REPORT ON VALUE-BASED PURCHASING METHODOLOGIES FY

Transcription

1 Agency for Health Care Administration REPORT ON VALUE-BASED PURCHASING METHODOLOGIES FY June East Northern Avenue, Suite 100 Phoenix, AZ Phone Fax

2 1. Executive Summary Background and Purpose of the Report The Florida Agency for Health Care Administration (AHCA) contracted with Health Services Advisory Group, Inc. (HSAG) to provide external quality review (EQR) services for the State s Medicaid managed care program. The EQR contract is composed of 11 key activities. One of the activities is to develop a methodology for a value-based purchasing initiative, which includes incentives for superior performance. The goals for the initiative include the recognition of demonstrated excellence in focused areas, the promotion of continuous quality improvement, and ultimately, improvements in the health outcomes of Medicaid managed care enrollees. These goals are in-line with the Value-Driven Health Care initiative, a national initiative currently implemented by the Centers for Medicare and Medicaid Services (CMS). Finally, a value-based purchasing initiative provides support and direction for state s quality strategies, which are mandated by CMS. Scope of Activities In developing the value-based purchasing methodology, HSAG considered a variety of performance data available to AHCA, as required by the Invitation to Negotiate (AHCA ITN 0601). HSAG considered all data available to AHCA during fiscal year This report focuses on the HMOs, which among the various model types in Florida, currently have the most data available to evaluate health plan performance. Although the proposed approach for value-based purchasing focused on the Florida Medicaid HMOs, it is not limited to the HMOs. The concepts and methods developed may be expanded to include other MCO types, including reform- HMOs, provider service networks (PSNs), nursing home diversion programs (NHDPs), and prepaid mental health plans (PMHPs). Summary of Findings HSAG reviewed 14 state approaches to value-based purchasing, and categorized the approaches into three different methods. The first method is comparative benchmarking, while the second method uses the Quality Improvement System for Managed Care (QISMC), a nationally recognized method. A hybrid QISMC approach was identified as the third method. All three methods can be used to provide incentives and/or corrective action plans for HMOs. HSAG evaluated the advantages and disadvantages of each approach and determined that the hybrid QISMC approach would be the best fit for the Florida Medicaid managed care program. HSAG also developed a method for defining superior plan performance. HSAG recommends using a two-tiered approach that may both define superior performance, and provide incentives for HMOs to improve, regardless of whether or not the HMO is a top performer. The model allows for superior performance to be based on the current year across all activities, while other awards may Report on Value-Based Purchasing Methodologies Page 1-1

3 EXECUTIVE SUMMARY be given based on continued improvement over time. Specific details regarding the model can be found in Section 3 of this report. Conclusions and Recommendations HSAG developed a proposed methodology for value-based purchasing, which is described in detail in Section 4 of this report. The methodology incorporates a variety of performance data, including performance measures, survey data, results of compliance monitoring activities, focused study results, and Child Health Check-Up (CHCUP) participation rates. The model includes a recommended weighting of the components, as well as proposed scoring algorithms. A discussion of potential incentives and disincentives is also included in Section 4. The model was purposely developed to allow AHCA flexibility in selection of performance activities to include in the valuebased purchasing initiative, so that the model can be modified as the needs of the Florida Medicaid managed care program grow and change. Next steps for implementation of the proposed model include the organization of a workgroup to determine the final components of the initiative, the solicitation of HMO and key stakeholder feedback, the finalization of the methodology, and official kick-off of the initiative. Considerations for successful implementation are discussed, including when incentives or disincentives are incorporated, and what type of non-monetary incentives can be implemented during the initial phase of implementation and in subsequent years. Report on Value-Based Purchasing Methodologies Page 1-2

4 2. Introduction Purpose There are a variety of methods that are being used by state Medicaid agencies for evaluating performance. These methods are typically used to reward high performance and/or penalize poor performance, using both financial and non-financial means. Creating an incentive program for high performers, in conjunction with corrective actions for poor performers, is the most popular method for evaluating health plan performance. The purpose of this report is twofold: To provide a method for AHCA to use to define superior or high HMO performance, as well as low HMO performance, and To provide AHCA with potential incentives and/or corrective actions that may be initiated as a result of HMO performance levels. Scope This report focuses on the HMOs, which among the various model types in Florida, currently have the most data available to evaluate health plan performance. Although this proposed approach for value-based purchasing focuses on the Florida Medicaid HMOs, it is not limited to the HMOs. The concepts and methods presented within this report may be expanded to include other MCO types, including reform-hmos, provider service networks (PSNs), nursing home diversion programs (NHDPs), and prepaid mental health plans (PMHPs). Methodology HSAG began with a review of the current evaluation method included in the July 2004 Medicaid HMO contract. After reviewing this information, HSAG conducted a scenario analysis and evaluated the method based on its potential to accurately reflect the performance levels for the HMOs. The scenario analysis can be found in Appendix A. HSAG discussed the findings based on the scenario analysis with key AHCA staff, and was informed that the evaluation method was not currently in use. HSAG also interviewed AHCA staff regarding areas of consideration to include in the proposed value-based purchasing methodology. Following the review of AHCA s information, HSAG conducted a literature review, which included 14 state approaches to value-based purchasing, and categorized the approaches into three different methods. Based on the available literature and state information, HSAG determined that these three methods appear to be the most common approaches used by states to objectively evaluate performance. All three methods can be used to provide incentives and/or corrective action plans for HMOs. Report on Value-Based Purchasing Methodologies Page 2-1

5 INTRODUCTION HSAG considered the following criteria for selection of data for evaluating levels of performance: 1. Evidenced based: The data used for measuring performance must be based on evidence documenting the linkage between the process being measured and the health outcomes. 2. Reliable: The data used must be reliable. Reliability means the data collected correctly identify the event that has been targeted for measuring and the results are reproducible. 3. Valid: The data used must be valid. Valid data make sense logically and capture meaningful aspects of care. 4. Comparable: The data used should have comparable data sources, data collection methods, and precise specifications. 5. Meaningful: The data used should measure an aspect of care that is meaningful to the AHCA, the managed care organizations (MCOs), the beneficiaries, and other interested stakeholders. 6. Controllability: The data used should measure an aspect of care that is within the control of the MCOs. Many stakeholders prefer to evaluate quality in terms of service and customer satisfaction, which are relatively easy to understand and measure. The Health Plan Employer Data and Information Set (HEDIS ) has become an industry standard in both the public and private sectors. HEDIS and the Consumer Assessment of Healthcare Providers and Systems (CAHPS ) measures are frequently used to take advantage of the standard calculations and audit methodologies as well as the wealth of benchmarking information available. However, HEDIS and CAHPS do not necessarily represent all the aspects of care. For example, HMOs may have excellent processes and efficiency, but have low HEDIS rates, or vice-versa. Therefore, HSAG considered a wide range of HMO activities to evaluate performance. Organization of the Report The Executive Summary (Section 1) presents the background information on the activity, the purpose of the report, the scope of the activity, a summary of findings, and a brief description of the proposed methodology. Section 2 provides an introduction to the report, which includes the purpose, scope, methodology, and organization of the report. Section 3 describes the current AHCA practices for evaluating performance, a review of other state s approaches to value-based purchasing, and an approach to defining superior health plan performance. Section 4 presents the proposed approach for value-based purchasing. HEDIS is a registered trademark of the National Committee for Quality Assurance (NCQA) CAHPS is a registered trademark of the Agency for Healthcare Policy and Research (AHRQ) Report on Value-Based Purchasing Methodologies Page 2-2

6 INTRODUCTION Section 5 presents a description of the next steps for implementation. Appendix A includes a summary of the results of the scenario analysis conducted by HSAG on the AHCA methodology included in the HMO contracts for evaluation plan performance. Report on Value-Based Purchasing Methodologies Page 2-3

7 3. Discussion of Findings Current Agency Practice In 2005, AHCA proposed a method for calculating annual composite scores for each health maintenance organization (HMO) to aid in assessing performance across five different plan service areas. The five service areas included: Consumer Assessment of Healthcare Providers and Systems (CAHPS ); Child Health Check-Up (CHCUP) participation rates; Onsite Audit Compliance results; Quality Improvement and Performance Indicators; and Accreditation status. The method, entitled Plan Service Performance was included in the HMO contract. This method, however, was not operationalized as a method to evaluate health plan performance. To date, AHCA uses various health plan data and monitoring results to evaluate health plan performance, including those areas listed above, although this process does not currently include a formal methodology. Review of Other State s Approaches Following the review of AHCA s information, HSAG reviewed 14 state approaches to value-based purchasing, and categorized the approaches into three different methods. The first method is comparative benchmarking, while the second method uses the Quality Improvement System for Managed Care (QISMC), a nationally recognized method. A hybrid QISMC approach was identified as the third method. HSAG determined that these three methods appear to be the most common approaches used by states, and are briefly described below. All three methods can be used to provide incentives and/or corrective action plans for HMOs. Method One: Comparative Benchmarking One common approach to evaluating health plan performance is by setting benchmarks, and then comparing HMO performance to those benchmarks. This method concentrates on performance of the HMO for the year under review, and is the least complicated method. The benchmarks are usually set using national performance levels, if available. When national data are not available, then states typically use other sources, such as state laws, or federal requirements. Often, the CAHPS is a registered trademark of the Agency for Healthcare Research and Quality (AHRQ). Report on Value-Based Purchasing Methodologies Page 3-1

8 DISCUSSION OF FINDINGS average for the state is calculated, and then high and low performance thresholds are established by using one or two standard deviations from the average score. Table 3-1 displays the advantages and disadvantages of using the comparative benchmarking method. Table 3-1 Advantages and Disadvantages of Comparative Benchmarking Advantages Disadvantages Least complicated. No trending of results. Removes potential bias of looking at May not, independently, provide an prior year. incentive for HMOs to improve performance (if HMO already exceeded New measures can be easily added without making adjustments to the method. Measures can be modified or deleted without making adjustments to method. high performance threshold). Differences between performance levels may appear unreasonable if there are too few HMOs and/or there is not a significant variance in the range of the rates (e.g. the difference between low performance and high performance may be less than one percent if all the rates are close together, which may appear unreasonable to the low performing HMOs). Method Two: Quality Improvement System for Managed Care (QISMC) The Quality Improvement System for Managed Care (QISMC) methodology, developed by the Centers for Medicare and Medicaid Services (CMS), uses a reduction of the difference between the current measurement and the desired outcome in moving towards a specified goal. In general, QISMC seeks to gain ten percent over the previous rate; therefore the following formula is applied: (Goal Rate) x Rate = Interim Goal for Performance For example, if the goal is 100 percent and the current rate is 60.0 percent, then the QISMC model produces ( ) x = 64.0 percent as the interim goal for the next measurement period. One feature of the QISMC method is that it is usually more difficult to improve rates as HMOs approach the goal. For example, it is generally easier to improve a rate of 20 percent than a rate of 80 percent. Using QISMC, an HMO with a rate of 20 percent would have to improve by eight percentage points, while the HMO already at 80 percent would need to gain two percentage points. This feature has a slight drawback; it may be difficult for HMOs at or near the goal to demonstrate improvement, much less a statistically significant improvement. Report on Value-Based Purchasing Methodologies Page 3-2

9 DISCUSSION OF FINDINGS Table 3-2 displays the advantages and disadvantages of using the QISMC method. Table 3-2 Advantages and Disadvantages of the QISMC Method Advantages Disadvantages Allows for trending of results The same measures must be used for trending, and any differences (e.g. a change in measure specifications) may invalidate results Can provide incentive for HMOs to Focus is on improving trends rather than improve results Nationally recognized system for quality improvement actual current performance New measures need a baseline rate before scoring HMOs at or near the goal may have difficulty in realizing improvement Method Three: Hybrid QISMC The hybrid QISMC method uses the QISMC method approved by CMS in conjunction with HEDIS percentiles, state goals, and/or other benchmarks. Several states, such as California and Michigan, have successfully utilized the hybrid QISMC approach for evaluating superior performance for their Medicaid programs. The hybrid QISMC model is highly adaptable, and may use goals other than 100 percent (e.g., HEDIS Medicaid 90th percentiles), may choose more significant gains (e.g., 15 percent gain), may set minimum performance levels (MPLs) and high performance levels (HPLs), and can be incorporated to evaluate other Florida Medicaid programs, such as PSNs, NHDPs, and PMHPs. The comparative benchmarking method can also be incorporated as a component of the hybrid QISMC method. The main emphasis of the hybrid QISMC method is to improve rates for low and average performing HMOs. The HMOs that are performing below the MPL are required to reach the MPL, while average performing HMOs (i.e., HMOs performing between the MPL and HPL) are required to demonstrate QISMC-type improvements in their rates. The hybrid QISMC methodology typically uses the HEDIS 90th percentile for the goal or the HPL. Once the goal or HPL is achieved, the HMO no longer needs to show a 10 percent increase from the rate, but should strive to maintain the rate above the HPL. Therefore, if the HPL is 80 percent and the current rate is 50 percent, the QISMC component of the approach identifies (100-50) x = 55 percent as the interim goal for the next measurement period. If the HMO s rate reaches 84 percent the next year, then the hybrid QISMC method would simply require the HMO to remain above 80 percent (the HPL) for the next measurement period. Report on Value-Based Purchasing Methodologies Page 3-3

10 DISCUSSION OF FINDINGS Table 3-3 displays the advantages and disadvantages of using the QISMC method. Table 3-3 Advantages and Disadvantages of the Hybrid QISMC Approach Advantages HMOs at the goal are only required to maintain their rate above the goal, but not necessarily improve the rate Provides incentives for HMOs to improve results Allows for trending of results Focus is on current performance and trending patterns Disadvantages The same measures must be used for trending, and any differences (e.g. a change in measure specifications) may invalidate results New measures need either a baseline rate or established benchmarking before scoring Defining Superior Performance Superior performance may be defined within an activity (e.g., performance measures), and/or across all activities. Within activities, superior performance is easily defined as the HMOs that receive the highest possible score in a particular activity. For example, a CHCUP participation rate of 97 percent would be in the highest scoring category and would receive 5 points based on the table presented for that activity. For activities with more than one measure (e.g., using five performance measures), there are two basic choices: 1) either all measures must achieve the highest score, or 2) a combined score for all the measures must exceed a threshold. HSAG recommends the latter approach and can provide AHCA with guidance for the scoring threshold based on the actual number of performance measures selected for inclusion in the model. There are several disadvantages, however, to defining superior performance simply within activities. It is possible to be a top performer for the annual compliance review, but be the lowest performer in every other activity. In addition, HMOs may begin to focus on one particular activity to ensure their status as a superior performer. HSAG recommends using a two-tiered approach that may both define superior performance, and still provide incentives for HMOs to improve, regardless of whether or not the HMO is a top performer. This is presented in more detail in Section 4 of this report, in the discussions regarding incentives. The model allows for superior performance to be based on the current year across all activities, while other awards may be given based on continued improvement over time. HSAG recommends using all of the activities to designate superior performance as described below: Score each activity as previously described. Add the points within each activity to obtain the total score for the activity. Report on Value-Based Purchasing Methodologies Page 3-4

11 DISCUSSION OF FINDINGS Divide the total score for each activity by the total possible points for that activity by HMO some HMOs may have fewer possible points. Multiply the result by the assigned weight. Sum the total points across all of the activities to obtain a final, overall score. Performance levels may then be identified using the following algorithm displayed in Table 3-4: Table 3-4 Performance Level Algorithm Performance Level Overall Score Excellent or Superior 90% 100% Above Average 80% - 89% Average 70% - 79% Fair 60% - 69% Poor 0% - 59% Any score of a zero within an activity automatically disqualifies the HMO from receiving a superior performance designation, regardless of the overall score. This is an important concept as more measures are added, since a zero could be washed out with high scores. Report on Value-Based Purchasing Methodologies Page 3-5

12 4. Proposed Approach Some state Medicaid agencies use the hybrid QISMC approach on a select few HEDIS measures, and base the entire incentive program on improving those measures. The idea is that by focusing on three to five measures in various areas, the HMOs will actually improve on all the measures, as quality improvement efforts will trickle into similar measures and services. There is merit to this methodology, especially for states just beginning with value-based purchasing programs. Starting slowly with a few measures, and then expanding after the HMOs become more adept at improving rates, will allow the HMOs to obtain the necessary knowledge required to improve rates across all measures without being overwhelmed. Even if starting slowly, HMO performance should encompass more than HEDIS and CAHPS. Determining levels of performance using measurements that do not have national benchmarks for comparison can be difficult, but is necessary for a more robust picture of the HMO s performance. Therefore, this approach will consider scoring the following Florida Medicaid HMO mandatory activities: Performance Measures (HEDIS) CAHPS (Adult Survey) Annual Compliance Review Performance Improvement Projects (PIPs) Focused Studies CHCUP Participation Rate Scoring Methodology Based on the review of current approaches used by state agencies, HSAG proposes the following hybrid QISMC methodology for consideration by AHCA. The individual mandatory activities are described below along with the proposed scoring methodology. A weight has also been assigned to each activity to account for the relative importance of some activities compared to others. This scoring methodology and weight assignment is a recommendation, and may easily be adjusted. Additionally, the scoring methodology presented below may be expanded to include the reform- HMOs, PSNs, NHDPs, and PMHPs, at the discretion of AHCA. Report on Value-Based Purchasing Methodologies Page 4-1

13 PROPOSED APPROACH Performance Measures Weighting of 50% HSAG recommends the inclusion of HEDIS measures because of the standardized data collection process, appropriateness to the Medicaid population, and the richness of available benchmark data. The HMO s rate for each selected HEDIS measure will be compared to the NCQA national Medicaid percentiles, and points will be awarded as displayed in the table below. HEDIS measures that are identified as first year measures by NCQA will not be included for scoring, even if the HMOs are required to report on them. Table 4-1 displays the recommended scoring grid using national Medicaid HEDIS percentiles. Table 4-1 Scoring Grid for HEDIS Performance Measures Performance Level National Medicaid HEDIS Percentiles Points High Performance Level (HPL) 90 th percentile 4 75 th 89 th percentile 3 50th 74 th percentile 2 25th 49 th percentile 1 Minimum Performance Level (MPL) <25 th percentile or Not Report (NR) 0 Note: The HPL is shown at the 90 th percentile. This may be changed to be at or above the 75 th percentile, and the points can be adjusted accordingly. Similarly, the MPL can be set at the 10 th percentile. The following HEDIS measures were required for reporting by the HMO non-reform contract and are presented as an example for the purposes of scoring: Well-Child Visits in the First 15 Months of Life, Zero Visit Rate Well-Child Visits in the First 15 Months of Life, Six or More Visits Rate Well-Child Visits in the Third, Fourth, Fifth and Sixth Years of Life Adolescent Well-Care Visits Breast Cancer Screening Chlamydia Screening in Women, Years Chlamydia Screening in Women, Years Chlamydia Screening in Women, Combined Rate Use of Appropriate Medications for People with Asthma, 5 9 Years Report on Value-Based Purchasing Methodologies Page 4-2

14 PROPOSED APPROACH Use of Appropriate Medications for People with Asthma, Years Use of Appropriate Medications for People with Asthma, Years Use of Appropriate Medications for People with Asthma, Combined Rate Additional measures of interest may be added at the discretion of AHCA, such as Comprehensive Diabetes Care. HSAG recommends starting with three to five HEDIS measures from the set of measures reported by the HMOs. HMOs that do not have rates for specific measures due to insufficient population sizes will not be penalized. The overall rate will be based only on the HEDIS measures with a valid rate, and those measures with insufficient populations (e.g. under 30 cases) will not be included in the numerator or denominator for the HMO. However, a measure status of Not Report, or NR, is included and is counted as a zero. Not Report indicated that there was no valid rate produced for the measure. HEDIS measures are continually modified, added, or retired. AHCA is not limited to this specific list, and may choose to add (or remove) other Medicaid HEDIS measures in the future. HSAG recommends AHCA begins with a subset of the above measures, and allow the HMOs to provide input on additional measures to include in the following years. In addition, since the most costly health issues tend to be related to diabetes, asthma, depression, and heart disease, HSAG suggests AHCA review the current list of HEDIS measures and consider adding HEDIS measures that address these important aspects of care. CAHPS Weighting of 15% Member satisfaction is an important aspect of HMO performance and can be readily measured by CAHPS. Under the current EQRO scope of work, HSAG is evaluating the Florida Medicaid consumer satisfaction survey process. At this time, however, the CAHPS survey methodology used by the deviates from the standardized approach recommended by the Agency for Healthcare Research and Quality (AHRQ) and the National Committee for Quality Assurance (NCQA). Given these methodological differences, Florida's CAHPS results cannot be compared to national benchmarks and thresholds published by NCQA. However, if the State requires that the survey be administered in accordance with AHRQ and NCQA specifications, then the survey results would be comparable to NCQA benchmarks and thresholds, allowing for integration of this comparative evaluation into the overall value-based purchasing scoring algorithm. Once the methodology for CAHPS administration is standardized, AHCA may consider the proposed scoring algorithm. The CAHPS questions will be summarized by nine measures of satisfaction. These measures include four global ratings and five composite scores. The global ratings reflect overall satisfaction with Personal Doctor, Specialist, All Health Care, and Health Plan. The composite scores are derived from sets of questions grouped together to address different aspects of care: Getting Needed Care, Getting Care Quickly, How Well Doctors Communicate, Courteous and Helpful Office Staff, and Customer Service. The points for Adult CAHPS measures will be based on the National CAHPS Medicaid Percentiles for the global ratings and the composite questions. Table 4-2 displays the recommended scoring grid using national Medicaid CAHPS percentiles. Report on Value-Based Purchasing Methodologies Page 4-3

15 PROPOSED APPROACH Table 4-2 Scoring Grid for CAHPS Results National Adult CAHPS Medicaid Percentiles Points 90th Percentile 4 75th 89th Percentile 3 50th 74th Percentile 2 25th 49th Percentile 1 < 25th Percentile or Not Report (NR) 0 Annual Compliance Review Weighting of 10% According to the Balanced Budget Act of 1997 (BBA), 42 Code of Federal Regulations (CFR) , the state or its EQRO must conduct a review to determine the Medicaid managed care organizations compliance with standards established by the state for access to care, structure and operations, and quality measurement and improvement. The annual compliance review provides information on HMO processes for complying with state and federal requirements, and is important in the underlying performance of the HMO. Table 4-3 displays the scoring for this activity. According to the CFR, not all compliance standards need to be reviewed each year, and therefore, only those standards under review will be included. Any standards that need to be reviewed as a follow-up for corrective actions from the prior year s annual compliance audit will not be scored unless they are part of the current year s annual compliance audit for all of the HMOs. Table 4-3 Scoring Grid for Compliance Monitoring Results Compliance Monitoring Score Points 90% - 100% 4 80% - 89% 3 70% - 79% 2 60% - 69% 1 0% - 59% 0 Focused Studies Weighting of 10% The HMOs are required by AHCA to complete two focused studies each year. Focused studies typically target a service provided by the HMO, or concentrate on a specific disease or condition to direct improvement efforts. In general, focused studies should be included in the scoring. If the Report on Value-Based Purchasing Methodologies Page 4-4

16 PROPOSED APPROACH focused study is using a standardized measure with benchmarks (e.g., HEDIS measures), then the scoring would be similar to those found in the section on performance measures. For the EQR contract year , the focused studies topics were: Identification of People with Special Health Care Needs (SHCN), and Adolescent Well-Care Visits (AWC). The focused study for SHCN, however, cannot be used for this effort, because the study topic is not appropriate to use for comparisons of HMO performance. The final data for the AWC focused study will be available after July, The AWC focused study contains one HEDIS rate along with multiple other study indicators. AHCA may choose to score none, some, or all of the study indicators. To score the additional study indicators, HSAG recommends using the comparative benchmarking method (within this hybrid QISMC method) by awarding points based on the average score, plus or minus two standard deviations. Table 4-4 below shows the distribution of points based on the rates. Table 4-4 Scoring Grid for Compliance Monitoring Results Rate for Study Indicator Compared to HMO Average Points 2 Standard Deviations Higher 4 1 Standard Deviation Higher 3 Within 1 Standard Deviation 2 1 Standard Deviation Lower 1 2 Standard Deviations Lower 0 Performance Improvement Projects Weighting of 10% The purpose of the PIP is to achieve, through ongoing measurements and intervention, significant improvement that is sustained over time in both clinical care and non-clinical areas. This structured method of assessing and improving health plan processes is expected to have a favorable effect on health outcomes and member satisfaction. Additionally, as one of the mandatory EQR activities under the BBA, the state is required have at least one PIP that is conducted by its contracted HMOs to be validated. The points for PIPs will be based on the validation status of the PIP, along with the percentage of evaluation elements that received a Met status. PIPs that are relatively new and do not have baseline rates will not be scored. Only PIPs that have been validated will be eligible for scoring. Table 4-5 displays the recommended scoring grid for validated PIPs. Report on Value-Based Purchasing Methodologies Page 4-5

17 PROPOSED APPROACH Table 4-5 Scoring Grid for Validated PIPs Validation Status Percentage of Evaluation Elements with a Met Status Points Met 90% - 100% 4 Met 80% - 89% 3 Partially Met 70% or above 2 Partially Met 60% - 69% 1 Not Met Any Not Met Score 0 A statewide collaborative PIP eliminates the possibility that some HMOs may believe their PIP was more complicated or challenging than others (which may be true), and therefore should have special consideration, or not be scored at all. In addition, HMOs conduct more than one PIP, and the number of PIPs being conducted at a given time varies not all of the HMOs may be conducting the same number of PIPs. Therefore, a statewide collaborative PIP is recommended for scoring purposes to ensure all HMOs are evaluated on an equal basis. Otherwise, HSAG recommends using the highest scoring clinical PIP per HMO for the measurement period. CHCUP Participation Rate Weighting of 5% The CHCUP participation rate is a utilization measure designed to determine if Medicaid members under 21 years of age are receiving early, periodic, screening, diagnosis, and treatment (EPSDT) services, as required by CMS. However, low CHCUP rates may also be indicative of access-to-care barriers, member non-compliance, provider non-compliance, and/or low submission of encounter data from providers reimbursed through capitation payments. The CHCUP participation rate, therefore, is an important measure to consider in evaluation of performance across the Medicaid HMOs. Currently, the CHCUP participation rates as reported by the individual HMOs are required to be verified by an independent auditor. Although this verification occurs, there is not a formal, standardized method in place for the technical calculation of the rate. It is possible that the methods used to calculate the participation rates differ across HMOs, impacting the ability to compare performance. HSAG recommends the CHCUP rate should not be included as part of the valuebased purchasing program until it has been determined to be standardized and validated. Once the methodology for CHCUP has been validated and standardized, AHCA may consider the proposed scoring algorithm. Table 4-6 displays the recommended scoring algorithm for CHCUP participation rates. Report on Value-Based Purchasing Methodologies Page 4-6

18 PROPOSED APPROACH Table 4-6 Scoring Grid for CHCUP Participation Rates CHCUP Participation Rate Points 95% - 100% 5 90% - 94% 4 85% - 89% 3 80% - 84% 2 75% - 79% 1 0% - 74% 0 The remainder of this section includes a discussion of incentives and disincentives for consideration by AHCA, to accompany the scoring methodology. Incentives The two most common types of incentives used are ceremonial, non-monetary awards, and monetary incentives (e.g., bonus pools). A discussion of the most common types of incentives is described below. Non-Monetary Incentives Producing awards is relatively inexpensive, and provides the HMOs with an opportunity to shine. Publicly reporting performance results allows consumers to make more informed choices. HMOs will often use awards in advertising to attract more members, which has the added benefit of more members enrolling in the higher performing HMOs. Being recognized as a top performer may also increase commercial or Medicare enrollment for HMOs with other product lines. HSAG recommends using this type of award not only to recognize the superior performers, but to provide some level of rewards for HMOs that show considerable improvement from year-to-year. Honorable mentions may also be given to HMOs that come close to the superior performance level by demonstrating superior performance in several activities and average to above average in other activities. Reducing administrative requirements for those who meet or exceed performance standards is another potential, non-monetary incentive. An example of an administrative requirement may be a monthly report that must be created and sent to the State. Reducing the number and/or frequency of some reports may be beneficial to the HMO. AHCA should discuss reducing administrative requirements with the HMOs to determine which area could be most beneficial without creating additional issues within the various State departments. Auto-assignment of members can be used as both an incentive and disincentive for HMOs. Typically, auto-assignment is equally balanced among HMOs in the area. For high performing HMOs, the auto-assignment can be weighted, allowing more members to be assigned to the better Report on Value-Based Purchasing Methodologies Page 4-7

19 PROPOSED APPROACH performing HMOs in the area at a higher rate. Although this is not considered a direct monetary incentive, increased enrollment equates to an increase in the capitation payment for the HMO. Monetary Incentives Monetary incentives usually consider current rates compared to benchmarks or established goals (current measure), and improvement from the prior year s measurement (improvement measure). Current and improvement measures are described below in more detail. If three HEDIS measures were considered for the incentive payment, then the total payment would be comprised of six independent calculations. In addition, usually a maximum incentive payout amount is defined upfront. If the calculated payments exceed the maximum, then all payments will be proportionately reduced, such that the total equals the maximum. Current Measure for Monetary Incentives For each performance measure, the incentive will be paid in full if it is reported at or above the national Medicaid HEDIS 90th percentile. No incentive payment will be made if the reported amount is at or below the national Medicaid HEDIS mean. The incentive payment will be pro-rated if performance is reported between the national Medicaid HEDIS mean and 90th percentile. The most current national Medicaid HEDIS means and percentiles available at the time of the report will be used. The computation of the pro-rated amount will be a simple linear interpolation that compares the actual score to the mean and 90th percentile scores. For example, assume the payment incentive is based on a total of $0.20 per member per month, and that Utopia HMO obtains a HEDIS rate of 63.8 percent, which is between the HEDIS mean and the 90th percentile. The HEDIS mean is 59.9 percent and the HEDIS 90th percentile is 75.1 percent. The pro-rated amount would be calculated as follows: Calculation Example: National Medicaid 90th Percentile (75.1%) National Mean (59.9%) = 15.2% Current HMO rate (63.8) National Mean (59.9%) = 3.9% Difference between rate and mean (3.9%) / Difference between 90th percentile and the mean (15.2%) = 25.7% Therefore, the incentive award = x $0.20 = $0.05 per member per month. Improvement Measure for Monetary Incentives Improvement is computed by taking the nominal improvement (e.g., CY 2006 score minus the CY 2005 score) and dividing by the maximum possible improvement (100% minus the CY 2005 score). For each measure, an improvement of 10% or more will result in full payment of the allowable incentive. Failure to improve will result in no incentive payment. The incentive payment will be pro-rated if the improvement is positive, but less than 10%. Report on Value-Based Purchasing Methodologies Page 4-8

20 PROPOSED APPROACH A "full" bonus payment could only be achieved if both the 10% improvement and the 90th percentile benchmark were met for all measures. Partial achievement of the goals will result in a partial bonus payment. Disincentives The two most common methods of disincentives are the use of corrective action plans, and autoassignment of members. In most instances, the HMO submits a corrective action plan to the State for approval. The State reviews the corrective action plan and may request additional action, if necessary. The HMO is then routinely monitored by the state to ensure the corrective action plan is being followed. The other disincentive is based on how members may be auto-assigned if they do not choose an HMO. As mentioned previously under the incentives, auto-assignment is generally balanced among HMOs in the area. For poor performing HMOs, the auto-assignment can be weighted, allowing more members to be assigned to the better performing HMOs in the area at a higher rate. This may result in a loss of capitation payments to the HMO. Report on Value-Based Purchasing Methodologies Page 4-9

21 5. Next Steps for Implementation Throughout this report, there are noted areas that need to be discussed in order to assist AHCA in making important decisions. Inclusion or exclusion of certain activities, or measures within activities ultimately needs approval by AHCA. HSAG recognizes that there are both contractual and political ramifications to these decisions. Nevertheless, the overall concept of how to score, define, and award superior performance has been provided as a foundation. Upon agreement within AHCA to proceed with a value-based purchasing initiative, HSAG recommends the following steps: Step One Convene a Value-Based Purchasing Workgroup Upon agreement to move forward with the value-based purchasing initiative, AHCA should convene a workgroup to focus on discussing key decisions that will need to be made in order to prepare a preliminary methodology. The workgroup should include individuals from various bureaus within AHCA, in order to bring together a variety of skill sets, expertise, and work experience. AHCA should consider including at least one representative from the HMOs to assist in developing HMO buy-in. Additionally, key stakeholders should also be considered for inclusion, whether they will be involved in the actual decision making or not. A method for decision making must be identified, whether it is group consensus, or select individuals who are assigned final decision-making responsibilities. HSAG, as the EQRO, will be a co-leader of the workgroup, along with a designated co-leader from AHCA. The final output of the workgroup should be a proposed methodology and implementation plan. Step Two Solicit HMO and Key Stakeholder Input Once a proposed methodology and implementation plan have been developed by the workgroup, they should be provided to the HMOs and key stakeholders for review and an opportunity to provide feedback. AHCA may consider an open comment period, during which written comments are accepted and reviewed by the workgroup. AHCA should also consider hosting a meeting to discuss the methodology and implementation plan, solicit input, and answer any questions from the HMOs and key stakeholders. It is critical to allow the impacted HMOs and key stakeholders to provide input, which will foster HMO buy-in, an essential component of a successful value-based purchasing initiative. Step Three Finalize Methodology and Implementation Plan All comments and input received during the open comment period should be carefully and systematically considered by the workgroup. Final decisions on any modifications to the methodology should be made according to the decision-making method used by the group. AHCA should consider providing a response to all who submitted written comments, with an indication of whether the comment resulted in a change to the methodology, and if it did not, the rationale Report on Value-Based Purchasing Methodologies Page 5-1

22 NEXT STEPS FOR IMPLEMENTATION considered by the workgroup. The final result of this step should be a final methodology and implementation plan. AHCA may consider conducting a kick-off meeting to officially roll-out the initiative, although this step is optional. A kick-off meeting may further support HMO buy-in and enthusiasm for the initiative. HSAG has prepared a general implementation grid (Figure 5-1), which can be used as the basis for the development of the specific implementation plan. Figure 5-1 Implementation Key Steps and Considerations Year One Year Two Year Three Provide non-monetary rewards, including public recognition, awards ceremony, etc. Require immediate corrective action plans for low performers. Reduce selected items of administrative burden for high performers. Limit or increase autoassignment. Continue non-monetary rewards. Continue to require and monitor corrective action plans for low performers. Continue to reduce selected items of administrative burden for high performers. Consider expanding key measures of performance, including additional HEDIS measures. Implement monetary rewards. Continue non-monetary rewards, such as public recognition, award ceremony. When developing the specific implementation plan, the following items should also be considered: The value-based purchasing initiative should be rolled out slowly for a limited number of performance measures. HSAG recommends initially using the following three HEDIS measures: Well-Child Visits in the First 15 Months of Life (0 visits and 6+ visits), Breast Cancer Screening, and Chlamydia Screening in Women, Combined Rate. AHCA should review the current list of HEDIS measures and consider adding HEDIS measures that address important and costly aspects of care, such as diabetes, asthma, depression, and heart disease. HSAG recommends incorporating the statewide collaborative PIP into the initiative model to ensure all HMOs are on evaluated on an equal basis. An awards ceremony should be used initially; no monetary incentives should be provided in the first three years. HSAG recommends considering reducing administrative requirements on high performing HMOs after the first year, as an incentive for maintaining high performance. The amount Report on Value-Based Purchasing Methodologies Page 5-2

23 NEXT STEPS FOR IMPLEMENTATION and types of administrative requirements should be discussed further in conjunction with the HMOs. Corrective action plans can be implemented for low performing HMOs immediately. Typically this requires the HMO to submit a corrective action plan to the state for approval. The state monitors the HMO to ensure the corrective action plan is followed and effective. HSAG recommends low performing HMOs seek assistance through independent contractors or other entities to help improve performance. HSAG does not recommend using auto-assignment for incentives or disincentives for at least two years. CHCUP rate should not be included as part of the value-based purchasing initiative until it has been determined to be standardized and validated. The current CAHPS rates should not be included as part of the value-based purchasing initiative. However, if the State requires that the survey be administered in accordance with AHRQ and NCQA specifications, HSAG would recommend using the survey results in the value-based purchasing initiative. The workgroup should meet annually to reevaluate the methodology and make any necessary revisions. Report on Value-Based Purchasing Methodologies Page 5-3

24 Appendix A Scenario Analysis of Current AHCA Method of Calculating Plan Service Performance Plan Service Performance The following information was compiled from the Medicaid HMO Contract, and analyzed to determine if the current methodology provided a valid and appropriate method for awarding plan performance. Any potential issues are identified and discussed, along with recommendations on how to clarify, correct, or improve the methodology. The following table designates the weight assigned to each performance measure in the Agencydefined categories. Table 1 CAHPS CHCUP On-Site Audit (Minimum of 3) Quality Accreditation Compliance Improvement & Performance Indicators 15% 20% 15% 35% 15% The external quality review organization (EQRO) vendor may recommend how items will be scored to get these ratings. Overall scoring will apply until modified by the EQRO and would be measured as: Table 2 Rating Score Excellent % Commendable 81-90% Passing 71-80% Provisional 61-70% Failed <60% Potential Issue: The following issue is considered minor and would have negligible impact on the evaluation of plan performance. The overall scoring does not actually include 60 percent, as this is somewhere between failed (<60%) and provisional (61-70%). In this methodology, it is assumed a 60 percent would equate to the failed rating. Restructuring the ratings using , 80-89, 70-79, 60-69, and <60 would provide a more concise definition and align better with points awarded in the following sections. Scenario Analysis for AHCA s Current Method Page 1 HSAG

25 Appendix A Consumer Assessment of Health Plans Study (CAHPS) Survey CAHPS Survey results in yearly conducted survey data regarding the member s assessment of their satisfaction with health care services. Starting with the 2004 HMO report, the plans will be rated in five areas based on the latest CAHPS survey: Overall Plan Satisfaction Ease in Getting to See a Specialist Ease in Getting Needed Care, Tests, or Treatment How Well Providers Communicate with Members Getting Help from Customer Service Possible ratings are 0 to 5 stars in each area. The plan performance scoring is based on the value given by the CAHPS survey and is converted to stars. The CAHPS survey stars will be equated to points. The points are totaled and averaged into a overall composite rating for this section. For example, if two categories scored five stars (=2 x 15 points), two categories score four stars (=2 x 12 points), and one scored three stars (=1 x 8 points), the total score is 62 points averaged by the number of categories (rounded to the nearest whole number). The example s composite performance score in this section is 12 points. Based on Choosing A Quality Health Plan: Florida HMO Report 2005, and the Medicaid HMO Contract, the scores are converted based on standard deviations from the average CAHPS score for each question as follows in Table 3: Table 3 Stars Description Points Awarded ***** 1 or more standard deviations above the average score 15 **** 0.5 standard deviations above the average score 12 *** Within 0.5 standard deviations (above or below) the average 8 score ** 0.5 standard deviations below the average score 4 * 1 or more standard deviations below the average score 1 No Stars Undetermined assume measure is Not Reportable (NR) 0 Potential Issues: The following issues potentially could have significant impact on the evaluation of plan performance. Although the methodology states the plan performance is based on the value given by the CAHPS survey, it does not specify how the values are converted to stars. In addition, the CAHPS survey was not conducted using the NCQA standardized methodology. Scenario Analysis for AHCA s Current Method Page 2 HSAG