July 11th, 2014
John Mathewson, executive vice president of Health Care Services for Children with Special Needs (HSC) – a Medicaid managed care plan in D.C. for children on Supplemental Security Income (SSI) – recently spoke at the Association for Community Affiliated Plans (ACAP) CEO Summit before the July 4 Recess.
Mathewson described what he has dubbed The Kitten Paradox: When HSC examined environmental factors for children with asthma, it found that the presence of pets in the house was a common thread, not too far behind having a smoker around. Yet, it turns out the value a cat brings by protecting from mice or spawning a litter for sale outweighs any financial costs to the family associated with an ER visit, which are often free or carry a low copayment. Thus the paradox.
An awardee at the conference, Hennepin Health, catalogued the evidence showing that reliable housing can improve health outcomes, including improving mental health and lowering emergency room and inpatient hospital utilization.
The focus of these sessions was the social determinants of health, and a lot of these safety net health plan leaders’ heads were nodding throughout. The plans, which disproportionately serve Medicaid enrollees and thus ‘dual eligible’ seniors in Medicare, know something about the importance of social determinants that the health policy community – at least in Washington – is only now slowly waking up to.
Social factors, not the least of which is socioeconomic status, are key drivers of health outcomes. Many of these health plans are taking action to address these factors because they are confident doing so will be cost-effective in the long run. But for now, prevailing Federal health policies are part of the problem, not the solution.
If you want to know the next big health policy question facing Washington, look no further.
MA Star Ratings
The quality of care provided under the private insurance option in Medicare – referred to as Medicare Advantage (MA) – is measured under the Star Ratings program. Plans are evaluated and scored on a variety of metrics that are combined to produce composite scores for their medical benefit (the “Part C” side) and their drug benefit, if they offer one (the “Part D” side).
The subset of MA plans that focus exclusively on dual-eligible Medicare beneficiaries, called Dual Eligible Special Needs Plans (D-SNPs), have historically underperformed traditional MA plans by a substantial margin. A comprehensive Ingenix Consulting analysis of 2010 Star Ratings found that the most common score for D-SNPs was a full point below the mode score for standard plans.
The solution? Plans have offered a few. Measures of improvement could be better emphasized over those that they are less able to control, such as medication adherence. Like plans (i.e., D-SNPs) could be compared to each other, rather than to those targeting broader, generally healthier segments of the senior population. Ultimately, measures could be risk adjusted to account for socioeconomic factors that impact health outcomes.
Otherwise, they contend, plans that focus on higher-risk, lower-income Medicare beneficiaries will continue to be penalized, which – in a paradox of its own – reduces their ability to invest in the coordinated, patient-centered care these patients need. They will also slowly exit the space, whether by choice or for the consequences of poor performance on the Star Rating program.
This issue is about to get more acute. Come fall, the Centers for Medicare and Medicaid Services (CMS) plans to exercise its authority to terminate MA plans that have not yet achieved three stars on both their C and D composite scores. Over 150,000 Medicare beneficiaries are currently enrolled in D-SNPs that are in jeopardy of termination.
So far, the Administration isn’t budging. Many in the higher echelons, as well as some of their key supporters on Capitol Hill, are afraid Medicare might adopt anachronistic proxies that mask the true factors driving health disparities. But recently, other officials have suggested that a related fear must also be reckoned with: that policies on the books today are having the unintended consequence of exacerbating these disparities.
While plans may have a bird’s eye view of the impact of socioeconomic factors on health, providers may see the impact closer up. The Medicare Hospital Readmission Reduction Program (HRRP) was launched to incent hospitals and their partners to focus on keeping patients out of that setting after their initial release. It encourages the adoption of clearer discharge planning, patient follow-up and coordination among sites of care to ensure safe, healthy recoveries.
But last year, Kaiser Health News reported that 77 percent of hospitals treating a disproportionately high number of low-income beneficiaries were subject to penalties under the program, while only 36 percent of those with a low number of such patients were. Subsequent studies published in peer-reviewed literature, including an article in May’s edition of Health Affairs by Hu et al., have drawn similar conclusions.
Congress is taking note. In March, Congressman Jim Renacci introduced the bipartisan Establishing Beneficiary Equity in the Hospital Readmission Program Act. In brief, that bill would require CMS to adjust penalties under the HRRP based on the proportion of dual eligibles a provider serves.
Last month, Senator Joe Manchin followed suit. While emphasizing different factors, his bill proposes to solve the same problem as Mr. Renacci’s: that the HRRP is skewed to punish hospitals that care for those most in need.
Though not as aggressive as the Renacci or Manchin bills, the Finance Committee, led by Senator Ron Wyden, included direction to study the impact of socioeconomic factors on Medicare quality metrics in the permanent Sustainable Growth Rate (SGR) repeal bill they passed out of committee last December.
In its June 2013 Report to Congress, MedPAC delved into the HRRP in great detail, noting as one of the program’s four key issues the correlation between socioeconomic status and readmission rates.
Performing its own analysis using eligibility for SSI as the marker for low income, MedPAC confirmed the findings of others: hospitals treating the highest number of SSI-eligible patients were four times as likely to face readmission penalties as hospitals treating the fewest SSI-eligible patients. And their penalties were steep, averaging more than double the typical penalty amount.
MedPAC ran retrospective tests on Medicare data using other factors, including race. It found that SSI was the most effective predictor of readmissions among these.
In the report, while not a fully vetted Commissioner recommendation, MedPAC suggests a “potential solution”: use cohorts to evaluate hospital performance relative to similarly-situated peers.
All hospitals would continue to report publicly on readmissions, without adjustment for socioeconomic factors, so no contributing causes would be “masked.” But penalties would be calculated based on the hospitals’ performance relative to others serving patients with similar income profiles.
This approach would not only likely make for a more equitable HRRP, it would diminish the existing incentive providers may otherwise have to avoid treating low-income patients.
MedPAC tested its concept to determine what readmissions penalties would be if it were adopted. Predictably, the variation in these penalties among hospitals treating different patient cohorts declined considerably.
A recent Health Affairs article by Nagasako et al. evaluated potential performance of Missouri hospitals under a model that controlled for factors such as poverty, educational attainment, and housing vacancy. The authors found that this model substantially reduced disparities in calculated readmission rates.
While the Commission noted that this issue warrants further inquiry, they appear to be on the path to a solution that can address the parallel fears held by CMS and others that have stalled progress on this issue.
NQF Up Next
On March 18, the National Quality Forum (NQF) issued a draft report titled Risk Adjustment for Socioeconomic Status or Other Sociodemographic Factors. In the report, which was commissioned by CMS, NQF’s Expert Panel articulated broad consensus around eight recommendations to support “making correct conclusions about quality of care and prevent unintended consequences such as the worsening of healthcare disparities.”
The report states that “carefully done” adjustments for sociodemographic factors would more accurately reflect performance without obscuring the quality of patient outcomes. It notes that “peer group comparisons, such as the approach recommended by MedPAC,” could be used as part of a solution to this complicated challenge.
A small minority of the panel, the report notes, preferred “not to adjust for sociodemographic factors based on concerns about masking disparities or accepting different standards.”
A raft of over 600 public comments from 150+ organizations and individuals ensued. According to a memo prepared for the Expert Panel, the vast majority supported its recommendations. The memo notes that “most” commenters representing consumers and purchasers did not support the recommendations, however. CMS opposed them as well.
On April 29, in response to a flurry of media coverage of their groundbreaking – and, to some, contentious – report, NQF issued a press release that distanced the broader organization from the Expert Panel’s draft while emphasizing limitations around its contents.
For example, the organization noted that the suggestions in the report apply to the NQF measure endorsement process, “not Federal policy” – a nuance for sure, because NQF measures are typically deployed in Medicare and other Federal programs. (CMS, after all, commissioned the report.) The NQF press release said it will pursue a “rigorous, multi-phased standard review process” and that the recommendations “may change as a result.”
The final NQF report is expected before the end of this month.
Where We’re Heading
So, whether we like it or not, the interplay of social factors and health outcomes is an issue that’s coming for us. Congress, the Administration, and the health policy community are going to have to look underneath the hood of existing quality measures and make an honest assessment of the factors that drive plan and provider performance on them.
While it’s going to be a difficult journey, we can share a confidence that the vast majority of actors here share the same goal: measuring performance accurately while tackling head-on the persistent challenge of health disparities.
For many of the plans in the audience at ACAP’s Summit and their enrollees, the clock is ticking.
Editor’s note: Billy Wynne represents several plans and providers that serve a high number of low-income Medicare beneficiaries.Email This Post Print This Post
Don't miss the insightful policy recommendations and thought-provoking research findings published in Health Affairs.