Diagnostics for Patient Safety and Quality of Care

Transcription

1 Session L17 The presenters have nothing to disclose Diagnostics for Patient Safety and Quality of Care Carol Haraden Kate Jones Pat O Connor Orlando, FL December 09, 2012

2

3 Objectives P3 Describe the current diagnostic approach of their organization Discuss the use of the data gained from diagnostic tools to prevent, detect, and mitigate harm to patients Appraise the usefulness of the presented diagnostic tools to improve the safety and quality of their organization over time

4 Agenda 4 Introduction to learning lab 1:00-1:05 Diagnostics introduction 1:05-1:10 Mortality Diagnostic1:10-1:45 GTT- 1:45-2:45 Break 2:45-3:00 Waste tool- 3:00-3:45 Using What is Learned from the Tools 3:45-4:25 Wrap up 4:25-4:30

5 5 What data do you have about how people are harmed and why they die? So what do you do with all this information?

6 Once we have information 6 Organization of information, display, prioritization and plan of action.

7 7 GTT Clinical data Mortality review Waste tool Concurrent review Risk management information

8 8 Mortality review Waste tool GTT deaths HARM Risk Management information Concurrent review What are important and recurrent issues? High volume and high impact with success.

9 How can we learn about our system performance?

10 Diagnostic Journey People die and are harmed unnecessarily every day in our hospitals. In order for us to understand this, we need a diagnostic journey that moves out of a model for judgment and into a model for learning.

11 11 Mortality Diagnostic

12 The Mortality Diagnostic 2x2 Matrix Review most recent 50 consecutive deaths. Place them into a two by two matrix based on: - Was the patient admitted for palliative care? - Was the patient admitted to the ICU? Focus your work initially on boxes that have at least 20% of your mortality.

13 Diagnostic The 2 x 2 Matrix Admitted to the ICU? Yes No Admitted for Palliative Care Only? Yes No Box #1 Box #2 Box #3 Box #4

14 The Mortality Diagnostic - Failure to Recognize, Plan, Communicate Analyze deaths in box 3 and 4 for evidence of failure to: recognize, communicate, plan. This will help you understand the local environment.

15 Recognize, Communicate, Plan Failure to Recognize: Any situation in which a patient has died and there was evidence that an intervention could have been made anytime prior to the patient s death Failure to Plan, such as: diagnosis, treatment, or calling a rescue team. Failure to Communicate: Patient to staff, clinician to clinician, inadequate documentation, inadequate supervisor, leadership (no quarterback for the team), etc.

16 The Mortality Diagnostic - The Impact of Care Evaluate ALL deaths in box 3 and box 4 to assess the estimated impact of our care on mortality *As you review the deaths in box 3 & 4, ask yourself the questions honestly (focusing on learning, not judgment): Was perfect care rendered? If perfect care wasn t rendered, could the outcome of death have been prevented if the care had been better?

17 The Mortality Diagnostic - Evidence of Adverse Events Analyze deaths in box 3 and 4 for evidence of adverse events using the Global Trigger Tool. This will give some further direction to local problems.

18 US 2X2 Table Aggregate 64 Hospitals ICU Admission Comfort Care 86/3175 3% (0-14%) Non Comfort Care 1161/ % (10-72%) No ICU Admission 402/ % (0-40%) 1526/ % (18-76%)

19 Unknown unknowns There are known knowns. These are things we know that we know. There are known unknowns. That is to say, there are things that we know we don't know. But there are also unknown unknowns. There are things we don't know we don't know. Donald Rumsfeld

20 Knowing what might go wrong...

21 Reduces the chance of error

22

23

24

25 Reliability First review (all deaths) 6/53 identified by 1 st Assessor 5/53 confirmed by 2 nd Assessor Second review (unexpected deaths) 9/37 identified by 1 st Assessor 9/37 confirmed by 2 nd Assessor

26 Mortality Review May cases (from Jan 2011) 14/53 excluded (death anticipated) 8/14 from Nursing Homes 37/39 death unexpected VTE prophylaxis prescription rate Time to first Consultant review Time to first dose antibiotic (if sepsis) Data collection form revised further

27 Case Study hrs 84 year old was admitted with diarrhoea and vomiting hypoglycaemic. treated with IV fluids and oral glucose. Arterial blood gases were checked and showed a partially compensated metabolic acidosis with base excess -11. No explanation was documented Blood results showed acute kidney injury grade 2. The blood gas results handwritten in the notes and a serum lactate 9.0 but no comment was made as to the significance. The drug chart confirms that no antibiotics were prescribed hours after arrival at hospital the patient suffered a fatal cardiac arrest. A discussion with the family was documented which states they were happy for a death certificate to be issued and she was not referred for a Post Mortem. The cause of death on the death certificate was pulmonary thromboembolism.

28 Case Study 1: Key points This case represents a failure to recognise severe metabolic acidosis due to lactate representing hypoperfusion most likely due to septic shock. Early aggressive fluid resuscitation and IV antibiotics are essential treatments

29 Case Study 2 An 84 year old man with previous pulmonary thromboembolism was admitted by his GP with acute breathlessness and chest pain, suspected further PTE. On arrival his observations were stable, SEWS = 0. The working diagnosis was?acs,?pte. His first ECG was performed after 2.5 hours and there were no diagnostic changes. 4 hours after arrival he was given Aspirin, Clopidogrel and Fondaparinux. At that time he was noted to appear unwell and routine observations were repeated for first time since admission SEWS now = 7. Medical staff reviewed now noted to be severely hypoxic, given stat dose Dalteparin and commenced CPAP although patient rapdily deteriorated with ventricular arrhythmias and fatal cardiac arrest.

30 Case Study 2: Key points When the GP letter suggests pulmonary embolism the diagnosis should be given strong consideration and if in doubt treated. A 4 hour delay between initial observations is too long in an emergency admission, even if the initial observations are normal. In addition a 2.5 hour delay in performing an ECG in suspected ACS is unacceptable.

31 Case Study 3 An 87 year old lady blind and lived at home, semi-independent with support from relatives developed a sore wrist and was prescribed a regular NSAID by her GP Two days later (a Public Holiday) her pain remained uncontrolled and oramorph was prescribed. The next day she was admitted to hospital with severe acute kidney injury, serum creatinine 422, potassium 6.6, and noted to be receiving Ibuprofen 400 mg tds and also on Lisinopril 5 mg daily. The AKI recovered with rehydration and stopping nephrotoxic drugs and she appeared to be making reasonable progress until she had a sudden acute deterioration on day 7 with dyspnoea and hypoxia and died within 24 hours. The presumptive diagnosis was a pulmonary embolus. Review of the drug chart showed that DVT prophylaxis had not been prescribed although there was no reason stated in the medical notes.

32 Case Study 3: Key points NSAIDs are a high risk prescription in the elderly and co-prescription with angiotensin-blocking drugs can cause severe acute kidney injury. All medical admissions are at risk of venous thromboembolism and should be prescribed DVT prophylaxis unless there is a contra-indication.

33 Local use Understand and track our own mortality rate Develop local specialty team reviews of cases of concern Recruit more Second Assessors Develop comparison indicators Address recording, coding and data quality issues Use established service meetings to share Case Studies and highlight themes.

34 Reporting Monthly reporting to from ward to Board trustees Ensure transparency of methodology Develop appropriate presentation, publication and practical dissemination of results Meet needs of different audiences in relation to interpretation, use and context ( public, Board, Chief Executive,C suite, Healthcare Professionals

35 35 Global Trigger Tool

36 Why use trigger tools Traditional reporting of errors, incidents or events does not reliably occur Voluntary methods underestimate events Trigger tool does not rely on technology Establishes baseline of adverse events for hospital

37 Global Trigger Tool Review chart for triggers that are sensitive and specific for harm Find a trigger- was there harm? Not all triggers mean there was harm!

38 Global Trigger Tool Modules Cares (General) Critical Care Medication Surgery L&D ED

39 Definition of Harm Unintended physical injury resulting from or contributed by medical care that requires additional monitoring, treatment or hospitalisation or results in death

40 Category of Harm E Temporary harm, intervention required F Temporary harm, initial or prolonged hospitalization G Permanent patient harm H Life sustaining intervention required I Contributing to Death

41 C1 C2 C3 C4 C5 C6 C7 C8 C9 C10 C11 C12 C13 C14 C15 Cares Module Triggers Transfusion or use of blood products Any Code or arrest Dialysis Positive blood culture X-Ray or Doppler studies for emboli Abrupt drop of greater than 25% in Hg or Hemtocrit Patient fall Decubiti Readmission within 30 days Restraint use Infection of any kind In hospital Stroke Transfer to higher level of care Any procedure complication Other

42 20 notes selected by medical records

43 20 minutes per case by one reviewer

44 Steps in review process LOOK FOR TRIGGERS Coding summary Discharge summary Lab results Radiology reports Procedure notes Nursing/multidisciplinary notes

45 STOP! STOP Ask yourself: Ask yourself Would I want it to happen to me? Would I want it to happen to me? An adverse event is an event that results in harm from the patient s point of view

46 Endoscopy Example of a trigger: Transfer to higher level of care Post procedure somnolent and hypotensive (BP 80) transferred to ICU Placed on Bi-Pap Received standard Demerol and Versed for procedure Given Romazicon; stayed in unit 12 hours.

47 Global Trigger Tool Examples Readmit within 30 days with recurrence of abscess right hip. Readmit next day w/ileus s/p exp lap for tumor. Stopped lasix-acute renal failure. Readmitted in 30 days for wound revision due to incisional seroma. Volume Depletion with altered mental status caused by Lasix -resulted in hospital admission. ARF due to nephrotoxicity due to combination of ACE and NSAIDS taken at home. Ischemic colitis had rt hemicolectomy. Unresponsive, coded. Decreased loc & sats on Morphine PCA. Rec'd Narcan.

48 Consecutive Adverse Events 1-Iatrogenic pneumothorax 2-Sternal wound infection 3-Thrombophlebitis 4-Post Surgical bleed 5-ICU delirium 6-Nosocomial pneumonia 7-Theophyline toxiciy/arrythmia 8-GI bleed 9-Iatrogenic pneumothorax 10-ICU delirium 11-Fluid overload 12-Oversedation 13-Urinary obstruction 14-ICU delirium 15-Rash 16-Aspiration pneumonia 17-Nausea 18-Pulmonary embolus 19-Nosocomial pneumonia 20-Sternal wound dehiscence 21-Dialysis induced hypotension 22-Severe hypotension with NTG 23-Renal failure post surgery 24-ICU delirium 25-Sternal wound infection

49 Cost Analysis Variable Favorable/(Unfav) Charge Net Revenue Direct NOI Pt. # Impact Impact Impact Cost Impact Impact entire stay $57,484 $15,525 $16,700 ($1,175) extra hospital days $3,428 $0 $1,170 ($1,170) extra ICU days $10,422 $0 $2,650 ($2,650) extra ICU days $7,930 $0 $2,500 ($2,500) Total ICU costs $1,502 $0 $865 ($865) Total Hospital Costs $21,500 $3,958 $6,430 ($2,472) extra ICU days $6,592 $0 $2,695 ($2,695) Indwelling Cath, 8 vent hours, 1 critical care day $8,768 $0 $3,245 ($3,245) extra ICU days $9,180 $0 $2,345 ($2,345) days ICU care $13,756 $0 $4,485 ($4,485) No additional cost n/a n/a n/a n/a extra ICU days $19,341 $0 $7,150 ($7,150) extra ICU days $19,032 $0 $3,730 ($3,730) 2 extra ICU days and return to OR $16,436 $0 $5,125 ($5,125) extra ICU days $15,090 $0 $4,408 ($4,408) no additional cost n/a n/a n/a n/a extra ICU days $4,086 $0 $1,619 ($1,619)

50

51 Trigger Tool Drill Down 51 When you find harm reoccurring and you want to understand it more deeply, Pull an enriched sample of patients: from the population of patients who may have a condition predisposing them to a particular harm: diabetics: hyperglycemia; patients with atrial fibrillation: anti-coagulation Using codes to find patients with particular harm like pressure ulcers and falls Other?

52 Look at the enriched sample 52 What can you learn about: - the real occurrence rate? - the clinical care processes underlying the harm - particular characteristics of the patients that suffer that harm?

53 Waste Identification Tool

54 Design Elements Identifies waste from perspective of the shop floor Engages front line staff Use of the Tool is simple Provides an infrastructure for continuous and deliberate waste identification and reduction sensitive to the political and economic environment

55 Modules Ward waste related to bed utilization Patient Care - unnecessary patient care Diagnosis unnecessary diagnostic tests and procedures Treatment treatments given that are not supported by scientific evidence and therefore may be unnecessary (protocols, pathways, guidelines) Patient waste from the patient s perspective

56 Key Questions Ward module Hospital bed use related to flow represents a huge opportunity for wasted beds. Patient in a bed with a completed discharge waiting to go. Bed being held for a patient (medical or surgical) Bed being held for any kind of admission or transfer

57 Ward Module continues Imaging procedures not able to be done or delayed Surgery delayed due to blood tests not done Consultation not done within your institutions period of time

58

59

60 Waste Identification Tool Worksheet Ward Module Clinical Waste Unit ICU) Reviewers Date Wed, April 10 Bed ID Waste Waste Waste Streams Yes No Nosocomial Infection Adverse Drug Event Procedure Complication Unnecessary Hospitalization Flow Delay Clinical Care Delay Comments T-1 X Awaiting PICC IR T-2 X X Lap chole comp T-4 X X Futility EOL,family T-5 X X T-7 X X No Plan T-8 X X No drip on floor. Pt walking around ICU T-9 X T-10 X X No (insulin) drip on floor B-S X X No OR til Fri B-T X X X Card. Consult, no family meeting, EOL B-O X X Pt fell, No OR til Fri B-S X C-M X X End of Life (EOL) C-A X C-J X X Trach Collar trial not done V-R x X Awaiting trach & G-Tube V-A X X X Inf & EOL futility V-A X X X Pneumothorax, & EOL futility V-P X X X Graft inf. & hematoma Total # of wasted beds the numerator 16 Total # beds reviewed the denominator 19 % waste 84%

61 Operational Waste Examples Bed is empty, but no demand Bed empty for reasons such as equipment failure/bathroom not working/no hot water Patient in a private room for medical purposes and the other bed or beds are not able to be used (ex: isolation room with 2 beds in one room, but no one occupying the other bed) Room is dirty and waiting to be cleaned Patient rooms used as storerooms or offices Patients in bed for outpatient services such as administration of chemo or blood

62 I Day, 6 Units (10 mins each Unit) Medical floor Unit 1 24 patients -4 waste criteria Unit 2 21 patients -5waste criteria Unit 1 24 patients -2 waste criteria Unit 1 22 patients -5 waste criteria Unit 1 24 patients -3 waste criteria Unit 1 24 patients -2 waste criteria Opportunity for efficiencies 21! Lost hours organisational/patient waiting/delays 312Hrs

63 Patient Module Interview 54 year old male with recent hip replacement: EKG done the day of surgery when had been done in internists office two days before surgery. Sequential compression devices which kept falling off and did not ever seem to work. Physical therapy continued walking patient even after he was walking on his own without difficulty. Portion sizes for meals continued to be large even though he requested smaller portions at least three times.

64 Customizing the Waste Tool Customization for any setting as long as: Front-line approach is used. Evaluation is qualitative (responding yes or no to whether waste is present at the time of review). Waste types are clearly identified and defined.

65 Example of a Customization A large health care system customized Tool for use in its office practices. Waste types tested include: room contains non-functioning equipment room used inappropriately patient in room but MD is not in clinic patient in room dilating patient in room waiting for diagnostic testing patient in room longer because initial paperwork is incomplete

66 Using What is Learned From the Tools

67 After you use the tools Tools can be used to help us understand where we are with system level outcomes We need to then measure and improve the processes related to the harms and problems (waste) we are finding

68 A Story

69 The journey. What are our problems seen from the diagnostics? How would we prioritise these? How reliable are we? What are the issues? How do they relate to our strategic plan? How do we deploy the work in a way that we meet our most important aims?

70 The journey. What are our problems as seen from the diagnostics? How would we prioritise these? How reliable are we? What are the issues? How do they relate to our strategic plan? How do we deploy the work in a way that we meet our most important aims?

71 The journey. What are our problems seen from the diagnostics? How would we prioritise these? How reliable are we? What are the issues? How do they relate to our strategic plan? How do we deploy the work in a way that we meet our most important aims?

72 Tools 1. Pareto 2.? 3.?

73 The journey. What are our problems seen from the diagnostics? How would we prioritise these? How reliable are we? What are the issues? How do they relate to our strategic plan? How do we deploy the work in a way that we meet our most important aims?

74 What do we mean by process reliability? Chaos; failure in more than 20% of opportunities 80-90% reliability the process fails less than one in ten times 95% reliability - five in 100 times the process fails. 99% reliability - one in 100 times the process fails well designed system with low variation and co-operative relationships

75 A quick diagnosis of reliability? Less than 80% process reliability Five frontline users can not easily articulate the process 80-90% reliability less than1 in 10 times the process fails. 95% reliability - 1 in 20 times the process fails Five frontline users can easily articulate the process 99% reliability - 1 in 100 times the process fails, a well designed system with low variation and cooperative relationships

76 The journey. What are our problems seen from the diagnostics? How would we prioritise these? How reliable are we? What are the issues? How do they relate to our strategic plan? How do we deploy the work in a way that we meet our most important aims?

77 Making the links Large System Aims Meso-System Projects

78 The journey. What are our problems seen from the diagnostics? How would we prioritise these? How reliable are we? What are the issues? How do they relate to our strategic plan? How do we deploy the work in a way that we meet our most important aims?

79 The Intuitive Structure Large System Tier 1 Meso- System Meso- System Meso- System Tier 2 Project Project Project Project Project Project Project Tier 3 Project

80 Example: System Medication Safety SYSTEM Medication Safety Tier 1: Big Dot Hospitals Rehab Offices Tier 2: Portfolio Med. reconciliation Standardized dosing Self med Family Capacity Correct list Patient capability Drug availability Tier 3: Projects

81 Example: Hospital Medication Safety Hospital Medication Safety Tier 1: Big Dot Med Surg Medication process Pharm Tier 2: Portfolio Med. reconciliation IV pumps Prescribing Administration Admix Standardized dosing Tier 3: Projects Drug Availability

82 So what next? What do we think we need to improve in order to improve our outcomes? Where is the greatest opportunity? What process do we need to make more reliable?

83 Patient Observation Chart Audit Audit Guidelines: Collect a sample of observation charts for patients on the ward and answer the questions on the checklist. Sample size: Frequency: Scoring and individual chart: If all items are answered 'YES' then the chart receives a score '1'. If any item is answered 'NO' then the chart receives a score '0'. Calculating the overall audit score: Sum the individual chart scores. Convert to a percentage: Sum of chart scores Number of charts audited x 100 Chart Answer YES or NO only 1 Standard Has the chart got an addressograph or full name and hospital number? 2 Is the page numbered at the top of the chart? 3 Have doctors requested frequency of Observations? 4 Does the chart indicate the frequency of observation? 5 Is the temperature recorded correctly and as requested? 6 Is the BP recorded correctly and as requested? 7 Is the pulse recorded correctly and as requested? 8 Are respirations recorded correctly and as requested? x 9 Are O2 SATs recorded correctly and as requested? 10 Has the patient triggered in TPR or BP? 11 If yes, has the appropriate action been taken? 12 Was action taken within 1 hour? 13 Has pain assessment been documented? 14 Is urinalysis recorded? 15 Is bowel action recorded? CHART SCORE 0 1 Ward: Date: AUDIT SCORE Audit carried out by: