Inside Angle
From 3M Health Information Systems
Why CMS needs to get the hospital-acquired complication policy right
In our March blog, we highlighted some of the challenges with the CMS Hospital Readmission Reduction Program (HRRP). In that blog, we alluded to similar design flaws that are present in the Hospital-Acquired Condition Reduction Program (HACRP). The HACRP is, like the HRRP, a penalty-only performance measure. A one percent payment penalty is assessed on the worst performing hospitals (set at 25 percent of them) and, in keeping with other CMS pay-for-performance policies, the penalty is based upon an administratively-determined amount rather than any relationship with the costs of the relatively poor performance. HACRP results are determined by summing rankings on narrowly-defined measures (Central Line-Associated Blood Stream Infection, CLABSI and Catheter-Associated Urinary Tract Infections, CAUTI) into a composite score, after which minimal performance variation is leveraged into the single, relatively large payment reduction of one percent. A more detailed analysis of the HACRP can be found in our recent article, “Is the CMS hospital acquired condition reduction program a valid measure of hospital performance?”
HACRP is not the first CMS foray into policy for improving performance on hospital-acquired conditions (HAC). In October 2008, the HAC Present on Admission (HAC POA) payment adjustment was introduced by CMS. The HAC POA policy (which is still in effect) disallows a limited range of post-admission HACs from inclusion in MS-DRG assignment. Despite significant initial comment, the policy appears to have almost disappeared from public conversation instead of the dialogue continuing about it being reformed. This lack of attention is likely the result of the small amounts of money (less than 0.02 percent of payments)1 that are reduced from payment. It appears that the lesson learned after this initial effort to impact hospital complications was that a small base from which to apply a penalty requires a disproportionately large penalty in the HACRP in order to drive change and maintain focus. We believe that this is fundamentally the wrong response.
Unlike the CMS approach , we believe that HACs can identify all preventable complications (with adequate risk adjustment) and that they represent sufficiently significant costs to power a policy without resorting to leveraging a “multiplier” for a penalty relative to the measured cost of the performance2. Our approach to measuring HACs is contained within our Potentially Preventable Complication (PPC) logic. Moreover, we have seen excellent results from states, such as Maryland, that have deployed the PPC logic, incorporating an incentive structure to decrease PPCs within a continuous quality improvement framework. The Maryland HAC reduction initiative has resulted in an approximately 50 percent reduction in PPCs over the past five years 3,4. Availability of the clinical logic, which drills down right to the patient-level information, together with financial incentives is the key to this program’s success. Through an open communications process with hundreds of clinicians meeting several times a year to engage in detailed clinical conversations, best practices are learned and suggestions for refining the PPC logic are shared. This level of clinical detail provides the basis for clinicians impacted by the measurement to help refine the logic.
Considerable focus has been placed on reducing hospital-acquired complications over the last decade, but there is considerably more work to be done. The CDC, despite reporting reductions in the frequency of hospital-acquired infections (HAIs), estimates that 1 in 25 patients on any given day has an HAI and that the majority of these infections occur outside of an ICU. The HACRP policy is driven by a composite score of which 50 percent is derived from CLABSI and CAUTI ICU rates. The rate-based approach embedded within PPCs allows for transparent comparison adjusted for patient mix. We believe that the CMS should move away from its composite score of narrowly-focused HACs toward a more expansive assessment of hospital performance in managing HACs. A systematic refocusing of effort to reduce all hospital complications is what’s needed to see a drop in rates—this meets the ultimate goal of better outcomes for patients and reduced costs.
Richard Fuller, MS, is an economist with 3M Clinical and Economic Research.
Norbert Goldfield, MD, is medical director for 3M Clinical and Economic Research.
1Fuller RL, McCullough EC, Averill RF. A new approach to reducing payments made to hospitals with high complication rates. Inquiry. 2011;48(1):68-83. http://www.ncbi.nlm.nih.gov/pubmed/26348621. Accessed July 12, 2011.
2Fuller RL, McCullough EC, Bao MZ, Averill RF. Estimating the costs of potentially preventable hospital acquired complications. Health Care Financ Rev. 2009;30(4):17-32. http://www.ncbi.nlm.nih.gov/pubmed/19719030. Accessed November 8, 2011.
3Patel A, Rajkumar R, Colmers JM, Kinzer D, Conway PH, Sharfstein JM. Maryland’s Global Hospital Budgets–Preliminary Results from an All-Payer Model. N Engl J Med. 2015;373(20):1899-1901. doi:10.1056/NEJMp1508037.
4Calikoglu S, Murray R, Feeney D. Hospital pay-for-performance programs in Maryland produced strong results, including reduced hospital-acquired conditions. Health Aff (Millwood). 2012;31(12):2649-2658. doi:10.1377/hlthaff.2012.0357.
Thanks for the article series.
I do not understand why HAC “penalties” should work. The additional (self-inflicted) costs of a HAC far outweigh the potential loss for the cc/MCC reimbursement bump. Somehow, losing (say) $10k in reimbursement is expected to be more “motivating” than an increase in cost of 2x or 3x or 10x that amount. Strange.
Thank you for your comment.
To restate, the question raised is why a penalty, which may be computed as less than or equal to the added cost of an HAC, should motivate a provider to improve more than the loss that may be sustained from the direct HAC cost relative to some fixed payment (i.e. DRG payment). This is a good point but makes two assumptions. First, that the total costs of HACs in terms of excess frequency and individual cost are known. Generally these are not except in those situations where a regulator or payer identifies the cost and frequency of the excess HACs (such as in the example of Maryland ). Second, that those reading reports circulated to them will both believe and act upon them. This is why the penalty is effective. It serves as a guarantee that, believe it or not, unless you reform you will incur this amount of financial loss (or hopefully reward with symmetrical incentives).
Thanks for your input. I didn’t realize that costs are not generally known. I did this study here (major academic medical center) many years ago, using both our own costing analysis (TSI) and UHC data. I had hoped to make a big impact on HAC awareness though cost knowledge to the provider community here. Nothing. Not only were the direct costs higher, but the increase in LOS also prevents us from seeing “the next patient,” thereby reducing total patient throughput (and hence, even more lost revenue).
Sadly, my initiative simply had no traction here. But when the penalty was announced, it was like we were under attack and all the bells and whistles went off. All hands on deck.
Still doesn’t make sense to me. Oh well.
The experience you raise is not uncommon. Your study is typical of those that show the opportunity for gains yet get sidelined as the returns are “not proven” or may accrue over a longer time horizon. In short, maintaining the status quo is preferred to action. There is an old adage in behavioral psychology that “losses loom larger than gains” hence the threat of a penalty suddenly frees up money and momentum to realize the gains you had already identified. Couple that with the newfound ability of quality folks to insert definitive ROI amounts on budgets and presto you have an initiative. Saying that we understand why this happens is not the same as saying that it makes sense! Thanks again for sharing your expertise.