TABLE C-1 Comparison Among Nonspecialist Groups’ Pay-for-Performance Position Statements
Design Issue |
American Medical Association |
American College of Physicians |
American Academy of Family Physicians |
Participation (Voluntary/Mandatory) |
Voluntary only; nonparticipation should not threaten economic viability of physician practices; must not favor participation of particular specialties, or groups of particular sizes or information technology (IT) capabilities; not linked to participation in other programs. |
Not applicable. |
Rewards for voluntary measurement and reporting; minimum number of encounters per patient per year; minimum number of patients/physician per year before data are considered valid. |
Unit of Accountability |
Physician practice groups and/or across health care systems (rather than individually) when feasible; no financial penalties based on factors outside of the physicians’ control. |
Physician office. |
Develop methodology to allow physicians to receive payments for achieving systemwide Medicare savings attributable to individual physicians, physician group practices, or physician-guided chronic care coordination. |
Improvement/ Excellence |
Both. |
Primary goal of pay for performance must be to promote continuously improving quality of care across the health care delivery system. |
Reward both improving performance and meeting performance targets. |
Weighting |
Quality focused. |
Data should be fully adjusted for case-mix composition. |
Start with strongest weight for structural measures, followed by process measures and finally clinical outcome measures in a phased implementation. |
Rewards/Penalties |
Rewards only. |
Should be directed at positive rather than negative rewards. |
Reward-based (positive incentives instead of withholds and penalties). |
Absolute or Tournament |
Absolute only (no comparative rankings). |
In the early (i.e., reporting) stage, rewards should be absolute; as program progresses to pay based on actual performance, rewards should be balanced between rewarding high performance and rewarding substantial improvement over time. |
Incentive payments should reward both performance improvement and meeting performance targets and not limited by tournament-type incentives. |
Payment |
Minimize potential financial and technological barriers, including costs of start-up; reimburse physicians for administrative costs. |
Rewards must be greater than the cost of the physician’s participation in a P4P program. |
Financial rewards should cover administrative costs of participation and should increase proportionately based on the number of dimensions of care, time, and costs associated with documentation/IT. |
Funding |
New funds. |
Ideally, new funds; if savings result, physicians should benefit from these savings; oppose withholds. However, due to the current fiscal environment, ACP recognizes that a redistribution of funds across and within geographic locations and specialties and between physicians and hospitals or other health care providers may be necessary. |
New money and/or redistribution (e.g., from Part A); positive sustainable growth rate updates as a floor; no penalties for potential volume increases; support funds by setting aside a portion of the recommended inflation updates for 2006/7. |
Phasing |
Programs phased in to include all voluntary physicians (after pilot testing). |
Supports incentives for IT—especially in the beginning for small practices. |
Payment for structural (e.g., IT use) improvements, followed by payment for reporting, leading to payment for performance using validated measures (e.g., National Quality Forum [NQF], Ambulatory care Quality Alliance [AQA]). |
Design Issue |
American Medical Association |
American College of Physicians |
American Academy of Family Physicians |
Measure Development and Selection |
Evidence-based; prospectively defined; allow for variation based on physician’s patient-specific clinical judgment and patient preferences; developed collaboratively across specialties; risk-adjusted, stable for 2 years; aimed at areas with significant promise for improvement; pilot-tested, analyzing for patient de-selection; physicians review/comment on the accuracy and validity of both the data and analysis before use (see their preliminary ratings and adjust practice before public release). |
Generally agrees with MedPAC; measures should be evidence-based, valid, and reliable; relevant to physician’s clinical responsibilities; practical; relate to clinical conditions of highest priority; selected with stakeholder consensus. |
Clinical measures validated by multi-stakeholder process, with full transparency; utilize valid peer groups, evidence-based statistical norms, and/or evidence-based clinical policies; incentives for adoption/utilization of IT, implementation of systems to improve quality of care and safety, access to timely care, patient acceptability and satisfaction with care. |
Administrative vs. Chart Data |
Both; medical record data collection must not be burdensome. |
Does not support MedPAC’s recommendation for submitting laboratory test values on claims. Chart abstraction data should be used only when the clinical/ public health benefit clearly outweighs the burden and disruption of this kind of data collection. |
Supports combination approach with phased implementation as described above and disclosure of data source(s) (e.g.,administrative, chart audits, surveys, pharmacy). |
Care Coordination |
Encourage collaboration across all members of the health care team. |
Fundamental reform of physician payment system needed; reimbursement should not be based on volume and episodes of acute illnesses, but on patient-centered, physician-guided care coordination and quality performance based on evidence-based clinical measures. |
Supports care management fee for patient’s chosen “personal medical home” and/or chronic care management fee for patients with selected chronic diseases. |
TABLE C-2 Comparison Among Specialist Groups’ Pay-for-Performance Position Statements
Design Issue |
Alliance of Specialty Medicine |
Society of Thoracic Surgeons |
American College of Cardiology |
Participation (Voluntary/ Mandatory) |
Voluntary through pilots. |
Voluntary. |
Not applicable. |
Unit of Accountability |
Not applicable. |
Cardiac Thoracic surgeon group/ hospital. |
Physician groups rather than individual measurement; encourage collaboration between physician groups and across specialties (e.g., specialty and primary care groups). |
Improvement/ Excellence |
Not applicable. |
Primary focus is on continuous quality improvement through compliance with process measures linked to quality and tracking of risk-adjusted outcome measures. Secondary focus is on achievement of excellence as defined by credible evidence-based thresholds for performance measures. |
Excellence based on agreed-upon baseline standards. |
Weighting |
Not applicable. |
Based on National Quality Forum (NQF) endorsed measures as either single measures or roll-up (composite) measures created with appropriate statistical modeling and risk adjustment for all outcome measures. Composite measures should include four domains of care: perioperative medical care, operative care, postoperative risk-adjusted mortality, and postoperative risk-adjusted morbidity. |
Not applicable. |
Design Issue |
Alliance of Specialty Medicine |
Society of Thoracic Surgeons |
American College of Cardiology |
Rewards/ Penalties |
Rewards only. |
Rewards; no penalties. Rewards recognize continuous improvement in the quality of patient care and achievement of scientifically credible and achievable performance thresholds. |
Rewards only. |
Absolute or Tournament |
Not applicable. |
Absolute. |
Absolute. |
Payment |
Not applicable. |
Tiered incentive for achievement of structural, process, and risk-adjusted outcome measures. Advocates three phases: pay for participation in a clinical database, pay for demonstration of improvement over historical baselines, pay for performance through achievement of performance thresholds. Increased incentives for level of attainment of thresholds. |
Rewards should correlate with investments made to improve care and sustainability of performance (e.g., time, training, technology). |
Funding |
New funds should be available if necessary. |
Budget-neutral framework should not be tournament style (reducing all physician fees to create incentive pool), but should be based on shared savings generated through reductions in complications realized through continuous quality improvement and tracking of process measures linked to quality and risk-adjusted outcome measures. |
Not applicable. |
Phasing |
Should be phased in for physicians willing to participate. |
Begin with payment for structural measures (pay for participation in a clinical database/pay for reporting) and move toward payment for process measures linked to quality and achievement of risk-adjusted outcome measures demonstrated to reduce costs. With feedback on performance, physicians should be encouraged to employ continuous quality improvement and engage in the creation of quality-focused cost containment. |
Not applicable. |
Measure Development and Selection |
Evidence-based measures developed by physicians pertaining to issues that physicians can control; specialty specific; data collection and reporting should not be burdensome to practices. |
Clinically relevant, scientifically valid and credible measures developed by physicians and their respective specialty societies. All measures should eventually be vetted through the NQF. Supports adoption of a consensus set of structural measures; process measures linked to quality and risk-adjusted outcome measures. Measures should be consistent with the principles and criteria recommended in the Institute of Medicine’s 2006 Performance Measurement report. Attribution must be set to address systems of care and not individual physician measurement where appropriate. |
Evidence-based and developed with a credible methodology. Risk-adjusted and clearly defined measures with consistent definitions; all data should be audited; providers should have the ability to comment on the data. |
Design Issue |
Alliance of Specialty Medicine |
Society of Thoracic Surgeons |
American College of Cardiology |
Administrative vs.Chart Data |
Supports the Centers for Medicare and Medicaid Services’ (CMS) Physician Voluntary Reporting Program that uses administrative data. |
Clinical data must be used to drive quality improvement. Administrative data should be linked to clinical data through a blending of specialty-specific databases and CMS financial data only to evaluate costs and efficiency, but not determine quality. |
Chart data preferred over administrative data. |
Care Coordination |
Not applicable. |
Encourages collaboration among providers and across disease conditions and care settings. Attribution (individual physician, physician group, hospital) must be appropriate inorder to encourage effective hand-offs and transitions of care within and across episodes of care. Pay-for-performance programs must evaluate and reward systems of health care delivery. |
Encourages collaboration among providers. |
TABLE C-3 Comparison Among Purchaser and Consumer Groups’ Pay-for-Performance Statements
Design Issue |
Alliance of Community Health Plans |
National Buisiness Group on Health |
National Patient Advocacy Foundation |
Participation (Voluntary/ Mandatory) |
Mandatory. |
Voluntary |
Voluntary. |
Unit of Accountability |
Health plans, physicians, hospitals, and other providers. |
Not applicable. |
Not applicable. |
Improvement/ Excellence |
Both; favors excellence. |
Both. |
Not applicable. |
Weighting |
Favors clinical performance. |
Rewards specifically related to the quality of care provided, with a focus on efficiency. |
Not applicable. |
Rewards/ Penalties |
Rewards. |
Rewards. |
Incentives that encourage improved quality of care delivery that will benefit both patients and providers; reimbursement systems should not be punitive. |
Absolute or Tournament |
Not applicable. |
Not applicable. |
Not applicable. |
Payment |
Not applicable. |
Not applicable. |
Allow provision for assisting physicians in obtaining the technology needed to participate. |
Funding |
New, dedicated stream of funding (add-on). |
Not applicable. |
Not applicable. |
Design Issue |
Alliance of Community Health Plans |
National Buisiness Group on Health |
National Patient Advocacy Foundation |
Phasing |
Start with Medicare Advantage plans and extend to physicians and hospitals (favor clinical). |
Not applicable. |
Not applicable. |
Measure Development and Selection |
Continue to evaluate Medicare plans; use evidence-based measures. |
Performance measures should be incorporated in addition to structure and process measures; Medicare measures should be those developed by nationally recognized quality measurement organizations such as the National Committee for Quality Assurance or the National Quality Forum. |
Not applicable. |
Administrative vs. Chart Data |
Both. |
Possible financial assistance to providers and physicians in low-income urban areas to help with the purchase of software or systems needed. |
Both, not to be overly burdensome or costly. In addition include interoperable electronic health infrastructures. |
Care Coordination |
Support care coordination (MedPAC). |
Care coordination, hand-offs, and the team approach should be a part of pay for performance and measured appropriately, rather than solely single practitioners or facilities being measured for their portion of the treatment of an episode of care. |
Collaboration among all stakeholders to include participation from the patient, physician and provider communities. |
TABLE C-4 Ambulatory care Quality Alliance
Design Issues |
Ambulatory care Quality Alliance (AQA)a |
Participation (Voluntary/Mandatory) |
Voluntary. |
Unit of Accountability |
Physicians and physician groups. |
Improvement/Excellence |
Not applicable. |
Weighting |
Alignment or linkage of quality and cost of care measures. |
Rewards/Penalties |
Not applicable. |
Absolute or Tournament |
Not applicable. |
Payment |
Not applicable. |
Funding |
Not applicable. |
Phasing |
No discussion to date. |
Measure Development and Selection |
AQA does not develop measures; it approves measures developed by National Committee for Quality Assurance, Joint Commission on Accreditation of Healthcare Organizations, American Medical Association PCPI, and other medical specialty organizations that meet AQA-defined parameters. |
Administrative vs. Chart Data |
Implementation of measures should be as least burdensome as possible. While the AQA Performance Measurement Workgroup acknowledges that administrative data should be considered as the logical starting point, there is interest in moving beyond claims and other administrative data as soon as is practicable. As appropriate, measures derived from medical chart review should not be excluded. |
Care Coordination |
Not applicable. |
aThe AQA does not specifically address or take a position on pay-for-performance programs; rather, its work focuses on evidence-based, valid, reliable performance measures for quality improvement and accountability. AQA encourages implementation of standardized performance measures endorsed by AQA for physician assessment, one component of which may be pay-for-performance programs. |