getting the value (roi) from laboratory software: outcomes ...€¦ · getting the value (roi) from...

Getting the Value (ROI) from Getting the Value (ROI) from Laboratory Software: Laboratory Software:

Outcomes, Outcomes, Outcomes, Outcomes, OutcomesOutcomes

Kenneth Blick, Ph.D., ABCC, FACBProfessor

Department of PathologyUniversity of Oklahoma Health Sciences Center and OU Medical Center

Oklahoma University Health Oklahoma University Health Sciences Center (OUHSC)Sciences Center (OUHSC)

The first faculty and students, c. 1895

Oklahoma University Health Oklahoma University Health Science Center/OU Medical Science Center/OU Medical

Center Center

OU Medical CenterOU Medical Center

Pneumatic tube systemPneumatic tube system

Observations:Observations:Market forces are driving tremendous change Market forces are driving tremendous change in healthcare…many hospitals need “lifein healthcare…many hospitals need “life--support.”support.”Critical care demands for realCritical care demands for real--time laboratory time laboratory services increasing.services increasing.Physician’s perceive laboratory results have Physician’s perceive laboratory results have more “historic” value than immediate more “historic” value than immediate diagnostic and therapeutic significance.#diagnostic and therapeutic significance.#The word is out….hospitals are dangerous The word is out….hospitals are dangerous places to be, especially for patients. places to be, especially for patients.

#Kost, G.Am J Clin Path,1995; 104:S2

Our customers can’t afford our products!

Daily Oklahoman, October 16, 2003Physicians can’t afford to practice…malpractice insurance up nearly 4 fold in last two years

NoNo “Help on the Way”!“Help on the Way”!

Sources of PreSources of Pre--Analytical ErrorAnalytical Error

Missing Missing PredrawPredraw instructions, no instructions, no postdrawpostdrawinstructions, illegible collection data on instructions, illegible collection data on label, no collection data on label, wrong label, no collection data on label, wrong person drawn, wrong container used, person drawn, wrong container used, collected at the wrong time, drawn in the collected at the wrong time, drawn in the wrong order, missing label, missing order, wrong order, missing label, missing order, wrong label, label misaligned, lost wrong label, label misaligned, lost specimen, no tracking, etc., etc.specimen, no tracking, etc., etc.

Clinical Lab News, Oct 2002

Not ready for prime time!

CHF: Congestive “Hospital” CHF: Congestive “Hospital” FailureFailure

“Full House” Causes ED Bypass, Lost Admissions

10 miles

CCU

No CCU beds available to take ED patients

Stepdown

No stepdownbeds available to take ED patients as floor is full

Neighboring Hospital

Source: Cardiovascular Roundtable Advisory Board

Overcrowding #1 problem Dr. WF Peacock

75 % EDs Operating Overcapacity!

HEALTHCARE HEALTHCARE DILEMMADILEMMA

Long ED wait

Patient risk

Inappropriate admissions &

discharges

Telemetry Backup

Fewer budget dollars

ED diversion

Rapid Testing

Essential

Imperfect information

flow

Bed Shortage

OUMCOUMC

Consists of multiple freestanding Consists of multiple freestanding hospitalshospitals–– Two adult hospitalsTwo adult hospitals-- each with each with chemchem lablab–– Children’s hospitalChildren’s hospital–– Outpatient clinicsOutpatient clinics-- with own with own chemchem lablabTogether, more than 2.5 million Together, more than 2.5 million chemistry tests per year/2,000 chemistry tests per year/2,000 chemistry samples per day.chemistry samples per day.

Getting enough capacity: Getting enough capacity: Eliminating batch processes Eliminating batch processes

whenever possible?whenever possible?

Old Paradigm: Increased demand for batch Old Paradigm: Increased demand for batch laboratory services has been met with laboratory services has been met with increased staff. increased staff.

New Paradigm: Evidenced based medicine New Paradigm: Evidenced based medicine requires realrequires real--time services. Since old solutions time services. Since old solutions cannot succeed, technology is now required to cannot succeed, technology is now required to change the clinical laboratory to a realchange the clinical laboratory to a real--time time laboratory information infrastructure.laboratory information infrastructure.

Giving the lab back to the Giving the lab back to the physicians…team effort!physicians…team effort!

ED Lab Cath Lab Pharmacy CriticalCare

SHORTNESS OF BREATH

CHEST PAIN

STROKE

It takes a multi-disciplinary team to• enhance Quality of Care• improve Cost Management

New SOB panel: BNP, D-dimer, Troponin-I, CK-MB, Myoglobin 6/30/04

In an effort to increase efficiency, we began In an effort to increase efficiency, we began a project in July 2003 to consolidate all a project in July 2003 to consolidate all chemistry testing into an automated core chemistry testing into an automated core laboratory. laboratory.

In March 2004, we went live on our total In March 2004, we went live on our total laboratory automation track system for laboratory automation track system for chemistry.chemistry.

RealReal--time time QueuelessQueuelessLaboratoryLaboratory

New assignment for staff: Keep the New assignment for staff: Keep the information flowinginformation flowing

The The QueuelessQueueless Core Laboratory: No Batch Core Laboratory: No Batch ProcessesProcesses

Decentralized RealDecentralized Real--Time Specimen Time Specimen Collection:Collection:

1.1. Order Entry/Label printing on the Order Entry/Label printing on the floors and clinics floors and clinics

2.2. Collection/labeling by nurses and Collection/labeling by nurses and housestaffhousestaff

3.3. Lab Phlebotomy for Lab Phlebotomy for preadmitspreadmits, etc., etc.Decentralized Specimen TransportDecentralized Specimen Transport IntraIntra-- and and InterhospitalInterhospital Pneumatic Tube Pneumatic Tube

Transport Transport Centralized Processing/Testing/Core Centralized Processing/Testing/Core LaboratoryLaboratory

Elimination of “STAT” LaboratoriesElimination of “STAT” Laboratories

Automatic Specimen Receive in the Automatic Specimen Receive in the LaboratoryLaboratory

Track/Total Automation SystemTrack/Total Automation System

Automated Specimen ProcessingAutomated Specimen Processing Track/Total Automation SystemTrack/Total Automation System

Automated “Load Balanced” TestingAutomated “Load Balanced” Testing Track/Total Automation SystemTrack/Total Automation System

Autoverification/Release of ResultsAutoverification/Release of Results DataLinkDataLink (BCI) and Meditech LIS(BCI) and Meditech LIS

Specimen Repository/Add on OrdersSpecimen Repository/Add on Orders Track/Total Automation Track/Total Automation System/Specimen StockyardSystem/Specimen Stockyard

BCI, Beckman Coulter ; LIS, Laboratory Information System

OUMC ChemistryOUMC Chemistry--LeanLean

Centered around a vendor TLA (total Centered around a vendor TLA (total laboratory automation) systemslaboratory automation) systems–– 3 high speed 3 high speed walkawaywalkaway chemistry analyzerschemistry analyzers–– 2 high speed 2 high speed walkawaywalkaway hematology analyzershematology analyzers–– 2 2 walkawaywalkaway automated automated immnoassayimmnoassay

instrumentsinstruments–– Track specimen handling middleware by Track specimen handling middleware by –– Result reporting/control by a core lab Result reporting/control by a core lab

middleware solution (middleware solution (ChemChem, Hematology, , Hematology, etc.)etc.)

Legacy Solutions: Impact on CURRENT PROCESS Legacy Solutions: Impact on CURRENT PROCESS


order enteredin LIS

labels printedin phleb. area

phleb.monitors order

activity

phlebdispatched for

collection

specimencollected

labeled andbagged

return to lab

SpecimensRECEIVED in

LIS

IN-HOUSESPECIMENS

ER specimenpnuematic

tube

Park TubeOrder? No

LabelAttached ?

YesLabel Printed

Label Applied

No

Yes

ERSPECIMENS

RequestDelivered to

Lab

ADD ONTESTS

LocateSpecimen

Tech and Timeof Draw

Entered in LIS

Yes

REFERENCETESTING

Samplesdelivered

PatientRegistration

Order entryand labelprinting

Label Tubes

Load toTransport

Racks

centrifuge ?

load/balancecentrifuge

spin

load togeneric racks

label aliquottubes

racksequenciallyby accession

aliquot ?

pour off toprepared

aliquot tube

NO

load togeneric racks

NO

ORDERREVIEW

Deliver toAliquoting

Bench

STAT?

BATCH TESTSHEMATOLOGYCOAGULATION

CHEMISTRY

AlertTechnologist

YES

Take toSections

RXL

Load reagents,calibrate

analysis

unloadanalyzer

specimen

manual reviewof results

OK ?

data

RELEASEresults

yes

Fix Problem

no

Transfer togeneric racks

SpecimenOK ?

Load analyzerand begin run

Adjust Rackfor Tube Size

Decap andPour into Cup

Sort tubes forstorage

Transfer toCold Storage

ARCHITECT


analysis

unloadanalyzer

specimen


OK ?

data

RELEASEresults

yes

Fix Problem

no


SpecimenOK ?



Day Shift?

Immunolyte


analysis

unloadanalyzer

specimen


OK ?

data

RELEASEresults

yes

Fix Problem

no

SpecimenOK ?Redraw NO



Sort toInstrument

Store for DayShift

NO

YES

NO

AxSYM

TDMor

DAU


analysis

unloadanalyzer

specimen


OK ?

data

RELEASEresults

yes

Fix Problem

no

Pour offUrine

DAU

TDM

SpecimenOK ?



Redraw orUltrafuge

7 56

121110

8 4

21

9 3PROCESSWAIT STATES

BIOHAZARDEXPOSURE

PATIENTSAFETY

7 56

121110

8 4

21

9 3

7 56

121110

8 4

21

9 3

7 56

121110

8 4

21

9 3

7 56

121110

8 4

21

9 3

7 56

121110

8 4

21

9 3

7 56

121110

8 4

21

9 3

7 56

121110

8 4

21

9 3

7 56

121110

8 4

21

9 3

7 56

121110

8 4

21

9 3

7 56

121110

8 4

21

9 3

7 56

121110

8 4

21

9 3

7 56

121110

8 4

21

9 3

7 56

121110

8 4

21

9 3

7 56

121110

8 4

21

9 3

7 56

121110

8 4

21

9 3

7 56

121110

8 4

21

9 3

7 56

121110

8 4

21

9 3

7 56

121110

8 4

21

9 3

7 56

121110

8 4

21

9 3

7 56

121110

8 4

21

9 3

7 56

121110

8 4

21

9 3

7 56

121110

8 4

21

9 37 56

121110

8 4

21

9 3

7 56

121110

8 4

21

9 3

7 56

121110

8 4

21

9 3

7 56

121110

8 4

21

9 3

7 56

121110

8 4

21

9 3

7 56

121110

8 4

21

9 3

7 56

121110

8 4

21

9 3

7 56

121110

8 4

21

9 3

7 56

121110

8 4

21

9 3

7 56

121110

8 4

21

9 3

7 56

121110

8 4

21

9 3

7 56

121110

8 4

21

9 3

7 56

121110

8 4

21

9 3

AUTOMATION SOLUTIONAUTOMATION SOLUTION


ADD-ONrequest from

LIS

TO IMMUNO AND SPECIALS SECTION

INLET

VERF. BARCODE READ

HEM OUTLET

DE-CAPPER

LXCONNECTION

LXCONNECTION

OUTLET

TO HEMATOLOGYSECTION

auto rackedspecimens

auto rackedspecimens

DATALINKAUTO REVIEWOF RESULTS

OK ?

RELEASE

yes

DATALINKrequest RE-RUN

no

CENTRIFUGE

MANUALREVIEW

OK ? yesno

DE-CAPPER

REFRIGSTOCKYARD

CAPPER

Automated Process

order enteredin LIS

labels printedin phleb. area

phleb.monitors order

activity

phlebdispatched for

collection

specimencollected

labeled andbagged

return to lab

SpecimensRECEIVED in

LIS

IN-HOUSESPECIMENS

ER specimenpnuematic

tube

Park TubeOrder? No

LabelAttached ?

YesLabel Printed

Label Applied

No

Yes

ERSPECIMENS

Tech and Timeof Draw

Entered in LIS

Yes

REFERENCETESTING

Samplesdelivered

PatientRegistration

Order entryand labelprinting

Label Tubes

Load toTransport

Racks

PROCESSWAIT STATES

BIOHAZARDEXPOSURE

PATIENTSAFETY

7 56

121110

8 4

21

9 3

OUMC Core Lab

Command Central, Command Central, software module which software module which operates on a DL/operates on a DL/RemisolRemisol2000 console.2000 console.Consolidates the Consolidates the DL/DL/RemisolRemisol 2000 functions 2000 functions and instrument monitoring and instrument monitoring into a single workstation. into a single workstation. Monitors and alert the user Monitors and alert the user on the status of the on the status of the attached analyzers screens attached analyzers screens to conditions that require to conditions that require intervention.intervention.The alerted user will have The alerted user will have complete access to complete access to analyzer that requires analyzer that requires intervention.intervention.

Example of Instrument Example of Instrument ConnectionsConnections

Data CollectionData Collection

Monitored turnMonitored turn--around times for three around times for three analytesanalytes(potassium, (potassium, troponintroponin, thyroid stimulating , thyroid stimulating hormone) between July 2003 and August 2004hormone) between July 2003 and August 2004

Data subdivided by the hospital of origination Data subdivided by the hospital of origination and whether the test was ordered and whether the test was ordered ““statstat”” from from the emergency departmentthe emergency department

Also examined the staffing needs and testing Also examined the staffing needs and testing volumes over this time periodvolumes over this time period

Other data mining outcomes since goOther data mining outcomes since go--live live March 04?March 04?

Closed Stat laboratory at Presbyterian and Children’s Closed Stat laboratory at Presbyterian and Children’s Hospital Towers: TAT improved or remained the same for Hospital Towers: TAT improved or remained the same for Stat and routine. Saved $1.5 M annuallyStat and routine. Saved $1.5 M annuallyEliminated Stat testing altogether: FIFOEliminated Stat testing altogether: FIFOStaffing Core Lab: 70 FTEs to 59; 10 per diem to 1Staffing Core Lab: 70 FTEs to 59; 10 per diem to 1Productivity of entire Core Lab: paid FTE/TestsProductivity of entire Core Lab: paid FTE/Tests-- .13 to .13 to .10.10Test volume increase over 6 %; overtime from 3.5 % to Test volume increase over 6 %; overtime from 3.5 % to 1.3 %.1.3 %.Telephone calls reduced substantially: lab and customer Telephone calls reduced substantially: lab and customer support areas; 1 FTE eliminatedsupport areas; 1 FTE eliminatedBrought in 15 tests which had been referredBrought in 15 tests which had been referredEliminated staff turnover and minimized stress on Eliminated staff turnover and minimized stress on employeesemployeesError reduction efforts…no mined data as yetError reduction efforts…no mined data as yet

Benefits of LeanBenefits of Lean

More free tech timeMore free tech time–– To deal with problem specimensTo deal with problem specimens–– Perform labor intensive test not capable of Perform labor intensive test not capable of

running on instruments connected to the running on instruments connected to the automated lineautomated line

Brought back inBrought back in--house over 20 previous sendhouse over 20 previous send--out reference testsout reference testsThereby, reducing our reference bill (over Thereby, reducing our reference bill (over $22,000/month) and making results available to $22,000/month) and making results available to clinicians more rapidly clinicians more rapidly

Benefits of LeanBenefits of Lean

Increased efficiencyIncreased efficiency–– Over the past year the number of techs Over the past year the number of techs

employed decreased by 16%employed decreased by 16%–– Over the same time billable tests per Over the same time billable tests per

month has increased by over 6%month has increased by over 6%–– Equates to a 35% improvement in our Equates to a 35% improvement in our

productivity index (a ratio of billable tests productivity index (a ratio of billable tests per techper tech--hour)hour)

–– Productivity ratio 0.288 to 0.145 in the last Productivity ratio 0.288 to 0.145 in the last 4 years.4 years.

Intangible BenefitsIntangible Benefits48 processes eliminated with over 90 high 48 processes eliminated with over 90 high volume tests on line and available 24/7volume tests on line and available 24/7Redundant design with no downtimeRedundant design with no downtimePatient safety issues addressed…positive Patient safety issues addressed…positive sample ID with rapid results available for sample ID with rapid results available for critical carecritical careLess stress with less errorsLess stress with less errorsGetting real value out of investments in Getting real value out of investments in technology, space and stafftechnology, space and staffHaving more funHaving more fun

What Expert Systems Do?What Expert Systems Do?

Reflex testingReflex testingCancels testingCancels testingData event alerts to technologistsData event alerts to technologistsPhysician notification/critical resultsPhysician notification/critical resultsEmail, fax, remote print, coded commentsEmail, fax, remote print, coded commentsStanding ordersStanding ordersGather data for prospective studiesGather data for prospective studiesLogs for outcomesLogs for outcomesEtc.Etc.

Reflex Testing/Other Examples:Reflex Testing/Other Examples:

Order manual differential/make the slideOrder manual differential/make the slideReflex to UA microscopic (Atlas/Reflex to UA microscopic (Atlas/SysmexSysmex UFUF--100)100)Track Load Balance/Specimen deferralTrack Load Balance/Specimen deferralSelective Processing/SortingSelective Processing/SortingManual diff when CSF WBC <5Manual diff when CSF WBC <5Neonatal Neonatal BiliBili when HI >= 8when HI >= 8Order Order creatininecreatinine (if not ordered) when (if not ordered) when gentamycingentamycin results are entered results are entered

DL2000: Typical Boolean Rules DL2000: Typical Boolean Rules (Anion Gap Check):(Anion Gap Check):

IF (INRANGE(AGAP) THEN VALIDIF (INRANGE(AGAP) THEN VALIDIF (AGAP<AGAP,VALIDLOW)IF (AGAP<AGAP,VALIDLOW)THEN THEN REFLEX(CO2);REFLEX(K);RFLEX(NA)REFLEX(CO2);REFLEX(K);RFLEX(NA);REFLEX(CL);REFLEX(CL)

Play Once Active

DL2000: Neonatal DL2000: Neonatal Bilirubin:TechnologistBilirubin:Technologist AlertAlert

IF(HEMOL)>=8 and (EXIST(TBIL) AND IF(HEMOL)>=8 and (EXIST(TBIL) AND (AGE<30D)(AGE<30D)THEN THEN REFLEX(TBE);MSG (RUN TBE and result REFLEX(TBE);MSG (RUN TBE and result with coded comment HEMBIL)with coded comment HEMBIL)

Play Once Active

Unacceptable SpecimensUnacceptable Specimens

Performs IndicesPerforms IndicesAlerts technologistAlerts technologistOrder to recollect if appropriateOrder to recollect if appropriateSuppression of resultsSuppression of resultsAppends commentAppends commentRSP for critical patients RSP for critical patients Logs data eventLogs data eventEmail to supervisorEmail to supervisor

Critical ResultsCritical Results

Involves rules at four levels: instrument, track, Involves rules at four levels: instrument, track, DataLink2000, Meditech 4.9.1 LISDataLink2000, Meditech 4.9.1 LISInstrument automatically rechecks results; age Instrument automatically rechecks results; age and sex matched rangesand sex matched rangesAlerts the technologist on Alerts the technologist on DataLinkDataLink with filterwith filterProvides the physician name and phone Provides the physician name and phone number or gives special instructions, LISnumber or gives special instructions, LISLogs the event, LISLogs the event, LISDocuments the call/Documents the call/ReadbackReadback, LIS, LISEmails supervisor on weekends, NetworkEmails supervisor on weekends, NetworkRSP/RSP/autofaxautofax, LIS, LIS

Other Indicators of Laboratory Other Indicators of Laboratory PerformancePerformance

Mean and standard deviation of TAT can Mean and standard deviation of TAT can be adequate measures of laboratory be adequate measures of laboratory performanceperformanceA better method would be setting clinically A better method would be setting clinically based TAT goals and monitoring success based TAT goals and monitoring success in achieving these goalsin achieving these goals

Major Problems with POCT Major Problems with POCT ComputerizationComputerization

Connectivity issues with POCT devices Connectivity issues with POCT devices and associated DMSand associated DMSIT challenged, primitive POCT devicesIT challenged, primitive POCT devicesUnique nature of POCT when compared Unique nature of POCT when compared with traditional “core” lab testingwith traditional “core” lab testingUnique database requirements for POCTUnique database requirements for POCTManagement support for plethora of Management support for plethora of remote POCT devicesremote POCT devices

Many types of laboratory billingMany types of laboratory billing

Inpatient hospitalInpatient hospital--DRG, Part ADRG, Part AOutpatientOutpatient--hospitalhospitalOutpatient/reference LabOutpatient/reference LabContract billingContract billingManaged care billing/trackingManaged care billing/trackingProfessional billing, Part BProfessional billing, Part BCP Inpatient Professional Billing (CP Inpatient Professional Billing (ProfeeProfee; 7; 7--10%)10%)Facility and test modifiersFacility and test modifiersUB92s or HCFA (CMS) 1500UB92s or HCFA (CMS) 1500Electronic versus manual claimsElectronic versus manual claims

July 2004 Relationship between Lab % Outliers and ED Patient Turnover in hours ?

2.6%

5.0%

6.9%

2.67

3.18

3.81

0.0%

1.0%

2.0%

3.0%

4.0%

5.0%

6.0%

7.0%

8.0%

9.0%

10.0%

Averaged 4 BEST GC Division AVG (12facilities)

Averaged 4 WORST2.00

2.20

2.40

2.60

2.80

3.00

3.20

3.40

3.60

3.80

4.00GC Lab % OutliersER Patient Turnover in Hours

OUMC Indicators of Laboratory OUMC Indicators of Laboratory PerformancePerformance

We have set the following goalsWe have set the following goals–– Potassium TAT Potassium TAT << 40 minutes40 minutes–– Core Lab Core Lab TroponinTroponin TAT TAT << 60 minutes60 minutesThrough implementation of our automated Through implementation of our automated chemistry system, we have drastically chemistry system, we have drastically improved our success in meeting these improved our success in meeting these goalsgoals

0%

5%

10%

15%

20%

25%

30%

35%

0 5 10 15 20 25 30 35 40 45 50 55 60 65

Minutes

Perc

ent o

f Spe

cim

ens

Jul 03Aug 04

Comparison of TAT Comparison of TAT distributions for K from ER distributions for K from ER

mean = 30.3SD = 13.4H/W = 0.45mean = 27.3SD = 9.1H/W = 1.04

Comparison of TAT distributions for Comparison of TAT distributions for TroponinTroponin from ERfrom ER

0%

5%

10%

15%

20%

25%

30%

35%

0 10 20 30 40 50 60 70 80 90Minutes

Perc

ent o

f Spe

cim

ens

Jul 03

Aug 04mean = 48.4SD = 21.9H/W = 0.36mean = 48.4SD = 9.0H/W = 1.73

STAT Tests No Longer NeededSTAT Tests No Longer Needed

30 28

3832

48 50

60

48

0

10

20

30

40

50

60

70

7/03 8/04

Ave

rage

TA

T (m

in)

K ERK not ERTroponin ERTroponin not ER

““The laboratory is no longer The laboratory is no longer an issue.”an issue.”

John Stuemky, M.D.John Stuemky, M.D.Medical Director, EDMedical Director, EDOU Medical CenterOU Medical Center

Unpredictable LAB Service Effects on ED Unpredictable LAB Service Effects on ED LOSLOS

Correlation PP-- valuevalue

Before AutomationBefore AutomationK Outliers > 18 %K Outliers > 18 %

0.980.98 0.010.01 Highly CorrelatedED LOS depends on LabED LOS depends on Lab

Before AutomationBefore Automation# of ED Patients# of ED Patients

0.530.53 0.280.28 ED LOS depends more ED LOS depends more on lab delays than on on lab delays than on patient volumepatient volume

After AutomationAfter AutomationK Outliers < 5 %K Outliers < 5 %

0.540.54 0.460.46 With improved TAT the With improved TAT the ED LOS is not dependent ED LOS is not dependent on Lab delayson Lab delays

After AutomationAfter Automation# of ED Patients# of ED Patients

0.880.88 0.060.06 Highly CorrelatedHighly CorrelatedED LOS now depends ED LOS now depends more on patient volume more on patient volume than on Lab delaysthan on Lab delays

Before automation: Nadm_LOS=+2.79[%K] + 78.8

Benchmarking by percent Benchmarking by percent outliersoutliers

0%10%20%30%40%50%60%70%80%90%

100%

Jul Aug Sep Oct Nov Dec Jan Feb Mar Apr May Jun Jul Aug

Perc

ent m

eetin

g go

al

Potassium not ERPotassium ERTroponin not ERTroponin ER

PatientArrives AdmitDisposition

?TreatmentDiagnosis

Home

• Optimal Tx, faster, can reduce IP LOS

• Reduction in variable cost / increase in revenue

IMPACT ASSESSMENT at OUMCIMPACT ASSESSMENT at OUMC

Reduce ED LOS

Optimize Treatment

Decrease Unnecessary Admissions

Reduce IP LOS

EXAMPLE: HOSPITAL “A” DATA EXAMPLE: HOSPITAL “A” DATA Hospital DataHospital Data CommentsComments

Total ED Visits/yrTotal ED Visits/yr 104,439104,439 2002 annual2002 annual

Avg. ED LOS all ptsAvg. ED LOS all pts 4:06 hours4:06 hours 246 minutes246 minutes

#,% admitted, all pts#,% admitted, all pts 19%19%

Annual ED OT $Annual ED OT $ $45,883$45,883

Annual #, % LWOTsAnnual #, % LWOTs 4.1%4.1%4,253 patients4,253 patients

Annual #, % DivertsAnnual #, % Diverts 510.22 Hours/yr510.22 Hours/yr =21.26 day increments=21.26 day increments

# ED Pts presenting w/ # ED Pts presenting w/ Chest Pain, SOBChest Pain, SOB

9,3629,362 ~9% of patients~9% of patients

Avg. ED LOS CP, SOB Avg. ED LOS CP, SOB ptspts

CP 5:01 hoursCP 5:01 hoursDIB 4:08 hoursDIB 4:08 hours

Average weighted timeAverage weighted time4:37 hours4:37 hours

% ACS/AMI M/care% ACS/AMI M/care 70.6%70.6%

% CHF M/care% CHF M/care 81.2%81.2%

ConclusionsConclusionsAutomated/Expert systems can handle most Automated/Expert systems can handle most routine and nonroutine and non--routine data eventsroutine data eventsAutomated/Expert systems facilitate handing Automated/Expert systems facilitate handing problems in realproblems in real--time and more predictably than time and more predictably than humanshumansBy elimination of batch testing system through By elimination of batch testing system through use of automation and expert systems, use of automation and expert systems, laboratories have great potential to improve the laboratories have great potential to improve the overall quality and predictability of laboratory overall quality and predictability of laboratory services. Such improvements are now services. Such improvements are now essential for the effective practice of Evidenced essential for the effective practice of Evidenced Based Medicine especially for the critically ill.Based Medicine especially for the critically ill.