Quality Improvement (QI) HOW DOES IT WORK? Dr S Narayanan Neonatal Consultant Watford General Hospital
Outline of the talk Background Definitions QI What? Why? When? Where? How? Case study Discussion & Questions
Progress Background We get it right 54% of the time. -Brent James, MD, MStat Executive Director, Intermountain Health Care Time Patient care 1 in 10 in hospital at risk of patient safety incident
Background Inefficient health care Poor patient experience Patient harm Increase in healthcare costs Drop in staff morale Hospital reputation
QUALITY IMPROVEMENT Quality: The degree to which health services increase the likelihood of desired health outcomes and are consistent with current professional knowledge
How can we improve quality? Extrinsic Central initiatives Economic drivers (CQUIN) Professional standards Regulatory frameworks Intrinsic Organizational commitment Leadership Staff engagement Patient participation
Other factors that improve quality Sustained focus on continuous improvement Emphasis on internal motivators Align quality at every level Redefine ways of engaging with service users and team Learn from other sectors
METHODS UNDERSTAND THE PROBLEM [POWER OF DATA] PROCESSES AND PATHWAYS DEMAND/CAPACITY/WORK FLOW CHOOSE APPROPRIATE TOOLS CHANGE ASSESS EVALUATION/MEASUREMENT
QI: QUALITY IMPROVEMENT TOOLS Better patient experience and outcomes achieved through changing provider behaviour through systematic analysis of performance and conscious efforts to improve it BUSINESS PROCESS RE-ENGINEERING COLLABORATIVES LEAN Plan, Do, Study, Act (PDSA) Six Sigma
The Improvement Model What are we trying to accomplish? How will we know that a change is an improvement? What changes can we make that will result in improvement?
PDSA- Plan Do Study Act Plan Always includes a prediction Do--test the change Study Did my prediction hold? What assumptions need revision? Act Adapt Adopt Abandon
PDSA Cycles Traditional Approach to Activities P D S A Suggested Approach Time P D S A Time saved Which is typical of your organisation?
How do we go about changing the system? Establish a future plan Act Plan Do 5-Step Process for Improvement 1. Select the opportunity for improvement 2. Study the current situation 3. Analyze the causes 4. Develop a theory for improvement 5. Select the team Model for Improvement AIM What are we trying to accomplish? CHANGES What change can we make that will result in improvement? MEASURES How will we know that a change is an improvement? Study the results Study Implement the Improvement ACT STUDY PLAN DO Ideal Future ACT PLAN ACT PLAN STUDY DO Present Situation STUDY DO
Measurement for improvement Measurement is for improvement not judgement. D. Berwick
1 Decide Aim 2 Choose Measures 3 Define Measures 7 Steps to measurement 4 Collect Data 7 Repeat steps 4-6 6 Review Measures 5 Analyse & Present
Step 1 Decide Aim
20 Step 2 Choose Measures Driver diagrams Process maps Fishbone analysis Five whys? Pareto charts Scatter diagrams
Driver Diagrams 21 O Outcome measure P Process measure Aim Driver Intervention O1 Ward round starts on time Arrive on time All information handy Chair/Facilitator P1 P2 P3 Handover MUST finish on time EWTD compliance Display handover times Diary monitoring Night staff to leave on time P4 P5 P6 Effective work flow Sign posting/prompts Cons/Reg for the week P7 P8
Process maps Start? Decision Point? Handover? End?
Fishbone analysis
25 Step 3 Define Measures An operational definition is a description, in quantifiable terms, of what to measure and the steps to follow to measure it consistently Are we measuring the same thing?
Advice on creating measurement definitions 26 Repeatability Can you, who created the definition, understand it and repeat it? Also known as test-retest error, used as an estimate of short-term variation Reproducibility After repeatability, try seeing if the definition that you have created can be reproduced by other people?
27 Step 4 Collect Data Practical considerations: 5 W s and 1H
28 Step 4 Collect Data Decisions, decisions What - All patients or a sample? Who took the data? (what role?) When When was the data taken - real time or retrospective? Where is the data from? How was the data taken? (What process?) Turn the data into a different unit (hours into days)
Step 5 Analyse & Present 29
Step 6 Review Measures It is a waste of time collecting and analysing your data if you don't take action on the results
Human factors Change management/influencing change
Summary Quality improvement is key Facets of QI PDSA small tests of change Power of data Tools Map processes Measuring outcomes Human dimensions of change
Summary (contd ) QI EVERYONE S JOB A3 ANYONE ANYTIME ANYWHERE
MEDIREST SERVICES Quality improvement initiative: To improve family experience by facilitating early discharge of neonates treated for suspected early onset infection MEDICAL TEAM Dr S Narayanan [Project Lead] Dr S Sukhani Dr K Barnes Dr R Mardare LABORATORY TEAM J Paske A Hawkins
Introduction NICE Guidance (CG149, Aug 2012) Infants with suspected early onset infection Stop abx at 36hrs Clinically well and blood cultures are negative Need to establish hospital systems to provide blood culture results at 36hrs after starting abx
Survey of practice in regional neonatal units [n = 59 ] (ADC 2014) 40 % of units compliant to 36 hour reporting of blood cultures NICE GDG should consider practical challenges and effective feedback mechanisms & be responsive for reasons behind non compliance
Audit data Of 32 randomly selected babies on treatment: Only one baby had cultures available at 36 hrs Delay in culture reporting 18 babies on abx Implications: Infant pain/distress from cannulation, risk of sideeffects/drug toxicity Family mother/infant separation, feeding and family experience Provider logistical and financial What are we trying to accomplish?
QI process: understand cause for problem Identification of barriers to delivery of high quality service 1. Fishbone analysis root causes for delay in culture processing and reporting 2. Process mapping journey of blood culture 3. Audit time taken for each stage and frequency it occurs
Delay in obtaining baby s hospital number Delay in porter collecting sample Delay in porter delivering sample to lab Delay in culture being sent to the lab Delay in getting baby to NICU eg feeding, bonding with mother Delay in culture being taken by doctor Reduced capacity on NICU eg space, staff, physical space Lack of awareness / recognition patient needs culture and treatment If barcode tampered with or obscured by patient label Delay in culture being processed by lab No lab staff available out of hours to process sample Delay in culture being reported by lab Delay between sample being received by lab and actually being processed Babies discharged 36 hours after starting antibiotics No lab staff available out of hours to report culture result
Process mapping on NICU Decision to treat Cultures taken Obtain Hosp no Call porters Leave NICU to go to lab
Process mapping in lab Out of hours: Culture taken to collection area In hours: Culture booked in Cultures processed onto BACTEC analyser Cultures automatically read every 10 mins Preliminary cultures: 48 hours Result automatically updates Win path and ICE system
Proforma for data collection Name DOB NHS no Time of birth Time of decision to treat if available Date & time culture taken Date & time hospital number obtaine d from midwive s Porters called? If so, what time Date& time culture left NICU Date & time culture received by lab Date & time culture processe d by lab Date & time 48hr result released Date & time abx stopp ed No of hrs delay Reason for delay in stopping abx eg high CRP, delay in getting culture result, clinically unwell
Results 20 randomly selected babies with suspected early onset sepsis One baby did not have cultures taken so n= 19 9 born during daytime hours (09.00-16.00) 10 born out of hours (16.30-09.00)
Time mapping on NICU Decision to treat Cultures taken Obtain Hosp no Call porters Leave NICU for lab Av 79m In hrs:31m, OOH 109m Av: 49m In hrs: 51m, OOH:49m Av: 25 m, In hrs: 22m, OOH 26m Av: 151m, In hrs: 105min, OOH: 183min
Time mapping in lab Average: 4 hours In hours: 3 hrs OOH: 5.5 hrs Cultures leave NICU In hours: Culture booked in Cultures processed onto BACTEC analyser Av: 90m In hrs:83m OOH: 95m Cultures automatically read every 10 mins Preliminary cultures: 48 hours Result automatically logged onto Winpath and ICE Average : 51 hrs In hours: 42hrs OOH: 52 hrs
Overall time mapping Average : 51 hrs In hours: 42hrs, OOH: 52 hrs Cultures taken Taken to the lab Booked in Culture Processed 48hr result Av: 4 hours Average 59 hrs
Conclusions Multiple steps required for culture processing Involves resources & staff from multiple teams Data showed significant delays occurred at the following stages: Cultures leaving NICU booked in by lab (4 hrs) From culture booking in to 48hrs result (51 hrs)
IDENTIFYING BOTTLENECKS NETWORKED SATELLITE BLOOD CULTURE ANALYSER BEDSIDE TRANSIT LAB BOOKING PROCESSES FINAL RESULT
CHANGE SATELLITE ANALYSER STUDY ACT QI CYCLE SMALL TESTS OF CHANGE & CONTINUOUS IMPROVEMENT AUDIT PROBLEM AIM MEASUREMENT P D S A
BENEFITS Early discharges ( 99 % of blood cultures negative, > 75 % babies clinically well with normal infection markers) Improved family experience Other benefits: Breast feeding Less nosocomial infections Less antibiotic side effects Less complaints
0.7 bed days saved = ( 420 per bed day* 0.7 * 4 babies/wk * 52) Cost of implementing change: Analyser lease = 6000 per annum/3 year lease Consumables = 2000 per year Savings: 61, 152 18000 = 37, 152/ year
Summary Quality improvement is NOT an accident Needs concerted efforts Everybody s business Understanding of methodology PDSA Change management Continual improvement
Thank you! Questions?