Years ago, I was on a panel where some people were arguing that paying docs for quality was the silver bullet we were waiting for. As you can imagine, I was the skeptic on the panel. After three other people talked about how offering docs bonuses for quality metrics would spur them to change their behavior, I presented them with the following thought experiment:
Let's say that you can see 20 patients in a day. You are told that from now on, 5% of your pay will be tied to quality. So you can invest in a new EMR that will help identify patients in need of support, you can pay for a case manager in your clinic to help manage them, you can invest in a social worker to try and reduce barriers to improving care, you can collect data and track these metrics, hold countless meetings to discuss and change your clinic's operations, and maybe you'll get that 5%.Or, you could do none of those things, and just try and squeeze in one more patient and just make up the 5% on volume. What do you think docs will do?
I'm not sure how popular I was at the conference, and the other panelists seemed to feel I didn't think very highly of doctors. But I'm swayed by the fact that financial incentives do drive physician behavior. That's why some think "paying for quality" will work. I'm not so sure. A recent study supports my viewpoint, though. "Does Compensating Primary Care Providers to Produce Higher Quality Make Them More or Less Patient Centric?":
Both payment reform and patient engagement are key elements of health care reform. Yet the question of how incentivizing primary care providers (PCPs) on quality outcomes affects the degree to which PCPs are supportive of patient activation and patient self-management has received little attention. In this mixed-methods study, we use in-depth interviews and survey data from PCPs working in a Pioneer Accountable Care Organization that implemented a compensation model in which a large percentage of PCP salary is based on quality performance. We assess how much PCPs report focusing their efforts on supporting patient activation and self-management, and whether or not they become frustrated with patients who do not change their behaviors.
Researchers used mixed methods to interview and survey primary care physicians who were part of a Pioneer Accountable Care Organization, and who had a large percentage of their salary based on quality. Specifically, the researchers wanted to look at how primary care docs were engaging in efforts to support patient activation and change their self management, versus other strategies.
The first survey asked them what they intended to do over the next six months to increase their income, before a lot of changes occurred. Only 10% reported that they planned to increase patient activation and/or increase patient self-management skills. About 27% reported that they planned to improve team effectiveness. But more, or 28%, reported that they planned to increase their relative value production (RVU). Almost a third (32%) said they planned to build their panel size.
They planned to make it up in volume.
A year later, they were surveyed again to see what they had actually done. Only 15% had increased efforts to activate patients or support their self-management to "a large or very large extent". About 60% of them said they'd made little or no change to support this.
When asked about the obstacles to improving quality, a quarter of doctors responded that they didn't know how to support patients effectively in making behavior change. But almost half cited a lack of resources, such as coaches, nurses, and educators - staff positions that have been cut or reduced by many hospitals and clinics in recent years. Two-thirds said that they lacked the time to spend with patients on these efforts. But the most common response - given by 70% of physicians - was that they felt that patients were unwilling to change their behavior.
The most commonly reported frustrations with this payment model had a similar flavor. They did not like that patients' lifestyle behaviors influences their salaries (36%), they felt the models were too complex (36%), and that quality metrics were not good indicators of quality care (35%). From the Discussion:
Although PCPs acknowledge that patient behaviors largely determine quality metrics, most PCPs did not put their efforts into increasing patient activation and patient self-management. Furthermore, the results indicate that the PCPs did not become more patient centered, as evidenced by no increase in the use of patient-partnering behaviors, after the introduction of the incentive. And yet it is these patient partnering behaviors that are likely to improve self-management and ultimately improve clinical outcomes.
Touting proclamations that we will now "pay for quality and not quantity" sound great, but are likely not enough. Systems that support patients in changing behaviors and management outside the medical setting are likely needed to achieve the goals we all want. Unfortunately, the system invests in this infrastructure in sub-optimal ways. Physicians may compensate for this by trying to increase volume in other ways, which would subvert the intent of the policy in the first place.