By ricHard A. norling
and Stephanie Alexander
Regardless of how it’s funded, pay for performance, or value-based purchasing, is coming.
Congress has mandated that the Centers for Medicare and Medicaid Services develop a plan by late 2008 for hospital value-based purchasing. Recently, the Institute of Medicine urged CMS to gradually phase in P4P nationwide as a way to accelerate quality improvement.
CMS is hard at work developing that plan, and its Hospital Quality Incentive Demonstration project with the Premier healthcare alliance will be one model they examine closely.
Since October 2003, Premier and CMS have been conducting the HQID project with more than 260 hospitals from 37 states. The first national project of its kind, HQID is designed to determine if economic incentives are effective in improving inpatient care quality. Participating hospitals report process and outcome measures in five clinical areas – acute myocardial infarction, congestive heart failure, coronary artery bypass graft, pneumonia and hip and knee replacement.
Over the first two years of the project, more than $17 million has been awarded to top-performing hospitals. The average composite quality score has risen 11.8 percent in that time, meaning better care and outcomes for more than 800,000 patients. For example, data on the reported mortality rate for AMI suggests that more than 1,200 patient lives have been saved. In February, CMS officially extended the project for three more years.
The program created executive focus on quality improvement, elevating the discussion to hospital executives and the board, which has been a critical success factor in performance improvement. The process identified top performers and rapidly disseminated best practices, all of which produced improved performance.
The question of whether the incentives are driving improvements misses the point that this is a real-world project designed to improve patient outcomes. The success of the project lies not just in the fact that it has proven that pay-for-performance can work, but in the improved care patients are receiving.
To determine the impact of pay-for-performance on patient care if such standards were implemented nationally, Premier used the following methodology. Reliable care was measured in this study by the percent of recommended care processes a patient actually received. If a patient received all of the care process measures, the patient was included in the “High” reliable care category. Similarly, if the patient received less than half of the measures, he or she was included in the “Low” category. Otherwise, they were grouped in the “Medium” category. After patients were grouped, the average hospital cost, mortality rate, readmission rate, complication rate and average length of stay were calculated for patients in each category.
The analysis showed that if patients in the Low and Medium groups had received High reliable care, it could have resulted in nearly 5,652 fewer deaths, 6,000 fewer complications, 10,000 fewer re-admissions and 1 million fewer days in the hospital. In addition, hospital costs could have been as much as $1.4 billion lower.
An important lesson of the past several years, confirmed by the demonstration project, is to focus on the bundle of measures rather than individual indicators. The benefit to patients is greatest when hospitals can reliably deliver a bundle of care rather than a single measure in isolation.
The HQID project also indicated that P4P programs need to address, among other aspects, undue fragmentation, duplication and after-the-fact inspection, all of which result in suboptimal effectiveness and efficiency. The programs also need to negate complications and errors, which are strongly associated with high cost of care, re-admissions and mortality or disability.
Incentives coupled with transparency are strongly preferable to penalties to create systemic improvement, and hospitals should be able to share savings with other stakeholders, particularly physicians.
To accelerate a rising tide of improvement, new P4P programs should focus on building the productive capacity of the healthcare delivery system, improving the reliable execution of evidence-based medicine, managing handoffs between care levels and sites, removing financial and regulatory barriers to integrated care for beneficiaries and measuring return on investment via population-based efficiency and effectiveness measures.
It’s best for hospitals and patients if there is greater consistency and connection among the plethora of P4P programs available today. The administrative burden on hospitals becomes too great when every insurer has a unique set of measures and data collection methods, for example.
To achieve this, Medicare, as the largest payer, should lead the way. That includes identifying appropriate data elements to track and prompting research into best practices. Other payers should follow Medicare’s lead, realizing that an all-payer approach, not a patchwork of dozens of programs, is best for hospitals and patients.