Post Operative Nausea and Vomiting (PONV) Rescue Outcomes After Amisulpride Treatment

Deidentified Electronic Health Record (EHR) information related to treatment of Post Operative Nausea and Vomiting (PONV) will be collected from academic and community hospitals throughout the United States. Deidentified EHRs for patients receiving treatment for PONV will be harvested in two cohorts:

Cohort 1, Rescue treatment according to the standard of care (SOC) at each study site prior to the introduction of IV amisulpride into clinical practice

Cohort 2, Rescue treatment IV amisulpride.

There is no planned investigational agent, prescribed treatment regimen, or mandated intervention, and treating clinicians will determine therapy for PONV according to clinical judgment. The projected inclusion over the course of this observational study is to enroll a total of 20,000+ patient EHRs evenly divided among those that received SOC and amisulpride treatment. Effectiveness and safety outcomes will be analyzed using standard statistical methods for observational studies. Comparison of deidentified patient EHRs in Cohort 1 and Cohort 2, including response rates, will be compiled and analyzed by an independent medical data analytics company.

In regard to a Quality Assurance (QA) Plan that addresses data validation and registry procedures, including any plans for site monitoring and auditing, ArborMetrix performs QA as part of regular business. QA involves both automation testing and manual testing strategies including Failure Mode and Effects Analysis (FMEA), bounds testing and exploratory testing. Monitoring and auditing is done in line with their continued work toward HITRUST Compliance.

Regarding data checks to compare data entered into the registry against predefined rules for range or consistency with other data fields in the registry, data validation is built into the registry configuration based on predefined rules as part of the data analytics and measures development. When receiving data in a batch, legal values, legal relationships, primary keys, constraints, duplicates, range, format, length, data type, parent-child relationship, variable name are all checked at both import and staging. The analytic process checks naming, ticket #, unit test (Statistical Analysis System (SAS), Structure Query Language (SQL)), integration, rules, range, consistency, outliers, missing and distribution. As part of the Risk and Reliability (RR) Adjustment, ArborMetrix checks missing, unique values, Observed to Expected (OE) ratio, C-stats, R squared and adjusted R squared. Report validation in the UI (User Interface), which in this case means the registry platform, compares reporting dictionary to analytic definition.

Regarding source data verification and assessing the accuracy, completeness, or representativeness of registry data, data for PROMPT is being pulled directly from the EMR system (i.e., external data sources) into the registry. No source data verification will be conducted.

Regarding Standard Operating Procedures to address registry operations and analysis activities, such as patient recruitment, data collection, data management, data analysis, reporting for adverse events, and change management - there is no patient recruitment for PROMPT. It is an observational study. Information on data collection/management/analysis is answered in other questions. The change management system involves a Change Management Board, Change Management Request form outlining event spurring the change, requested date/time, engineers assigned, devices and/or systems being changed, a description of the change including technical details, impact, risk validation method and backout plan. The Request is peer reviewed and approved by leadership and technical personnel prior to implementation.

Plan for missing data to address situations where variables are reported as missing, unavailable, non-reported, uninterpretable, or considered missing because of data inconsistency or out-of-range results - As part of the data integration process, if variables necessary to meet measure criteria are deemed not available, Acacia will determine the next steps (i.e., remove measure, change measure criteria, etc.).