The Analytic Information Warehouse (AIW): a platform for analytics using electronic health record data.

OBJECTIVE: To create an analytics platform for specifying and detecting clinical phenotypes and other derived variables in electronic health record (EHR) data for quality improvement investigations. MATERIALS AND METHODS: We have developed an architecture for an Analytic Information Warehouse (AIW). It supports transforming data represented in different physical schemas into a common data model, specifying derived variables in terms of the common model to enable their reuse, computing derived variables while enforcing invariants and ensuring correctness and consistency of data transformations, long-term curation of derived data, and export of derived data into standard analysis tools. It includes software that implements these features and a computing environment that enables secure high-performance access to and processing of large datasets extracted from EHRs. RESULTS: We have implemented and deployed the architecture in production locally. The software is available as open source. We have used it as part of hospital operations in a project to reduce rates of hospital readmission within 30days. The project examined the association of over 100 derived variables representing disease and co-morbidity phenotypes with readmissions in 5years of data from our institution's clinical data warehouse and the UHC Clinical Database (CDB). The CDB contains administrative data from over 200 hospitals that are in academic medical centers or affiliated with such centers. DISCUSSION AND CONCLUSION: A widely available platform for managing and detecting phenotypes in EHR data could accelerate the use of such data in quality improvement and comparative effectiveness studies.

A two-pronged quality improvement training program for leaders and frontline staff.

BACKGROUND: A unique two-pronged QI training program was developed at Emory Healthcare (Atlanta), which encompasses five hospitals and a multispecialty physician practice. One two-day program, Leadership for Healthcare Improvement, is offered to leadership, and a four-month program, Practical Methods for Healthcare Improvement, is offered to frontline staff and middle managers. KNOWLEDGE ASSESSMENT: Participants in the leadership program completed self-assessments of QI competencies and pre- and postcourse QI knowledge tests. Semistructured interviews with selected participants in the practical methods program were performed to assess QI project sustainability and short-term outcomes. RESULTS: More than 600 employees completed one of the training programs in 2008 and 2009. Leadership course participants significantly improved knowledge in all content areas, and self-assessments revealed high comfort levels with QI principles following the training. All practical methods participants were able to initiate and implement QI projects. Participants described significant challenges with team functionality, but a majority of the QI projects made progress toward achieving their aim statement goals. A review of completed projects shows that a significant number were sustained up to one year after program completion. Quality leaders continue to modify the program based on learner feedback and institutional goals. CONCLUSIONS: This initiative shows the feasibility of implementing a broad-based in-house QI training program for multidisciplinary staff across an integrated health system. Initial assessment shows knowledge improvements and successful QI project implementations, with many projects active up to one year following the courses.

Use of a mandatory declination form in a program for influenza vaccination of healthcare workers.

OBJECTIVE: To evaluate the utility and impact of using a declination form in the context of an influenza immunization program for healthcare workers. METHODS: A combined form for documentation of vaccination consent, medical contraindication(s) for vaccination, or vaccination declination was used during the 2006-2007 influenza season in a healthcare system employing approximately 9,200 nonphysician employees in 3 hospitals; a skilled nursing care facility; a large, multisite, faculty-practice plan; and an administrative building. Responses were entered into a database that contained files from human resources departments, which allowed correlation with job category and work location. RESULTS: The overall levels of influenza vaccination coverage of employees increased from 43% (3,892 of 9,050) during the 2005-2006 season to 66.5% (6,123 of 9,214) during the 2006-2007 season. Of 9,214 employees, 1,898 (20.6%) signed the declination statement. Among the occupation groups, nurses had the lowest rate of declining vaccination (13.2% [393 of 2,970]; P < .0001), followed by pharmacy personnel (18.1% [40 of 221]), ancillary personnel with frequent patient contact (21.9% [169 of 771), and all others (24.7% [1,296 of 5,252]). Among the employees who declined vaccination, nurses were the least likely to select the reasons "afraid of needles" (3.8% [15 of 393], vs. 9.1% [137 of 1,505] for all other groups; P < .001) and "fear of getting influenza from the vaccine" (13.5% [53 of 393], vs. 20.5% [309 of 1,505]; P = .002). Seven pregnant nurses had been advised by their obstetricians to avoid vaccination. When declination of influenza vaccination was analyzed by age, 16% of personnel (797 of 4,980) 50 years of age and older declined to be vaccinated, compared with 26% of personnel (1,101 of 4,234) younger than 50 years of age (P < .0001). CONCLUSIONS: Implementing use of the declination form during the 2006-2007 influenza season was one of several measures that led to a 55% increase in the acceptance of influenza vaccination by healthcare workers in our healthcare system. Although we cannot determine to what degree use of the declination form contributed to the increased rate of vaccination, use of this form helped the vaccination program assess the reasons for declination and will help to focus future vaccination campaigns.

Leveraging derived data elements in data analytic models for understanding and predicting hospital readmissions.

Hospital readmissions depend on numerous factors. Automated risk calculation using electronic health record (EHR) data could allow targeting care to prevent them. EHRs usually are incomplete with respect to data relevant to readmissions prediction. Lack of standard data representations in EHRs restricts generalizability of predictive models. We propose developing such models by first generating derived variables that characterize clinical phenotype. This reduces the number of variables, reduces noise, introduces clinical knowledge into model building, and abstracts away the underlying data representation, thus facilitating use of standard data mining algorithms. We combined this pre-processing step with a random forest algorithm to compute risk for readmission within 30 days for patients in ten disease categories. Results were promising for encounters that our algorithm assigned very high or very low risk. Assigning patients to either of these two risk groups could be of value to patient care teams aiming to prevent readmissions.

Impact of surgical care improvement project inf-9 on postoperative urinary tract infections: do exemptions interfere with quality patient care?

BACKGROUND: The Surgical Care Improvement Project (SCIP) Inf-9 guideline promotes removal of indwelling urinary catheters (IUCs) within 48 hours of surgery. OBJECTIVES: To determine whether a correlation exists between SCIP Inf-9 compliance and postoperative urinary tract infection (UTI) rates and whether an association exists between UTI rates and SCIP Inf-9 exemption status. DESIGN Retrospective case control study. SETTING: Southeastern academic medical center. PATIENTS: American College of Surgeons National Surgical Quality Improvement Program (NSQIP) and SCIP Inf-9 compliance data were collected prospectively on randomly selected general and vascular surgery inpatients. Monthly UTI rates and SCIP Inf-9 compliance scores were tested for correlation. Complete NSQIP data for all the inpatients with postoperative UTIs were compared with a group of 100 random controls to determine whether an association exists between UTI rates and SCIP Inf-9 exemption status. MAIN OUTCOME MEASURE: Postoperative UTI. RESULTS: In 2459 patients reviewed, SCIP Inf-9 compliance increased over time, but this was not correlated with improved monthly UTI rates. Sixty-one of the 69 UTIs (88.4%) were compliant with SCIP Inf-9; however, 49 (71.0%) of these were considered exempt from the guideline and, therefore, the IUC was not removed within 48 hours of surgery. Retrospective review of 100 random controls showed a similar compliance rate (84.0%, P = .43) but a lower rate of exemption (23.5%, P < .001). The odds of developing a postoperative UTI were 8 times higher in patients deemed exempt from SCIP Inf-9 (odds ratio [OR], 7.99; 95% CI, 3.85-16.61). After controlling for differences between the 2 groups, the adjusted ORs slightly increased (OR, 8.34; 95% CI, 3.70-18.76). CONCLUSIONS: Most UTIs occurred in patients deemed exempt from SCIP Inf-9. Although compliance rates remain high, practices are not actually improving. Surgical Care Improvement Project Inf-9 guidelines should be modified with fewer exemptions to facilitate earlier removal of IUCs.

Is the incidence of heparin-induced thrombocytopenia affected by the increased use of heparin for VTE prophylaxis?

BACKGROUND: The increased exposure to heparin products for thromboprophylaxis against VTE in hospitalized patients raises concerns for an increase in the incidence of heparin-induced thrombocytopenia(HIT). METHODS: We analyzed, among 90,875 patients exposed to heparin products between 2005 and 2009, the number of hematologic consultations for thrombocytopenia, requests for heparin induced antibodies by enzyme-linked immunosorbent assay, and cases given a diagnosis of HIT by the hematology consult service. RESULTS: We observed that despite a doubling in the number of patients receiving pharmacoprophylaxis with heparin, there was no significant increase in the number of consultations for thrombocytopenia,the number of requests for HIT tests, the number of positive HIT test results, or the number of HIT diagnoses. The number of cases of HIT was low and represented < 0.1% of patients exposed to heparin. CONCLUSIONS: We conclude that concerns about HIT should not be a limiting factor for the systematic implementation of heparin-based VTE prophylaxis.

A Web-based program to ensure compliance of medical staff providers with mandated health care facility requirements.

BACKGROUND: We designed a Web-based program for educating medical staff providers on infection control and prevention measures and for ensuring that such providers comply with mandated immunizations, influenza vaccination, and tuberculin skin testing. METHODS: Our program was designed in a health care system employing approximately 2,699 medical staff providers and 9,700 other health care workers in 3 hospitals; a skilled nursing care facility; and a large, multisite, faculty practice plan. We developed a program for educating medical staff providers on infection control and antibiotic stewardship and ensuring compliance with institutional immunization, tuberculin skin testing, and influenza immunization requirements. A post-test required a 100% correct score to pass. The modules allowed access from any computer with Internet connectivity. RESULTS: In total, 2,666 of the 2,699 (98.8%) medical staff providers completed the program, and 76.4% of the 2,666 received influenza vaccination. CONCLUSION: Entering the data into an electronic database allowed for ease in monitoring compliance, and the requirement for successfully passing a post-test following the educational modules ensured that medical staff providers had read the material. Entry into the electronic database also allowed for analysis of reasons for varied compliance with influenza immunization that will be useful in focusing future educational efforts.