scholarly journals A nonparametric approach to weighted estimating equations for regression analysis with missing covariates

2012 ◽  
Vol 56 (1) ◽  
pp. 100-113 ◽  
Author(s):  
An Creemers ◽  
Marc Aerts ◽  
Niel Hens ◽  
Geert Molenberghs
Keyword(s):  
Regression Analysis ◽  
Estimating Equations ◽  
Missing Covariates ◽  
Nonparametric Approach

scholarly journals A24 RATE OF GASTROINTESTINAL ENDOSCOPY USE AND CONCOMITTANT ANESHESIOLOGY ASSISTANCE AND PREDICTORS OF ANESTHESIOLOGY ASSISTANCE: A POPULATION-BASED ANALYSIS

2020 ◽  
Vol 3 (Supplement_1) ◽  
pp. 28-30
Author(s):  
H Singh ◽  
S Derksen ◽  
M Sirski ◽  
S McCulloch ◽  
L M Lix
Keyword(s):  
Logistic Regression ◽  
Regression Analysis ◽  
Logistic Regression Analysis ◽  
Generalized Estimating Equations ◽  
Time Trends ◽  
Gastrointestinal Endoscopy ◽  
Estimating Equations ◽  
Income Quintile ◽  
Rate Of Increase ◽  
Generalized Estimating

Abstract Background There are limited Canadian data on time trends of gastrointestinal endoscopy (GIE) use and concomitant anesthesiology assistance. Aims To test the time trends and variations in concomitant anesthesiology assistance during GIE in the central Canadian province of Manitoba. Methods Physicians claims reimbursement data from the provincial health ministry were used to ascertain upper & lower GIE procedures and anesthesiology assistance annually from 1984 to 2016. The annual provincial population was determined from health insurance registration files. Generalized linear models with generalized estimating equations were used to test the linear trend over time and differences amongst age groups, health regions and income quintiles. Logistic regression analysis with generalized estimating equations was used to assess predictors of anesthesiology use. Results 410, 685 individuals had at least one procedure in the observation period. Approximately half of the procedures were performed outside the recommended CRC screening age group of 50–74 years, with 38% among those less than 50 years. The average annual rate of increase was 4.6% (95% CI 4.4–4.8%). The rate of increase stabilized among the > 75 years group over the last 5 years of the study. There were marked regional variation in GIE procedures, with the smallest increase in Winnipeg region (the largest urban region):2.9% (95%CI: 2.5–3.3) vs. 5.2% (95%CI: 4.8–5.6) in Southern Health; RR: 2.2(figure 1 A). There were no differences in GIE rates by income quintile. Similar patterns were seen in analysis stratified by upper and lower GIE. Concomitant anesthesiology use ranged from 1% in Winnipeg region to 80% in the southern rural region (figure 1B). In the logistic regression analysis, independent predictors of concomitant anesthesiology use in rural regions included younger age, female sex, lower income quintile, and higher comorbidities of patients, GP endoscopist, region of physician practice, and lower volume endoscopy physician. Conclusions Marked regional variations in GIE use and concomitant anesthesiology use were observed in a universal health care system. Efforts to standardize care are needed to reduce variations. Funding Agencies Manitoba Health

Sign in / Sign up

Export Citation Format

Share Document