Background: Publically reported data is increasingly important to a hospital’s reputation and bottom line. Programs such as CMS Hospital Value Based Purchasing (HVBP) place 2% of a hospital’s base payments at risk, while the CMS Star Rating is frequently used by consumers to select a hospital for elective care. Both programs incorporate claims data to calculate 30-day mortality rates for conditions including acute myocardial infarction (AMI), congestive heart failure (CHF), chronic obstructive pulmonary disease (COPD), pneumonia (PNA), and acute ischemic stroke (CVA).
Purpose: Tertiary referral centers—particularly those with an accept-all policy—are often at a disadvantage with mortality metrics due to caring for the highest acuity patients. These centers need to implement innovative processes to ensure that publically reported data accurately represents the quality of care delivered. The purpose of this initiative is to leverage clinical documentation improvement (CDI) nurses and certified coders for quality by partnering with clinicians to concurrently review charts for accurate documentation and coding.
Description: A physician reviewing mortality records to screen for quality of care concerns found that charts were frequently coded inaccurately due to poor documentation, leading to the incorrect attribution of mortalities. For example, a patient with a Type-A aortic dissection and subsequent CVA was incorrectly coded as a CVA mortality, while a patient with septic shock and demand ischemia was incorrectly coded as an AMI mortality. A multi-disciplinary committee was convened to review CMS mortality cases once monthly. However, this retrospective review often led to the need to re-bill when a change in principal diagnosis was recommended. In January 2017, a new process was initiated. When an eligible mortality occurs, the case is reviewed concurrently by a physician, a CDI manager, and an inpatient coding manager. If all agree on the principal diagnosis, the bill is released and the mortality is recorded internally. If disagreement exists, further discussion often leads to a query to the discharging physician to obtain clarification on the principal diagnosis.
Conclusions: During the first 12 months of the mortality review process, overall CMS condition specific mortality decreased from a 3 year average of 5.73% to 3.9%. Similarly, AMI mortality decreased from 10.27% to 6.7%, CHF mortality decreased from 4.97% to 3.4%, COPD mortality decreased from 3.23% to 2%, PNA mortality decreased from 3.53% to 2.2%, and CVA mortality decreased from 6.63% to 4.8%. These significant reductions in mortality are estimated to yield HVBP gains of $402,000 at end of the 3 year reporting period and sustained gains of $120,000 per year going forward, assuming maintenance of current performance. Improvements in hospital star rating are also anticipated. This initiative demonstrates that personnel such as CDI nurses and coders, traditionally housed within health information technology or financial departments, can be leveraged to improve the quality of publically reported data, especially when paired with a physician who can provide clinical insight.
.jpg)