Researchers focus on various quality metrics to improve care

A comprehensive in-hospital disease management program improves quality of care and adherence to performance measures, while reducing length of stay for patients hospitalized with acute decompensated heart failure, according to a poster presentation at the 2008 American College of Cardiology meeting.

Two other poster presentations that focused on quality issues found that a continuous quality improvement program decreased mortality after cardiac surgery and that a taxonomy for diagnostic errors in pediatric echocardiography helps target quality improvement initiatives.

Vimal Rabdiya, MD, and colleagues at Saint Francis Hospital, Hartford, Conn., analyzed the prospectively collected data of 1,707 patients admitted to their hospital for acute decompensated heart failure (ADHF) over one year.

Researchers divided patients into three groups based on whether the admission was to the heart failure floor (DMP) and who was the primary physician responsible for patient care (cardiology and medicine).

These groups were compared for the average length of hospital stay and rates of conformity with the Joint Commission core performance measures:

  • discharge instructions
  • assessment of left ventricular function
  • use of ACE inhibitors or angiotensin receptor blockers, and
  • smoking cessation counseling.

Group A (DMP) consisted of 608 patients; group B (cardiology) consisted of 761; group C (medicine) consisted of 338. Researchers found no significant difference in baseline patient characteristics among the groups. They found that group A was associated with a shorter length of stay and a higher rate of conformity with Joint Commission measures.

In another presentation, Sotiris C. Stamou, MD, and colleagues at the Carolinas Heart Institute, Charlotte, N.C., investigated the effects of a continuous quality improvement (CQI) program and goal-oriented, multidisciplinary protocols had on outcomes after cardiac surgery.

Researchers divided patients into two groups: those who had surgery (CABG, valve, or both) after the establishment of the CQI (n=922) and those prior to the institution of the CQI (n=1289).

The protocols included standardized communication tools and goal sheets, sedation monitoring, respiratory protocols for early extubation and best pulmonary practices bundles, computerized euglycemia management, blood management and infection control programs, and multidisciplinary ICU rounds

Stamou and colleagues found a lower operative mortality in the CQI group compared to the non-CQI group (2.6 vs. 5 percent, respectively. The difference was significant. 

Significant independent predictors of higher mortality after cardiac surgery included
diabetes, congestive heart failure, unstable angina, age over 75 years and prolonged pump time. Non-significant predictors were chronic renal insufficiency, previous cardiovascular operation and prolonged operation. The use of CQI was significantly associated with lower mortality after cardiac surgery.

Oscar J. Benavidez, MD, and colleagues at Children's Hospital Boston theorized that a taxonomy for diagnostic errors in pediatric echocardiography could assist in targeting quality improvement initiatives.

Researchers identified diagnostic errors by quality improvement initiatives and voluntary reporting. They examined each case to identify and categorize contributing factors by reviewing medical records, diagnostic images and conducting interviews. The taxonomy was applied prospectively from December 2004 to August 2007.

During the study period, approximately 32,000 echocardiograms were performed. Among the 87 diagnostic error cases identified, 61 percent were false negative, 20 percent false positive and 20 percent discrepant diagnosis (wrong pathologic diagnosis).

Regarding severity, 30 percent were minor and had no clinical impact; 63 percent moderate and impacted clinical management or the patient was at risk or experienced a transient adverse event; 6 percent major, experienced permanent adverse event; and 1 percent catastrophic, the error contributed to patient demise.

One-third of errors were preventable (the correct diagnosis was apparent on the study); 46 percent possibly preventable and 21 percent not preventable. Sixty-nine percent of preventable errors were moderate severity or greater.

Top contributors to diagnostic errors were: poor imaging windows, 17 percent; image misinterpretation, 13 percent; modality limitation, 11 percent; incomplete examination, 10 percent; and under appreciation of a finding, 10 percent.

“These findings illustrate the utility of a diagnostic error taxonomy to identify targets for quality improvement such as preventable diagnostic errors of moderate severity,” Benavidez said.