Tu1140 REGIONAL VARIATIONS AND TRENDS IN QUALITY INDICATOR (QI) PERFORMANCE IN BARRETT’S ESOPHAGUS (BE): RESULTS FROM A POPULATION-BASED STUDY USING A NATIONAL BENCHMARKING REGISTRY

Abstract

QIs in BE have been defined with the goal of improving patient outcomes. While adherence to the Seattle biopsy protocol (4-quadrant q2 cm) and surveillance intervals in non-dysplastic BE patients (NDBE) of 3-5 yrs are proposed QIs, population-based data addressing regional variations are not available. Using a national benchmarking clinical registry, to evaluate regional variation in adherence to the Seattle biopsy protocol and adherence to recommended surveillance intervals in NDBE and low-grade dysplasia (LGD). We analyzed data from the GI Quality Improvement Consortium (GIQuIC) Registry, a national data repository of endoscopy quality measures. Upper endoscopy data include procedure indication, demographics, endoscopy findings, and pathology results. EGD records from 1/2012-9/2017 were assessed. Patients (pts) with an indication of BE screening or surveillance, or an endoscopic finding of BE, were included. Adherence to Seattle protocol was defined by dividing the BE length by number of pathology jars, with a ratio of ≤2.0 with rounding down (lenient definition) or rounding up (stringent definition) for odd BE lengths considered adherent. Adherence to surveillance for NDBE was calculated as the proportion recommended to undergo EGD between 3-5 yrs (non-adherent if EGD recommended at 3 months–2 yrs from the index EGD) for NDBE and the proportion recommended an EGD at ≤1 year for LGD. Cochran-Armitage test for trend was applied to detect improvement in adherence. 786,712 EGDs were assessed, and 58,709 (7.5%) EGDs in 53,541 pts met inclusion criteria [mean age 61.3 yrs (SD 12.6), 60.4% male, 90.2% Caucasian]. Majority of cases were performed by GIs (92.3%) with propofol (78.7%), and represented 263 practices nationwide and 1457 unique providers. Distribution of cases based on US census region was: Northeast (n=14735, 27.6%), South (n=16376, 30.7%), Midwest (n=10,068, 18.8%) and West (n=12,152, 22.8%). Mean BE length was 2.3 cm (SD 2.3). No significant regional variations were noted in demographics, proportion of pts undergoing biopsies and confirmed BE and BE pts diagnosed with dysplasia (Table 1). There are marked variations in adherence with Seattle protocol by region (Table 2), with the highest rates of adherence noted in the Northeast and the lowest in the Midwest. In time-trend analysis, there was an improvement in adherence rates in the Midwest and a decline in the South. Among NDBE pts, non-adherence rates with the 3-yr surveillance endoscopy threshold ranged from 24-33% with the highest rates in the Northeast (Table 2). Endoscopists in the Northeast are most likely to be compliant with adherence to the Seattle biopsy protocol, but least likely to be compliant with recommended surveillance intervals. Further research to understand the factors influencing non-adherence to guidelines is warranted.Tabled 1Table 1: Regional variations in patient demographics and endoscopy-based variablesVariableNortheastSouthMidwestWestOverallDemographics (patient-based)Mean age, yrs (SD)61.2 (12.6)61.5 (12.4)60.9 (13.0)61.4 (12.5)61.3 (12.6)Males (n,%)8772 (59.5%)9921 (60.6%)6188 (61.4%)7350 (60.4%)32,346 (60.4%)Caucasians (n,%)89.9%89.2%93.1%89.4%90.2%Endoscopy-based variablesGIs*13,198 (95.1%)15,406 (89.8%)9272 (91.7%)11,671 (93.6%)49,654 (92.3%)Propofol sedation6475 (83%)4775 (79.6%)3283 (65.1%)5686 (83%)20,264 (78.7%)Adverse event rate0.01%0.02%0.02%0.03%0.02%Mean BE length, cm, SD2.05 (1.97)2.4 (2.38)2.5 (2.46)2.3 (2.44)2.3 (2.3)Proportion of patients with biopsies (n, %)15,792 (97.3%)16,594 (92.7%)10,007 (90.3%)12,332 (92.9%)54,943 (93.6%)Confirmed BE (n, %)8413 (55%)8745 (54.9%)5709 (57.5%)7012 (56.8%)29,978 (55.8%)Pathology results (n, %)NDBE7311 (92.2%)7561 (91.7%)4782 (88.5%)6202 (92.7%)25,945 (91.5%)IND/LGD499 (6.3%)565 (6.8%)506 (9.3%)376 (5.6%)1951 (6.8%)HGD113 (1.4%)117 (1.4%)111 (2.1%)107 (1.6%)449 (1.6%) Open table in a new tab