risk and Cost health plans whose Medicare contracts are effective for any period during calendar year 1996, must report those HEDIS 3.0 measures whose specifications can be met, as detailed in the final HEDIS 3.0 document. The reporting year is 1996, although some measures require data from earlier years as well. HCFA recognizes that plans, whose initial contract began in 1996, will not be able to report all of the required HEDIS measures. The Medicare relevant measures in HEDIS 3.0 which plans must report are listed in Attachment I.
Please note the following. Plans which have Section 1876 Risk contracts with cost enrollees remaining from previous contracts shall include only their risk members in calculating the performance measures. Plans which have expanded the service area of a given contract at any point during 1996 must include information regarding those beneficiaries from the entire contract service area who meet the sampling or denominator specifications for a given measure.
The HEDIS 3.0 Volume 2, Technical Specifications clearly provides, for each measure, the following: Description of the Measure; and Administrative Data Specification and Hybrid Method Specification, both of which describe Calculation, Denominator and Numerator, where appropriate. Based on this information, plans can determine those measures which they must report.
New Contractors: Plans whose initial contract is effective on or after January 1, 1997, will not report during 1997, since the HEDIS 3.0 measures are for care provided in 1996. However, these plans must have systems in place to collect HEDIS information so that they can provide reliable and valid HEDIS data in 1998, for the reporting year of 1997.
Regarding Medicare CAHPS: All Section 1876 Risk and Cost health plans whose Medicare contracts were in effect on or before January 1, 1996, must comply with this survey requirement during 1997.
Medicare CAHPS does not apply to plans which received a contract effective after January 1, 1996. However, such plans may be required to undertake an enrollee satisfaction survey during 1997 to comply with the HCFA regulations on physician incentive plans (Vol. 61, Federal Register, 13430, March 27, 1996). Plans may wish to use Medicare CAHPS for this purpose.
Please refer to paragraph C.2 (page 8) for more information on the relationship of CAHPS to the physician incentive regulations.
No Minimum Size Requirement: There is no minimum size requirement for plans to comply with reporting HEDIS 3.0 data and participating in Medicare CAHPS.
However, in determining whether to report a specific HEDIS measure, plans must follow the General Guidelines for Data Collection and Reporting in the final HEDIS 3.0 Volume 2, Technical Specifications. Discussion regarding the size of the denominator is found under Specific Guidelines for Effectiveness of Care Measures (pages 6-9), Access/Availability of Care Measures (pages 10-11) and Use of Services Measures (pages 12-14).
Even though health plans do not meet specifications for all the measures, they must establish systems for collecting and reporting the data, and should be able to use the results of the measures for quality improvement purposes.
The two surveys (Health of Seniors and CAHPS) will include all members if the plan has less than the sample size identified later in this policy letter.
Demonstration Projects: Plans participating in either the Social HMOs or the Medicare Choices demonstrations funded by HCFA's Office of Research and Demonstrations (ORD) must report HEDIS 3.0 and participate in Medicare CAHPS if they are Section 1876 Risk or Cost plans. They must also adhere to additional requirements contained in their agreements with ORD.
2. Implications for Failure to Comply
HCFA expects full compliance with the requirements of this OPL. Health plans must meet the time lines, provide the required data and give assurances that the data are accurate. Plans which do not comply may be subject to sanctions as provided for in Section 1876(i) of the Social Security Act, as amended by Section 215 of P.L. 104-191.
3. Sampling and Reporting Unit
Plans' contracts are identified, by HCFA, for sampling, collecting and reporting purposes, as "Contract Reporting" or "Market Area Reporting." The latter category applies where the contract service area includes more than one "market area", that is it covers more than one major community or city. This alternative approach will provide more meaningful information to beneficiaries and to HCFA. In these situations, plans will report two or more sets of data for a given contract.
HCFA has assessed each HMO's contract service area to determine whether the Market Area approach is necessary. HCFA already notified those plans whose contracts warrant this approach. If you have not received notice, reporting for your plan will be by contract. The Market Area approach will apply to Medicare CAHPS and the HEDIS 3.0 measures, including the Health of Seniors survey.
A description of the reporting categories follow:
Contract Reporting:
This is for plans' contracts which have not been identified as reporting under the Market Area approach. Most plans' contracts will be reported this way.
Market Area Reporting:
Plans' contracts identified by HCFA to be in this category will report two measures under the Cost of Care Domain and three measures under the Health Plan Descriptive Information Domain by contract. The remaining measures will be sampled and reported by Market Areas.
Attachment I identifies the HEDIS 3.0 domains/measures by how plans will sample, collect and report summary data for their contracts. Specifically:
* There is one measure in the Health Plan Stability Domain which all plans will report by legal entity. If a plan has more than one contract, data for all contracts will be aggregated in this measure.
* There are two measures in the Cost of Care Domain and three measures in the Health Plan Descriptive Information Domain which all plans will report by contract for each of its contracts.
* All remaining measures will be reported by contract for plans' contracts which are in the Contract Reporting Category, or by Market Area for plans' contracts which are in the Market Area Reporting Category.
4. Use of Data
Data reported to HCFA under this requirement will be used in a variety of ways. The primary audience for the summary data is the Medicare beneficiary. These data will provide comparative information on health plans to beneficiaries to assist them in choosing among plans. Where applicable, HCFA expects plans to use these data for internal quality improvement. Each plan's summary HEDIS and Medicare CAHPS data will be arrayed and returned to them. These data should help the plans to identify some of the areas where their quality improvement efforts need to be targeted.
Further, these data will provide HCFA and the Peer Review Organizations information useful for monitoring quality of, and access to, care provided by the plans. HCFA may target those areas of the plans that warrant further review based on these data. For example, those plans which score significantly below the mean for all of the effectiveness of care measures will be subject to increased scrutiny and further actions by HCFA, as appropriate.
5. Release of Data
HCFA intends to release summary data for HEDIS 3.0 and the beneficiary satisfaction survey in late 1997. The information will identify each plan and array measures in a readily understandable fashion for Medicare beneficiaries. For individual measures which have an inadequate sample size, HCFA will determine an appropriate way to report such results that recognize both compliance with the measure, and the inadequate sample size. In addition, a notation will be made of those plans which are Section 1876 Cost contractors.
HCFA will solicit input from stakeholders (including health plans and beneficiaries) regarding availability and flow of data, and how data will be reported to the public.
6. Use of Vendors to Administer the Surveys, Sampling, and Cost Estimates
Initially, there will be a competitive process to select one vendor to administer the Health of Seniors HEDIS measure and another vendor to administer the Medicare CAHPS. In subsequent years, HCFA will select more than one vendor from which plans can choose to have the survey executed; under this approach, one vendor could conduct the survey for all of a plan's purchasers.
The per beneficiary cost will be held to the lowest possible level. HCFA estimates a cost ranging from $15 to $20 per completed survey each for Medicare CAHPS and for the Health of Seniors measure. For each survey, the final cost to each plan will be the prorated amount of the total cost of administering the survey, based on the number of completed surveys. When the completed survey data have been reviewed and approved by HCFA, the vendors will be authorized to collect funds from each of the participating plans which will in turn remit payment directly to the vendor.
HCFA will provide the sample lists of beneficiaries to each vendor. HCFA will coordinate the two lists so that the same beneficiary will not be surveyed under Health of Seniors and Medicare CAHPS.
B. Specifics for Implementing HEDIS 3.0
1. The Time Line for Reporting HEDIS Measures to HCFA
By June 30, 1997, plans must report summary data for all required HEDIS 3.0 measures, except for the Health of Seniors measure (see below). Submission of HEDIS 3.0 data by June 30, 1997, gives HCFA sufficient opportunity to assure the validity of the data collected, before publication, and to make the data available in a timely manner for beneficiary information and plan quality improvement activity.
In addition, by October 31, 1997, plans must provide the patient identifiers (previously referred to as raw data) used to calculate the summary data. Beneficiaries shall be identified by their individual health insurance claim (HIC) number. In addition, for some of the measures, plans must also provide the member month contribution of each beneficiary to the calculation of the denominator. Attachment II lists the four clinical Effectiveness of Care process measures (excluding the Health of Seniors measure) and five Use of Services measures for which patient identifiers must be provided, and the measures for which member month contributions must be provided. The specific information to be reported is described. Having data with patient identifiers will allow for analyses of national policy and research interests. These analyses will not be used for public plan-to-plan comparisons.
The patient identifier data will be protected in accordance with the Privacy Act of 1974.
The Health of Seniors results will not be reported to the public in 1997; the first reporting of the change in health status will be in 1999.
2. Where Data Should be Sent
The summary data and patient identifier data for all plan-generated HEDIS 3.0 measures must be sent to NCQA, HCFA's contractor for the collection of data. A software package will be developed for HEDIS 3.0; it is expected to be available in late April. Plans must use this program to report these data.
Please note that the HEDIS measure, Language Translation Services, under the Informed Health Care Choices domain, is a descriptive measure and will not be included in the software package. Therefore, it must be reported directly to HCFA, to your regional office managed care specialist.
Plans must retain data used for reporting for three years.
3. Certification of Data Validity
Because of the critical importance of ensuring accurate data, there will be a process of external validation of the HEDIS measures following the submission of the data to HCFA and before public reporting. In addition, the plan's senior executive officer and its director of medical affairs will be required to provide written attestation to the validity of the plan generated data.
4. How the Health of Seniors Measure is Different from Other Effectiveness of Care Measures
The Short Form (SF) 36 will be used to solicit self-reported information from a sample of Medicare beneficiaries for the HEDIS functional status measure, Health of Seniors. This measure is the first "outcomes" measure for the Medicare population. Because it measures outcomes rather than the process of care, it is primarily intended for population-based comparison purposes, by plan. The Health of Seniors measure is not a substitute for assessment tools that plans are currently using for clinical quality improvement.
An independent vendor will be selected to administer the SF 36 in 1997. In 1997, 1,000 beneficiaries per plan will be surveyed with a targeted response rate of at least 70 percent. If the plan has fewer than 1,000 eligible members, all will be surveyed.
Since the Health of Seniors measure looks at health status over a two-year period, results from this survey will not be publicly released in 1997. The survey in 1999 will assess the same beneficiaries' health status compared to two years ago. Beneficiaries will be categorized into those who are better, the same or worse over the two year period. Each plan's score, the percent of beneficiaries who are better, the same or worse, will be reported in late 1999.
C. Specifics for Implementing Medicare CAHPS
Medicare CAHPS will collect information on beneficiary access, utilization and satisfaction. It includes the CAHPS core items, as well as a set of questions developed specifically to address the health care concerns of Medicare beneficiaries in managed care plans. The survey will be administered to a sample of members by an independent vendor in late spring.
1. Information Regarding the Sample
The Medicare CAHPS will require a sample size of six-hundred (600) members. However, if the plan has fewer than 600 eligible members, all will be surveyed. Members must be continuously enrolled in the plan for one year by the time the survey is administered in order to be included in the sample. The survey will aim to achieve at least a 70 percent response rate. The survey will be administered by mail with the necessary telephone follow-up in order to achieve this response rate.
2. Relationship to Survey Required by the Physician Incentive Regulations
On March 27, 1996, HCFA published a final regulation regarding physician incentive plans in managed care settings. These regulations, codified at 42 CFR 417.479, require plans whose providers have been determined to be at substantial financial risk to conduct a survey of current and recently disenrolled (within the past 12 months) members. Medicare CAHPS will satisfy the requirement for current enrollees. The current Medicare CAHPS does not contain a module for disenrolled members; a standardized disenrollment module will be available as part of Medicare CAHPS in 1998. In the meantime, in order to satisfy the physician incentive requirements for a survey of disenrollees, HCFA will supply a standardized disenrollment survey and sampling specifications for plans to self-administer in 1997. They will be available this spring.
Plans that are not required to do a survey under the physician incentive regulations, but are required to do the Medicare CAHPS under the terms of this OPL, are not required to conduct a disenrollment survey.
Contacts:
For technical questions regarding HEDIS specifications on sampling, data collection and reporting, please contact the NCQA Technical Advice Line at 202-955-1737.
If you have other questions, please contact: Steve Balcerzak (410-786-9281) or Mervyn John (410-786-1141), both of OMC's Operations and Oversight Team, or Connie Forster (410-786-1036) of OMC's Quality and Performance Standards Team. The fax number for OMC is 410-786-5010.
cc: OMC All
RO Managed Care Contact, Regions I - X
Regional Administrators, Regions I - X
Policy and Program Improvement Team
Attachment I
HEDIS 3.0 Domains/Measures by Category of Reporting
For Summary Data
ALL PLANS TO REPORT BY LEGAL ENTITY:
Health Plan Stability:
* Performance Indicators
ALL PLANS TO REPORT BY CONTRACT
Cost of Care:
* High-Occurrence/High-Cost DRGs
* Rate Trends
Health Plan Descriptive Information
* Provider Compensation
* Total Enrollment
* Enrollment by Payer (Member Years/Months)
CONTRACT REPORTING CATEGORY TO REPORT BY CONTRACT;
MARKET AREA REPORTING CATEGORY TO REPORT BY MARKET AREA
Effectiveness of Care
* Breast Cancer Screening
* Beta Blocker Treatment After A Heart Attack
* Eye Exams for People with Diabetes
* Follow-up After Hospitalization for Mental Illness
* The Health of Seniors
Access to/Availability of Care
* Adults' Access to Prevention/Ambulatory Health Services
* Availability of Primary Care Providers
* Availability of Mental Health/Chemical Dependency Providers
* Availability of Language Interpretation Services, Part II
Health Plan Stability
* Years in Business/Total Membership
* Disenrollment
* Provider Turnover
Use of Services
* Frequency of Selected Procedures
* Inpatient Utilization - General Hospital/Acute Care
* Ambulatory Care
* Inpatient Utilization - Non-Acute Care
* Mental Health Utilization - Inpatient Discharges and Average Length of Stay
* Mental Health Utilization - Percentage of Members Receiving Inpatient, Day/Night and Ambulatory Services
* Readmission for Specified Mental Health Disorders
* Chemical Dependency Utilization - Inpatient Discharges and Average Length of Stay
* Chemical Dependency Utilization - Percentage of Members Receiving Inpatient, Day/Night and Ambulatory Services
* Readmission for Chemical Dependency
* Outpatient Drug Utilization (for those with a Drug Benefit)
Informed Health Care Choices
* Language Translation Services
Health Plan Descriptive Information
* Board Certification/Residency Completion
* Preventive Care and Health Promotion
Attachment II-1
Measures Which Require Patient Identifier
The health plan must provide patient identifier information (the HIC number) for those in the numerator and those in the denominator for the Effectiveness of Care measures and two of the Use of Services measures.
* Breast Cancer Screening
* Beta Blocker Treatment After A Heart Attack
* Eye Exams for People with Diabetes
* Follow-up After Hospitalization for Mental Illness
* Readmission for Specified Mental Health Disorders
* Readmission for Chemical Dependency
In addition, for the following Use of Services measures, the HIC number must be provided for all members. (The plan membership is basically the denominator.) The member month contribution (MMC) will be necessary for each HIC number. For these measures, the plan must flag each beneficiary who received the specific service. Attachment II-2 provides a descriptive chart of the necessary data for the following three Use of Services measures.
* Frequency of Selected Procedures
* Mental Health Utilization - Percentage of Members Receiving Inpatient, Day/Night and Ambulatory Services
* Chemical Dependency Utilization - Percentage of Members Receiving Inpatient, Day/Night and Ambulatory Services
Attachment II-2
Use of Services Patient Identifier Information*
Plan Membership |
Procedures/Services (Numerator) - Each selected procedure has a column |
||||||||||
HIC # |
MMC |
Proc 1 |
Proc 2 |
Proc 3 |
Proc 4 |
Proc 5 |
Proc 6 |
Proc 7 |
Proc 8 |
Proc 9 |
Proc 10 |
HIC 1 MM HIC 2 MM HIC 3 MM HIC 4 MM
HIC100 MM |
|
|
|
|
|
|
|
|
|
|
|
Plan Membership: HIC # MMC |
Mental Health Utilization:
Any I/P Day/Night Amb |
Chemical Dependency Utilization:
Any I/P Day/Night Amb |
||||||
HIC 1 MM
HIC 2 MM
HIC 3 MM
HIC 100 MM |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Plan Membership:
All Medicare members, by HIC number. HCFA will aggregate by age/gender, as necessary. Member month contribution (MMC) for each HIC number (plan member).
Selected Procedures/Service:
Flag each member who received any procedure/service identified by the columns.
* This is the information which HCFA will need; the software design may be different but will incorporate the above information.
Addendum to OPL #47 REVISED
QUESTIONS AND ANSWERS
HCFA has received a number of questions concerning Operational Policy Letter #47. OPL#47 REVISED addresses some of the questions we received; others are answered below. We anticipate that we will have additional addenda regarding other questions and updates on this important and evolving requirement on Medicare contractor reporting.
PLEASE NOTE: Information on Medicare requirements in the OPL, supplemented by this Addendum (and future addenda), supersedes information in the HEDIS document if it is inconsistent.
General Reporting Issues:
1. Please describe what HCFA means by legal entity.
A. The legal entity is the licensed organization which contracts with HCFA. This entity submits the balance sheet and other financial reports, for the company, to HCFA as required by federal regulations.
2. Have all plans that must report data by market areas within a contract been notified?
A. Yes, only 15 Medicare risk or cost contracts have been identified as needing to report by market area. The health plans were all notified in January.
3. For a plan with multiple Medicare contracts and fewer than 5,000 or 10,000 Medicare enrollees in the combined contracts, may they produce HEDIS results for the entire plan rather than for each contract?
A. No, HEDIS reporting and both surveys apply to each Section 1876 contract. In this way, the major purposes of performance measurement reporting and satisfaction surveys will be better served. Information used for beneficiaries choice, plans' internal quality improvement activities and HCFA oversight will be more pertinent since it will relate to each specific contract.
4. With the recent approval of "portability" for some HMOs, can you clarify which health plan, the host or home plan, would be held accountable for the results of relevant measures when the care was provided at the host plan.
A. The home plan must report the data related to services received by its members when out of the plan's service area. As part of the Visitor Program/Affiliate Option (portability), the host plan is treated as another health care provider under the home plan's contract with HCFA. The home plan is responsible for assuring that the host plan fulfills the home plan's obligations under the contract. Therefore, the responsibility for reporting HEDIS data and participation in the surveys is that of the home plan.
5. What patient identifiers should health plans use to provide raw data to HCFA?
A. Plans will use the Health Insurance Claim (HIC) number as the identifier. The HIC number is the number assigned by HCFA to the beneficiary when he/she signs up for Medicare. Health plans use this number for accretions/deletions.
6. Must a plan which has a Section 1876 Social HMO (SHMO) contract and a Section 1876 risk or cost contract provide HEDIS data for each contract?
A. Yes, HEDIS data must be reported separately for each Section 1876 contract held by the same plan.
7. A critical issue is access to raw data under the FOI provisions. Even without patient identifiers, the data would be valuable to market competitors. An example was presented in which a hospital requested access to the data reported by health plans. Although HCFA's stated limited access would be to researchers, how will HCFA distinguish between different types of research and what is valid access to the data for research purposes and what puts proprietary market information at risk?
A. HCFA is currently working with its Office of Freedom of Information and Privacy. We will provide a response to this question as soon as possible.
8. There is reference to protection of raw data in accordance with the Privacy Act of 1974. Please state the specific provision of the Privacy Act of 1974 that protects the data we are requested to send.
A. The entire Privacy Act, Title 5, U.S.C., Sec.552a will apply.
9. The consumer "report cards" to be released later this year could be positioned as a potential market disadvantage for new Medicare risk products. There is a critical difference between not reporting data and not being able to report data due to the technical specifications of HEDIS, an inadequate number of enrollees for statistical significance and the enrollment period required for CAHPS. In fact, some plans with new products in 1996 may not be able to report on several measures until 1998. The many products being contracted in 1997 may not have a report until 1999. With regard to the Health of Seniors, some plans will not have data until 2000.
How will HCFA represent these differences externally to consumers and other interested parties to avoid possible inadvertent negative market impact for health plans which do not meet the technical specifications for specific measures, as opposed to health plans not willing, or not able to product the data?
A. Although we are not planning to publish "report cards," HCFA clearly intends to address the issues raised in the question above. We will engage interested parties representing plans and consumers to work with us in the best way to present performance data to address the parties' needs. Issues related to plans not meeting technical specifications, not having adequate sample size, and a variety of other collection and reporting issues of national interest will be addressed by the group.
10. As it is expected that Medicare products will be reported on NCQA software, will HCFA require that Medicare data be included in Quality Compass?
A. No. NCQA is HCFAs contractor for purposes of collecting data. HCFA will release the data following validation. At this point, the data will be public. There will be nothing to preclude NCQA, and others from using it in a variety of formats. We anticipate that it will be available to the public in such formats as the Internet, printed materials, and CD ROM. We will be looking to ICAs (Information, Counseling, and Assistance) of state offices of aging and other advocacy organizations to help reach the beneficiary population.
11. To determine continuous enrollment eligibility for specific measures, should the plan count enrollees in a risk-based or cost-based Section 1876 (of the Social Security Act) contract who were previous members with the plan under an HCPP (Health Care Prepayment Plan, Section 1833) contract? Similarly, should a health plan count current risk-based members who moved from their previous 1876 cost-based contract?
A. No. For measures with a continuous enrollment requirement, enrollees who were receiving care through the health plan's HCPP contract should not be counted for continuous enrollment purposes until the point they became enrolled in the health plan's risk-based or cost-based 1876 contract. Similarly, those who are current risk-based members, but previously were Section 1876 cost-based members with the same plan should be excluded from the measure.
HEDIS ISSUES
1. Are health plans required to submit data for measures for which the number of enrollees meeting the technical specification is not adequate to meet the sample size required in the HEDIS 3.0 technical specifications?
A. Under HEDIS, all plans, whose contracts were effective prior to January 1, 1997, will have to report HEDIS measures in accordance with the final HEDIS 3.0 Volume 2, Technical Specifications. See the OPL, No Minimum Size Requirements, on page 4.
2. If health plans were approved in 1996, what measures are required for these plans?
A. Plans should review Attachments I and II of this OPL #47 REVISED which delineate the measures which plans must report. Health plans must review HEDIS 3.0 Volume 2, Technical
Specifications for each required measure to determine whether they meet the specifications for reporting.
For both the Health of Seniors and CAHPS surveys, all members who meet the eligibility specification will be surveyed if the plan has fewer than the sample size. As stated elsewhere, CAHPS will only apply to plans with contracts effective on or before January 1, 1996.
3. What happened to the measures addressing flu shots for older adults?
A. This measure and the measure addressing "advice to quit smoking," which were in earlier versions of HEDIS, were added to the CAHPS survey. Reporting on these measures will be solely through the survey. Therefore, plans should not provide data under the Effectiveness of Care Domain for these two measures.
4. How should health plans report on HEDIS measures required by HEDIS 3.0?
A. Health plans must report using a software package which NCQA will develop as part of a contract with HCFA. Each plan will receive a copy of the software as soon as it is available, along with directions on where to send the disk. In the meantime, health plans should obtain the final HEDIS 3.0 Volume 2, Technical Specifications. This will enable plans to organize resources they need to properly collect HEDIS information.
5. If the issuance of the software is delayed, at what point will HCFA consider delaying the submission date?
A. NCQA plans to have its own software package available by late April. If it is not available in time, HCFA will assess whether to delay submission of data.
6. NCQA notes that health plans must supply HEDIS data by June 2 while HCFA notes that health plans must supply data by June 30. How do these two dates work together for submission of HEDIS data?
A. The earlier date, June 2, was established by the Committee on Performance Measurement representing public and private purchasers, health plans and consumers. HCFA elected a later report date to allow plans to stagger the collection and reporting of HEDIS data. The OPL requires that health plans report HEDIS data by June 30, 1997 for their Medicare population.
7. Attachment I lists required domains/measures by category for reporting. It is inconsistent in a few places with the domains/measures matrix in NCQA's HEDIS 3.0 Volume 2, Technical Specification.
A. OMC's Policy Letter (including this addendum and future addenda) take precedence over the HEDIS 3.0 document. See page 2 which states, "Please note that HCFA's requirements are different... we are not requiring reporting on all the HEDIS measures identified as applicable to Medicare in the specifications of each measure and in Appendix 3..."
8. Did HCFA add the following measures that are not part of HEDIS 3.0:
* Access of Mental Health/Chemical Dependency Providers and Availability of Language
Interpretation Services, Part II under the Access to/Availability of Care Domain;
* Enrollment by payor under Health Plan Descriptive Information Domain?
A. All three of these measures are part of HEDIS and are found on the following pages of the final HEDIS 3.0 Volume 2: Availability of Mental Health/Chemical Dependency Providers is found on page 77, Availability of Language Interpretation Services, Part II is found on page 99, and Enrollment by Payor is on page 247. All were in the DRAFT document. These measures have been identified in the OPL Attachment I as measures to be reported, thus, plans must report them. The software package developed by NCQA will include them.
9. Several HEDIS 3.0 measures were not in requirements for HCFA reports. Were the following requirements deleted for HCFA measures?
A) Domain-Informed Health Care Choices: New Member Orientation
B) Domain-Health Plan Descriptive Information:
-Physicians under capitation
-Utilization Management
-Risk Management
-Case Management
-Recredentialing
-Quality Assurance and Improvement
-Arrangements with Public Health and Social Services Entities
A. HCFA made a decision to not require all measures in HEDIS 3.0; those mentioned above were excluded.
10. Will health plans receive raw data (patient identifier data)?
A. HCFA will receive patient identifiers (i.e., HIC numbers) for numerators and denominators for certain measures (defined in Attachment II of this OPL) from the health plans. Since the plans are submitting this information, HCFA will not return the data to the plans. Patient identifier data will allow HCFA to match the data base to others for special projects and research which, among other issues, could assess whether certain groups (e.g. ethnic, racial, gender, geographic) are receiving fewer or more services than others. These data will not be used for plan to plan comparison purposes. We anticipate that research results (not plan specific data) will be disseminated to plans and as appropriate, plans could initiate QI projects based on the results in a given geographic area, such as a region or nationally.
11. Must a health plan which does not have an outpatient prescription drug benefit report the beta blocker measure?
A. Yes, assuming that it has more than 29 members who meet the specifications for the denominator. Specifications for the numerator are those health plan members included in the denominator who have received a prescription for beta blockers within seven days after discharge from the hospital with a diagnosis of AMI, or within 30 days prior to the hospitalization for AMI. Plans may have to use the hybrid method to collect the necessary data.
12. For which measures in Attachment II must the denominator listing include the member month contribution.
A. This OPL Attachment II, Measures Which Require Patient Identifier, is revised to require that the member month contribution for each beneficiary (HIC number) in the denominator be provided ONLY for the following Use of Services Measures:
* Frequency of Selected Procedures
* Mental Health Utilization - Percentage of Members Receiving Inpatient, Day/Night and Ambulatory Services
* Chemical Dependency Utilization - Percentage of Members Receiving Inpatient,Day/Night and Ambulatory Services
13. There is reference to October 31 as a date for plans submission of raw data used to calculate summary data, i.e., list of beneficiaries (by HIC number) in the denominator and numerator for the four clinical and five use of services measures. There are two questions:
(a) Should the data reported June 30 to HCFA (or its designee) be "frozen" in order to control the numerator and denominator numbers and member identification numbers submitted October 31? If we do not "freeze" the data set, the numerators and denominators submitted in October will be different from those submitted June 30?
A. The patient identifier data submitted by October 31 must relate to the summary data submitted by June 30. The purpose is to provide the numerator/denominator information for the specific summary data provided.
(b) For the raw data, we need the data specifications by February 15, 1997.
A. We anticipate that under contract with NCQA, NCQA will provide the standardized electronic format for patient identifier data. Attachment II provides specific information on what is to be reported.
HEALTH OF SENIORS:
1. What is the status of the selection of a vendor to conduct this survey?
A. The fielding of the Health of Seniors survey will be carried out in conjunction with NCQA and the expert panel. HCFA will advise health plans of opportunities for input as stakeholders related to vendor cost, payment and plan feedback issues.
2. Will the health plans have access to the members' data before 1999, as this could be key in quality improvement efforts?
A. Health of Seniors data will be plan-specific and in aggregate form (i.e., no patient-level information will be provided). Aggregate data will be provided after the 24 month follow-up survey in 1999. It has not yet been decided whether aggregate data will be provided after the first baseline survey in 1997. As part of a contract with NCQA, NCQA will convene a Health of Seniors technical expert panel which will suggest an approach for the release of aggregate data.
3. Do we have an estimate of the range of cost for the Health of Seniors survey?
A. Yes, the anticipated range is $15 - $20 per completed survey.
4. Can you clarify the 1,000 cohort drawn for the Health of Seniors survey? In year two (1998) of the evaluation, will additional sampling be drawn if less than 1,000 enrollees can be contacted? And, will this occur again in 1999? When will the next cohort of 1,000 enrollees be drawn, in other words, will Health of Seniors reports come out each year after 1999 (e.g. requiring a new cohort be drawn in 1997 for reporting in 2000)? Due to the cost per survey, this is a particular budget issue for plans which may expect to pay for more than 1,000 surveys per year depending on the reporting schedule.
A. The diagram below indicates the current sampling strategy for the Health of Seniors measure. Replacement of the 1,000 persons surveyed at baseline will not be necessary, as the 24 month follow-up does not anticipate a 100% response rate (the goal is at least 70%). It is anticipated that an entirely new cohort will be drawn every year so that the measure may be reported annually.
1997 1998 1999 2000
Baseline 24 Mo Follow-up 24 Mo Follow-up
1,000 1,000 Cohort I Cohort II
Cohort I Cohort II
New Sample of New Sample of
1,000 Drawn for 1,000 Drawn for
Cohort III Cohort IV
Since the technical expert panel will review the proposed specifications for the measure, this could change.
5. How will enrollee mortality be handled in reporting the results of the SF-36?
A. Beneficiaries who die between the baseline survey and 24 month follow-up are categorized in "worse" condition for reporting purposes.
6. Will HCFA risk adjust in its surveys?
A. We anticipate, as approved by the technical expert panel, that the Health of Seniors measure will be risk adjusted for age, race, sex, social support, co-morbid conditions, and normal expected decline in health status based on age.
CAHPS:
1. HEDIS 3.0 information notes that CAHPS will include questions on flu shots for seniors. Will the information from the CAHPS survey get back to the health plans so they can include this information to properly report this measure on HEDIS or should the health plans also collect this information?
A. The draft flu and smoking cessation effectiveness of care measures were moved to the CAHPS survey so plans should not collect and report the information separately. It is our intent to provide CAHPS data back to the plans for their use in quality improvement activity.
2. When will the CAHPS data be returned to the health plans?
A. We expect that the data will be available to plans sometime in the fall.
3. In what format will the data be returned to the plans? In the OPL it is noted that "summary" data will be provided. The CAHPS survey can be an important source of quality improvement initiatives. Can the data be returned via disc from the vendor so that the individual plans can analyze the array of responses on questions and even work to determine which questions are best predictors of satisfaction, which populations have the most satisfaction issues, etc.?
A. HCFA intends to release summary data to the plans via electronic format.
4. If a plan does not meet the specifications (outlined in OPL #47 REVISED) for participation in the CAHPS survey, may it voluntarily contract with the vendor to administer the CAHPS survey, and then use the data for comparisons for marketing purposes or for quality improvement activity.
A. As a business decision, a plan which does not meet the specifications for inclusion in the CAHPS survey could contract with the CAHPS vendor to survey some or all of its members. However, HCFA would not include the results of this survey in the comparison information it releases, since that plan's data likely will not meet the one year continuous enrollment criteria. Thus comparisons can not be made. Such survey data may be used in marketing materials, only if consistent with HCFA's national marketing guidelines. We encourage use of such data for quality improvement activity.
5. Regarding Medicare CAHPS, the OPL states: "All Section 1876 Risk and Cost health plans whose Medicare contracts were in effect in on or before January 1, 1997 must comply with this survey requirement during 1997." Please clarify what we must comply with in 1997. We understand that HCFA will administer the survey.
A. Medicare CAHPS applies to all Section 1876 contractors whose Medicare contract were in effect on or before January 1, 1996 [please note the year is 1996, NOT 1997]. HCFA will use an independent contractor to administer the CAHPS survey to plan members under each contract. The sample will include enrollees who have been with the plan for at least one year prior to the implementation of the survey.
6. Will HCFA risk adjust in its surveys?
A. There is potential for limited risk adjustment in the CAHPS survey based on the following demographic data to be collected: gender, age, race, ethnicity and self-reported health status.
DATA VALIDITY AND AUDIT
1. The OPL notes that attestations will need to be submitted with the data and before reporting. Will the attestation associated with the NCQA software submission be sufficient?
A. HCFA, in coordination with NCQA, will design an attestation process that is as similar as possible to the one NCQA uses, but which will meet the needs of HCFA. There will be a mechanism for NCQA to inform HCFA if data are submitted without the attestation.
2. Does HCFA require that the data reported by plans be audited by a third party? What is the process for data validation?
A. HCFA is in the process of selecting one or two PROs whose contract(s) will be modified to validate HEDIS data. Medicare measures subject to desk review audit will include the four required effectiveness of care measures and three of the utilization measures chosen randomly. All plans will be subject to desk audit. In addition, the PROs will conduct approximately 60 field audits of plans. The health plans undergoing the on-site reviews will include at least 30 randomly selected and 30 whose data are designated as outliers during the desk audits.
The primary goal of HEDIS data validation is to quantify the accuracy and completeness of the data collected. Another aspect of this review is for HCFA to assess the accuracy of the data
definitions for the HEDIS measures. For example, it may be found that all or most health plans have problems with a measure; the problem could be with the data definition, not the health plans.
3. If a health plan currently undergoes an external audit of its data, can it be deemed to substitute for the HCFA audit?
A. No. However, HCFA may revisit this issue for future years.