predictors of asp recommendation disagreement 809
was<.05, where the P value was calculated from an F-distribution based on the likelihood-ratio test. After obtaining the initial model, variables were added and removed iteratively to find alternative models. Models were ranked by Akaike information criterion (AIC) to possibly account for other variables that could be considered more parsimonious. The new models that were developed all had comparable AIC values but had added variables that were not statistically sig- nificant. The final model was the best fitting model. Statistical analyses were conducted using R statistical package, version 3.2 (R Foundation for Statistical Computing, Vienna, Austria).
results
Of the 4,727 antimicrobial audits performed during the 2-year study period, 1,323 (28%) resulted in a PAFR (Figure 2). Vancomycin and piperacillin-tazobactam had the greatest number of PAFRs, accounting for 279 of 304 broad-spectrum gram-positive antibiotics (92%) and 140 of 519 broad- spectrum gram-negative antibiotics (27%), respectively. Of 1,323 PAFR, 525 (~40%) were for the infectious problem of suspected or proven sepsis. The most common type of recommendation was to stop the antimicrobial (46%) and the majority of recommendations (85%) were communicated to a UBP. Of the 1,323 PAFR, 1,046 were followed (including 34
recommendations with an alternative approach agreed upon), resulting in a 79% acceptance rate. After exclusion of PAFR that were not followed for a reason other than disagree (n=90), there were 187 PAFR with a recommendation not followed. Therefore, the incidence of PAFR disagreement was
15%. Univariate comparison identified the following statisti- cally significant patient-level variables associated with PAFR disagreement: patient age, infectious problem, medical service, presence of central line, days from admission to audit, total intensive care unit length of stay, total length of stay, antimicrobial type, and recommendation type (Table 1). In the adjusted analysis, the following patient-level, programmatic, and provider-level variables were statistically significant predictors of PAFR disagreement: recommendation type, infectious problem, time from admission to audit date, medical service, and years of attending experience (Table 2). Compared to bacteremia, providers were 5- to 6-fold more likely to disagree with PAFR pertaining to intra-abdominal infection, febrile neutropenia, or skin and soft-tissue infection. Disagreement with PAFR was more likely when PAF was performed 31–90 days into a patient’s hospital admission compared to the first 30 days. The PAFR were more likely to be followed if the recommendation type was to clarify the antimicrobial plan or to optimize the antimicrobial dose or frequency compared to stopping the audited antimicrobial. The PICU had a 2.7-fold higher probability of not following PAFR compared to the NICU. For every year of experience following completion of postgraduate training, attending providers were 2.4% less likely to follow PAFR.
discussion
figure 2. Total number of audits performed during the study period and study cohort identification based on exclusion of recommendations not followed for reasons other than disagreement.
In the face of widespread antimicrobial utilization, emerging resistance, increasing regulatory requirements, and limited healthcare resources, an understanding of how to maximize both the efficiency and the impact of ASP strategies is critically important. Prospective audit and feedback have been shown to affect positive changes in antimicrobial utilization; however, to maximize the success of this strategy, the drivers of provider acceptance of recommendations must be understood. To our knowledge, this is the most comprehensive study evaluating the predictors of PAFR disagreement, including patient-level, antimicrobial, programmatic, and provider-level factors. Building on previously identified predictors of disagreement, we found that several additional factors were associated with provider disagreement with PAFRs at our hospital. There is no standard approach to the PAF process, and the literature reveals a variety of strategies.3,5,7,9,10 Unique aspects of our PAF program include the auditing of all injectable medications, communication of PAFR to the UBP in most cases, and documentation of PAFR in the electronic medical record. Despite these differences, our analysis reveals some remarkable similarities across pediatric PAF programs. As has been reported elsewhere, our recommendations were most commonly made for broad-spectrum antibiotics such as piperacillin/tazobactam and vancomycin.8 Similar to prior studies, we also found that recommendations were most commonly made for patients with suspected or proven sepsis, and the most common recommendation was to stop the antimicrobial.8,9 Given that this finding has been reported in
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52 |
Page 53 |
Page 54 |
Page 55 |
Page 56 |
Page 57 |
Page 58 |
Page 59 |
Page 60 |
Page 61 |
Page 62 |
Page 63 |
Page 64 |
Page 65 |
Page 66 |
Page 67 |
Page 68 |
Page 69 |
Page 70 |
Page 71 |
Page 72 |
Page 73 |
Page 74 |
Page 75 |
Page 76 |
Page 77 |
Page 78 |
Page 79 |
Page 80 |
Page 81 |
Page 82 |
Page 83 |
Page 84 |
Page 85 |
Page 86 |
Page 87 |
Page 88 |
Page 89 |
Page 90 |
Page 91 |
Page 92 |
Page 93 |
Page 94 |
Page 95 |
Page 96 |
Page 97 |
Page 98 |
Page 99 |
Page 100 |
Page 101 |
Page 102 |
Page 103 |
Page 104 |
Page 105 |
Page 106 |
Page 107 |
Page 108 |
Page 109 |
Page 110 |
Page 111 |
Page 112 |
Page 113 |
Page 114 |
Page 115 |
Page 116 |
Page 117 |
Page 118 |
Page 119 |
Page 120 |
Page 121 |
Page 122 |
Page 123 |
Page 124 |
Page 125 |
Page 126 |
Page 127 |
Page 128 |
Page 129 |
Page 130 |
Page 131 |
Page 132 |
Page 133 |
Page 134 |
Page 135 |
Page 136 |
Page 137 |
Page 138 |
Page 139 |
Page 140 |
Page 141 |
Page 142 |
Page 143 |
Page 144
