Quality of care for ascites, the most common complication of cirrhosis, was found to be suboptimal at several Veterans Affairs medical centers, reported Dr. Fasiha Kanwal and colleagues in the July issue of Gastroenterology.
"In general, care targeted at diagnosis and treatment was more likely to meet standards than preventive care," wrote Dr. Kanwal, of the Michael E. DeBakey Veterans Affairs Medical Center, Houston.
"We also found a trend towards improved outcomes in patients who met recommended quality indicators," added the investigators, although these findings "can only be regarded as preliminary."
The authors studied records from 774 patients (mean age 54.7 years, 99% male) in a database comprising 3 VA medical centers and 15 affiliated clinics in the Midwest (Gastroenterology 2012 [doi: 10.1053/j.gastro.2012.03.038]).
All patients had at least two ICD-9 codes for cirrhosis or at least one code for cirrhosis with either a code for complications of cirrhosis or an aspartate aminotransferase to platelet ratio greater than 2. The patients were seen between January 2000 and December 2007.
The authors compared data on these patients to a set of class 1 ascites care quality indicators (QIs). These indicators were derived by using the RAND/University of California, Los Angeles (UCLA), Appropriateness Method, which had been previously published elsewhere (Clin. Gastroenterol. Hepatol. 2010;8:709-17).
If a patient had been hospitalized more than once, only the first hospitalization was assessed. The rate of adherence to each QI was expressed as a percentage of subjects who received the recommended care, among those who were eligible for the QI.
The first QI assessed the percentage of new-onset ascites patients who underwent abdominal paracentesis within 30 days of diagnosis. On this measure, the VA scored 50.6%. The second indicator was whether known ascites patients admitted with either ascites or hepatic encephalopathy underwent abdominal paracentesis during the index hospitalization. Just over half (57.6%) of patients met this criterion.
The next QI was more likely to be met: 89.3% of patients who underwent abdominal paracentesis received ascitic fluid cell count and differential. Another indicator that was met for a high percentage of patients addressed whether ascites patients with normal renal function received diuretics within 30 days of diagnosis – 82.8% met this criterion.
Similarly, among hospitalized patients with spontaneous bacterial peritonitis (SBP), 72.0% received antibiotics within 24 hours before or after ascitic fluid analysis.
However, just 30% of patients with SBP who survived and were discharged from the facility received long-term outpatient antibiotics (for secondary prophylaxis) within 30 days. And just under half (49.2%) of patients admitted with a GI bleed received antibiotics during the index hospitalization.
The final QI was associated with the worst compliance rate: just 22.2% of patients with ascitic fluid total protein levels less than 1 g/dL and serum bilirubin of greater than 2.5 mg/dL received long-term outpatient antibiotics (for primary prophylaxis) within –3 to 30 days of that test result.
Next, the authors assessed which demographic or other independent factors were associated with higher QI compliance. In general, they reported that better care was inversely related to a worsening liver disease. More specifically, they found that patients who saw a gastroenterologist received higher-quality care than those who did not (odds ratio, 1.33), as did patients who were seen at a VA facility with academic affiliation, versus unaffiliated centers (OR, 1.73).
Finally, in two exploratory analyses, the authors examined how adherence to the ascites QIs affected patient outcomes.
Not surprisingly, "we found that after adjusting for age, liver disease severity, and comorbidity, patients receiving suboptimum care had 37% higher odds of death and 35% higher odds of readmission during the 12-month follow-up compared to patients who received optimum ascites care," although these figures did not reach statistical significance.
This study was supported by the 2008 American Society of Gastrointestinal Endoscopy Quality of Care Award and by the 2009 American College of Gastroenterology Clinical Research Award. The authors stated that they had no personal conflicts of interest.