Hospital Groups Back NQF Report on Patient Sociodemographics

The American Hospital Association wants CMS to adjust measures in pay-for-performance programs to reflect factors such as race and income, even though doing so "would be a complex undertaking" for everyone involved.

Two hospital associations are providing a ringing endorsement of a National Quality Forum draft report that recommends risk adjusting quality measures for hospitals' patient mixes to reflect socioeconomic factors such as insurance coverage, race, and income.

The American Hospital Association on Monday sent a letter to NQF and the Centers for Medicare & Medicaid Services urging them to act quickly on the report's recommendations.

"A large body of evidence demonstrates that sociodemographic factors such as income and insurance status affect many patient outcomes, including readmissions and costs," AHA Executive Vice President Rick Pollack writes in the letter.

"Sociodemographic adjustment allows for all providers to be fairly and accurately assessed on the quality of care they provide and their contribution to patient outcomes while mitigating negative unintended consequences of measurement. Identifying appropriate sociodemographic adjustments also may help to highlight the impact of those factors on patient outcomes, allowing them to be addressed."

AHA wants CMS to use the NQF guidelines to adjust measures in quality reporting and pay-for-performance programs on markers such as readmissions, even as Pollack concedes that adopting the recommendations "would be a complex undertaking" for everyone involved.

'Really Important Things to Consider'
"The negative unintended consequences of failing to adjust measures for sociodemographic factors are substantial," Pollack said in the letter, which was sent at the close of the 30-day comment period. "Moreover, we urge NQF to place a high priority on working with CMS to rapidly address its measures."

"Lastly," the letter says, "we concur with the panel's recommendation that NQF expand its role by developing more detailed implementation guidance for measures, and clarifying for what uses a measure is endorsed."

Nancy Foster, AHA's vice president of quality and safety policy, says clinical factors are already baked into hospital outcome measures to reflect hospitals that serve sicker or older populations. The NQF recommendations take the process one step further.

Reduce Readmissions by Activating Patients to Do 'Self-Care'

"This is a report that says beyond the clinical factors there are really important things to consider when you are looking at outcomes, particularly outcomes that you are measuring, and a considerable amount of times measuring post hospitalization," Foster says.

"Things like a patient's ability to follow the instructions that were given to him or her upon discharge, whether they can get to a rehab facility or they can get to physical therapy or they can get out and exercise in a safe environment or they can find appropriate food in their local grocery store. All of those things play into whether or not a patient actually recovers as quickly and as well as we'd all want, and certainly the patient would want," she says.

AHA and other hospital groups have been calling for a socio-demographic metric for years and Foster says the federal government may now be responding to research.

"Enough questions and enough research has been published that Health and Human Services thought it was important for NQF to look at this question and determine whether there was something really there," Foster says. "We have reason to believe that the sociodemographic factors are significant and outside the control of providers, and perhaps should be adjusted for."

"There have been any number of stories," says Foster, "usually published about readmission measures, and because they have significant penalties attached to them and because the data are out there now one can actually do the analyses."

Safety Net Hospitals Also Supportive
Beth Feldpush, senior vice president for policy and advocacy at America's Essential Hospitals, said the safety net hospital lobby "strongly supports the NQF panel's recommendations."

"We think that there is a large body of emerging evidence showing the sociodemographic factors can influence health outcomes," Feldpush says. "That includes these factors in risk-adjustment models where there is evidence to do so will really improve the science of performance measurements."

Practically speaking, Feldpush says bringing sociodemographic data such as income, race, and access to insurance will create "a more accurate picture of performance, which means that some institutions that are doing very well now may not do as well, and some institutions that are not doing as well now may do better. We think the overall effect is that it will improve accuracy and really shed light on the true performance for various hospitals."

Currently none of that is taken into account now. "For individual patients for example with readmissions measures, their medical history is taken into account and I guess some very common socio-economic demographics such as age and gender are used," Feldpush says. "But you don't get any sense as to the individual's income level or education level or anything related to the community in which they live."

Feldpush says she can't predict how safety net hospitals or any entire class of providers will be affected if CMS takes up the NQF recommendations. "We think it will really improve the clarity of the picture and you're certainly going to see a difference in performance across the field," she says.

Data Collection 'Will Take a Little Time'
"But we believe that because of the patients that we serve, having this information in the risk adjustment models will be very helpful to our providers because they do tend to serve a disproportionate number of low-income individual. But there is no way you could say at this point that all safety net hospitals do better or all or some other type of hospitals will do worse."

"It's not easy to collect the data on these sociodemographic factors and that is one reason that some have argued that you shouldn't do this—that it is too challenging," Feldpush says.

"We would push back and say that just because something is hard work doesn't mean we should back away from it. But it will take a little time to make sure we are collecting the socio-demographic information in an accurate way. Although I would also say that providers and hospitals are engaged in collecting race and ethnicity and language preference data now, and this recommendation—particularly if it was going to be put into use for quality measures—would certainly put momentum behind those efforts already underway to collect the data."