Table 2

Changes in clinical outcomes from EMR before and after change point in RA-CUSUM curve analysis for endoscopist proficiency gain (A) non-cancer patients and (B) patients with oesophagogastric cancer

OutcomeChange pointOverall rateBefore change pointAfter change pointp Value
(A) Non-cancer patients
 Emergency interventionNo PGC
 30-day mortality2 cases1.0% (0.8% to 1.2%)1.8% (1.4% to 2.3%)0.5% (0.4% to 0.7%)<0.001
 90-day mortality2 cases2.2% (2.0% to 2.5%)3.6% (3.1% to 4.3%)1.3% (1.1% to 1.7%)<0.001
(B) Patients with oesophagogastric cancer
 Emergency intervention1 case0.6% (0.3% to 1.3%)1.5% (0.7% to 3.2%)0.1% (0.0% to 0.8%)0.009
 30-day mortality4 cases3.1% (2.2% to 4.3%)4.5% (3.2% to 6.3%)0.3% (0.1% to 1.5%)<0.001
 90-day mortality2 cases6.8% (5.4% to 8.4%)10.8% (8.4% to 13.6%)2.6% (1.5% to 4.3%)<0.001
 Re-intervention43 cases23.6% (21.1% to 26.2%)24.6% (22.0% to 27.3%)10.8% (5.4% to 19.4%)0.005
 Oesophagectomy/gastrectomy14.6% (12.9% to 16.8%)15.3% (13.2% to 17.6%)5.3% (2.1% to 12.8%)0.017
 Repeat EMR9.0% (7.4% to 10.9%)9.3% (7.6% to 11.2%)5.3% (2.1% to 12.8%)0.238
  • Emergency intervention was defined as an acute emergency operation or endoscopic stent insertion performed within 14 days of the primary EMR procedure.

  • Re-intervention was defined as repeat EMR, oesophagectomy or gastrectomy for the treatment of cancer at least 30 days following the initial EMR procedure.

  • EMR, endoscopic mucosal resection; PGC, proficiency gain curve; RA-CUSUM, risk-adjusted cumulative sum.