qotd: Safety-net hospitals are inappropriately penalized by P4P programs

Annals of Internal Medicine
September 7, 2015
The Financial Effect of Value-Based Purchasing and the Hospital
Readmissions Reduction Program on Safety-Net Hospitals in 2014
By Matlin Gilman, BA; Jason M. Hockenberry, PhD; E. Kathleen Adams, PhD;
Arnold S. Milstein, MD; Ira B. Wilson, MD, MSc; and Edmund R. Becker, PhD

Background:

Medicare's value-based purchasing (VBP) and the Hospital Readmissions
Reduction Program (HRRP) could disproportionately affect safety-net
hospitals.

Measurements:

Safety-net hospitals were defined as being in the top quartile of the
Medicare disproportionate share hospital (DSH) patient percentage and
Medicare uncompensated care (UCC) payments per bed. The differences in
penalties in both total dollars and dollars per bed between safety-net
hospitals and other hospitals were estimated with the use of bivariate
and graphical regression methods.

Results:

Safety-net hospitals in the top quartile of each measure were more
likely to be penalized under VBP than other hospitals (62.9% vs. 51.0%
under the DSH definition and 60.3% vs. 51.5% under the UCC per-bed
definition). This was also the case under the HRRP (80.8% vs. 69.0% and
81.9% vs. 68.7%, respectively). Safety-net hospitals also had larger
payment penalties ($115 900 vs. $66 600 and $150 100 vs. $54 900,
respectively). On a per-bed basis, this translated to $436 versus $332
and $491 versus $314, respectively. Sensitivity analysis setting the
cutoff at the top decile rather than the top quartile decile led to
similar conclusions with somewhat larger differences between safety-net
and other hospitals. The quadratic fit of the data indicated that the
larger effect of these penalties is in the middle of the distribution of
the DSH and UCC measures.

From the Discussion

Although the payment penalties that safety-net hospitals are receiving
under VBP and the HRRP were usually small, supplemental analysis reveals
that approximately 1 in every 10 safety-net hospitals in the top
quartile of DSH definition are receiving payment rate reductions
totaling 1.0% or greater in 2014. Because safety-net hospitals are known
to have had historically low margins even before the 2008 economic
recession, losing 1.0% or more of Medicare inpatient payments could have
a significant effect on these hospitals' financial conditions for 2
reasons. First, if they were to receive persistent annual reductions in
payment rate despite quality improvement efforts, the accumulation of
these small penalties on overall financial position would further
disadvantage these hospitals. Second, the transition from DSH to the
Medicare UCC payments mandated under the ACA will mean less revenue as
the proportion of uninsured patients decreases and mandated reduction in
the total pool of funds increases. As such, it is not clear that these
reductions will be fully offset from revenue by newly insured patients
covered by the ACA's insurance expansion, given earlier evidence from
insurance expansions in Massachusetts. When safety-net hospitals close,
patients in these communities are negatively affected, and close
monitoring of the financial condition of these institutions is warranted
as the stakes in VBP and the HRRP increase.

http://annals.org/article.aspx?articleid=2434617

***

Annals of Internal Medicine
September 7, 2015
Editorial
Collateral Damage: Pay-for-Performance Initiatives and Safety-Net Hospitals
By Steffie Woolhandler, MD, MPH and David U. Himmelstein, MD

In this issue, Gilman and colleagues document a side effect of
Medicare's pay-for-performance (P4P) initiatives: They divert money from
underresourced safety-net hospitals that are mainstays of care in many
minority communities to hospitals serving more affluent patients.

Medicare's P4P program, which does not adjust for patients'
socioeconomic status, assumes that bonuses and penalties will prod
substandard providers to improve or see their patients migrate to
higher-quality options. However, when quality problems are due to a
hospital's financial distress and patients cannot go elsewhere,
penalizing low scorers may well punish patients and exacerbate quality
disparities. Prescribing a starvation diet for safety-net hospitals that
are strapped for cash and are quality challenged makes no sense unless
the goal is to close them.

Are P4P's benefits worth the risk? The evidence is surprisingly slim. A
few small, randomized, controlled trials in outpatient settings have
shown improvement on surrogate measures, but most have found no
improvement, and none have demonstrated reductions in death or
disability rates.

Safety-net providers spend less on administration and may lack the
administrative resources to keep up in the gaming arms race — the health
care equivalent of teaching to the test. Expensive computer systems and
consultants facilitate aggressive upcoding and meticulous documentation
of comorbid conditions that can exaggerate patients' severity of illness
and inflate risk-adjusted quality scores. Costly administrative efforts
can also ferret out exceptions and ensure that boxes are checked (for
example, "Yes, we counseled against smoking"), increasing
process-of-care scores, regardless of whether they improve care.

Tethering physicians' rewards to box checking and redundant
documentation risks both substituting insurers' priorities for patients'
goals and demoralizing physicians. Pay for performance can crowd out
intrinsic motivation that keeps us doing good work even when no one is
looking.

Paying for quality has strong intuitive appeal. However, as with other
medical interventions, intuition may mislead, and adopting everywhere
policies that have been proven nowhere puts millions at risk for
unintended side effects.

http://annals.org/article.aspx?articleid=2434618

***


Comment by Don McCanne

Now that the Affordable Care Act has withstood constitutional challenges
and five years of implementation, attention has turned to efforts to
improve quality while controlling costs - paying for quality rather than
volume through various pay-for-performance schemes (P4P) including
Medicare's value-based purchasing (VBP) and the Hospital Readmissions
Reduction Program (HRRP). Those closely following the reform process
realize that such efforts to date have provided only negligible, if any,
improvements in quality and cost containment.

Today's report shows that these programs also can have adverse
consequences. Because of factors such as chronic underfunding, high
pressures placed on clinical and administrative staff, and patient
populations that are less capable of interacting efficiently with the
health care system, safety-net hospitals do not and are not expected to
perform well on these measures. As a result, these underfunded
institutions suffer further financial penalties which can only compound
their difficulties in trying to meet the needs of the vulnerable
populations they serve.

The lesson here is that we are diverting our resources and talent to
efforts that will fall far short of our goals on quality and
affordability, primarily because we passively accept the fact that our
dysfunctional system under the Affordable Care Act is here to stay.
Instead we should be working on introducing a system that actually will
meet our goals, not only for quality and affordability, but also for
universality, efficiency, comprehensiveness, accessibility, and,
especially, equity - a single payer national health program.