National and Regional Conformity to the 2007 ACG/AASLD Practice Guidelines for Prevention and Management of Gastroesophageal Varices and Variceal Hemorrhage in Cirrhosis

Purpose: Clinical practice varies by region; practice guidelines (PG) try to minimize this. PG are evidence-based recommendations to assist practitioners with appropriate healthcare for specified problems. We surveyed national and regional clinicians' conformity with the ACG/AASLD's PG for prevention and management of gastroesophageal varices and variceal hemorrhage in cirrhosis, (Am J of Gastro; 2007, Sept 102 [9]). We hypothesized compliance with the PG would vary by region despite equal access to evidence. Methods: We evaluated 11 of 24 PG. A national cohort (NAT) was surveyed during a symposium at the ACG Scientific Meeting, Oct 2007. A Northeastern Ohio cohort (NEO) of gastroenterologists obtained from societies' databases was surveyed through three mailings, Nov 2007- Jan 2008. Descriptive statistics were computed for 14 questions-medians, 25th and 75th percentiles for ordinal variables and frequencies for categorical factors. Pearson's chi-square tests assessed associations between compliance with the PG and categorical factors. Results: NAT had 159 respondents of 400 estimated symposium attendees. NEO had 63 of 160 surveyed (38.8% responded). Fifty-four percent of respondents showed compliance with the PG. Overall there was no difference between NAT and NEO compliance (NAT-58.3%, NEO-50.0%; P= 0.16). Individual questions had significant differences including primary prevention for large varices in low risk patients (NAT- 61.3%, NEO- 78%; P= 0.022), beta-blocker titration (NAT- 19.1%, NEO- 6.8%; P= 0.027), and antibiotics in variceal hemorrhage (NAT- 69.3%, NEO- 46.6%; P= 0.002). Academic physicians (ACD) versus private physicians (PRV) had significantly higher self-reported compliance for recommendations on initial variceal screening (ACD- 83.6%, PRV- 70.1%; P= 0.032), octreotide duration in hemorrhage (ACD- 59.4%, PRV- 41.5%; P= 0.015), antibiotics in variceal hemorrhage (ACD- 76.8%, PRV- 55.6%; P= 0.003), and controlling gastric varices (ACD- 97.1%, PRV- 87.1%; P= 0.023). Subjects practicing >40% hepatology (HEP) had significantly higher self-reported compliance for beta-blocker titration (HEP- 40%, Other-11.4%; P= < 0.001), antibiotics in variceal hemorrhage (HEP- 86.7%, Other- 58.9%; P= 0.004) and prevention of rebleeding (HEP- 93.3%, Other-76.0%; P= 0.033). Conclusion: PG compliance is low. ACD and HEP physicians are more often in compliance. Regional differences are apparent. Education is needed nationally and regionally to ensure exposure to validated measures for management of gastroesophageal varices. Studies are needed to understand non-compliance, regional differences and morbidity/mortality associated with lower compliance. A study is planned to evaluate changes in physicians' compliance over time.