CUSP: Learning from Defects. CUSP Boot Camp March 2013

Transcription

1 CUSP: Learning from Defects CUSP Boot Camp March 2013

2 Objectives Participants will be able to: 1. Describe how defect analysis is a foundational element of CUSP. 2. Discuss latent and active failures and how to assess for both during a defect analysis. 3. Describe the five steps of learning from defects. 4. State at least four probing questions to get the most from a defect analysis.

3 How do you identify defects on your unit? Incident reports Staff reports/ feedback Patient/family reports Monitors Safety rounds Near-miss reports Gemba walks Physician feedback Staff Safety Assessment survey What else?

4 What is a Defect? Defect: Any clinical or operational event or situation that you would not want to happen again. May include events that you believe actually caused harm or put patients at risk for significant harm. Johns Hopkins s Learning from Defects tool from CUSP Toolkit at Accessed Oct 8, 2012.

5 Definitions Defect Analysis: -the process of analyzing a defect to determine its root cause. Defect Prevention: -the process of addressing root causes of defects to prevent their future occurrence. Lanier K. Experiences In Root Cause Analysis and Defect Prevention Methods. Presentation publically available at Accessed April 2010.

6 Basic steps in the L-from-D Process 1. What happened? 2. Why did it happen? 3. How will you reduce the likelihood of it happening again? 4. How will you know the risk is reduced? 5. With whom should you share the learning?

7 Why bother? Reimbursement systems are being reengineered to reward the best performers (no margin, no mission, no salary, no job ) Growing transparency of outcomes Defects are expensive in many ways Excellence in care provided - learning from each defect is the best way to prevent it from happening again The Golden Rule-it s what we d want done for us or the person we love most!

8 It s All About Prevention Learning from defects reduces the likelihood of event occurring again and causing harm. Learning may occur in two areas: o System design o Human behavior Using a tool to aid learning provides structure and consistency Learning from each CAUTI (or any type of harm) allows learning from common cause variations. Learning from defects that might contribute to CAUTIs (any type of harm) allows prevention further upstream from the harm to make care even safer.

9 Background Concept #1 Normalized deviance: (Workarounds, short-cuts) Long-term phenomenon in which individuals or teams repeatedly accept a lower standard of performance until that lower standard becomes the norm. I.e. what starts out as a deviation from SOP becomes routine practice and is normalized. It becomes acceptable and may no longer even be recognized as deviation. (Just Culture terms for this: behavioral drift, at-risk behavior) Allowed because we get away with it most of the time

10 Challenger Tragedy NASA team accepted a lower standard of performance on solid rocket booster O-rings. Got comfortable with seeing occasional O-ring damage. By the time Challenger launched, O-ring damage was considered a marginal problem. In 1986, O-ring damage had become an accepted normalized deviance. On 1/28/86, Challenger exploded soon after take-off, killing all 7 crew members.

11 What are some common healthcare normalized deviances? Inconsistent hand hygiene practices. Not disinfecting aspiration ports on Foleys when obtaining specimens. Not scrubbing the hub on CVC ports when accessed. Hanging the Foley bag on the walker or gait belt when ambulating the patient or throwing it up on the stretcher during transport or hanging it from the side-rail. Diapering patients with diarrhea. Not giving pre-op antibiotic in a timely manner to ensure tissue concentration is at peak when incision made. What else?

12 Excuses to Deviate The rule(s) is stupid May not appear to have value Rule or reason for it is unknown Not valued New technology Complexity of work Rule does not apply to this patient Rule(s) doesn t apply to me I know a better way Healthcare worker is not the one at risk in most situations. Competing priorities: o Cost and time, o Culture (politics, hierarchy, lack of communication openness, etc.) It won t hurt anything this one time Assumed or minimal consequences.

13 Background Concept 2: Errors in Complex Systems

14 Active Failures Active failures: errors that occur at the point of contact between a human and some aspect of a larger system (e.g., a humanmachine interface). They are generally readily apparent (e.g., pushing an incorrect button, not securing a Foley) and almost always involve someone at the frontline or sharp end.

15 Latent Failures Latent Failures: less apparent failures of organization or design that contribute to the occurrence of errors or allowed them to cause harm to patients. Occurrences are at the blunt end but may result in active failures on the sharp end. AKA: Accidents waiting to happen. Reason JT. Human Error. New York, NY: Cambridge University Press; 1990.

16 Swiss Cheese Model

17 Background Concept 3: Analysis of a defect should occur as soon as possible after the defect is recognized. The best light is often that from a burning bridge. ~Don Henley

18 A Review: What are the 5 Steps in the LfD Process? 1. What happened? 2. Why did it happen? 3. How will you reduce the likelihood of it happening again? 4. How will you know the risk is reduced? 5. With whom should you share the learning?

19 Tips For HAI LfDs It is rare to find the exact source in a nonoutbreak-related infection. HAIs are usually the culmination of many infection prevention breaches. (More than one hole lined up in the Swiss cheese ) Your job in a defect analysis is to search for the contributing factors that were most likely to have culminated in the infection in this particular patient.

20 Learn about the organism and how it s typically transmitted, if there has been an organism identified. Tap into your Infection Preventionist to help the group process through the chain of infection to help in search for contributing factors. Tip for HAIs

21 Need to learn: What is organism? Is this particular organism a MDRO? Where is it commonly found in nature? How is it transmitted? How common is it in your organization? How long can it live in the environment?

22 Step 1: What Happened? Use a standard event investigation format. No perfect form so use what is familiar and accepted that accomplishes the goal. Must establish a chronological order of events and data related to the event. Usually requires some research and walking the process (gemba). The same person should not do all the research/walking the process. Include assessment for active and latent failures.

23 Case Study 44 y.o. female, Jenny, w/ colon CA admitted to your med/surg ICU on 7/12/12 with diarrhea and severe dehydration. She is S/P partial colectomy and has a colostomy. She was in process of completing her 3rd round of chemotherapy prior to admission. CL inserted 7/12 in Rt. subclavian for hydration and to monitor CVP. On 7/19, she became hypothermic w/a temp of 95 7 F and hypotensive. BC Gram stain + for Gram positive cocci. BCs + for Enterococcus faecalis on 7/20. CXR negative for pneumonia. Other cultures. CLABSI diagnosis made and antibiotic started.

24 Case Study The IP notified the unit manager immediately upon identifying the BSI. The unit manager began the LfD process by reviewing the chart, interviewing staff, and pulling the unit s CUSP team together. For this particular defect, UM has added: o Two RNs that had been caring for Jenny o Two CNAs that had been caring for Jenny o Intensivist o Risk Manager o Infection Preventionist

25 Polling Question 3 Assess the process so far. Is there anything you would do differently? A. Nothing looks like everything has been done appropriately so far. B. I would have more than one person seeking information on the case. C. I would add additional key stakeholders D. Answer B & C

26 What have we learned from Event Report Tool? 1. E. faecalis grown in blood from BC done 7 days after admission and 7 days after CL was inserted. 2. Patient was immunocompromised: leukopenia (WBC 2 on adm.) 3. Patient was having severe diarrhea 4. Pool nurse staffing used more than desired 5. MD does not participate in discussion for daily review of CL necessity. 6. All CL insertion and maintenance protocols consistent w/evidencebased science and followed appropriately. UM states, Guys, I think this is a case where we ve done everything humanly possible to prevent this infection. She s just a very sick patient and is immunocompromised.

27 Polling Question 4 Do you agree with the UM? A. No, they ve not looked at everything they need to yet. B. Yes. All CL prevention strategies were in place, there is nothing more they could have done to prevent it. C. I m not sure.

28 Dig deeper Remember the Chain of Infection

29 Other issues to explore Heavy contamination of environment w/feces occurs multiple times per day with bouts of diarrhea and EVS not present to clean after each episode. How does this get cleaned up? Does it? Bath basins as a reservoir? Frequency of linen changes? Patient/visitor hand hygiene? What are hand hygiene and scrub the hub practices really like, inside the patient rooms? Etc., etc.

30 Additional Information Needed Gemba walks for ground truth: HH, scrub the hub, bathing practices, equipment and environmental cleaning practices Focused literature review on interventional personal hygiene for best practice and compare to current practice Review of chart by another person to see through a different set of lenses Interview EVS, review EVS policies with them, invite them to join the team Interview patient (if able) and visitors to learn what they do/ what they ve been taught.

31 Step 2: Why Did It Happen? 1. Review events from event report with unit CUSP team 2. Identify any gaps 3. For each gap identified, ask Why? and then for each reason generated, ask Why? at least four more times or until the group feels the root cause has been found. (Keep pushing to the 5 Whys to ensure root cause is determined and latent failures are recognized if present.) Further research/lit review may be necessary to gather ground truth, determine why it happened, and if there were any gaps with best practice. This is another time when gemba walks, staff and/or patient interviews, and literature reviews may be indicated.

32 Step 2: The 5 Why s an Example Gap: Direct patient-contact environmental surfaces are not cleaned in occupied beds. 1. Why? EVS policy prohibits. 2. Why? Risk of cleaning chemicals causing respiratory and/or topical issues with patients. 3. Why? Direct or close contact w/hospital-grade chemicals can be toxic and patients are already compromised. 4. Why? There are no physical or spatial barriers between the patient and the surface disinfectants currently used. 5. Why? Not possible, given the physical proximity. Would need to protect the patient s entire body and airway. 6. Why? Spray disinfectants are currently the primary type used and thus are not usable in close proximity to the patient. 7. Why? Cost of other types of products prohibitive.

33 Step 3: How can we reduce the chances of it happening again? 1. Prioritize most important contributing factors. 2. Develop interventions (countermeasures) to defend against the most important contributing factors. 3. Then, rate each countermeasure on its ability to mitigate the root cause and on the team s belief that the countermeasure will be executed. 4. Make an action plan for 2-5 of the highest scoring countermeasure. Must have many different perspectives. Must engage leadership.

34 Contributing Factors Importance to current event, 1 (low) to 5 (high) Importance to future events 1 (low) to 5 (high) Ease of Resolution 3 =easy, 2 = fairly easy 1=hard Total Score Asst does not wear max barriers because it is not part of protocol =8 Unit s HH compliance low due to? =11 All staff unable to verbalize correct drsg change procedure due to inconsistent orientation for regular and nursing pool staff. Swabbing hub is not consistently done because of perceived time constraints and alcohol wipes not always readily available. Direct patient-contact environmental surfaces are not cleaned in occupied beds due to cost of non-aerosol surface disinfectant products = = =12

35 Countermeasures Consider safe design principles: Standardize eliminate steps when possible Create independent checklists Learn when things go wrong (analyze defects) Safe designs apply to technical and team work. Brainstorm strategies and consider: 1. Ability to mitigate error. 2. Strength of countermeasures to prevent error. 3. Ease of implementation. (Resources and will necessary to implement.)

36 Rank Order: Strength of Error Reduction Strategies 7. Forcing functions and constraints Designed so that errors are virtually impossible or very difficult to make. Example: removing potassium chloride for injection from all patient care areas. Lessons human fallibility by limiting reliance on memory. 6. Automation and Computerization Example: Placing alcohol wipe in each Pyxis drawer with IV medications so wipe issued with each dose of medication. Standardizes processes/materials/resources to promote awareness of evidence-based practice and increase consistency between providers. It makes defect detection possible also. 5. Standardization and Protocols Example: Standardized CL insertion kit/cart. Standardized CL dressing change kits. Creates redundancies in system to look for errors 4. Checklists and double-check systems Example: Checklist to guide central line dressing changes. Requirement for an RN to recalculate doses of heparin in pediatrics behind pharmacist. Sets standards and expectations. Helps define what defects are. 3. Rules and policies Example: CL maintenance policy based on current science to prevent CL complications such as infection. 2. Education/Information Education and sharing of information Example: Staff education on CL maintenance policy. 1. Instructions to be more careful, vigilant. Admonitions to be more careful to prevent error. This is very traditional in the mental model where individuals have total control of their actions. Example: Management counsels an individual to be more careful after staff member has made a med error.

37 Countermeasures (interventions to reduce the risk of the defect) Direct patient-contact environmental surfaces are not cleaned in occupied beds due to cost of non-aerosol surface disinfectant products. Ability to mitigate the contributing factor, 1 (low) to 5 (high) Team believes the countermeasur e will be implemented and executed, 1 (low) to 5 (high) Strength of Strategy, 1 (low) to 7 (high) Total Perform cost-benefit analysis of cost of non-aerosol cleaning wipes compared to cost of CLABSI. Trial disinfectant wipes on unit for three months, monitoring HAI outcomes, staff and patient satisfaction. Engage EVS to implement more rigorous cleaning protocols. Educate staff on importance of environmental hygiene and how long organisms can live on inanimate surfaces

38 Step 4: How will we know the risk is reduced? Assess outcome and process data (HAI rates, compliance to cleaning protocols, monitoring of specific process changes, etc.) Talk to staff to get their perspectives. Talk to patients/families to get their perspectives. Do gemba walks for direct observation. Feedback from patient safety rounds.

39 Step 5: How do we communicate our findings and to whom? Internal communications Other units Medical staff Risk Management Environmental services External communications PSO Other hospitals/healthcare settings Collaborative networks

40 Avoid Common Pitfalls in LfD and DO Use a scientific investigation approach Make sure more than one person does the investigation. Consider the medical record as only one source of information. Involve direct care providers that cared for the specific patient. Look for active AND latent errors. Involve physicians. Involve patients and/or family members at the very least to gain more information from their perspectives. Avoid getting caught in analysis paralysis. Implement stronger countermeasures than just educate and, admonishments to providers to Be more careful. Learn from individual defects as well as in aggregate. Share what you ve learned with others.