Author + information
- Vivek T. Kulkarni,
- Nancy Kim,
- Ying Dai,
- Kumar Dharmarajan,
- Behnood Bikdeli,
- Kyan Safavi and
- Harlan Krumholz
Noninvasive positive pressure ventilation (NIPPV) has been used in acute heart failure (AHF) for 20 years with no consensus as to its role. Little is known about hospital practice patterns of NIPPV use in AHF, including the timing and setting of therapy, and the relationship between hospital NIPPV use and in-hospital mortality.
The Perspective® database from Premier, Inc., includes a date-stamped log of all billed items for admissions at over 400 hospitals. We included all hospitals with 25 or more AHF admissions in 2005-2010. We calculated the percent of AHF admissions at each hospital receiving primary NIPPV (without or before intubation). The main outcome was risk-standardized NIPPV rate, estimated with hierarchical generalized linear models (HGLM) using patient age, sex, Elixhauser comorbidities, and a random hospital effect. We also examined the percent of primary NIPPV at each hospital that was given as early therapy (days 1-2), and the percent given in non-intensive care settings. We used HGLM to estimate in-hospital risk-standardized mortality rate (RSMR) for each hospital's AHF patients. Kruskal-Wallis tests were used to test associations.
395 hospitals (median beds: 266) had 696,739 AHF admissions (median annual volume: 274). Median risk-standardized NIPPV rate was 9.3%; the 10th, 25th, 75th, and 90th percentiles were 1.1%, 4.7%, 12.9%, and 16.6%, respectively. A median of 79% of primary NIPPV was given as early therapy; a median of 63% was given in non-intensive care settings. Hospitals with more beds and higher annual volume had higher risk-standardized NIPPV rates, less early therapy, and more non-intensive care use (p<0.05 for all associations). Median RSMR was 5.3%; RSMRs did not relate to NIPPV use, even between the highest and lowest 10% of NIPPV users (p=0.38).
The median hospital uses primary NIPPV in 9.3% of AHF patients, but rates vary substantially. Larger hospitals use primary NIPPV more overall but tend to use it less as early therapy and to use it more in non-intensive care settings, suggesting significant institutional effects on practice patterns. Even at extremes of use, primary NIPPV was unrelated to in-hospital mortality.
Poster Sessions, Expo North
Sunday, March 10, 2013, 9:45 a.m.-10:30 a.m.
Session Title: Improving Heart Failure Outcomes II
Abstract Category: 28. Quality of Care and Outcomes Assessment
Presentation Number: 1200-99
- 2013 American College of Cardiology Foundation