The Challenge of Adjusting Health Measures for Social Risk Factors

The landmark report, To Err Is Human, came out around the time that the National Quality Forum (NQF) was founded, and in that time, there have been tremendous improvements in hospital-based outcomes, fewer infections, and better deliveries. However, one area where more needs to be done is incorporating social risk factors into healthcare measurements, explained Shantanu Agrawal, MD, CEO and president of NQF, during the day 1 plenary at the fall meeting of the ACO & Emerging Healthcare Delivery Coalition® in Nashville, Tennessee.

Today, one of the big foci in healthcare is in readmissions, he said, which gets to the health equity issue, and more than other quality measures, readmissions “clearly interact with the community setting,” Agrawal said. A hospital’s ability to reduce the readmission rate is reliant on transitions of care and resources in the community.

He explained that when he was going through training, he participated in ride alongs with paramedics in West Philadelphia, and on one occasion they visited a patient’s home during the summer. Agrawal was familiar with the patient because she had been a frequent flyer at the emergency department (ED). The patient, who had significant heart failure, was living in a house had no air conditioning during 90-degree weather with just a small table fan. Seeing that situation, Agrawal knew it was impossible for the patient to get up and take her medications or cook.

The temperature was so high in the house, that Agrawal felt the patient might have been better off in the ED, where the temperature was much cooler and the environment was much safer for her.

“If we miss the home environment—and we were missing the home environment—we miss basically the entire picture of her care,” Agrawal said.

For more than a decade, NQF has been involved in health equity through measure selection and endorsement, measurement frameworks, best practices, and implementation guidance.

In 2016, NQF published the report Improving Population Health by Working with Communities, though which NQF worked with 10 communities around the country to assess what elements a population health program needs to actually affect the health of the population.

“I think the moment you start getting into a population health frame of mind, you instantly start thinking of social disparities; the everyday social risk factors that people, citizens, patients have that contributor to the health picture that you then see in the delivery system,” Agrawal said.

From 2015 to 2017, NQF explored social risk adjustment in measurement through a pilot trial. The organization considered risk adjustment for 303 measures and determined if there was some conceptual basis for risk adjusting the measure for various social measures. At that point, NQF was able to whittle the list down to 65 measures. Then, NQF asked the measure developer and the clinicians promoting the measure to run calculations including social risk factors to determine if the risk factor impacted the model. By the end of the pilot, NQF endorsed 17 measures with social risk adjustment.

“Which meant there was a huge gap between where stakeholders, measure developers, and others felt that there could be adjustment to where there actually could be adjustment,” Agrawal said.

So why so few measures? NQF learned that data sources in this area are highly limited. The most commonly used data source across the 65 risk factors was dual eligibility for Medicare and Medicaid. Dual eligibility was considered a proxy for other risk factors, although it wasn’t perfect.

Other data sources that were available were country of origin and an urban versus a rural residence. Other social risk factors commonly thought of didn’t have a comprehensive data source, which limited the ability of clinicians and measurement developers to demonstrate what they thought to be true regarding if social risk factors impacted the measure.

NQF has also recently released a disparities roadmap, which provides a vision for the new Health Equity Program made up of 4 major pieces:

1. Identifying populations at risk
2. Ensuring there is a performance measurement approach that incentivizes addressing disparities
3. Be clear about the best practices and implementing them across the country
4. Consider various payment models that can incentivize providers and others to address equity issues

The ideal state for healthcare would be one that has high-quality, timely data on social risk factors and disparities, with all stakeholders working together systematically to reduce disparities through the use of clear interventions demonstrate progress toward health equity, and new payment models that approach the issue directly.

“I think we have to take a very practical approach to dealing with this problem, because our patients actually need that assistance today,” Agrawal said.