Prospective antimicrobial audit and feedback did not decrease case fatality : Experiences from a hospital in northern Taiwan

Introduction: Although a prospective antimicrobial audit and feedback is an effective strategy in an antibiotic stewardship program, previous researchers have not adequately demonstrated a successful impact on patient outcomes. In this study, the causes of fatalities associated with a prospective antimicrobial audit and feedback were analyzed. Methodology: Between June and September 2014, applications for 16 target parenteral formulas (including ceftriaxone, ceftazidime, cefepime, piperacillin/tazobactam, vancomycin, teicoplanin, ertapenem, imipenem/cilastatin, meropenem, levofloxacin, moxifloxacin, ciprofloxacin, tigecycline, linezolid, daptomycin, and amikacin), which were not approved by infectious diseases (ID) specialists, were followed up until patients were either discharged or passed away. Results: Of the 292 cases studied, 193 (66%) were male, with a mean age (standard deviation) of 65.5 (19.3) years. There were five reasons for rejection, including dosage adjustments (37%), no evidence of bacterial infection (28.8%), modifications according to antimicrobial susceptibility (18.8%), target pathogens not being covered (7.2%), and redundant therapy (4.1%). Multiple logistic regression analysis demonstrated that an age greater than 75 years (odds ratio [OR]: 2.58; 95% confidence interval [CI]: 1.32–5.50; p = 0.005) was associated with significant mortality, while urinary tract (OR: 0.26; 95% CI: 0.09–0.70; p = 0.013) and soft tissue/bone infections (OR: 0.18; 95% CI: 0.05– 0.61; p = 0.006) were associated with survival. Adjustments according to ID physicians’ recommendations were not statistically significant (OR: 0.53; 95% CI: 0.27–1.06; p = 0.074). Conclusions: Antimicrobial adjustments according to ID physicians’ recommendations showed only marginally preventative effects against fatalities.


Introduction
Antibiotic resistance is an increasingly urgent global emergency [1][2][3].Powerful selective pressure from antibiotic overconsumption may be an important contribution to antibiotic resistance [1,3].For instance, Houvinen et al. [4] highlighted the development of lowlevel resistance in streptococci to erythromycin in Finland, because of the limits placed on new macrolide use.In the contrast, the superbug New Delhi Metallo-β lactamase 1-producing (NDM-1) Enterobacteriaceae emerged in communities where broad-spectrum antibiotics were freely sold over the counter [5].In medical institutions, alarmingly high rates of resistant organisms are prevalent in intensive care units [6], particularly among ventilator-dependent patients [7], since the burden of broad-spectrum antibiotic usage is heavy.At the individual level, patients with higher levels of antibiotic consumption show higher rates of acquiring resistant organisms [8].More specifically, in Taiwan, patients who acquired penicillin-resistant Streptococcus pneumoniae had a 15-day prior history of exposure to antibiotics [9].
In this regard, judicious use of antibiotics may be an important strategy to preserve the effectiveness of antimicrobial agents.The Infectious Diseases Society of America (IDSA) and the Society for Healthcare Epidemiology of America (SHEA) have put forward guidelines for the development and implementation of antibiotic stewardship programs [10,11].In late 2011, the Center for Disease Control, R.O.C. (Taiwan CDC) made a national commitment to the containment of antibiotic resistance, and a nationwide antimicrobial stewardship task force was consequently established [12].Two core strategies were recommended: formulary restriction/preauthorization for certain antibiotics, and prospective antimicrobial audits and feedback.Although the latter was endorsed by expert societies, previous researchers had not adequately demonstrated an impact on patient survival.In this study, the causes of fatalities associated with a prospective antimicrobial audit and feedback were analyzed.

Patient samples
Between June and September 2014, antibiotic use at Taoyuan General Hospital was reviewed.Taoyuan General Hospital is a 900-bed regional referral hospital in northern Taiwan.The formulary listed 64 parenteral and oral antibiotics in total, which were monitored monthly for their defined daily dose (DDD).From the antibiotic lists, 16 target parenteral formulae (ceftriaxone, ceftazidime, cefepime, piperacillin/tazobactam, vancomycin, teicoplanin, ertapenem, imipenem/cilastatin, meropenem, levofloxacin, moxifloxacin, ciprofloxacin, tigecycline, linezolid, daptomycin, and amikacin) were prospectively audited by three infectious diseases (ID) physicians.The above antibiotics were prescribed by patients' primary physicians and reviewed by ID specialists within 24 to 48 hours of being in the hospital information system, to verify their rationalities.Prescriptions that were not recommended by ID physicians were sent back, with suggestions noted on computers.Patients' outcomes were followed for up to 30 days, until patients were discharged or passed away.
The study was approved by the institutional review board of this hospital and informed consent was waived.

Statistical analyses
Demographic data are presented as mean ± standard deviation (SD) for continuous variables, and percentiles for discrete variables.Chi-square tests and student's ttest were used when feasible.Covariates with p < 0.2 in the univariate analyses were included in the multivariate logistic regression analyses to determine which covariates predicted fatalities.All statistical analyses were conducted using SAS version 9.3 (SAS Institute, Cary, USA).

Discussion
There are several commonly adopted strategies in antimicrobial stewardship: formulary restrictions, preauthorization, prospective antimicrobial audits with feedback, and antimicrobial cycling [11][12][13][14].Until now, it remains unclear which were the best interventions.It is likely that the effectiveness of each strategy varies because of different settings, sociocultural contexts, and the addition of various parts of care bundles into antibiotic stewardship programs [15].
In this study, 14.1% of prescriptions were not recommended by ID physicians, and among these, 19.1% of their recommendations were not accepted by primary physicians (data not shown); this is similar to a previous study in Taiwan in which the acceptance rate for ID physicians' suggestions was 80.6% [16].Prospective antimicrobial audits and feedback systems have also shown benefits such as cost-effectiveness [17], reductions in the inappropriate antibiotic prescriptions [10], decreases in antimicrobial consumption [18][19] and "bug-drug" mismatch [20].
Evaluating patients' outcomes in an antibiotic stewardship program is full of challenges.In previous research, a decrease in resistant pathogens, such as Clostridium difficile [18], extended-spectrum βlactamase-producing E. coli and K. pneumonia [19], and methicillin-resistant Staphylococcus aureus (MRSA) [21] was noted.Decreases in hospital stays [22] and improvements in successful treatments [20] have also been reported.
Unfortunately, there is a dearth of information about the impact on patient survival.Wang et al. [16] revealed that blood culture-guided, on-line antibiotic stewardship did not impact patient survival.Norwak et al. [23] showed that a prospective antimicrobial stewardship did not dampen patients' clinical outcomes.This current study also shows only marginally protective effects (OR: 0.53; 95% CI: 0.27 to 1.06; p = 0.07) on patient survival.There are a number of possible explanations for this.First, the hospitalized population was very old and often had common comorbidities; 14% of patients and their families had accepted hospice care, so the benefits of optimizing antibiotics may have waned due to the above factors.Second, the sample size used in this study was small, and the study was conducted for only a short period of time.
The limitations of this study also warrant discussion.For instance, this was a retrospective observational study, so the results were less powerful for its study framework.Furthermore, this study was conducted in a single referral hospital, and thus any generalizations should be made with caution.Finally, the survival rates of patients who received proper antibiotics from the beginning were not analyzed, so a comparison of the contribution of adequate and inadequate choices of antibiotics to patient survival was not possible.

Conclusions
Because of old age and the ceasing in some patients of aggressive treatments, antimicrobial adjustments according to ID physicians' recommendations showed only marginal effects in reducing fatalities.Under such circumstances, it should be more prudent to prescribe antibiotics.

Table 1 .
Demographic characteristics of patients who were given inadequate antibiotics in a regional hospital in northern Taiwan.
ICU: intensive care unit; ESRD: end-stage renal disease; HIV: human immunodeficiency virus

Table 2 .
Reasons and antibiotics not recommended by infectious diseases specialists in a regional hospital in northern Taiwan.

Table 3 .
Factors related to case fatalities using multiple logistic regression analyses adjusted by sex.