Development, Validation and Real-World Application of Comprehensive Metrics to Improve Hospitals' Antibiotic Prescribing

点击翻译按钮获取中文摘要

Background: Antimicrobial resistance is an urgent public health problem and a threat to patient safety. Antimicrobial use is the strongest risk factor for the emergence of antimicrobial resistance By reducing unnecessary antimicrobial use and improving necessary use, antimicrobial stewardship programs (ASPs) are an important tool in the battle against antimicrobial resistance. ASPs are mandated both in and outside VHA. Significance: There is not a universally accepted metric to assess ASP success. A hospital-level metric, known as the Standardized Antimicrobial Administration Ratio (SAAR), is being widely implemented across US hospitals, including in VHA. However, the SAAR has not been validated and has several weaknesses, including lack of patient-level risk adjustment and no consideration for antimicrobial spectrum of activity. Innovation & Impact: This proposal will target a critical research gap by developing, refining and validating novel metrics that address the limitations of the SAAR. These novel metrics include: 1) a risk-standardized days of therapy (DOT) ratio (RSDTR), which differs from the SAAR in its adjustment for differences in patient case-mix; 2) a Days of Antimicrobial Spectrum Coverage (DASC) score, which differs from the SAAR in its capture of both antimicrobial use and spectrum of activity, and 3) a risk-standardized DASC. Through our aims, we will assess 3 types of validity (construct, content and predictive) for the SAAR and the 3 novel metrics. Specific Aims: A1: Identify non-modifiable factors that contribute to appropriate variation in antimicrobial use across VHA acute-care hospitals by evaluating potential patient-level, hospital-level, and environmental effects. H1: Non-modifiable factors can be identified by consensus-building exercises and mathematical techniques. A2: Evaluate whether hospitals that have implemented more robust stewardship processes, as measured by a mandatory VHA survey, perform better on the SAAR and the three novel metrics. H2: The association between the robustness of stewardship and a hospital's metric performance will be strongest for the risk-standardized DASC score. A3: Using qualitative methods, assess the content validity of the SAAR and the three novel metrics among key stakeholders across ten hospitals, including high and low-performing sites. H3: Key stakeholders across all sites will confirm the validity of the RSDTR and risk-standardized DASC and will have concerns about the validity of metrics that do not involve patient-level risk-adjustment. A4: Evaluate the longitudinal relationship between a hospital's performance on the four metrics and the emergence of healthcare-associated antimicrobial resistance. H4: The novel metrics will have a stronger association with future antimicrobial resistance than the SAAR. Methodology: Our proposal will leverage both quantitative and qualitative methods. In Aim 1, we will use a modified two-stage Delphi method, which will integrate data-driven and human-driven selection strategies, to identify non-modifiable risk factors that should be used to risk-standardize DASC and also DOT (e.g. the RSDTR metric). In Aim 2, we will use hospitals' responses to a mandatory survey to measure the association between the intensity of stewardship activities and metric performance (construct validity). In Aim 3, we will perform semi-structured interviews and focus groups with key stakeholders to assess the content validity of the four metrics. In Aim 4, we will leverage 17 years of VHA data to determine whether a hospital's performance on each metric is associated with the future development of antimicrobial resistance (predictive validity). Next Steps/Implementation: Any metric we validate would inform health policy and could be incorporated into our team's existing stewardship dashboard. Our findings could be used in a multicenter, audit-and-feedback trial that leverages the dashboard and the validated metric(s) to improve inpatient antimicrobial use.