Adverse Event Rates among Medicare Patients with Common Medical Conditions

A study of the adverse event rates among Medicare patients with common medical conditions discovered a decline in patients hospitalized for acute myocardial infarction (MI) or congestive heart failure (CHF), while no change was seen for patients hospitalized for pneumonia or conditions requiring surgery.

Medical records of a random sample of hospitalized patients ≥65 years of age for acute MI, CHF, pneumonia, or conditions requiring surgery were obtained from the Medicare Patient Safety Monitoring System between 2005 and 2011 across all states, including Washington, DC, and Puerto Rico. The 21 identified events for which a patient was at risk during hospitalization were divided into 4 clinical categories: (1) adverse drug event; (2) general events; (3) hospital-acquired infection; and (4) post-procedural event. The composite outcomes for any of the 21 events were determined to be the rate of occurrence of adverse events for which patients were at risk, the proportion of patients with ≥1 adverse events, or the number of adverse events per 1000 hospitalizations.

The final study group included 11,339 patients with acute MI, 15,374 with CHF, 18,269 with pneumonia, and 16,481 with conditions requiring surgery, totaling 61,523 participants within 4372 hospitals. 

Observed adverse event rates among patients with acute MI or CHF decreased dramatically for all 3 outcomes. For acute MI there was a decline in the rate of occurrence of adverse events for which patients were at risk from 5% to 3.7% (95% confidence interval [CI], 0.7-1.9), the proportion of patients ≥1 adverse events from 26% to 19.4% (95% CI, 3.3-10.2), and in the number of adverse events per 1000 hospitalizations from 401.9 to 262.2 (95% CI, 90.6-189); CHF decreased from 3.7% to 2.7% (95% CI, 0.5-1.4), 17.5% to 14.2% (95% CI, 1-5.5), and from 235.2 to 166.9 (95% CI, 39.9-96.7) for each outcome, respectively.

No change was observed in the rate of occurrence of adverse events in patients with pneumonia or with conditions requiring surgery. Patients with pneumonia had slight increases, while patients with conditions requiring surgery had slight decreases in the 2 other outcomes. When patient characteristics and geographical differences were taken into account for these conditions, no significant changes were observed. 

Although in-hospital adverse event rates overall declined among patients with acute MI or CHF for all 3 outcomes, an increase in several events, including contrast-induced nephropathy, was observed in patients with acute MI, pneumonia, and patients with conditions requiring surgery. According to the study’s authors, the “increased adverse event rates among surgical patients indicates the continuing challenge and identifies an important target for patient safety initiative.”

Additionally, the researchers acknowledged a number of limitations that existed within the study.  The observed association between adverse events and longer hospitals stays may have simply been a reflection of the increased opportunity for adverse events as a result of a longer stay. The researchers were unable to measure the daily risk of adverse events due to missing data for some cases and were unable to account for events that went undocumented. There was also a change in the sample design during the study that altered the sample of patient discharges gathered from acute care hospitals.

Other factors that may have affected the study’s results included more restrictive transfusion practices causing a decrease in the ability to detect anticoagulant-related events and increased attention to certain complications, leading to increased frequency of testing and detection independent of any change in actual rates.

While a decline in the rates of in-hospital adverse events for patients with acute MI or CHF was observed, this reduction was not seen across the board. The researchers concluded the study provides much needed current data on the rates of adverse events, which will hopefully increase efforts to prevent in-hospital patient harm.