Discussion - Part 1 (Root Cause Analysis)

IDENTIFYING THE CAUSES OF ADVERSE EVENTS

The mechanism commonly used to investigate serious adverse events is root cause analysis (RCA). RCA will be discussed only briefly in Case-A8. 

• Background information
• Typical steps
• Further reading

BACKGROUND 

RCA is a structured retrospective investigation that has gained  prominence with the rise of the quality movement,  hemovigilance*, and the requirement to analyse sentinel events*.

* Hemovigilance: systematic monitoring of adverse events of transfusion, for example:

• Canada: TTISS

• European Haemovigilance Network

• UK SHOT Program

• USA: MERS-TM

* Sentinel events: unexpected events involving death or serious injury, or the risk thereof

Purpose. RCA's purpose is to identify the causes, including basic and contributing factors, of an adverse event or "near miss," and the actions necessary to prevent them from happening in the future. Put another way, the goals of RCA investigations are to identify

• what happened (unsafe act)

• how it happened (human or technologic behavior)

• why it happened (contributory factors)

• how to prevent it from happening again

Selection. Not all adverse events merit RCA. Some merit a routine investigation in which basic facts are gathered and others merit an expanded investigation requiring more time and resources.  An expanded investigation using RCA is based on risk to the patient and risk to the organization. See assessing risk from MERS-TM, a no-fault event reporting system for transfusion medicine developed in the USA.

ROOT CAUSE CHARACTERISTICS

Root causes have certain common characteristics.

Cause and effect. To be a root cause, the putative cause must show cause and effect, i.e., show a clear link from the root cause to the adverse event being investigated. Suppose that a root cause is identified as insufficient training. The root cause should demonstrate that a specific training deficiency caused the error. For example:

• The nurse administered blood to the wrong patient due to not having been trained in procedures for checking patient identity prior to initiating  transfusions.
• The physician failed to prescribe irradiated blood due to not having been trained in current transfusion guidelines for preventing graft-versus-host disease.

Duty to act. Failure to act in a certain way is causal only if there is a pre-existing duty to act. For example:

• Doctor's failure to prescribe irradiated blood products can only be causal if there was a requirement  to prescribe irradiated products in the first place. 

• Technologist's failure to look up the patient's history in the LIS can only be causal if there was a requirement  to do a history check. 

The duty to act usually arises from professional standards and practice guidelines.

Other characteristics of RCA

• RCA is interdisciplinary and involves those closest to the process.

• RCA asks what and why until all aspects of the process are reviewed and all contributing factors are identified. Digging deeper into causes requires investigators to give up preconceived biases deriving from past experiences and the natural tendency of busy people to find a reason quickly so as to conserve time. Like icebergs, many true causes of  transfusion errors are below the surface and not immediately apparent.

• RCA focuses on systems and processes, not individual performance, which fits with the quality system blame-free approach to error management. 

Most  medical errors are not due to incompetent workers but rather due to flaws in the system. These flaws are referred to as "latent errors" and represent breakdowns in the system that made the error more likely to happen.  Since the system contributes to most medical errors, the tendency to blame the individual healthcare provider diminishes.

• RCA involves quality improvement by identifying changes that could be made in systems and processes  to improve performance and reduce the risk of adverse events or near misses recurring. Once identified and implemented, follow-up is needed to assess effectiveness of the corrective action.

TYPICAL STEPS 

RCA is a formal process involving highly structured reporting mechanisms, such as those used in the MERS-TM system. For an overview of the MERS-TM project, see 

• Presentation by Dr. Harold S. Kaplan (Principal Investigator)

In MERS-TM,  root causes fall into standardized categories:

• technical (equipment, software, forms)
• organizational (policies, procedures and protocols)
• human causes (knowledge-based, rule-based, and skill based -see Table 1)

Using standardized categories  facilitates being able to identify and compare

• adverse events regionally, nationally, and internationally

• trends in significant types of adverse events and their causes

• preventative measures

Regardless of the system used, RCA generally includes the following steps:

1. Gather information. May involve interviews; review of patient charts, nursing notes, medical notes, blood transfusion request forms, transfusion service worksheets and related specimens and donor units, information systems (hospital and laboratory); and more. 

2. Chart the causal factors (factors that if not present would have prevented the adverse event or reduced its severity).  Often there is a proximate cause (immediately visible) and a series of contributing factors or antecedent events. Involves charting the sequence of events that led to the adverse event and the conditions surrounding them.  Hypothetical example:

• Patent requiring irradiated RBC receives receives un-irradiated blood.  

  • Proximate cause: Technologist failed to look up patient's history in the laboratory information system

  • Contributing factors: Error occurred on evening shift when staffing was short due to last-minute absence of ill staff member; shift was particularly busy due to coping with urgent blood requests for a patient with a ruptured aortic aneurysm; technologist was a new staff member working first shift on her own

Identifying causal factors can be done using tools such cause and effect diagrams, flowcharts, why-why diagrams, etc. See

• Examples of RCA tools
• MERS-TM causal tree

3. Identify the root causes. Involves identifying the underlying causes for each causal factor. 

4. Recommend corrective actions. Depending on the root causes, examples of corrective action many include individual retraining, group education sessions, developing mistake-proofing tools, revising work flow patterns, and improving communication mechanisms.

5. Implement recommendations. 

6. Assess effectiveness of corrective actions at appropriate intervals. 

7. Consult guidelines and experts as appropriate throughout the investigation.

TraQ self study questions  

1. Why is it important to use standard terminology for root causes?

2. Suppose that a RCA revealed that a transfusion reaction occurred for the following cause: The nurse failed to notice that a unit of RBC had been out of the refrigerator and left sitting at room temperature for 3 hours and started to transfused it instead of returning it to the transfusion service.  How may such a root cause be coded using the MERS-TM codes?

MORE DISCUSSION...

• Part 1: RCA: <--You are here
• Part 2: Testing errors
• Part 3: Risk management
• Part 4: Supervision of students
• Part 5: Regulations and standards

FURTHER READING

Transfusion

Callum JL, Kaplan HS, Merkley LL, Pinkerton PH, Rabin Fastman B, Romans RA, et al. Reporting of near-miss events for transfusion medicine: improving transfusion safety. Transfusion 2001;41:1204–11. 

Kaplan HS, Callum JL, Fastman BR, Merkley LL. The medical event reporting system for transfusion medicine: Will it help get the right blood to the right patient? Transfus Med Rev 2002 Apr.;16(2):86-102. 

MERS-TM website

Myhre BA, McRuer D. Human error-a significant cause of transfusion mortality. Transfusion 2000 Jul;40(7):879-85.

Siegenthaler MA, Schneider P, Vu DH, Tissot JD. Haemovigilance in a general university hospital: need for a more comprehensive classification and a codification of transfusion-related events. Vox Sang 2005 Jan;88(1):22-30

General

Barach P, Small SD. Reporting and preventing medical mishaps: lessons from non-medical near miss reporting systems. Br Med J 2000;320: 759.

Institute of Medicine. The chasm in quality: select indicators from recent reports  

Chassin MR, Becher, EC. The wrong patient. Ann Intern Med 4 June 2002 Jun2;136(11):826-33.

Tools & Resources

• Canada: TTISS (Transfusion Transmitted Injuries Surveillance System)

• European Haemovigilance Network

• MERS-TM

  • MERS-TM causal tree 
  • Presentation by Dr. Harold S. Kaplan (Principal MERS-TM Investigator)
  • Tutorial on assessing risk to determine if RCA is needed (MERS-TM)

• Examples of RCA tools

• UK SHOT Program

Copyright © 2005 - 2010, TraQ Program of the British Columbia Provincial Blood Coordinating Office, a program of the Provincial Health Services Authority. All rights reserved.