Patient Safety and Public Health Informatics

Patient Safety and Public Health Informatics Iona Thraen, ACSW Patient Safety Director Agenda  DATA  Past - designed specifically for patient safety purposes and can be easily changed  Present - designed for other’s purposes but patient safety can use – not easily changed  Future - not currently designed for patient safety purposes but could be created, redesigned and/or repurposed  The Message  Integrating a patient safety paradigm into public health surveillance strategies and perspectives provides:  multiple opportunities    to identify the scope to improve the system to assure patient safety The Past Data designed for patient safety reporting Voluntary  - Self Report of Sentinel Events The 2000 IOM report methodology (retrospective chart review) estimates at the lower end a death rate due to adverse events of 1.3/1000 admissions Utah had 268,652 hospital discharges in 2005   An estimated 350 deaths (most conservative) would be due to adverse events Average # SE reported since 2001 has been between 30-40/year – a tenfold under-reporting   New rule revisions by users group expanded 8 general categories to 32 specific categories to be consistent with NQF, CMS, and JCAHO Sentinel Events Hospitals/Ambulatory Surgical Centers Sentinel Events 10/15/2001-12/31/2006 60 50 40 30 20 10 0 37 7 2001 2002 48 39 46 36 2003 2004 2005 2006 Total number reported by year Number of Occurances Q 0 1 2 3 4 5 6 Utah Wrong Site Surgeries 2002-Q2/2007 Utah Wrong Site Surgery Initiative C³ Time Periods 420 Q 01 220 Q 02 42 Q 002 220 Q 03 420 Q 03 220 Q 04 420 Q 04 220 Q 05 42 Q 005 220 Q 06 420 Q 06 220 07 Baseline INTV The Present Data designed for other’s purposes that patient safety can use  Hospital Discharge Data – Administrative claims data     Misadventures Adverse Events Adverse Drug Events AHRQ-Patient Safety Indicators     Medical Examiners prescription based overdoses Perinatal clinical chart reviews Vital Records (Death and Birth certificates – ICD 10) Other UDOH registries currently in place Misadventures Rate of Misadventures per 100 Inpatient Discharges in Utah Acute Care Hospitals, 1999-2004 Data Notes Adverse event ICD-9-CM codes can be in any of up to 9 reported diagnosis codes including ecode(s).ICD-9-CM codes: E870-E876, 998.2, 998.4, 998.7. Utah Adverse Event Classes, 2001 Version. Data Sources Utah Inpatient Hospital Discharge Data, Office of Health Care Statistics, Utah Department of Health; Adverse Events – Inpatient Hospitalizations Adverse Drug Events – Inpatient Hospitalizations AHRQ - PSIs AHRQ-PSIs Adverse Drug Events Prescription Drug Overdoses 300 250 200 150 100 50 0 illicit_only not_illicit both 19 91 19 93 19 95 19 97 19 99 20 01 20 03 20 05 Prescription medication overdose deaths* by implicated medication Utah 1997-2005 140 120 100 80 60 40 20 0 1997 1998 1999 2000 2001 2002 2003 2004 2005 Methadone Hydrocodone Oxycodone Fentanyl The Future Data systems in need of creation, redesign or repurposing       Health Associated Infections (CLA-BSI, Influenza vaccination rates) ME/DOPL controlled substance DB – prescriptive patterns Peri-natal mortality chart review (extended to morbidity, use of IHI triggers, etc.) Clinical informatics (pharmacy, episodes of care, labs, etc.) BRFSS – Behavior Risk Factor Surveillance System Other UDOH registries – AHRQ Registries for Evaluating Patient Outcomes (Ch 9 – Detecting AEs)  Traumatic Brain Injury  EMS pre hospital data base  MDS/OASIS  Vital records – death certificates (ICD-10) Health Associated Infections (CLA-BSI, Influenza vaccination rates) User group formation  CDC definitions – ICU  Voluntary WEB reporting  Terminology reconciliation (JCAHO/CDC)  Immunization (hospitals/nursing homes)  ME/DOPL controlled substance DB – prescriptive patterns Repurposing from a criminal justice approach to a public health intervention  Definition of alerts  Algorithm development  Pattern recognition  Peri-natal mortality chart review (extended to morbidity, use of IHI triggers, etc.) Chart review database  Mortality and Morbidity traditional review process  IHI triggers and chart review data  Intervention development  Clinical Informatics - RHIOs  Pharmacy data  Selected medications  Number of prescriptions  Mixture of prescriptions  Dosages  Other  Laboratory data  Outliers  Ranges  Timing BRFSS – Behavior Risk Factor Surveillance System Construction of controlled substance use questions - process  Question examples  Testing of questions  Process of implementation  Data capture – establishing a baseline  Adverse Event Detection, Processing and Reporting using Registries Establish a process with an oversight group  Define scope of detections   Device related  Medication related  Process related  Procedure related  Other Adverse Event Detection, Processing and Reporting using Registries  Formalized and systematized definitions and applicability to source of data  Clinical  chart information and clinical expert opinion  Existing codified tools (ICD 9, ICD 10) Predictive values  Triggers  Others  Validation and verification Adverse Event Detection, Processing and Reporting using Registries Specification of protections  Use of information   Public disclosure  Trend analysis  Intervention  Other