Background Incident reports suggested that patients werereaching the operating room (OR) without completion of nec-essary preparatory tasks. Incidents included Near Misses withpotential for harm. Parallel processing and inadequate commu-nication among preoperative nurses, anesthesia providers, andOR nurses were determined to be root causes of many ofthese failures.Objectives Significantly increase the number of days-betweenpreoperative task-completion (PTC) failures.Methods Incident reports were analyzed for root causes. Pre-operative process flow was mapped. Process flow at otherinstitutions was observed. Multidisciplinary bedside handoffsutilizing a task-completion checklist were tested, adapted, andadopted as a new preoperative process (figure 1). Days-between PTC failures were plotted on an XMR chart as theprimary metric. First case procedure start times were plottedon XMR charts as a balancing measure. Qualitative data werecollected about process issues identified by the handoffprocess.Results After introduction of bedside handoffs days-betweenPTC failures reaching the OR increased from every 5 daysto >40 days (figure 2). The average procedure start time wasdelayed by 8 min (figure 3). A majority of PTC failures thatwere stopped from reaching the OR were surgeon-specific

(figure 4A). Unavailability of nurses was a reported barrier toprocess success. (figure 4B).Conclusions Bedside handoffs inclusive of preoperative nurses,anesthesia providers, and OR nurses increased the daysbetween PTC incidents reaching the OR. This safety interven-tion had the tradeoff of a slight decrease in efficiency asmeasured by procedure start times. Interventions targetingnurse availability and earlier surgeon task completion are stillnecessary to optimize efficiency.

IHI ID 17 IMPROVING TIMELY RECOGNITION OF SEPSISIN PEDIATRIC INTENSIVE CARE UNIT PATIENTS

Rhea Vidrine, Matthew Zackoff, Stephen Pfeiffer, Zachary Paff, Brandy Seger,Jessica Walden, Carrie Schnieder, Taryn Stumpf, Cecilia Smith, Erika Stalets, Maya Dewan.Cincinnati Childrens, USA

10.1136/ihisciabs.17

Background Sepsis is a leading cause of pediatric mortality.Prior research shows that patients who receive antibioticswithin 6 hours of sepsis recognition have decreased in-hospitalmortality. While we had demonstrated improvements in recog-nition of sepsis for newly admitted patients, delayed

Abstract IHI ID 16 Figure 3 First-case start timesIndividuals (XMR) chart depicting the average start time for first cases in the operating room. The chart is annotated for important time points in thestudy. Special cause is illustrated by points / connectors in red and by points above the upper control limit. The shift upward of the centerline afterspecial cause was met in the upper chart illustrates the average start time becoming significantly later after ‘Go-Live’ of the new handoff process.The widening control limits illustrate increased variation in the start times after introducing the new process. Dashed red line = upper control limits(UCL) and lower control limits (LCL); Light blue line = centerline depicting the mean for each value

Abstract IHI ID 16 Figure 4 Pareto charts of issues identified by qualitative bedside handoff dataPanel A: Tasks requiring completion that were caught by bedside handoffs prior to transferring patient to the operating room. Green bars representcategories requiring surgeon presence to completePanel B: Issues that interfered with or needed completion at the time of bedside handoffs. Red bar and orange bars identify the issues with thehighest count. Yellow bars represent categories related to nurse availability

Abstracts

BMJ Open Quality 2018;7(Suppl 1):A1–A36 A21

on August 8, 2020 by guest. P

rotected by copyright.http://bm

jopenquality.bmj.com

/B

MJ O

pen Qual: first published as 10.1136/ihisciabs.17 on 1 D

ecember 2018. D

ownloaded from

Abstract IHI ID 17 Figure 1 Clinical decision support sepsis identification tool. This tool flags high-risk pediatric patients enabling a nursingassessment and prompt calling of a sepsis huddle if applicable

Abstract IHI ID 17 Figure 3 T-chart outlining

Abstract IHI ID 17 Figure 2 Pathway from prompt recognition via clinical decision support tool or clinical assessment to initiation of sepsistreatment

Abstracts

A22 BMJ Open Quality 2018;7(Suppl 1):A1–A36

on August 8, 2020 by guest. P

rotected by copyright.http://bm

jopenquality.bmj.com

/B

MJ O

pen Qual: first published as 10.1136/ihisciabs.17 on 1 D

ecember 2018. D

ownloaded from

recognition of sepsis in the pediatric ICU (PICU) remains asignificant vulnerability.Objectives To increase and sustain the days between delayedsevere sepsis recognition in the PICU by 50% within twoyears.Methods Using the Improving Pediatric Sepsis Outcomes data-base from January 1, 2016 to July 31, 2018 we identifiedepisodes of delayed recognition of severe sepsis in patientsadmitted to the PICU. We define delayed severe sepsis recog-nition as time to antibiotics 6 hours after obtaining blood cul-tures. We utilized several PDSA cycles to improve the sepsisidentification process within our unit including use of multi-disciplinary sepsis huddles for high risk patients and an auto-mated clinical decision support (CDS) tool (figure 1 and 2).The primary outcome of delayed sepsis recognition wastracked using a days-between t-chart.Results After implementation of sepsis huddles in May 2017,the days between episodes of delayed severe sepsis recogni-tion improved from our baseline of 9 days to 28 days. Sincethe implementation of our automated CDS sepsis screeningtool in May 2018, we have shown sustained improvement(figure 3).Conclusions Implementation of a sepsis huddles and anautomated CDS tool in the PICU has led to an improve-ment in the days between cases of delayed severe sepsisrecognition.

IHI ID 18 DEVELOPMENT OF AN EMS PROTOCOL TOSYSTEMATICALLY REDUCE PREVENTABLEAMBULANCE RESPONSES

Allison Infinger, Don Robinson, Patricia Dowbiggin, Jon Studnek. Mecklenburg EMS Agency,USA

10.1136/ihisciabs.18

Background Requests for Emergency Medical Services (EMS)are generated by patients, caregivers, bystanders, or alliedagencies. It was theorized that many requests initiated byallied agencies were preventable, resulting in response cancella-tions, patient refusals, or transport for non-life threateningconditions.Objectives The objective of this study was to reduce prevent-able responses to requests initiated by allied agencies by37%.Methods This study was conducted in Charlotte, NC from1/27/2017 to 8/7/2018 with a single EMS agency and a firstresponder service who responds to most EMS requests. Adescriptive analysis identified the number of requests initi-ated by allied agencies and categorized the outcome of theresponse (patient transport, refusal, or cancellation). Adriver diagram and PDSA ramps identified a promisingchange concept: delayed EMS dispatch to eligible requestsfor service (figure 1). Exclusion criteria were developed

Abstract IHI ID 18 Figure 1

Abstracts

BMJ Open Quality 2018;7(Suppl 1):A1–A36 A23

on August 8, 2020 by guest. P

rotected by copyright.http://bm

jopenquality.bmj.com

/B

MJ O

pen Qual: first published as 10.1136/ihisciabs.17 on 1 D

ecember 2018. D

ownloaded from