Nearly 60% of Deaths for CMS Mortality Measures Happen Outside the Index Stay


Get Your Hospital’s AI Insights and Reports on CMS Star Rating, Readmissions, Mortality and VBP


This article is part of the Dexur Quality Sights Newsletter, which is focused on Hospital Quality and Safety topics and is read by CEOs, CMOs, CNOs, CQOs, Quality, Safety, Compliance, Analytics & Performance teams. Please email dexurqualitysights@dexur5.com to join this newsletter and receive updates (not more than once per week).

This article is part of a series that highlights challenges that hospitals face by not having access to out-of-hospital data and the patient journey. The rest of the articles in this series will be shared at the bottom of this article when they are published.

Nearly 30% of Readmissions are to a Different Hospital From the Index Stay

20-40% of Deaths for CMS Pneumonia, HF, AMI and COPD 30 Day Mortality Measures have an Hospice Interaction


Hospitals have a challenge in tracking CMS 30-Day Mortality measures because they try to monitor more real-time information using internal hospital data, which does not have out-of-hospital information used for 30-day mortality and readmission measures. However, Dexur is an approved purchaser of Medicare claims data and, therefore, can analyze the full patient journey that impacts measures such as 30-day mortality, 30-day readmissions, MSPB, and EDAC.

To show the gap in the potential internal hospital data, Dexur analyzed in-hospital mortality at the index stay compared to the full 30-day mortality. The data shows that nearly 60% of deaths for CMS Mortality measures happen outside the index stay. The below table shows the summary stats by the measure at the national level, comparing observed rates in In-Hospital Mortality Rate at index stay vs. Total 30-day mortalities. Among all CMS conditions, heart failure has the highest percentage of deaths occurring outside the index stay with 64.64%, while CABG (Coronary Artery Bypass Grafting) has the lowest percentage at 7.52% for FY 2024 (01, Jul 2019 to 30, Jun 2022).

CMS Mortality Measure In-Hospital Mortality Rate at Index Stay- Observed (%) 30-Day Mortality Rate- Observed % of Total Mortalities outside of Index Stay
All CMS Mortality Measures 12.29% 5.18% 57.83%
COPD 8.59% 3.69% 57.06%
Pneumonia 16.79% 6.97% 58.49%
CABG 2.30% 2.13% 7.52%
AMI 9.97% 6.04% 39.44%
Heart Failure 11.37% 4.02% 64.64%

How Dexur platform helps in tracking in-hospital mortality rate and out-of-hospital mortalities

Since Dexur is an approved purchaser of Medicare claims data, Dexur’s detailed Readmission and Mortality reports show the breakdown by In-Hospital Mortality vs. 30 Day readmission rate.

Hospitals have little room for error because the difference between the 10th & 90th percentile performance is small. Therefore, Dexur recommends the following actions to account for these out-of-hospital mortality measures:

  1. Set Multiple Goals for Mortality Measures: For example, a more real-time goal for observed in-hospital mortality and a more delayed (4- 5 months delayed) goal based on the complete patient journey from a partner like Dexur and a final CMS-calibrated risk-adjusted published goal. Dexur’s goal-setting features allow for multiple goals based on different data sets for the same measure.

  2. Implement Technology and Reporting platforms for Multi-level data + Goal tracking for a single measure. Most technology platforms have a 1-1 mapping between data source and measure. However, in complex CMS measures, data feeds (e.g., Internal Hospital Data & External 30 Day Mortality) and the associated goals would be multi-level for the same measure. Dexur’s Measures and Data feature allows for these complex multi-source data, goals, and interventions for a single measure.

  3. Interventions should account for In-Hospital Mortality vs. Out-of-Hospital Mortality since the practices differ. For example, checking for real-time conditions in the hospital is different compared to a daily post-discharge check-in. Dexur’s interventions feature allows for seeing the impact of interventions between In-Hospital vs. Out-of-Hospital rates.