Postoperative Nausea and Vomiting in Adults: METHODS
A retrospective chart review was performed for 300 randomly selected adults undergoing inpatient and outpatient procedures from March 2003 to February 2004. Initially, 100 patients were in the dolasetron group, 99 were in the ondan-setron group, and 101 were in the generic granisetron group. As a result of having received combination therapy at the outset, 18 patients were excluded from the study. Of the 282 patients in the study, 89 received dolasetron, 93 received ondansetron generic, and 100 received canadian granisetron.
Men and women who were undergoing various types of orthopedic, intra-abdominal, and gynecological surgeries were enrolled. The study focused on the percentage of patients in each group who needed further rescue medications and the antiemetic agent that was administered in the event of breakthrough nausea and vomiting.
prescription drugs from canada
We also performed a cost-effectiveness analysis comparing different therapeutic strategies using the three agents from the perspective of an institutional setting.
All continuous variables were expressed as the mean ± standard deviation. We assessed differences in rates for categorical variables among patient groups by using Fisher’s exact test or the chi-square test for trend. Because of the skewness of the length of stay (LOS) data, we evaluated differences in LOS using the Wilcoxon Rank-Sum test or the Kruskal-Wallis test.
These tests were also used for ordinal variables such as the number of anesthetic agents. If the Kruskal-Wallis result was significant, Tukey’s pairwise multiple comparison test (P < .05) was used to determine the source of significance. Continuous and ordinal variables were correlated with LOS via the Spearman correlation coefficient.
As with many retrospective studies, there are often group imbalances with baseline variables and these variables are also predictive of the outcome variables. Multivariate analyses were used to control for baseline variables in order to confirm the univariate analyses of drug group being predictive of outcome variables. Stepwise multiple linear regression models were used to determine which variables, in addition to drug grouping, were independent predictors of LOS. We used the log of LOS to create more normally distributed data. canadian pharmacy viagra
We performed a stepwise logistic regression to determine the variables that were independent predictors of needing a rescue drug. In the stepwise procedures, P < .05 was used as a cut-off point. SAS 9.1 (Cary, North Carolina) for Windows 2000 was used to perform all statistical calculations. Results were considered statistically significant with a P value of less than .05.
