PubMed Health⌕ Search

Biomedical subjects

Adrian Levy

Publications and source records attributed to Adrian Levy.

7 recordsLinked to original sources
PubMed2007

Bayesian sensitivity analysis for unmeasured confounding in observational studies.

We consider Bayesian sensitivity analysis for unmeasured confounding in observational studies where the association between a binary exposure, binary response, measured confounders and a single binary unmeasured confounder can be formulated using logistic regression models. A model for unmeasured confounding is presented along with a family of prior distributions that model beliefs about a possible unknown unmeasured confounder. Simulation from the posterior distribution is accomplished using Markov chain Monte Carlo. Because the model for unmeasured confounding is not identifiable, standard large-sample theory for Bayesian analysis is not applicable. Consequently, the impact of different choices of prior distributions on the coverage probability of credible intervals is unknown. Using simulations, we investigate the coverage probability when averaged with respect to various distributions over the parameter space. The results indicate that credible intervals will have approximately nominal coverage probability, on average, when the prior distribution used for sensitivity analysis approximates the sampling distribution of model parameters in a hypothetical sequence of observational studies. We motivate the method in a study of the effectiveness of beta blocker therapy for treatment of heart failure.

Adrenergic beta-Antagonists↗
PubMed2006

Does wait-list size at registration influence time to surgery? Analysis of a population-based cardiac surgery registry.

OBJECTIVE: To determine whether the probability of undergoing coronary bypass surgery within a certain time was related to the number of patients on the wait list at registration for the operation in a publicly funded health system. METHODS: A prospective cohort study comparing waiting times among patients registered on wait lists at the hospitals delivering adult cardiac surgery. For each calendar week, the list size, the number of new registrations, and the number of direct admissions immediately after angiography characterized the demand for surgery. RESULTS: The length of delay in undergoing treatment was associated with list size at registration, with shorter times for shorter lists (log-rank test 1,198.3, p<.0001). When the list size at registration required clearance time over 1 week patients had 42 percent lower odds of undergoing surgery compared with lists with clearance time less than 1 week (odds ratio [OR] 0.58 percent, 95 percent, confidence interval [CI] 0.53-0.63), after adjustment for age, sex, comorbidity, period, and hospital. The weekly number of new registrations exceeding weekly service capacity had an independent effect toward longer service delays when the list size at registration required clearance time less than 1 week (OR 0.56 percent, 95 percent CI 0.45-0.71), but not for longer lists. Every time the operation was performed for a patient requiring surgery without registration on wait lists, the odds of surgery for listed patients were reduced by 6 percent (OR 0.94, CI 0.93-0.95). CONCLUSION: For wait-listed patients, time to surgery depends on the list size at registration, the number of new registrations, as well as on the weekly number of patients who move immediately from angiography to coronary bypass surgery without being registered on a wait list. Hospital managers may use these findings to improve resource planning and to reduce uncertainty when providing advice on expected treatment delays.

Aged↗
PubMed2006

Unplanned emergency surgery in relation to length of wait lists at registration.

OBJECTIVE: To compare the cumulative incidence of emergency surgery between two groups of patients classified according to the length of wait lists at the time of their registration for coronary artery bypass grafting (CABG) and to test for significant differences in the risk of emergency surgery resulting from registration on a longer wait list. METHODS: A prospective study of all adult British Columbia residents who registered to undergo isolated CABG. We compared the time-dependent cumulative incidence for undergoing planned surgery through unplanned emergency admission before or during a certain wait-list week between two categories of wait-list size. The list size was a simple count of patients with higher or equal urgency to undergo CABG who were on a wait list at the time of registration of a new patient. RESULTS: Wait lists with one month or less of clearance time were observed in all urgent patients and were more prevalent in semi-urgent than non-urgent patients (79.1% vs 44.7%, respectively). The patients registered on a list with a clearance time of more than one month had a rate of unplanned emergency admission similar to those on a list with a clearance time of one month or less, OR = 1.07 (95% CI, 0.78-1.47) after adjustment for age, sex, comorbidity, calendar period, urgency and week on the list. During fifty-two weeks of the wait-list follow-up, an equal proportion of patients underwent unplanned emergency surgery after registration on lists in both clearance-time categories, OR = 1.03 (95% CI, 0.78-1.37) after adjustment. The number of patients who underwent CABG without having been registered on a wait list in the same hospital exerted no independent effect. CONCLUSIONS: The length of a wait list at registration had no effect on the probability that a semi-urgent or non-urgent patient would undergo CABG through unplanned emergency admission before or during a certain wait-list week.

Adult↗
PubMed2005

Intravenous proton pump inhibitors before endoscopy in bleeding peptic ulcer with high-risk stigmata: a multicentre comparative study.

BACKGROUND: It is not clear if starting intravenous proton pump inhibitors (IV PPI) before endoscopic therapy provides additional benefit over starting it afterward in patients with high-risk ulcer stigmata of peptic ulcer disease. METHODS: All patients who received IV pantoprazole bolus and infusion and underwent endoscopy in six Canadian hospitals over 20 months were reviewed. Only patients with high-risk ulcer stigmata (arterial bleeding, oozing, nonbleeding visible vessel or adherent clot) were included. Patients receiving IV PPI before endoscopy (before group) were compared with those who received it after endoscopy (after group) with respect to endoscopic findings and, secondarily, to patient demographics and clinical outcomes. RESULTS: The demographics and baseline characteristics of the before group (n=57) and the after group (n=109) were similar. The before group was more likely to have had IV PPI started outside of daytime hours, and median time to endoscopy in patients admitted with upper gastrointestinal bleeding was 24 h (interquartile range 9.5 to 35) in the before group and 11.3 h (interquartile range 3.7 to 17.2) in the after group (P<0.0001). At the time of endoscopy, 33% of patients in the before group had actively bleeding lesions (Forrest 1a or 1b) compared with 54% in the after group (P=0.01), but there were no significant differences in rebleeding, surgical rates, intensive care unit admission or death between the groups. CONCLUSION: IV PPI infusions before endoscopy may lower the proportion of actively bleeding peptic ulcer lesions at endoscopy, but this finding does not appear to affect rates of rebleeding, surgery or death.

2-Pyridinylmethylsulfinylbenzimidazoles↗
PubMed2004

Description of prescribing practices in patients with upper gastrointestinal bleeding receiving intravenous proton pump inhibitors: a multicentre evaluation.

BACKGROUND: Intravenous forms of proton pump inhibitors (IV PPI) are routinely used for patients with acute upper gastrointestinal bleeding, but a significant concern for their inappropriate use has been suggested. patients and METHODS: All consecutive patients who received IV PPI (pantoprazole) over 20 months in six Canadian hospitals were reviewed. Prescribing practices, endoscopic findings and outcomes were recorded. RESULTS: A total of 854 patients received IV PPI. Over 90% of patients were given IV PPI for treatment of known or suspected active upper gastrointestinal bleeding. Most patients (69%) underwent upper endoscopy, and 58% of these patients had peptic ulcer disease (PUD). The majority of patients who had endoscopy (57%) had IV PPI administered in advance of the procedure. Of the 334 patients who had IV PPI given in advance, 46 (13.8%) were found to have high risk bleeding PUD stigmata at endoscopy. The remaining 288 patients (86.2%) with advance IV PPI had low-risk PUD lesions or non-PUD lesions; IV PPI was continued after endoscopy in 164 (56.9%) of these patients. CONCLUSIONS: IV PPI is often used before endoscopy in suspected upper gastrointestinal bleed and maintained, regardless of endoscopic findings, after the endoscopy in many Canadian centres. Further study is required to support these clinical practices.

2-Pyridinylmethylsulfinylbenzimidazoles↗
PubMed2004

Specter orange.

Nearly 30 years after the Vietnam War, a chemical weapon used by U.S. troops is still exacting a hideous toll on each new generation in Vietnam. The dioxin (TCCD) that contaminated the herbicide Agent Orange is one of the most toxic molecules known to science. The contaminant persists in the soil. The United States has done nothing to combat the medical and environmental catastrophe that is overwhelming the country.

2,4,5-Trichlorophenoxyacetic Acid↗