Value-Based Care News

NQF Eyes Adding Social Risk Factors to Value-Based Purchasing

NQF endorsed 17 measures that value-based purchasing models can use to address the effect of social risk factors on provider performance and payments, but finding more measures could be challenging.

NQF endorsed 17 quality measures for value-based purchasing models to use to address the effect of social risk factors on provider performance and payments

Source: Thinkstock

By Jacqueline LaPointe

- In response to stakeholder concerns that value-based purchasing programs unfairly penalize providers who treat greater proportions of disadvantaged patients, the National Quality Forum (NQF) recently endorsed 17 quality measures that address social risk factors.

The newly approved measures demonstrate a significant association with social risk factors, such as race, ethnicity, Medicaid status, insurance type, education level, percent of households under the federal poverty line, and marital status. Some key quality measures that account for these social risk factors include: 

• Optimal vascular care with an insurance product variable

• Chronic Obstruction Pulmonary Disease (COPD) or asthma in older adults admission rate with a variable of percentage of households under the federal poverty level

• Standardized mortality ratio for dialysis facilities with a race and ethnicity variable

READ MORE: Exploring Quality Measures Under Value-Based Purchasing Models

• Family experiences with care coordination measures with respondent education as a variable

• Discharge to community with a marital status variable

• 30-day unplanned readmissions for cancer patients (Phase 3) with a dual eligible variable

“All stakeholders want to see the quality of care for the most vulnerable improve while ensuring a level playing field for providers in value-based purchasing programs,” the organization wrote. “The increased use of NQF-endorsed measures for payment purposes underscores the importance of ensuring accurate comparisons so that rewards or penalties are fairly distributed and based on true differences in performance.”

Some stakeholders have recently contended that value-based purchasing models that do not account for social risk factors in provider performance disproportionately penalize those who treat disadvantaged populations.

READ MORE: Best Practices for Value-Based Purchasing Implementation

A recent JAMA Cardiology study revealed that hospitals that serve higher volumes of African-American patients and those with more severe conditions were more likely to receive a value-based penalty under the Medicare Hospital Readmissions Reduction Program.

In addition, the HHS Office of the Assistant Secretary for Planning and Evaluation and National Academies of Science, Engineering, and Medicine also called on industry leaders and policymakers to incorporate social risk factors into quality measures in value-based purchasing programs to ensure incentive payment and penalty determinations are fair.

However, other healthcare leaders remain skeptical about adding social risk factors to quality measurement, arguing that it could mask care disparities and lead to different performance standards based on patient demographics.

After a two-year trial period, NQF settled on 17 out of 303 quality measures. However, the organization stated that the journey to including social risk factors in quality measurement for value-based purchasing programs is “feasible, but challenging.”

NQF explained that it faced two major challenges with finding quality measures that clearly linked social risk factors with provider performance on certain measures.

READ MORE: GAO: Quality Measure Misalignment Impedes Provider Improvement

First, quality measures that demonstrated a valid conceptual model did not necessarily have the empirical risk adjustment analyses to show a substantial effect of social risk factors.

“One of the most striking findings of the trial was that measures with a conceptual basis for adjustment generally did not demonstrate an empirical relationship of the social risk factors to the outcome measured,” the organization explained. “In some instances, the effects were significant with small effect size, did not improve model performance, or did not meaningfully change hospital results.”

The organization also experienced issues during the public comment and member voting processes when some measures that had a valid conceptual basis for risk adjustment did not include the specific social risk factors in their final risk adjustment models. Quality measures for hospital readmissions, cost, and resource use particularly fell into this category.

“Providers expressed concerns that some social risk factors were statistically significant but were not included in the final model,” the report stated. “Developers cited various reasons for not including these factors, including not improving the performance of the risk model and concerns about masking quality problems.”

Second, NFQ reported obstacles related to data availability. The organization and measure developers faced limited access to patient-level data on social risk.

While data on social risk factors such as dual eligibility, nativity, and urbanicity or rurality was readily available, the majority of social risk factors lacked available data and research. Stakeholders especially lacked information on how sexual orientation, housing, social support, living alone, and wealth impacted quality measurement and provider performance in value-based purchasing programs.

NQF also noted that some stakeholders voiced concerns with including race as a social risk factor in value-based purchasing programs. The Risk Adjustment Expert panel stated that race and ethnicity should not be used because factors like education, income, language, and insurance status exacerbated care disparities based on race and ethnicity.

“The Panel cautioned that because of concerns of bias or racism, careful thought and consideration and a clear rationale are necessary when adjusting measures for race and ethnicity,” the report stated.

Other committee members argued that some measures also may have included race as a social risk factor as a proxy for socioeconomic status.

Based on the trial period, NQF expects a long journey ahead with incorporating social risk factors into value-based purchasing programs to ensure equitable penalties and payments.

“NQF is carefully reviewing the results of this important trial with input from key experts and stakeholders as it considers options for measure evaluation and endorsement going forward,” the report concluded. “The trial period has elucidated important issues, such as availability of data on social risk that should be pursued as we collectively drive toward elimination of healthcare disparities.”