Author + information
- Rashmee U. Shah, MD, MS∗ (, )
- Josef Stehlik, MD,
- Stavros G. Drakos, MD and
- James C. Fang, MD
- ↵∗University of Utah School of Medicine, Cardiovascular Medicine, 30 North 1900 East, Room 4A100, Salt Lake City, Utah 84105
In a study recently published in the Journal, Dr. Stretch and colleagues (1) described increasing utilization of short-term mechanical circulatory support (MCS) in the United States. The investigators found a dramatic increase in the use of short-term MCS over the past decade. The NIS (Nationwide Inpatient Sample), used in this study, is an excellent resource to study temporal trends on a large scale and is easily available to researchers. The NIS, however, has limitations that affect the interpretation of this study. Hospitals are included to approximate a 20% sample of all hospitals in the United States (2). The weighting strategy used to create national estimates relies on the assumption that similar hospitals have similar experiences (3). This assumption is probably true for common diagnostic codes, such as pneumonia and urinary tract infection, but may not be true for rarer events, such as short-term MCS. In other words, all rural, large, government, nonteaching hospitals in the Northeast (1 of 60 hospital strata in the NIS) will not necessarily adopt MCS at similar rates. The national estimates, therefore, are prone to inaccuracy.
The cost methodology used in the NIS is on the basis of schemas based on primary International Classification of Diseases, Ninth Revision, or diagnostically related groups schemas developed by the Agency for Healthcare Research and Quality (4). The methodology does not directly account for length of stay and therefore does not account for outlier payments. MCS patients, however, may be more likely to have long, complicated hospitalizations and qualify for outlier payments. The investigators report “other” as the indication for hospitalization in 20% to 25% of encounters; the strategy used to calculate costs for this significant group of patients is unclear.
Anecdotal experience suggests that the findings of this investigation are true: MCS use is probably increasing. Still, investigators must ensure rigorous methods to avoid inaccurate trend assessment. Research using large datasets, such as the NIS, is important in understanding health care delivery trends on a broad scale. This type of research will become more prevalent and important as administrative data become more readily available. Therefore, an increased focus on methods is warranted.
Please note: Dr. Shah owns stock in Gilead Sciences. Dr. Drakos is a consultant for Abiomed and Heartware. Dr. Fang is a consultant for Maquet and Abiomed; and has received research support from Abiomed. Dr. Stehlik has reported that he has no relationships relevant to the contents of this paper to disclose.
- American College of Cardiology Foundation
- Stretch R.,
- Sauer C.M.,
- Yuh D.D.,
- Bonde P.
- ↵Agency for Healthcare Research and Quality. Changes in the NIS sampling and weighting strategy for 1998. Available at: http://www.hcup-us.ahrq.gov/db/nation/nis/reports/Changes_in_NIS_Design_1998.pdf. Accessed January 20, 2013.
- ↵Agency for Healthcare Research and Quality. HCUP Methods Series. Calculating Nationwide Inpatient Sample Variances. Report #2003-2. Available at: http://www.hcup-us.ahrq.gov/reports/methods/CalculatingNISVariances200106092005.pdf. Accessed June 6, 2011.
- ↵Sun Y, Friedman B. Tools for More Accurate Inpatient Cost Estimates With HCUP Databases 2009. Errata added October 25, 2012. HCUP Methods Series Report # 2011-04. U.S. Agency for Healthcare Research and Quality. Available at: http://www.hcup-us.ahrq.gov/reports/methods/methods.jsp. Accessed October 2, 2014.