Clinical Safety & Effectiveness Cohort # 8

Clinical Safety & Effectiveness Cohort # 8 1 IMPROVING THE TIMELINESS OF PARACENTESIS: IMPACT OF A PROCEDURE TEAM DATE Educating for Quality Improvement & Patient Safety

FINANCIAL DISCLOSURE Patricia Wathen, MD has no relevant financial relationships with commercial interests to disclose. Deborah Kendall-Gallagher, RN, JD, PhD has no relevant financial relationships with commercial interests to disclose.

The Team CSE Participants: Patricia I. Wathen MD School of Medicine, Dept of Medicine Deborah Kendall Gallagher RN, JD, PHD School of Nursing Team Members: Phoebe King MD Chief Resident, Internal Medicine Megan Freeman MD Internal Medicine Resident John Vizuete MS IV Medical Student Jamie Kohn Research Specialist Special thanks to David Dooley MD, Michelle Ryerson, John Walker, Amjed Baghdadi, Raj Seghal MD, David Schmit MD, George Crawford MD and the residents rotating on the Patient Safety/Procedure Service

Background: Educational goals Internal Medicine residents frequently perform invasive bedside procedures such as paracentesis (removal of fluid from the abdomen) The Internal Medicine residency at UTHSCSA has used simulation to standardize procedural training for the past two years Challenge To bring standardization in training and supervision to the bedside To incorporate bedside ultrasound in procedure training

Procedure team: PS/QI in action Veteran s Administration funded a chief resident position specifically to promote training in Patient Safety/Quality improvement Our proposal included having the PSQI resident supervise a procedure team to train first year residents how to safely perform invasive procedures Studying the impact of the procedure team will help determine the effectiveness of this approach

PS2: studying the impact Our project: to investigate the effect of the procedure team on the timeliness of paracentesis in patients admitted to Internal Medicine ward teams.

Paracentesis: A commonly performed procedure Ascites, an abnormal accumulation of fluid in the abdomen, is a common diagnosis on admission to the Internal Medicine service at the Audie Murphy VA Hospital and University Hospital Current guidelines support diagnostic paracentesis (removing fluid from the abdomen) to analyze the fluid for infection and other conditions. Patients often require removal of larger amounts of fluid (therapeutic paracentesis) to relieve symptoms such as abdominal pain and respiratory distress.

Paracentesis: A commonly performed procedure Over 500 paracenteses are performed annually at the AMVAH and the University Hospital, the majority on Internal Medicine services.

Steps to Paracentesis The patient usually presents to a clinic visit or ED with abdominal distension The patient is admitted to an internal medicine service for paracentesis The admitting team performs the paracentesis May be supervised by an attending or a resident who has performed > 5 paracenteses The patient may be sent to radiology for ultrasound marking to locate site of fluid See Flow Chart for specific steps to procedure

Pre-Admission ED/Clinic Patient presents with abdominal swelling, pain, bloating TIME 1 Vital signs Assessed by ED (incl. ordering, drawing, waiting for labs) Ascites? YES: Patient has ascites NO ASCITES: Exit algorithm Notify Triagist Ascites? Admit for tap? NO: Perform tap in ED Labs done? NO: Wait for labs UNCLEAR: Need US to evaluate for ascites Order US Pt transport to US Post-Admission Medicine Service US done Pt returns US read Prep & Procedure Supplies collected Communicate and coordinate with nursing staff Resident writes admit order Bedflow finds bed for patient Bedflow notifies triage staff of bed Triage calls transport Pt transported to the ward Albumin required? YES: Order Albumin Med resident assesses pt Resident signed off? YES: Proceed solo US requested? No US Obtain working COW Obtain consent Albumin sent Nurse calls transport Transport/US NO: Med resident contacts supervisor Order diagnostic studies Transport takes specimen to lab Do procedure US ordered in computer US service called? Pt transported to US US performed, sonomarked Pt transported back to floor Label, bag specimens, place in box TIME 2 Specimen received in lab

Problems with Usual Care Noted Prior to Implementation of Procedure Team Delays The procedure is often postponed while the admitting team performs other urgent tasks Need for ultrasound marking may delay procedure Technique Standard of care is bedside ultrasound immediately prior to the procedure, as location of fluid may shift after ultrasound marking Training and supervision Residents are signed off to supervise paracentesis after performing 5 procedures. Is this enough? Determination of competency to perform a procedure should not be based on a number, but on objectively observed competency

Problems with usual care Nursing not available

Usual practice vs. Procedure team Usual practice Procedure team Ward team performs procedure Supervision by attending or experienced resident Sonomarking performed by radiology, or procedure done without ultrasound Procedure team performs procedures (available 8 am-4 pm weekdays) Supervised by PSQI chief resident Bedside ultrasound by procedure team

AIM Statement To reduce the time from presentation with ascites to receipt of ascites specimen in the lab by 20% by Sept 15, 2011.

Methods Quality Framework for project: Institute of Medicine s six domains of quality (Safe, Timely, Efficient, Effective, Equitable, & Patient-Centered [STEEP]) IRB approval educational projects conducted within the CS&E course deemed exempt Initial Step Created flow diagram based on retrospective data from sample of 10 paracentesis charts

Data Collection Population of Interest Patients admitted to Internal Medicine service (IM) at the VA or UHS with a procedure code for paracentesis Exclusions ICU patients Patients who had paracentesis performed > 36 hours after admission (VA: n=2, UHS: n=1) Data Sources Chart review (flow chart and pre-implementation) Administrative (pre-implementation) Procedure Team logs (pre- and post-procedure) Variables (red = variables included in analysis) Time of initial presentation (ER, clinic) [vital signs] Time of initial evaluation by IM [hospitalist note] Time of procedure [procedure note] Time specimen received in lab [lab log] Percentage of patients US sonomarked [chart note] Service ordering US (ED vs. IM) [chart note] Time Period Baseline VA (9/2010-6/2011 n=39) UHS (3/2011-6/2011 n=48) Post-implementation (complete data available) VA (7/2011-8/2011 n=7) UHS (7/2011-8/2011 n = 11)

VAH: Baseline Time from arrival to lab: mean of 15.3 hrs with mean range of 7.4 hrs 44.5 Paracentesis (mean) From Triage to Lab < 36 hrs UCL 38.5 34.5 24.5 14.5 CL 15.3 4.5-5.5 LCL -7.8-15.5 From Triage to Lab < 36 hrs 9/30/2010 10/6/2010 10/8/2010 10/17/2010 10/19/2010 10/25/2010 10/26/2010 10/27/2010 11/5/2010 11/24/2010 12/2/2010 12/4/2010 12/10/2010 12/15/2010 12/16/2010 12/21/2010 12/27/2010 12/28/2010 12/29/2010 12/30/2010 12/31/2010 1/2/2011 1/3/2011 1/15/2011 1/25/2011 1/30/2011 1/30/2011 2/18/2011 2/19/2011 3/5/2011 3/8/2011 4/6/2011 4/7/2011 4/17/2011 4/19/2011 4/23/2011 5/3/2011 5/13/2011 5/23/2011 Sept 2010 - May 2011

UHS: Baseline Time from arrival to lab: mean of 14.7 hr with mean range of 9.9 hrs 49.6 Paracentesis - Time from ER to Lab < 36hrs 39.6 UCL 41.1 29.6 19.6 CL 14.7 9.6-0.5-10.5 LCL -11.7-20.5 3/1/2011 3/7/2011 3/7/2011 3/8/2011 3/8/2011 3/13/2011 3/14/2011 3/16/2011 3/17/2011 3/17/2011 3/18/2011 3/21/2011 4/3/2011 4/5/2011 4/7/2011 4/7/2011 4/10/2011 4/11/2011 4/12/2011 4/15/2011 4/17/2011 4/18/2011 4/20/2011 4/26/2011 4/26/2011 4/27/2011 5/2/2011 5/10/2011 5/11/2011 5/12/2011 5/13/2011 5/13/2011 5/16/2011 5/19/2011 5/21/2011 5/23/2011 5/24/2011 5/26/2011 6/2/2011 6/3/2011 6/4/2011 6/7/2011 6/8/2011 6/12/2011 6/15/2011 6/15/2011 6/17/2011 6/20/2011 ER to Lab < 36hrs March 2011 - June 2011

Procedure Team Activity July/August 2011 62 paracentesis 17 at VA (10 cases excluded, e.g. > 36 hrs, changed service, etc.) 45 at UH (34 cases excluded, e.g. > 36 hrs, incomplete data, etc.) Percentage of Internal Medicine Paracentesis Performed by Procedure Team VA Hospital: 41% UH: estimated 43% Additional procedures 18 lumbar punctures 28 thoracenteses 5 arthrocenteses

VAH: Post-implementation time from arrival to lab: mean of 11 hrs with mean range of 14.5 hrs 59.8 49.8 Paracentesis (Mean) Triage to Lab < 36 Hrs UCL 49.5 39.8 Triage to Lab (hrs) 29.8 19.8 9.8-0.3-10.3-20.3-30.3 CL 11.0 LCL -27.4-40.3 7/4/2011 7/6/2011 7/21/2011 7/28/2011 8/14/2011 8/16/2011 8/30/2011 July 2011 - August 2011

UHS: Post-Implementation time from arrival to lab: mean of 14.1 hrs with mean range of 3.5 hrs 30.0 Paracentecis (mean) ER to Lab < 36hrs 25.0 UCL 24.9 ER to Lab < 36hrs 20.0 15.0 10.0 CL 14.1 5.0 LCL 3.2 0.0 7/1/2011 7/4/2011 7/5/2011 7/25/2011 7/28/2011 8/2/2011 8/10/2011 8/13/2011 8/14/2011 8/16/2011 8/25/2011 July 2011 - August 2011

Return on Investment Assumptions 27 bedside paracentesis procedures/month (average) Replace location of sonograms from radiology to bedside: $312 savings/per procedure ($300/radiology, $12/transport) 2 hour decrease in observation time per patient For each patient, decrease observation time by 2 hours ($58/hr, uncomplicated patient) Hospital reimbursement rate of 35% Project Costs Covered by Educational Costs Annual cost of faculty: $108, 800 One time cost for portable sonogram machine: $40,000 Annual Savings 27 bedside paracentesis procedures/month = 324 procedures per year x $312 = $101,088 2 hour decrease in observation time per patient = 324 procedures x $116 = $37, 584 Total annual savings: $101,088 + 37,584 = $138,672 Cost savings over 3 years: $138,672 x 3 =$416,016 [varies by hospital reimbursement rate

Discussion Was Aim of 20% in reduction from time of presentation to receipt of specimen lab met? Unknown - Insufficient data to determine if decrease in time just normal variation or sustained decrease in time but trend encouraging What specific component(s) of paracentesis process flow did procedure team impact? Results of decreased mean hrs but increased range suggest impact of procedure team influenced by myriad of factors. Subject of future research discussed by Dr. Wathen. ROI Significant savings demonstrated Limitations Data validity accuracy, inconsistency admin/ chart Unable to assess monthly range variation due to limited data

IOM Domains of Quality Safety Improved training and supervision of residents in bedside procedures Incorporates bedside ultrasound Timeliness Team can perform procedures while ward team is busy with other duties Efficient Bedside ultrasound vs. transport to radiology Procedure team is working on improving consent process, availability of supplies Effective Improved training may improve success rate for bedside procedures Equitable Applying best practices across the healthcare system Patient Centered More expeditious completion of procedures decreases patient discomfort and anxiety. Research shows high levels of patient satisfaction with procedure teams.

Use of Bedside Ultrasound Bedside Ultrasound is emerging as an important tool for invasive procedures Significantly reduces complications of Central Line placement and Thoracentesis Sonomarking in Radiology department has not been shown to have the same benefits For paracentesis bedside ultrasound improves procedure success rates Sonomarking still commonly performed prior to paracentesis At University Hospital, 100% of patients were sonomarked prior to paracentesis At VA 88% were sonomarked Half of the ultrasounds were order by ED personnel

Future Directions Dissemination of bedside ultrasound training Transition from sonomarking to bedside sono Hospitalists to be trained in bedside ultrasound All residents will be trained by 2013 Additional QI projects Examining how efficiency and effectiveness for paracentesis can be impacted at multiple points in the system Examination of cost/benefit of dedicated NP to perform procedures at multiple sites within the system (outpatient clinic, ER) to reduce length of stay and admissions.