Predictors of Failed Conscious Sedation in Patients Undergoing an Outpatient Colonoscopy and Implications for the Adenoma Detection Rate.

Guidelines to triage patients to conscious sedation (CS) or monitored anaesthesia care (MAC) for colonoscopy do not exist. We aimed to identify the CS failure rate, predictors of failure, and its impact on the adenoma detection rate (ADR). Strict (based on patient experience) and expanded (based on doses of sedative medications) definitions of CS failure were used. Patient and procedure-related variables were extracted. Multivariable logistic regression identified predictors for CS failure and the ADR. Among 766 patients, 29 (3.8%) and 175 (22.8%) patients failed CS by strict and expanded definitions, respectively. Female gender (OR 3.50; 95% CI: 1.37-8.94) and fellow involvement (OR 4.15; 95% CI: 1.79-9.58) were associated with failed CS by the strict definition. Younger age (OR 1.27, 95% CI: 1.07-1.49), outpatient opiate use (OR 1.71; 95% CI 1.03-2.84), use of an adjunct medication (OR 3.34; 95% CI: 1.94-5.73), and fellow involvement (OR 2.20; 95% CI: 1.31-3.71) were associated with failed CS by the expanded definition. Patients meeting strict failure criteria had a lower ADR (OR 0.30; 95% CI: 0.12-0.77). Several clinical factors may be useful for triaging to MAC. The ADR is lower in patients meeting strict criteria for failed CS.

Do Consultants Follow Up on Tests They Recommend? Insights from an Academic Inpatient Gastrointestinal Consult Service.

BACKGROUND: Inpatient care is a fundamental part of gastroenterology training and involves the recommendation, performance, and interpretation of diagnostic tests. However, test results are not always communicated to patients or treating providers. We determined the process of communication of test results and recommendations in our inpatient gastroenterology (GI) consult service. METHODS: Test recommendations on 304 consecutive new GI consults (age 60.2 ± 1.0 year) over a 2-month period were recorded. Demographic factors (age, race, gender, zip code, insurance status) were extracted from the electronic medical record (EMR). Charts were independently reviewed 6 months later to determine results of recommended tests, follow-up of actionable test results, 30-day readmission rates, and predictors of suboptimal communication. RESULTS: Of 490 recommended tests, 437 (89.2%) were performed, and 199 (45.5%) had actionable findings. Of these, 48 (24.1%) did not have documented follow-up. Failure of follow-up was higher for upper endoscopy (31.9%) compared to colonoscopy (18.0%, p = 0.07). Women (p = 0.07), patients on Medicare (p = 0.05), and procedures supervised by advanced GI fellows (p = 0.06) were less likely to receive follow-up. Median income and identification of a primary provider did not influence follow-up rates; 30-day readmission rates were not impacted. Female gender, insurance (Medicare) status, and attending type remained independent predictors of failure of follow-up on multivariate regression (p ≤ 0.03). CONCLUSIONS: Failure to follow up test results on inpatient services at a large academic center was unacceptably high. Maximizing personnel participation together with diligence and technology (EMR) will be required to improve communication.