2.7. Study outcomes:
The primary outcome was the pre- and post-intervention carbapenem consumption as a direct measure of the ASP efficacy, while the secondary outcomes was the pre- and post-intervention prevalence of theKlebsiella isolates and their resistance.
The study protocol was approved by the institutional review board of the hospital. The requirements for informed consent were waived for this retrospective medical records review. Patient privacy and confidentiality of medical records and data were maintained all through the study.
2.8. Statistical methods :
Data were collected and coded before analysis using the statistical package of social science (SPSS version 25). Normal distribution of different variables was tested using the z-value of skewness and kurtosis between -1.96 and +1.96 (Doane & Seward, 2011) and the p-value of Shapiro-Wilk’s test > 0.05 (Razali & Wah, 2011; Shapiro & Wilk, 1965). The post-intervention was compared to pre-intervention carbapenems consumption using the independent sample t-test. Chi-Square test and odds ratio with 95% confidence interval (95% CI) were used for comparison of qualitative data. One-way ANOVA (Analysis of Variance) test was used to compare carbapenem consumption among the threeKlebsiella resistance grades. Two ordinal logistic regression models were used to study the effect of each of carbapenem consumption and the intervention program on Klebsiella resistance ordinal grades. Statistical significance was considered at p-value ≤ 0.05.
RESULTS:
A total of 472 patients with positive culture results were initially evaluated during the study period of 22 months. After excluding 267 patients with culture results with non-klebsiella organisms, the remaining 205 patients represented our study sample. The initial 11 months from April 2017 till February 2018 represented the pre-intervention control period and included 61 patients (29.8 %), while the following 11 months from March 2018 till January 2019 represented the post-intervention period and included 144 patients (70.2 %). Sputum samples were the most common representing 95 samples (46.3 %). Other samples included 38 for urine (18.5 %), 32 for wound swabs (15.6 %), 23 for blood cultures (11.2 %), and 17 for other samples (8 %).