Mobile Communications

Mobile Communications

Speakers Brett Moran, MD, BCIM, BCCI Associate Chief Medical Officer and CMIO About me Former Professor of Internal Medicine where he practiced academic medicine at UTSW for 19 years Deeply involved in quality and safety projects relating to improving chronic disease management and has spoken at the OIG Quality Round Table on the topic of clinical informatics Recent areas of focus have included addressing toxic In Baskets and identifying and reducing copy & paste

Speakers Judy Herrington, MSN, RN Vice President of Nursing Medicine Services for Parkland About me Over thirty years as a RN with 25+ years in an expanded leadership capacity Employed at Parkland for the past 28 years working in a variety of roles supporting nursing practice. Member of the original Epic build team for the transition to the EMR and have supported informatics on a variety of committees, taskforces and projects since Masters of Science in Nursing Administration from the University of Texas in Tyler, Tyler Texas

Speakers Joseph Longo, MBA, CHCIO Vice President of IT Enterprise Technologies for Parkland About me Joe has held positions on the vendor, owner, and consulting side of the healthcare IT business. He has been the steward of IT strategic plans and technology direction for major hospital systems and IT firms Obtained his Masters of Business Administration from Baylor University in Waco, Texas

Architecture

Mobile Communication Devices

BYOD Pros More cost effective Customers more satisfied Less risk of property loss Works better for people who work outside of the hospital Cons Heterogeneous Some incompatibilities End user install issues More risk of conflicting apps, OS

Enterprise Mobile Alert Details Alert Type Patient Name *****,******** MRN Date of Birth Admit date Diagnosis Timestamp Room:##### Code status Isolation status

Adoption Strategy Pilot 10W & 5N New Parkland (Go Live May 19, 2015) Secure Text Messaging Nurse Call Alerts Patient Monitoring Alarms Stat Order Notification Critical Lab Notifications Abnormal Report Availability Bed Management Asset Tracking Infant Security Alerts Patient Experience Manually Generated Operator Alerts Fire Alarm Notifications PRN Medication Effectiveness Reminder

Applications of Mobile Alert and Alarms Nursing Nurse Call secondary Alerts Patient communication Patient Monitor/Tele Alarms PRN Medication Effectiveness Reminder (Pain reassessment after pain meds) Code Blue for babies Infant Security Alerts In Development Critical Results out of Radiology (McKesson) Enhanced Sepsis warnings via predictive analytics Out of ICU Leveraging Parkland Data and ML to generate real-time messages to the appropriate care givers Coordinated Care Teams Secure messaging with patient context STAT orders Abnormal Results Bed Management (Environmental Services and ADT bed availability) Patient Flow Anesthesia

Nurse Call Alerts Patient calls from pillow speakers, pull cords, code or staff assist buttons, or staff terminal buttons Audible annunciation indicates the specific need Rules-based escalation until need is addressed Able to remotely talk directly to patient from the mobile Removes the middleman Reduces noise pollution Nursing Use Case: Nurse Call/Patient Communication

Nursing Use Case: Nurse Call/Patient Communication

Nursing Use Case: Patient Monitor Alarms Patient Monitor Alarms Bedside Philips monitors Wireless Telemetry Sample statistics of usage* >24,000 total alarms delivered through the system per day >12,000 unique alarms per day *based upon 3/31/2017 data

Nursing Use Case: Pain Reassessment Automated, asynchronous alert sent to the mobile device on the nurse assigned to a patient Nurse receives a reminder 1 hour after giving a prn pain medication to assess the effectiveness of the intervention Required by TJC and CMS Great patient satisfier The right thing to do ~30,000 opportunities per month

Nursing Use Case: Baby Blue Baby Blue 200-500 alarms per week Automated alert sent to all members of the NICU when code button is pressed in a room Initiates a specialized response TotGuard Alerts route to all hospital mobile devices as an immediate alarm effecting rapid action

Nursing Use Case: Baby Blue 1600.00 Pedi 1400.00 1200.00 1000.00 800.00 Pedi 600.00 400.00 200.00 0.00 September 2015 September 2016 September 2017

Secure, encrypted, HIPAA compliant Team members can be found by selecting the patient name SMS is preferred by many nurses and providers due to busy nature of patient care Excellent audit trail capabilities Rules-based escalation pathways for alerts/patient calls/alarms Coordinated Care Teams: Secure Care Team Communication

Coordinated Care Teams: STAT Orders and Critical Lab Results STAT Respiratory Therapy orders STAT Med Orders Critical Lab Results Routed based upon patient s care team in Epic Enhances communication Real-time exchange of important messages that are time-sensitive

Coordinated Care Teams: Patient Flow Automated Admission notifications to: The receiving units, when bed is assigned in BAM/Registration The sending units (30 minutes later, automated message sent to sending unit to say the bed is available for transfer) Patient transport messages are initiated from a staff terminal within the patient s room: When the transporter is in the room to pick up a patient When the transporter returns with the patient Button generates the following notifications: Level 1: An instant notification to the RN and PCA assigned to the patient Level 2: Two minutes after the 1 st notification, if not accepted or declined, a second notification is sent to the Charge Nurse

Transport TAT

Improved Care Through More Accurate Attribution Markedly improved ability to know who is caring for the patient at any given moment in time

Attribution Tracking

Coordinated Care Teams: Anesthesia Anesthesia: Reassessment orders Epidural removals

Evolving Functionality: Critical Results Collaboration between 3 vendors in innovating to create solutions where few are present Leveraging Luminary status with both relationships to transform offerings Semi-automated critical result notification for radiology studies Process whereby radiologists create critical result notification and the PACS vendor relays that to the EMR, which passes along to the mobile communication vendor

Next Steps: Sepsis Predictive Alerts Data model aggregates multiple patient specific variables in real time to estimate the risk of a patient being classified as septic Labs, Vitals, Orders Predictive Analytics Engine Empirically Tested Statistical Model Sample Risk Score Component Variables Vitals Arterial PCO2 INR Platelet Count WBC Lactate And several other lab related variables Sepsis Risk Score

Treatment Timeline: Sample Patient 8:12 PM patient arrives in the Parkland ED 8:28 PM labs ordered 8:33 PM labs drawn 9:14 PM subset of labs result 9:25 PM Sepsis alert model fires 9:27 PM alert sent to nurse, provider receives BPA 9:37 PM BPA accepted, order set submitted 9:38 PM Sepsis activation alert to ED Nurse, ED phlebotomy, ED Pharmacy 10:08 PM patient receives first dose of broad spectrum antibiotics 11:27 PM Order placed for hospital admission 1:28 AM Patient transferred to floor

First Year Results: Process Tracking

PHHS Sepsis POA Performance: 3 Hours Sepsis Cases Lactate Within 3HR IV Abx Within 3HR Culture Before Abx FY13 (Pre) 1445 54.6% 27.1% 84.3% 6/2-7/17 (Post) 120 64.2% 50.0% 80.8% 8/16 Relative Improvement 214 63% 74.1 49 +18% +85% -4%

Parkland Sepsis POA Performance: Combined Bundle Metric FY13 (Pre) 1445 Sepsis Patients 14.0% Bundle Compliance 6/2-7/17 (Post) 120 Sepsis Patients 29.2% Bundle Compliance FY16 2145 Sepsis Patients 31.6% Bundle Compliance >2x relative improvement Maintenance and continued improvement

Parkland Sepsis POA Performance: Length of Stay Early results suggested: Average and median length of stay reductions for Sepsis POA patients FY13 (Pre) 6/2-7/31 (Post) Absolute Reduction Relative Reduction 1445 Sepsis Patients 263 Sepsis Patients Mean LOS: 10.36 days Mean LOS: 8.13-2.23 days -21.5% Median LOS: 6.61 days Median LOS: 5.78-0.83 days -12.6%

Parkland Sepsis Performance: Mortality Sustained results suggest: Significant mortality improvement No length of stay reductions for Sepsis patients FY13 (Pre) FY16 (Post) Absolute Reduction Relative Reduction 1445 Sepsis Patients 2927 Sepsis Patients Mortality: 10.8% Mortality: 8.9% ~56 lives -17.7% Median LOS: 7.25 days Median LOS: 7.41 +2.2%

Closing Remarks