By HERB KUHN
Historically, the Centers for Medicare & Medicaid Services’ (CMS) stance on the influence that social determinants of health (SDOH) have on health outcomes has been equal parts signal and noise. In April 2016, the agency announced it would begin adjusting the Medicare Advantage star ratings for dual-eligibility and other social factors. This was amid calls for increased equity in the performance determinations from the managed care industry. At the same time, CMS continued to refuse risk-adjustment for SDOH in the Hospital Readmissions Reduction Program (HRRP) despite the research supporting the influence of these factors on the HRRP.
It wasn’t until Congress interceded with the 21st Century Cures Act that CMS conceded to adjusting for dual-eligibility under the new stratified approach to determining HRRP penalties beginning in fiscal year 2019. The new methodology compares hospital readmission performance to peers within the same quintile of dual-eligible payer mix. The debate surrounding the adjustment of incentive-based performance metrics for SDOH likely is to continue, as many feel stratification is a step in the right direction, albeit a small one. And importantly, the Cures Act includes the option of direct risk-adjustment for SDOH, as deemed necessary by the Secretary of Health and Humans Services.
SDOH are defined as “the conditions in which people are born, grow, live, work and age.” The multidimensional nature of SDOH reach far beyond poverty, requiring a systemic approach to effectively moderate their effects on health outcomes. The criteria used to identify SDOH include factors that have a defined association with health, exist before the delivery of care, are not determined by the quality of care received and are not readily modifiable by health care providers.
The question of modifiability is central to the debate. In the absence of reimbursement for treating SDOH, providers lack the resources to modify health outcomes attributable to social complexities. Therefore, statistical adjustments are needed to account for differences in these complexities to ensure risk-adjusted performance comparisons of hospitals are accurate.
The hospital community is deeply encouraged by the noise reduction that CMS has recently provided by signaling steps toward direct reimbursement for the treatment of SDOH. In February 2018, they announced a major policy shift, enabling added flexibility for MA plans through supplemental benefits that allow reimbursement for nontraditional goods and services, such as transportation, groceries and air conditioning. And, while early evidence shows that the uptake of the supplemental benefits by the MA plans has been limited, an expanded “whole person” model is being developed through the Center for Medicare & Medicaid Innovation. The model would allow for housing, utilities and nutrition assistance, among others, that eventually could be scaled to cover socially complex fee-for-service beneficiaries. As HHS Secretary Alex Azar stated in a recent speech, “What if we provided solutions for the whole person, including addressing housing, nutrition and other social needs?”
Expanded services for patients with social complexity will require more nuanced data than are currently available. Standardized administrative data sources typically are limited to information on race, ethnicity, disability and dual-eligibility. However, the actual dimensions of social complexity — and their known association with health outcomes — are far more expansive.
There is good news on the data side of the equation. The conversion to the 10th revision of the International Classification of Diseases (ICD-10) in October 2015 created an opportunity for physician and non-physician providers to identify, diagnose and document patients with social complexity in a uniform diagnostic and billing data system.
Recent analysis of Missouri Hospital’s codes by the Missouri Hospital Association found the distribution and predictive characteristics of the 87 ICD-10 SDOH codes suggests the potential for a large advancement in the identification and documentation of social complexity for clinical applications. This includes including filling informational gaps at the patient level for risk adjustment, clinical support and population health. However, significant work will be required to expand awareness and uniform application of the codes.
The signals are becoming clearer. With the full-throated engagement of the HHS Secretary, work by several HHS agencies and predictive properties that ICD-10 provides, new opportunities exist for CMS to further refine its signal. This will allow more reflective programs and payment systems in support of vulnerable patients — the patients hospitals see every day.