California Physician Performance Initiative (CPPI)
Project Summary
CPPI is a multi-stakeholder initiative to measure and report on the performance of Physicians throughout California. This work is being conducted by the California Cooperative Healthcare Reporting Initiative (CCHRI), which is a statewide collaborative of physician organizations, health plans, purchasers and consumers that are working collectively to help consumers and purchasers make informed health care decisions.
The California Physician Performance Initiative (CPPI), begun in 2006, has developed a system to measure and report the quality of patient care that is provided by individual physicians in California.
CPPI’s goal is to improve patient care and its affordability by:
- Reporting results to physicians to help them gauge how well care for their patients meets national standards of care.
- Applying the performance results in ways to help consumers and purchasers get better value when they choose and use health care; and
- Adopting performance measures and reporting methods using the best available science to set performance standards.
Physician Quality Performance 2009 Reports
California Physician Performance Initiative clinical quality results were mailed to more than 13,000 California physicians on July 24, 2009. CPPI's physician-specific results are derived from the medical claims data aggregated across California's three largest commercial PPO health plans (Anthem Blue Cross, Blue Shield of California and United Healthcare) and the Anthem Blue Cross and Blue Shield of California commercial HMO health plans.
An explanation and sample of the 2009 Physician Performance Report may be found here.
Physician Quality Report Review and Corrections
Upon mailing the CPPI Performance Reports to physicians, each physician was asked to review and, as needed, correct their demographic record or quality scores. Physician feedback was very important as some project data is incomplete or wrong due to errors in claims submission or processing.
During an eight-week process, from July 24, 2009 to September 18, 2009 physicians were encouraged to request and verify the list of patients attributed to them and the accuracy of the results per the designated services for the quality measures. Any corrections that were provided by September 18, 2009 were applied to correct the quality results before the information was provided to health plans. Nearly 1,200 physicians submitted corrections. A physician can request a copy of their corrected scores here.
Clinical Quality Measures Used in Report
CPPI assessed physician performance using clinical quality measures that are evidence-based, nationally standardized and endorsed by major standard-setting bodies (i.e., the National Quality Forum and the AMA's Physician Consortium on Performance Improvement). The measures address preventive care and chronic condition management. The measures were reviewed and approved by the CPPI Physician Advisory Group.
The 17 measures listed below are relevant to the commercial population and can be scored using the administrative claims data available in California. CPPI uses measure specifications from the National Committee for Quality Assurance (NCQA), Healthcare Effectiveness Data and Information Set (HEDIS), and the Physician Consortium for Performance Improvement measurement systems.
A list of clinical quality measures used in the report may be found here.
Adult Diagnostic and Preventive Care
Diabetes Care for Adults
Cardiovascular Disease Care
Medication Management
Chronic Obstructive Pulmonary Disease Care (COPD)
Musculoskeleton Conditions
How Patients Were Attributed to a Physician
Each patient who was eligible for a measure was attributed to the physician(s) of the relevant specialties with whom they have a visit during the time specified for that measure. The relevant specialties were:
- Internal medicine and family/general practice for all measures;
- Cardiology for the six cardiovascular measures, the monitoring medications measure, and the diabetes LDL screening test measure;
- Endocrinology for the three diabetes measures and the two coronary artery disease measures;
- Obstetrics and gynecology for the breast cancer screening and cervical cancer measures;
- Gastroenterology for the colorectal cancer screening measure;
- Nephrology for the diabetes nephropathy measure;
- Allergy/immunology and pulmonology for the pharmacotherapy management of COPD and spirometry measures; and
- Rheumatology for the disease modifying anti-rheumatic drug therapy for rheumatoid arthritis measure.
Different rules were applied to attribute patients to Primary Care Physicians (PCPs) and to sub-specialists.
A PCP is deemed a relevant specialist for all seventeen (17) measures. The attribution rule assigns a patient to a single PCP with whom the patient had the most ambulatory/outpatient evaluation and management (E&M) visits during the measurement year and one year prior. There is one exception to this rule - PPO patients are not attributed to a PCP if the patient only had a single visit with a PCP during the two-year interval. HMO patients can be attributed to a PCP based on a minimum of a single visit with that PCP.
Patients were attributed to any measure-relevant sub-specialist with whom the patient had at least one E&M visit during the measure's attribution period. Patients can be attributed to multiple, relevant sub-specialists for a given measure. For example, patients who qualify for the diabetes LDL screening measure could be attributed to an endocrinologist and a cardiologist if both types of physicians had an E&M visit with a patient during the attribution period. A physician's specialty is determined by their primary specialty only.
How Composite Scores Were Constructed
A set of CPPI 2009 quality measures were organized into topics and then physician-specific composite scores were calculated. These composite scores are a measure of performance that combine multiple individual, related measures. The advantages of composite scores include:
- improve reliability of measures to better distinguish performance
- are succinct: communicates results in a summary, easy to digest way
- help to generalize level of performance for an area of patient care by making inferences beyond the individual measures - e.g., a set of diabetes measures can represent overall care for patients with diabetes
Individual quality measures were organized into four (4) clinical domains: prevention, diabetes, cardiovascular and respiratory. The clinical domains organization helps to ensure that the individual measures are related to a common underlying quality construct (e.g., diabetes care). The composite score is the combined results of all patients who are attributed to the physician for the quality measures that are clustered within a given domain. More information on the composites and relevant specialties for each composite may be found here.
How These Performance Results Will Be Used
Blue Shield of California notified its network physicians of the plan's intent to use CPPI quality results to recognize top performing physicians. This recognition is displayed in the plan physician directory that is provided to health plan members. Physicians are recognized for results that affirm that a threshold number of patients have received the designated service for a given quality measure. The performance recognition is measure-specific - a physician can merit recognition for any of the CPPI measures for which they have reliable results.
The other two CPPI participating plans, Anthem Blue Cross and United Healthcare, have not announced any plans to use CPPI results at this time.
CCHRI recommended the following elements of a method for determining and designating “top performing physicians”:
- Use a single performance designation that recognizes a physician as a top performer or omits the physician from any performance recognition. There would not be gradations of performance (e.g., top, middle, bottom).
- Apply a statistical technique to distinguish the top performing physicians from those who are not recognized as top performers – such methods to be transparent and available to anyone.
- Seek to recognize physicians scored in the upper half of the performance distribution but the threshold may be somewhat higher than the 50th percentile to truly distinguish top performers.
Give physicians with uncorrected scores the opportunity to view their scores in the context of the CPPI performance distribution and/or actual designation (e.g., physician is or is not a recognized top performer); physician can weigh this information in deciding to pursue corrections. Such physician communications should explain that health plans may determine other business uses in the future and the physician should weigh that in the decision whether to pursue data corrections. Health plan should notify physicians in advance about its schedule for public/business use of the results as part of the “top performing physician” designation. A health plan could determine if corrections submitted after the close of the CPPI corrections period or the plan’s public/business use date would be incorporated in the top performing physician’s designation formula at some point in the future but this would not impede the scheduled public/business use of the data.
CCHRI also recommend the following general guidance about the uses of the 2009 quality measures.
- Uses should be based on reliably scored results only.
- The three respiratory measures should not be used unless measure specifications are validated.
- The two heart failure measures should not be used unless measure specifications are validated.
- Disclaimer/explanatory text should accompany any public reporting to explain data limitations.
- Plans should adhere to Patient Charter principles.
- CCHRI stakeholders should adopt a data improvements plan and milestones to evaluate progress throughout 2010 with a target of implementing the data improvements not later than January 1, 2011. These data improvement objectives to include the following:
a) Health plans adopt internal data validation procedures and produce results to be shared with CPPI stakeholders
b) Health plans revise claims/encounter policies/procedures aimed at ensuring data completeness
c) Medical groups/physicians submit complete HMO encounter datasets
d) Medical groups/physicians submit individual physician ID in rendering provider field on all claims and encounters
e) Coding conventions reviewed and code standards affirmed/adopted (e.g., use of physician NPI, place of service codes, any ambiguities in measures’ code sets, etc.)
f) Methods changes to lower patient sample sizes and reduce burden of future patient list reviews
g) CPPI staff to prepare and distribute best practices educational materials to support physicians and practices in implementing changes to best meet the CCHRI data improvement plan objectives.
- For any CPPI plan that pursues a CPPI 2009 quality measures or data business use in the future, the plan should inform CCHRI of its methodology in advance of its use. At that time, CCHRI could provide additional guidance to the plan per the proposed methodology.
Physician Specialties Included in Reporting
All of the CPPI measures apply to internal medicine and family/general practice physicians. In addition to primary care, some CPPI measures apply to various physician specialties including the following relevant specialties by measure:
- Internal medicine and family/general practice for all measures;
- Cardiology for the six cardiovascular measures, the monitoring medications measure and the diabetes LDL screening test measure;
- Endocrinology for the three diabetes measures and the two coronary artery disease measures;
- Obstetrics/gynecology and gynecology for the breast cancer and cervical cancer screening measures;
- Gastroenterology for the colorectal cancer screening measure;
- Nephrology for the diabetes nephropathy measure;
- Allergy/immunology for the pharmacotherapy management of COPD and spirometry measures;
- Pulmonology for the pharmacotherapy management of COPD and spirometry and measures; and
- Rheumatology for the disease modifying anti-rheumatic drug therapy for rheumatoid arthritis measure.
How Reliable Are the Scores?
A reliability statistic was used to affirm that the results for a physician's sample of patients were representative of the true results if all of a physician's patients were included. A minimum reliability of 0.70 (on a 0.0 to 1.0 scale) is used as the threshold to be sure that the patient sample was large enough to yield consistent results.
CPPI Decision-Making and Governance
The CPPI Physician Advisory Group provides clinical review and guidance related to the design of the program, selection of measures, review of findings, and presentation of results to physicians. Additionally, the CCHRI Executive Committee provides strategic oversight, including development and implementation of the CPPI communication plan. Executive Committee members represent physician groups and associations, health plans, purchasers and consumers. Data collection, analysis and reporting is managed by the Pacific Business Group on Health (PBGH) on behalf of CCHRI.
The California Physician Performance Initiative is sponsored by the California Cooperative Healthcare Reporting Initiative (CCHRI) which is a multi-stakeholder collaborative of purchasers, health plans, physician groups and associations, and consumers that are working collectively to improve patient care in California. Please see www.cchri.org for more information.
Background
Propelled by the national AQA Alliance, the mission of the CPPI and similar endeavors across the U.S. is to “improve health care quality and patient safety through a collaborative process in which key stakeholders agree on a strategy for measuring performance at the physician or group level; collecting and aggregating data in the least burdensome way; and reporting meaningful information to consumers, physicians, and other stakeholders to inform choices and improve outcomes.”
This measurement and reporting initiative, taking place in phases over several years, and with the involvement of many stakeholders, informs the development of comprehensive, evidence-based national standards to measure the quality and cost of care provided by individual physicians.
The CPPI will help to set national standards for measuring, scoring and reporting physician performance.
In 2006, the Centers for Medicare and Medicaid Services (CMS) provided funding to aggregate Medicare fee-for-service and commercial claims data to calculate and report quality measures as part of a national effort to establish physician performance standards. (The six-site pilot project was known as the Better Quality Initiative (BQI). The voluntary addition of data from California’s three largest commercial PPOs (Anthem Blue Cross, Blue Shield of California, and United Healthcare) provided a large enough pool to test the reliability of an initial set of 15 quality measures as well as methods for attributing patient based on claims data of patient care provided in 2007. You may read the 2008 BQI final report, Enhancing Physician Quality Performance Measurement and Reporting Through Data Aggregation here.
|
