Home Health & Hospice Week

Add socio-demographic adjusters, commenters urged CMS.

If Medicare doesn’t want a whole section of the home health-eligible population to face access problems — and drive up program costs with institutional stays and ER visits — it had better consider some changes to the Home Health Value-Based Purchasing program.

So said commenters on the Centers for Medicare & Medicaid Services’ 2017 Home Health Prospective Payment System proposed rule published in July.

“The risk adjuster overall in use with Home Health Compare and the Star Ratings models fall[s] far short of accuracy needed to reasonably support a payment model,” stressed the National Association for Home Care & Hospice in its comment letter on the rule. “NAHC strongly encourages CMS to devote any necessary resources to refining the risk adjuster, particularly in the areas of socio-economic status of patients and patients with multiple comorbidities or a terminal condition.”

MedPAC, Agencies Agree For Once

Multiple commenters suggested taking socio-demographic factors into account in the risk adjustment model. “I strongly support the study of sociodemographic factors to measure risk adjustment,” said one commenter from Wisconsin. “I have observed that hospital outcomes appear to be impacted by hospital demographic patient populations and I believe that home health is as well. I would like to believe that all patients with a similar diagnosis have the same opportunity for positive outcomes, though that is not what I have witnessed.”

Even the Medicare Payment Advisory Commission acknowledged the importance of considering a patient’s socio-demographic factors. In discussing newly proposed quality measures, “the Commission recognizes that socio-economic status (SES) factors can play a role in quality and resource use measures,” the influential advisory body to Congress said in its comment letter on the rule.

However, MedPAC doesn’t want CMS to include those factors in risk adjustment. Instead, “a better way to address any differences in outcomes is to compare rates … that have not been adjusted for SES across ‘peer’ providers that have similar shares of, for example, low-income beneficiaries,” it offered in the letter. Resulting comparisons are then fair because providers would be compared with other providers with similar beneficiary populations.

Bottom line: Failing to address problems with the risk adjustment model by “keeping the same equation creates an incentive for providers to ‘manage their risk score’, instead of focusing on performance improvement,” cautioned Jim Kazmer, Director of Analytics at Wilmington, Mass.-based software company HealthWyse. “Some providers will be tempted to ‘drive up their risk score’. This is not the behavior for which we … want to provide incentives.”

Note: See the proposed rule at www.gpo.gov/fdsys/pkg/FR-2016-07-05/pdf/2016-15448.pdf.