Results of subgroup analyses for vancomycin administration based on the initial timing (h) of the audit and monitoring intervention by the ICT pharmacists, categorised exploratively into four groups.

The main explanatory variables were the initial timing of the audit and monitoring intervention by the ICT pharmacists (h) categorised as follows: <24 h, 24 to 72 h, 72 to 120 h, ≥120 h. aThe odds ratios derived from the multivariable logistic regression were adjusted for age, sex, weight, eCCr, creatinine, AST, ALT, albumin, comorbidities (diabetes, cardiovascular diseases, renal failure, hypertension and dyslipidaemia), immunosuppressant use, ICU admission, vancomycin loading dose, concomitant drug use (aminoglycoside, NSAIDs and piperacillin-tazobactam) and interventions of clinical pharmacists in charge of the wards. bThe odds ratios derived from the multivariable logistic regression were adjusted for age, sex, eCCr, albumin, ICU admission, vancomycin loading dose and interventions of clinical pharmacists in charge of the wards. cThe odds ratios derived from the multivariable logistic regression were adjusted for age, sex, comorbidities (diabetes, cancers, cardiovascular diseases, COPD, hepatic diseases, renal failure, hypertension, and dyslipidaemia), immunosuppressant use, ICU admission and interventions of clinical pharmacists in charge of the wards. dThe odds ratios derived from the multivariable logistic regression were adjusted for age, sex, eCCr, albumin, vancomycin loading dose, concomitant drug use (piperacillin-tazobactam) and interventions of clinical pharmacists in charge of the wards. eThe odds ratios derived from the multivariable logistic regression were adjusted for age, sex and eCCr. fThe odds ratios derived from the multivariable logistic regression were adjusted for age, sex, comorbidities (diabetes, cancers, and cardiovascular diseases) and interventions of clinical pharmacists in charge of the wards. gThe estimate from the multiple linear regression was adjusted for age, sex, weight, eCCr, creatinine, AST, ALT, albumin, comorbidities (diabetes, cardiovascular diseases), ICU admission, loading dose of vancomycin, concomitant drug use (aminoglycoside, NSAIDs and piperacillin-tazobactam), and interventions of clinical pharmacists in charge of the wards. hThe odds ratios derived from the multivariable logistic regression were adjusted for age, sex, eCCr and interventions of clinical pharmacists in charge of the wards. iThe odds ratios derived from the multivariable logistic regression were adjusted for age, sex, comorbidities (diabetes, cancers, cardiovascular diseases, COPD, hepatic diseases, renal failure, hypertension, and dyslipidaemia), immunosuppressant use and interventions of clinical pharmacists in charge of the wards. jThe odds ratios derived from the multivariable logistic regression were adjusted for age, sex, weight, eCCr, creatinine, AST, ALT, albumin, comorbidities (diabetes and cardiovascular diseases), immunosuppressant use, ICU admission, vancomycin loading dose, concomitant drug use (aminoglycoside, NSAIDs and piperacillin-tazobactam) and interventions of clinical pharmacists in charge of the wards. kThe odds ratios derived from the multivariable logistic regression were adjusted for age, sex, eCCr, ICU admission and interventions of clinical pharmacists in charge of the wards. ICT, infection control team; OR, odds ratio; 95% CI, 95% confidence interval; SD, standard deviation; SE, standard error; VCM, vancomycin; eCCr, estimate creatinine clearance; AST, aspartate aminotransferase, ALT, alanine aminotransferase; ICU, intensive care unit; NSAIDs, non-steroidal anti-inflammatory agents; COPD, chronic obstructive pulmonary disease; IQR, interquartile range.
(XLSX)