Data marts Intensive care experience Vitaly Herasevich MD
- Slides: 47
Data marts. Intensive care experience Vitaly Herasevich, MD, Ph. D, MSc Assistant Professor of Medicine and Anesthesiology, Department of Anesthesiology, Multidisciplinary Epidemiology and Translational Research in Intensive Care (M. E. T. R. I. C. ) Jun 2012 herasevich. vitaly@mayo. edu
Why we need Datamart?
April 2011 56 Hospitals worldwide Including Mayo Clinic
http: //mayoweb. mayo. edu/it-operations/supported-systems. html © 2011 MFMER | slide-4
EHR
Data volume before and in ICU Microbiology, labs, medications, chest X-ray, Nurses flowsheet, Clinical notes (history and impression/plan) – Vitals excluded
“Datapoints” in acute setting • • • Ventilator monitoring Brain function monitoring Fi. O 2 PIP PEEP/CPAP Mean Airway Pressure Tidal Volume • Electroencephalography • Intracranial pressure Invasive hemodynamic monitoring • • • Laboratory blood • Hemoglobin • Serum electrolytes • Blood chemistry • • • Labs Drug Orders Natural contexts Demographic data Microbiology Chronic diseases history X ray Allergies Stress Vitals Pain Average data points per day Per Patient Per 24 bedded ICU 60 1440 10 240 2 48 1950 46800 Central venous pressure Arterial blood gases and p. H Pulmonary arterial pressure Oxygen transport variables Intra-arterial blood pressure Routine noninvasive monitoring • • • EKG Arterial blood pressure Heart rate Respiratory rate Temperature Routine cardiac monitoring • • Cardiac output Hemodynamic variables Blood volume Colloidal osmotic pressure Fluid balance • Fluid IN • Fluid OUT • Urine output Tissue perfusion / oxygenation monitoring • Pulse oximetry • Transcutaneous oxygen and carbon dioxide monitoring
ICU Datamart (METRIC Datamart)
CPOE ICU demographics Enterprise orders HRBS Nursing Flow Sheet Monitored data MICS Lastword Chart+ Emergency acute area SQL Clinical notes MCLS Lastword YES ed p stor Key Facts HL 7 Radiology Reports Fluids: in/out Chart+ ~ 15, 000 admissions per year • ~ 1, 000 vital recordsresper week ICD-9 Transfusion Orders u ed DSS • Data available from 2003 MYSIS c o r p d e Updated every hour for vitals)Reports r in average (15 min Microbiology Past • history to RIMS ures d roce • PPI • Near real-time SQ • ED APACHE L s HRBS critical area Surgical schedule APACHE Surgical Historical Drug orders REP HRBS OR Data mart ICU Data mart Labs HRBS
Anesthesia Datamart (OR Datamart)
Anesthesia locations included in OR Data mart
© 2011 MFMER | slide-14
Available data (Apr 2012) Near real time (5 min delay) 1998 …. . 2012 Yearly average Monthly average Total 1998 -2011 Demographics 750, 000 65, 000 5, 400, 000 Events 20, 000 1, 600, 000 144, 000 Procedures 300, 000 26, 000 2, 500, 000 Fluids 7, 900, 000 650, 000 87, 000 OR Medications 5, 000 400, 000 50, 000 Vital signs 190, 000 16, 000 1, 600, 000 Documents 3, 200, 000 250, 000 27, 300, 000 Procedures 330, 000 28, 000 2, 500, 000 Teams 620, 000 50, 000 4, 600, 000
Approach
Rule one: lego bricks
Rule two: UNIX- no user interface • No formal web/query Interface • ODBC connection allows query from any app (JMP, Excel, SAS…)
Rule three: raw data
Approach: technically • SQL server with institutional support • Tables divided by years • In “Current tables” only patients who in currently in anesthesia location or ICU • EAV (entity – attribute – value) structure • Continuously “Testing – production” • Test –> production DBs
Data integrity Currently: no institutional 24/7 support Statistical control Real time monitoring
Herasevich V, Pickering BW, Dong Y, et al. Informatics infrastructure for syndrome surveillance, decision support, reporting, and modeling of critical illness. Mayo Clin Proc 2010; 85(3): 247 -254. (PMID: 20194152) Herasevich V, Kor D, Li M, et al. ICU Data Mart: A Non-IT Approach. Healthcare Informatics 2011; 28(11): 42 -45. (PMID: pending)
Areas of implementation
APACHE replacement
Reports 1994 - 2009
APACHE replacement project
Free text search for medical admission diagnoses
Clinical reports
Effective management Joint Commission on Healthcare Organizations (JCAHO) measurement of ICU performance. • • • Mortality report Length of Stay Review ICU Death Review ICU admission Low Risk Monitor Review ICU Readmission Review
METRIC Reports 1. Hospital Length of Stay for ICU Graduates – Unadjusted 2. ICU Length of Stay – Unadjusted 3. ICU Length of Stay – Adjusted 4. ICU Readmission Rate 5. ICU Admissions 6. ICU Admission Source and Service 7. Duration of Mechanical Ventilation 8. ICU Mortality Rate – Unadjusted 9. Hospital Mortality Rate – Adjusted 10. ICU Admissions for Low-Risk Monitoring 11. ICU Census - Hourly Utilization • Monthly reports • Ad-hock reports • Customized reports
Reports
Dashboards
Value of this data?
1985 2010 © 2011 MFMER | slide-34
Dashboards The Stability and Workload Index for Transfer (SWIFT score) Daily APACHE
Dashboards: administrative and clinical E L X E P M A © 2011 MFMER | slide-36
Name: IHD CCF Hypertension Surgery-Stent Age: Source: HR MAP ST ∆ 118 SR 65 No Fluid Bolus NE a. VP 4 L 0. 80 0. 03 EKG ECHO Troponin HB Blood Loss Hct X Match IV Access Normal N/A 0. 02 10 g/d. L 0 ml 0. 27 Done 2 x 20 G Syndrome Advisory Possible Sepsis MAP HR T WBC <65 >90 >38. 4 >12 Treatment Goals MAP Sv. O 2 AB >65 >70 <1 hr Status Obese COPD Smoker Alcoholism RR 28 Sp. O 2 92 Pa. O 2/Fi. O 2 200 O 2 8. 0 ETT IMV 0. 6 Easy VC To C WBC 38. 9 15 Anti Microbial AB admin Cultures Taken Source Control CXR p. H p. CO 2 PO 2 HCO 3 Date 03. 08 Infiltrate 7. 28 55 80 24 Source Blood BAL Urine CVP Lactate Sv. O 2 AB Given X-Match Organism Pending 10 3. 0 68 Yes Done CVS Respiratory Renal Yes No AWARE CNS Hematologic GIT Sepsis
Sniffers
Sniffers – rule based DSS
Notable sniffers ALI VILI Herasevich V, Yilmaz M, Khan H, et al. Validation of an electronic surveillance system for acute lung injury. Intensive Care Med 2009; 35(6): 10181023. (PMID: 19280175) Herasevich V, Tsapenko M, Kojicic M, et al. Limiting ventilator-induced lung injury through individual electronic medical record surveillance. Crit Care Med 2011; 39(1): 3439. (PMID: 20959788) Septic Shock Herasevich V, Pieper MS, Pulido J, et al. Enrollment into a time sensitive clinical study in the critical care setting: results from computerized septic shock sniffer implementation. J Am Med Inform Assoc 2011. (PMID: 21508415)
Data retrieval for research
Olmsted county o c d e Olmst ssion admi METRIC datamart
Clinical studies • Enrollment to time sensitive trials • Retrospective studies for Quality Improvement an research
Future
Facilitate 3 distinct data use scenarios • Point of care novel user interfaces, alerts and decision supports • Reporting • Research
Team/contacts Department Chair Bradly J. Narr, MD Clinical leadership Daryl Kor, MD Medical informatics leadership Vitaly Herasevich, MD, Ph. D 2 FTE OR datamart programming Nageswar Madde Data requests Rahul Kashyap, MBBS Clinical Expertise Arun Subramanian, MBBS Data integrity Greg Wilson, RRT Vision/Mentoring Ognjen Gajic, MD SAS programming Andrew Hanson ICU/OR datamart programming lead Man Li, MD Clinical Expertise Brian Pickering, MD
When the machines that men invented over time would now even function: what pleasant life this would be! Kurt Tucholsky, German writer and satirist, 1890 -1935 http: //mayoresearch. mayo. edu/mayo/research/clinical-informatics-in-intensive-care/ herasevich. vitaly@mayo. edu © 2011 MFMER | slide-47
- Barth
- Vitaly attack
- Vitaly feldman
- Sap rapid marts
- Characteristics of data mart
- Eim architecture
- Primary secondary and tertiary health care
- Imprinting meaning psychology
- Early experience vs. later experience
- What is indirect experience
- Icu unit meaning
- Intensive care units
- Data intensive workloads
- Ogsa in cloud computing
- Health and social care level 3 unit 6
- Patient care experience
- Raw data of experience
- Duty of care outcome care certificate
- Magnetul este corpul care are proprietatea de a atrage
- Palliative care versus hospice care
- Prin ce se înmulțesc mamiferele
- Care sunt simturile prin care sunt evocate
- Care certificate standard 14 answers
- Hip fracture care clinical care standard
- Unit 2 equality diversity and rights
- Reflective and intensive pronouns
- Intensive and extensive properties
- Specialty channels of distribution
- Testing listening
- Intensive phase of tb treatment
- Reflexive vs intensive pronouns
- Kinds of reading skills
- Intensive properties examples
- What is intensive distribution
- Politiques de distribution
- Intensive verbs
- What is an intensive pronoun
- Difference between reflexive and intensive
- Indiana university intensive english program
- Sedum roof lambeth
- Extensive reading
- Croissance intensive
- Manufacturer sponsored retailer franchise system
- Intensive listening
- Style of reading
- Examples of responsive speaking
- Extensive examples
- How to find mass with density and volume