UCLA

Antibiotic resistance is a serious global public health threat and antibiotic prescribing is the key driver. Computerized antibiotic stewardship interventions have been implemented to facilitate physicians' decision making and promote optimal antibiotic selection at the point of prescribing. However, predictors of patient receipt of computerized antibiotic stewardship interventions have not been studied, and the clinical benefits of such interventions to individual patients remain unclear. This dissertation investigated physician and patient factors associated with physicians' acceptance or patients' receipt of computerized antibiotic stewardship intervention, the comparative effectiveness of computerized antibiotic stewardship intervention on individual patients' clinical outcomes, and the modification of these effects by patient factors.

We followed up an inpatient cohort in a 1500-bed tertiary care hospital in Singapore, with its homegrown antibiotic computerized decision support system (CDSS) that integrates antibiotic stewardship with electronic prescribing. In addition, we conducted a mixed methods study on physicians, to determine the psychosocial factors associated with physicians' acceptance of CDSS recommendations.

We observed that physicians' willingness to consult the antibiotic CDSS determined acceptance of its recommendations, and that physicians would choose to exercise their own or clinical team's decision over the CDSS recommendations in complex patient situations when the antibiotic prescribing needs were not met. The prescribing physician --but not the attending physician or clinical specialty-- accounted for some (13.3%) of the variation in patients' receipt of CDSS recommendations. Patients requiring intensive care (OR 0.38, 95% CI 0.21-0.66) and those with renal impairment (OR 0.70, 95% CI 0.52-0.93) were less likely to receive the intervention, as their complex clinical conditions might require a physician's assessment in addition to antibiotic CDSS.

We further observed that patients' receipt of CDSS recommendations halved the odds of mortality in patients (OR 0.54, 95% CI 0.26-1.10), with patients aged <= 65 years having a greater mortality benefit (OR 0.45, 95% CI 0.20-1.00). No appreciable increase in infection-related readmission (OR 1.16, 95% CI 0.48-2.79) was found in survivors.

Our findings can help healthcare institutions in the design of new antibiotic CDSSs and enhancements of existing ones to promote the optimal use of antibiotics in the global battle against antibiotic resistance.