What Happens When Quality Payment Incentives Stop?

You don't have to go very far to hear about new bonuses or incentives being planned or implemented for providers and hospital chiefs to promote quality care. Rewards and bonuses always get our attention. But what happens when those incentives go away? Will that quality remain—or go away as well?

Writing in an article that appears this week in the British Medical Journal, British researchers found a dearth of information on the topic. They wanted to know because in April 2011, eight clinical indicators will be removed from the United Kingdom's Quality and Outcomes Framework, which has used pay-for-performance for primary care since 2004 to incentivize providers for various procedures.

Under the framework, providers receive bonus points—worth about $189 each in American currency—for using quality clinical care related to such areas as diabetes care, cervical cancer screening, and consultation length. Providers can get up to 1,000 points a year—which could mean a very nice annual bonus.

But to promote care in new and different clinical areas using limited funding, British officials have decided to remove some of the older indicators from the pay-for-performance framework. But what will happen to those indicators when the bonus plug is pulled is unclear, and little empirical studies seemed to be available, according to the researchers.

Their quest for answers eventually brought them to Kaiser Permanente Northern California. They looked at four clinical indicators Kaiser has used to rate quality of care in the plan’s adult population between 1999 and 2007: two indicators (diabetes glycemic control and hypertension control) continued to receive bonuses and two (screening for diabetic retinopathy and for cervical cancer) had been discontinued.

The researchers noted that the types of incentives used at Kaiser Permanente differed from the British system. For achieving target goals on a select list of clinical quality indicators, incentives would be awarded to the plan's facilities rather than individual physicians and providers.

This bonus money could be used to fund core facility operations, staffing, and quality improvement. And while incentives were at an organizational level (and not individual level), "this created alignment of leadership and engagement in performance improvement and also resulted in major investments in redesign," the researchers noted.

Some of the indicators were removed by Kaiser Permanente because rates of usage were relatively high, incentives did not seem to be leading to further increases, and it was thought that there were better opportunities to improve care, such as greater focus on cardiovascular risk reduction, according to the researchers.

For two of the conditions for which bonus payments were retained, the results were good. When incentives were used for glycemic control among diabetic patients, those adults achieving good control rose from about 41% in 2001 to almost 70% in 2007. For control of hypertension, the proportion of those reaching their goals (systolic blood pressure below 140 mm Hg) rose from 59% to 70% during that period.

During the five consecutive years (1999-2003) that screening for diabetic retinopathy received a bonus, screening increased from 85% to 88%; when the bonuses were dropped, though, it went to 80% in 2007.

Similarly, during the initial two years when financial incentives were attached for cervical cancer screening (1999 2000), screening rates rose slightly from 77% to 78%. During the next five years when no financial incentives were attached, rates fell to 74%. When incentives were then reattached for two more years (2006 7), screening rates increased again.

So put simply, after incentives were removed, screenings for diabetic retinopathy declined on average by about 3% per year and for cervical cancer by an average of 2% per year.

In their study, the researchers acknowledge that their study area is small. However, the questions they ask are big: what eventually happens when you remove payment incentives? Should payment incentives be used like training wheels—to get a certain clinical practice up and running and to promote quality care?

The researchers correctly note that if their findings are confirmed across a wider range of indicators, providers will need to be aware that if financial incentives are removed, their focus may change—and "they may need to think proactively about how to maintain previous levels of patient care."

But in the long run, healthcare policymakers need to think about what will happen when they remove—for whatever reason—financially incentivized clinical indicators. Rather than a blanket removal, maybe consider a stepwise reduction of payments against indicators, as the researchers suggest.

But what we need is something that has gotten very little attention when it comes to pay for performance: a good exit strategy.