Patient safety

Patient safety is the prevention, reduction, and analysis of medical errors experienced by patients in the health care setting. Medical errors are events that expose patients to potential harm. Harm that occurs as a result of medical error is known as an adverse event. Medical errors often result from the interaction between system-associated risk factors (e.g., job-related factors or organizational factors) and person-associated risk factors (e.g., health care personnel-associated risk factors and patient-associated risk factors). Specific types of medical errors include diagnostic errors, medication errors, and communication errors. The prevention strategies that most effectively reduce medical errors include fostering a safety culture, ensuring usability of technology, implementing standardized protocols, and establishing incident reporting systems. Once a medical error has occurred, it should be disclosed to the patient and any other relevant individuals, and a medical error analysis should be conducted to identify the causes of the error and prevent a similar error in the future.

See also “Quality improvement” and “Navigating stressful situations in residency.”

• All medical settings globally: ∼ 5% of patients experience preventable harm. ^[1]
• Hospitals globally: Nearly 10% of patients experience adverse events, and approximately half of these are preventable. ^[2]
• Medical errors are the 3^rd leading cause of death (∼250,000 deaths a year). ^[3]

Hazards, risk, and risk factors ^[4][5]

Adverse events and medical errors ^[6][7][8]

Definitions

• Medical error: an action or failure to act that exposes a patient to possible harm in the context of medical care, regardless of whether or not harm actually occurs ^[6][9]
• Adverse event: an unintended harmful consequence of a medical treatment, which may or may not be the result of a medical error ^[8][10]

Classification of adverse events ^[6][7]

Classification of medical errors ^[6][8]

The categories of medical error listed here are not mutually exclusive; they may occur in conjunction with or complement each other.

Specific medical error types

General principles ^[24][25]

• Health care is a complex system in which the interaction between multiple factors influences the risk of patient harm.
• Complex systems (e.g., hospitals) consist of innumerable interacting elements (e.g., machines, staff, facilities).
  • The interaction of many individual elements introduces a certain degree of unpredictability (e.g., malfunction, illness) that makes these systems susceptible to failure (e.g., incorrect results, interruption of processes).
  • Characteristics of complex systems include:
    • Nonlinear processes
    • Multiple/circular causality
    • Multilevel cooperation
    • Open systems
    • Self-organization
    • Synergy

Workplace-associated risk factors ^[4][26]

Environment

• High noise level
• Poor lighting
• Improper room temperature (e.g., too hot or too cold)
• Severe weather
• Bad air quality
• Poor workspace design, e.g., as relates to:
  • Floor plan of wards (e.g., location of nursing station for optimal proximity to all patient rooms)
  • Available space
  • Furniture and room layout

Technology and tools

• Ineffective communication technology (e.g., nurse call systems)
• Malfunctioning information technology (e.g., electronic health systems)
• Limited availability of tools and equipment (e.g., a designated place with fresh gloves, bandages, syringes in every patient room)

Human resources and organizational factors

• Understaffing
• Scheduling errors
• Excessive workload (e.g., due to mismatched ratio of medical personnel to the number of patients)

Health care personnel-associated risk factors ^[4]

• Inexperienced clinician: a deficiency of practical skills and/or knowledge, predisposing to judgment and/or diagnostic errors
• Overcommitment ^[27]
  • Characterized by placing a large amount of time and effort into one's job, even when it might not be necessary
  • Has been associated with burnout
• Fatigue: Sleep deprivation impacts cognitive performance. ^[28]
• Alert fatigue: the tendency to become desensitized to and subsequently ignore alerts prompted by clinical decision support systems or medical equipment due to the excessive number or limited clinical relevance of the alerts in the past

Stress-related consequences ^[29][30]

• Ongoing stress can lead to:
  • Negative coping behaviors (e.g., alcohol and drug use)
  • Psychiatric illness (e.g., anxiety, depression)
  • Impaired cognitive function (e.g., decision-making abilities)
  • Negative impacts on patient care (e.g., medical errors)
• Examples of stress-related illnesses include: ^[31]
  • Burnout syndrome
  • Compassion fatigue, which develops following long-term exposure to patient distress and includes :
    • Physical and/or mental exhaustion
    • Reduced empathy (e.g., indifference to others)
    • Irritability and anger
    • Reduced work satisfaction

Cognitive biases ^[15][32][33]

• Confirmation bias (psychology): the tendency to favor evidence that supports preconceived notions and ignore potentially conflicting or problematic evidence
• Anchoring bias: the tendency to inappropriately rely on an initial perception or piece of information, which hinders later judgment when new information becomes available (e.g., favoring a diagnosis proposed earlier despite new evidence)
• Availability bias: the tendency to make judgments based on the availability of information from memory: (e.g., when a physician makes a premature diagnosis that comes to mind easily and quickly due to having seen several patients with a similar clinical presentation)
• Framing bias: the tendency to be influenced by how information is presented (e.g., the order of symptoms and/or emphasis placed on specific findings)
• Visceral bias (or affective bias): the tendency for clinical decisions to be influenced by positive or negative feelings towards the patient (e.g., doubts regarding symptoms when described by a patient with substance use disorder)
• Ascertainment bias (psychology): the tendency to base decisions on preset assumptions (e.g, gender bias, stereotypes)
• Gender bias: the tendency to base decisions on false assumptions about an individual's real or perceived gender
• Information bias (psychology): the tendency to collect more information than necessary for a decision ^[34]
• Aggregate bias (psychology): the tendency to assume that aggregated data (e.g., clinical guidelines) do not apply to the individual patient ^[34]
• Commission bias: the tendency to prefer action over inaction
• Omission bias: the tendency to prefer inaction over action
• Premature closure bias (search satisficing): the tendency of a clinician to prematurely accept a diagnosis without verifying it and/or considering alternative causes, which can lead to misdiagnosis and/or delayed diagnosis
• Overconfidence bias: the tendency of an individual to assume that their skills and knowledge are greater than they are in reality
• Zebra retreat bias: the tendency of a clinician to disregard a potential diagnosis because of its rarity, although evidence exists that supports the rare diagnosis
• Conjunction fallacy: the false assumption that the probability of a joint event (e.g., multiple clinical findings) is greater than the probability of any of the events occurring individually ^[35]
• Implicit bias
  • Subconscious attitudes and/or perceptions toward a specific group of people (e.g., based on age, weight, race, or gender) ^[36]
  • Can impact patient care (e.g., Hispanic and Black patients are less likely to receive appropriate analgesia compared to White patients, even when reported pain scores are identical) ^[37]
  • Solutions
    • Educate individuals and communities (e.g., implicit bias workshops)
    • Encourage individual and group awareness (e.g., small group debriefing, self-monitoring thought processes, considering others' perspectives).

Implicit bias impacts patient care and can perpetuate health disparities.

Patient-associated risk factors

• Extremes of age ^[38]
• Multiple comorbidities ^[39]
• Prolonged hospital stay ^[38]
• Cultural factors (e.g., religious rules that do not permit men to examine women or vice versa) ^[40]
• Some social determinants of health, including: ^[41][42][43]
  • Low level of health literacy and awareness, which is often associated with low socioeconomic status ^[12]
  • Non-White race
  • Limited English proficiency ^[44]
  • Medicaid insurance ^[45]

General principles ^[8][46]

• Error prevention is a core aspect of quality and patient safety.
• The goal of error prevention is to minimize medical errors and their effect on patients.
• Error prevention is most effective when its focus lies on system errors rather than individual errors.
• Fundamentals of error prevention include:
  • Hazard and risk awareness (see “Patient safety terminology”)
  • Personnel training (keeping up-to-date with current guidelines, standards, and procedures)
  • A safety culture that incentivizes openness in error reporting rather than focusing on disciplinary action against individuals who have made an error
  • Human factors engineering to minimize system-associated risk factors (e.g., usability testing to optimize functioning of equipment)
  • Systems for error monitoring and error reporting (see “Management of medical errors”)
• Specific strategies have been developed to address common categories of medical error such as medication errors, communication errors.

Person-focused strategies (e.g., education) are a weaker error prevention strategy than most system-focused strategies (e.g., forcing functions) as person-focused strategies usually rely on human memory to perform a task correctly and therefore have a higher likelihood of error. ^[47]

Safety culture ^[48][49][50]

• Definition: a workplace culture that promotes safety awareness, develops and implements measures for the maintenance of a safe work environment, and ensures that individuals can openly express safety concerns
• Goal: improving error prevention and error identification
• Key features
  • Creates awareness for risks and consequences of errors
  • Fosters a sense of responsibility in maintaining a safe work environment
  • Creates an environment in which employees are not afraid to report errors
  • Flattens steep hierarchies in order to:
    • Promote collaboration between different ranks and disciplines
    • Reduce the reluctance to speak up to superiors about concerns

Simulation-based training ^[51]

• Definition: a training method that uses simulations of real-life scenarios to teach skills through first-hand experience ^[52]
• Advantages
  • Hands-on practice and repetition
  • Safe learning environment
  • Real-time, actionable feedback
  • No risk to patients
  • Ability to simulate high-risk and/or rare scenarios
  • Allows interdisciplinary teams to practice working together
• Disadvantages
  • Expensive
  • Artificial simulation environment may not replicate normal working conditions ^[53]
• Resources
  • Low- to high-fidelity manikins
  • Anatomical models
  • Trained actors
  • Virtual reality
• Examples
  • Technical skills: sutures, airway management, urinary catheter insertion, central line placement
  • Nontechnical skills: communication, decision-making, teamwork
  • History taking and physical examination
  • Emergency response scenarios (e.g., BLS, ACLS, ATLS, obstetric emergencies, motor vehicle crashes)

Human factors engineering ^[50][54]

• Definition: a discipline that examines the interaction between individuals, their work environment, and the tools and processes they use for work, and uses that information to design safe and productive systems ^[4]
• Goals (in the context of patient safety):
  • Identify risk factors for medical error
  • Design systems that optimize performance and reduce the risk of error
• Methods
  • Forcing functions
  • Standardization
  • Universal protocol
  • Simplification
  • Usability testing

According to the human factors engineering approach, medical errors are not primarily the result of individual actions, but of the interaction between the various risk factors for medical error. ^[54][55]

Forcing functions ^[56][57]

• Equipment, process, method, or system features designed to prevent error by either:
  • Prohibiting certain actions
  • Requiring the performance of a particular action before another action
• A highly effective technique for minimizing adverse events because it inhibits a chain of action that causes or perpetuates error
• Examples include:
  • Anesthesia gas cylinders with gas-specific nozzles to prevent attaching the wrong gas tubing to a cylinder
  • Electronic medical record adaptions, e.g.:
    • Alerts that require the user to acknowledge potential adverse reactions or interactions before placing orders for certain medications
    • A requirement that the physician exits one patient's record before opening a second one
    • A requirement that the user enters a password and a reason for accessing confidential psychiatric records
• Excessive use can lead to alert fatigue and use of overrides; forcing functions should be reserved for actions that can cause serious harm to the patient. ^[58]

Strategies to reduce the risk of alert fatigue caused by forcing functions include reducing alerts that are clinically irrelevant and/or unhelpful, as well as assigning different levels of severity to alerts. ^[58][59]

Standardization ^[57]

• The development and implementation of standards that apply to various aspects of a process or system in order to improve reliability, efficiency, communication, and safety
• Examples of standardization:
  • Protocols and guidelines to help ensure a consistent level of quality and increase efficiency
  • Using the same equipment across a system by avoiding the need for workers to learn to use multiple different products
  • Checklists to help prevent common errors by making sure various steps of a process are completed

Universal protocol ^[60]

• A three-step process developed by the Joint Commission to prevent wrong site, wrong procedure, and wrong person surgery ^[23]
  • Should be performed before invasive medical procedures
  • Relies on active communication to provide an environment in which all members of a patient's care team can actively speak up about potential errors (see “Safety culture”)
• Steps: The first two steps should take place before the patient enters the operating room.
  1. Preprocedural verification of the patient's identity, the surgical site, and the surgical procedure
  2. Surgical site marking
  3. Time out: a preoperative pause conducted by the surgical team in the operating room
     • Goal: to prevent harm from a wrong procedure or a procedure at the wrong site or on the wrong patient
     • Method: the patient's identity, the procedure, and the surgical site are confirmed immediately prior to the start of the procedure or surgical incision

Simplification ^[4][8]

• Reduction in complexity of equipment, systems, and processes to increase efficiency and reduce the risk of error
• Examples include:
  • Computerization and automation of health information technology
    • Ordering laboratory tests electronically (i.e., CPOE)
    • Clinical decision support system (CDSS): a computer-based system designed to analyze data within electronic health records to assist health care professionals in clinical decision-making tasks (e.g., by reporting drug interactions). ^[61][62]
  • Readily available standard equipment (e.g., tongue depressors, latex gloves, scissors)
  • Multicompartment medication device

Usability testing ^[54]

• Testing the use of new equipment and systems with real users to proactively identify possible risks or negative consequences, including:
  • Incompatibilities between health care personnel and the equipment they use (e.g., surgical scissors designed with loops that are too small to handle comfortably and precisely)
  • Poor design that impedes workflow, forcing staff to adopt temporary solutions (workarounds) that may increase the risk of errors
• Examples include:
  • Testing a new format of CPOE to see if it increases confusion for ordering clinicians
  • Testing a new process for medication administration to see if it leads to the use of workarounds

• Estimated rate of inaccurate diagnoses: 10–15% ^[63]
• Diagnostic errors are more common in some patient groups, e.g. those of:
  • Older age ^[64]
  • Female gender ^[65][66]
  • Non-White race ^[67][68][69]

Older patients, female patients, and patients of color are at higher risk of experiencing diagnostic error. Thoughtful, deliberate thinking, with an awareness of potential biases, is especially important when caring for these patients. ^[64][65][67]

Communication errors can arise either from breakdowns in communication between health care providers and patients, or different members of the health care team. Effective communication in health care settings can substantially reduce the risk of medical errors. See “Communication during residency” for further information on improving communication. ^[14]

Communication errors between health care staff

See also “Workplace communication.”

Risk factors ^[74][76]

• Environmental distractions (e.g., loud noise)
• Hierarchical relationships (e.g., a junior resident communicating with a senior resident)
• Inappropriate quality or quantity of communication
• Transition of care between teams or health care settings.

Prevention strategies

• Communicate as clearly as possible.
  • Avoid medical abbreviations.
  • Provide written orders rather than verbal orders whenever possible.
  • Practice closed-loop communication. ^[77]
• For critical written results, in addition contact the clinician directly or by phone and utilize check-back to confirm results have been correctly understood. ^[77]
• Anticipate transition of care errors and proactively intervene.
  • Handoff-related errors ^[74][76]
    • Optimize handover environments and use standardized handoff tools.
    • See “Signouts” for further information.
  • Discharge errors ^[76][78][79]
    • Identify patients at high risk. ^[80]
    • Start early discharge planning with multidisciplinary team support.
    • Ensure patients and their primary care physicians receive clear and concise discharge paperwork.
    • See “Hospital discharges” for further information.

Communication errors between patients and health care staff

Poor communication between patients and health care staff can lead to medical errors and a reduction in treatment adherence.

Risk factors ^[81]

• Language or cultural barriers
• Limited health literacy
• Extremes of age

Preventive strategies

• Patient involvement in care, e.g.:
  • Shared decision-making
  • Patient access to electronic health records
  • Patient verification of identity and any further relevant information (e.g., site of surgery, planned procedure, consent) prior to administering drugs or inducing anesthesia
• Clear communication
  • Clarify the patient's preferred language and offer medical interpretation services.
  • Avoid medical jargon and long sentences.
  • Check understanding and encourage questions.
  • Provide written material that summarizes important information at a 5th–6th grade reading level.
  • See also:
    • “Key principles of communication and counseling”
    • “Communication during patient encounters”
• Ensuring patients understand the next steps when transitioning between care.
  • When patients are referred, ensure they understand:
    • The referral process
    • The expected time frame
    • What to do if symptoms worsen before the referral appointment
  • When patients are discharged:
    • Provide discharge counseling and a copy of their discharge summary (which should be concise and relevant).
    • Book a follow-up appointment.
    • Consider arranging a postdischarge follow-up phone call. ^[82]
      • A phone call within 2–3 days of discharge to promote continuity of care and support patients until their first follow-up appointment
      • May improve coordination of care, increase patient engagement, and decrease readmission rates
    • See also “Hospital discharges.”

Encourage questions from patients by asking “What questions do you have for me?” rather than “Do you have any questions?” ^[83]

Risk factors ^[12][84][85]

Medication errors may be caused by system-associated risk factors and health care personnel-associated risk factors, as well as the following patient-associated and medication-associated risk factors.

Patient-associated risk factors ^[12]

• Extremes of age
• Female gender ^[86][87]
• Polypharmacy
• Limited health literacy
• Cognitive impairment
• ICU patient ^[88][89]

Medication-associated risk factors

• High-alert medications: medications that carry a higher risk of significant patient harm when used incorrectly, e.g., ^[12][84]
  • Antithrombotics (e.g., warfarin, DOACs)
  • Insulin and sulfonylureas
  • Opioids
  • IV sedatives (e.g., lorazepam)
  • IV adrenergic agonists and antagonists (e.g., epinephrine and labetalol)
  • IV antiarrhythmics (e.g., lidocaine)
  • Neuromuscular blocking agents (e.g., rocuronium)
  • Parenteral nutrition
  • Hypertonic saline
• Commonly confused medications, e.g.: ^[85]
  • Amiloride and amlodipine
  • Bupropion and buspirone
  • Clonidine and clozapine
  • Diazepam and diltiazem

Prevention strategies ^[74]

Conservative prescribing practices ^[90]

• Consider nonpharmacological treatments.
• Assess for nonadherence before making any medication changes.
• Use caution when prescribing newly approved drugs; examine the evidence closely.
• Start new medications one at a time, if possible. ^[90]
• Adjust dosages cautiously, especially in vulnerable patients, e.g.:
  • Children
  • Patients with hepatic or renal impairment
  • Older adults (see “Principles of prescribing for older adults”) ^[91]
• Discontinue medications that are no longer necessary.
• Do not restart medications that did not provide therapeutic benefit in the past.

Assess medication adherence by first making a validating statement acknowledging that many patients have difficulty taking all their medications as prescribed every day, and then asking nonjudgementally about how many days in the past week the patient might have missed a dose of their medication. ^[83]

Safe prescription writing ^[92]

The following strategies are used to prevent transcription errors and dispensing errors.

• Standardized prescription order writing ^[92][93]
  • Use precise, simple language.
  • Name the drug, dosage, route, frequency, and duration of administration (e.g., nitrofurantoin, 100 mg, orally, twice a day, for 5 days).
  • Include the indication for the medication on the prescription as an additional safety checkpoint (e.g., prescribed for a lower urinary tract infection).
  • Avoid abbreviations (e.g., use of q.d., derived from Latin to mean daily) and acronyms that may be misinterpreted.
  • Where available, use computerized physician order entry (CPOE) to facilitate standardized prescription order writing and reduce transcription errors. ; ^[12][94][95]
    • A system in which electronically placed orders for medications, tests, procedures, and consults are directly transferred to the recipient
    • Can prevent errors due to poor handwriting or ambiguous abbreviations
• Tall man lettering: a strategy used to clearly differentiate medications that look or sound alike by capitalizing the unique letters of two similar drug names (e.g., DOBUTamine vs. DOPamine) ^[96][97]

The Joint Commission maintains a “Do Not Use” list of abbreviations in prescribing because of their potential for misunderstanding. This list includes U, IU, QD, QOD, MS, MSO4, MgSO4, trailing zeros (1.0 mg), and a lack of leading zeros (e.g., .1 mg). ^[98]

Medication reconciliation ^[74]

• Definition: a process performed during transitions of care (e.g., admission, transfer, discharge) to compare new medication orders with previous ones
• Goal: to avoid medication errors such as duplication, omission, dosing errors, and drug interactions
• Steps ^[74][99]
  • Identify previous medications.
  • Compare previous medications to medications that are currently ordered.
    • Identify discrepancies.
    • Assess whether each medication should be continued, discontinued, dose-adjusted, or restarted.
  • Create a new medication list for the patient, documenting the reasons for any changes made.
• Verifying home medications ^{[83][100][101]}
  • Try to obtain information from at least one source in addition to the patient (e.g., family member or pharmacy).
  • Start with open-ended questions.
  • Ask the patient to include:
    • All forms of medications (e.g., pills, injections, patches, creams, eye drops)
    • Over-the-counter medications, vitamins, and supplements
  • A simple method is to ask the patient to bring in everything they take (known as a brown bag medication review) ^[101]
  • For each item, determine:
    • The prescribed strength, dosage, route, and frequency (including any medications prescribed for use only as needed)
    • Any discrepancies between prescribed use and actual use; ask the patient why, when, and how they take it.
    • When the last dose was taken

When admitting a patient, ask when they last took their medications so that new orders can be placed for the correct time.

When transferring or discharging a patient, do not forget to look back at the home medication list obtained on admission and reconcile those medications with the current inpatient orders.

Patient education on medications ^[102]

Limited health literacy is associated with an increased risk of medication errors. See also “Communication errors.”

• Clearly explain using simple language:
  • Why the medication has been prescribed
  • How to take it
  • When dosages should be omitted or changed (e.g., sick day rules)
  • Potential adverse effects and drug interactions
• Check for understanding by asking the patient to summarize; correct any errors using the teach back method.
• Ensure patients know who to contact if they have any questions about their medications.
• Consider offering dose administration aids (e.g., pill boxes or blister packs) to improve medication adherence and reduce medication errors. ^[103][104]

Dispensing medications

• Consider using automated dispensing cabinets. ^{[105][106][107]}
• Do not store look-alike and sound-alike medications in close proximity.
• For high-risk scenarios , use independent manual double checks. ^[108]
• For medication use in the OR, preferentially use: ^[109][110]
  • Prepackaged single-use syringes or syringes prefilled/premixed by the pharmacy in an aseptic setting
  • Single-use vials where possible
  • Standardized preprinted labels stating the drug’s name and concentration
  • Standardized arrangement of trays and with medications placed so labels are visible

Prescribing cascade ^[111][112]

• Definition: a cycle that arises when symptoms resulting from an adverse reaction to a medication are treated with a new medication ^[113]
• Stages
  1. A prescribed medication (e.g., NSAIDs for musculoskeletal pain) causes an adverse effect (e.g., hypertension).
  2. The adverse effect is misinterpreted as a new condition, which leads to prescription of a new drug (e.g., an ACE inhibitor).
  3. The new medication causes new adverse effects (e.g., dry cough), triggering another cycle of the prescribing cascade.
• Prevention
  • Use conservative prescribing practices.
  • Perform periodic brown bag medication reviews.
  • Improve the identification of an adverse drug event.
    • Educate patients about potential adverse effects of their medications.
    • Proactively ask patients about new symptoms after a new medication is started.
    • Always include an adverse drug reaction in the differential diagnosis of a new symptom.
  • If an adverse drug event is identified, consider whether it may be appropriate to:
    • Discontinue or reduce the dose of the medication
    • Switch to an alternative medication, or to a nonpharmacologic intervention
• Examples ^[111][112]
  • NSAIDs → hypertension → antihypertensive treatment
  • Thiazide diuretics → hyperuricemia → gout treatment
  • Antipsychotics → parkinsonism → antiparkinson medication

• Definition: a condition that a patient develops while hospitalized (i.e., that was not present on admission), and which often represents a preventable adverse event ^[114]
• Epidemiology: ∼ 85 hospital-acquired conditions per 1000 discharges (as of 2017) ^[114]

Since 2008, the Centers for Medicare and Medicaid Services has reduced reimbursements to hospitals for the treatment of many hospital-acquired conditions in order to encourage evidence-based preventive measures. ^[117]

When a medical error has occurred, important next steps include mitigation of any patient harm, error reporting, transparency with the patient and family, and analysis of the error to reduce the risk of the error occurring again. Identified errors can then be used as the basis for quality improvement initiatives.

General principles ^{[8][118][119]}

• Individual providers should respond to adverse events immediately.
  • Implement corrective measures to stabilize patient and minimize additional harm.
  • Communicate the adverse event to the patient.
    • If an error has occurred, perform medical error disclosure.
    • If the cause of an adverse event is not immediately known, the clinician should inform the patient and maintain contact while investigations are being carried out.
• Institutions should encourage error reporting and provide systems for reporting and responding to adverse events.
  • Error reporting is more likely to happen in environments that focus on accountability and system errors rather than blame and punishing individuals.
  • Promote use of incident reporting systems.

Medical error disclosure ^[8]

• Disclose error to the patient and, if necessary, a supervisor and the administration.
• The following points should be considered for optimal error disclosure:
  • Clearly admit an error has occurred.
  • State the course of events leading up to the error.
  • Explain the consequences of the error, both immediate and long-term (if applicable).
  • Describe corrective steps.
  • Express personal regret and apologize.
  • Allow ample time for questions and continued dialogue.
• For a suspected error by a colleague: ^[120][121]
  • Discuss the circumstances of the potential error with the colleague.
  • If an error was made, encourage the colleague to report the error via the standard protocol in place.
  • If there is disagreement regarding whether an error occurred or whether it needs to be disclosed, consider discussion with other individuals at the institution (e.g., medical director, ethics committee) to determine next steps.
• If needed, provide psychological and emotional support for the consequences of medical errors, both for the providers and patients involved (see also “Navigating stressful situations in residency”).

Regardless of the outcome of a treatment, a physician must inform the patient immediately if an error has occurred and disclose the nature of that error.

When disclosing a medical error to a patient, apologize directly and without qualification. Don't say: “I'm sorry you feel like that,” “I'm sorry you took it that way,” “I'm sorry, BUT...”. Instead say: “I'm sorry this happened,” “I'm truly sorry for the distress caused,” “I'm sorry, we have learned that...” ^[122]

Incident reporting systems (IRS) ^{[8][123][124]}

• Overview
  • IRS provide a means of reporting errors, near misses, and other safety concerns.
    • Voluntary reporting systems: usually involve a confidential report from an individual involved in the event
    • Mandatory reporting systems: state-based systems for reporting of deaths or other serious adverse events by health care organizations
  • Analysis of the reports collected helps to identify risks within organizations.
• Goal: developing and implementing strategies to address identified risks and prevent further errors
• Advantages
  • Useful in identifying commonly occurring and local systems errors (e.g., medication errors due to trailing zeros on labels), for which substantial data can be collected
  • Aggregation of data with the help of IRS facilitates the analysis of more severe adverse events (e.g., never events), for which only limited data exists.
  • Conclusions drawn from IRS data can be shared within and/or across organizations to help identify risks and prevent future error on a larger scale.
• Limitations
  • Voluntary reporting systems are biased in that they rely on self-reporting by health care personnel.
  • IRS cannot be used to measure safety in general and/or changes over time. ^[123]
  • Organizations cannot be compared with one another based on data from IRS.
  • Organizations might not have enough resources to thoroughly review the large number of reports that IRS generate.

Only a small number of errors (7%) are reported using an IRS. Encourage reporting by creating an easy-to-use, confidential IRS and fostering a culture that is open to learning from mistakes. ^[123][125]

For the legal consequences of medical error, see “Medical malpractice.”

Introduction ^[13]

• Description: Medical error analysis involves the investigation of known and potential causes of error in a system to prevent future errors.
• Goal: to minimize the number of system errors by implementing safety measures and checkpoints
• Types
  • Retrospective: an analysis of past errors done to gain knowledge of the types and causes of error (e.g., root cause analysis)
  • Prospective: an analysis of potential errors done to assess risks and hazards that may lead to errors in the future (e.g., failure mode and effects analysis)

Swiss cheese model ^[4][126]

• Every safety system is imperfect and will, therefore, have flaws that allow hazards to provoke errors and, potentially, harm.
• The Swiss cheese model illustrates how a multilayered safety system (multiple slices of cheese) can help prevent flaws (holes in each slice) from allowing hazards to pass through the entire system, i.e., if a hazard manages to pass one layer, the next will likely block it.
• Alignment of flaws in the individual layers will allow a hazard to pass through and allow the error to occur.

Root cause analysis ^[127]

Definition

The retrospective analysis of an error used to identify its (root) causes and develop measures to prevent its recurrence

Process

1. Identify the medical error: “What happened?”
   • Determine the circumstances of the error.
2. Determine the root cause of the error: “Why did it happen?”
   • Retrospectively analyze all possible risk factors for medical error that could have led to the error considering documentation, the equipment/drugs used, and the environment the patient was in.
   • Examples of root causes: slippery floor, lack of isolation rooms, nonadherence to a standardized safety protocol, malfunctioning equipment or technology
   • See “Methods” below.
3. Determine what could have prevented the error and develop measures to prevent it from occurring again in the future: “What can be done to prevent the error from recurring?”
   • Address root causes.
   • E.g., update technology, introduce double checks, use checklists, educate staff on new policies

Methods ^[128]

• Fishbone diagram (Ishikawa diagram or cause-and-effect diagram) ^[4]
  • A quality control technique used to break down the most significant factors that may have contributed to the error or problem into categories for improved visualization
  • Steps
    1. Identify the problem.
    2. Identify the main categories of factors contributing to the problem and draw branches off the main arrow to represent these.
    3. Repeat the method and draw branches off the main potential factors to identify more specific factors within each category.
    4. Use this data to specify the problem (if possible) and find possible multifactorial solutions.
  • Example
    • Problem: patient received the wrong medication
    • Main potential factors contributing to the problem: equipment, people, environment
    • Specific factors:
      • Equipment: malfunctioning medication barcode requiring a manual workaround
      • People: understaffing
      • Environment: multiple distractions
• Five Whys approach ^[4]
  • An interrogative technique used to identify the root cause of a problem by asking “why” five times.
  • Example: The patient received the wrong medication.
    • 1^st why: Why did the patient receive the wrong medication? → The nurse did not complete patient identification.
    • 2^nd why: Why was the identification information incomplete? → The patient did not have a wristband.
    • 3^rd why: Why did the patient not have a wristband? → The wristband was removed for a procedure and not replaced.
    • 4^th why: Why was the wristband not replaced? → The wristband printer was out of service.
    • 5^th why: Why was the printer out of service? → The IT staff did not have the resources to fix the printer (this would be the root cause).
• Process mapping
  • The visualization of processes using tools such as flowcharts to establish, analyze, and improve processes
  • Process maps outline each step within a process, identify task owners and relationships, and provide details regarding expected timelines. In the context of root cause analysis, the mapping of processes or the analysis of established process maps can help determine causal relationships between errors at different points of a process and identify the step in a process that requires correction to address an error.
  • Example
    • Anticoagulant blood testing process: patient arrives at clinic → patient information is processed during check-in → blood sample is taken → INR test is performed → patient is given results → clinic schedules a follow-up appointment with the patient
    • Error: A large number of patients never come to the follow-up appointment.
    • Root cause analysis: Determine which step of the process is responsible for the error (may be more than one).
    • Example causes: Patient email database is faulty and patients receive neither their results nor their invitation to schedule a follow-up appointment; a bottleneck in INR testing leads to excessive waiting periods, causing patients to seek a new provider.

Failure mode and effects analysis (FMEA) ^[4][128]

• Definition: : a prospective analysis used to identify potential risks and hazards by assessing the failure modes (i.e., the manner in which failure occurs) of a system
• Goal: : Determine the effects that potential failures may have and develop measures to prevent these failures from occurring.
• Steps
  1. Choose a specific system component.
  2. Perform process mapping.
  3. Identify potential failures (failure mode).
  4. Identify why it could go wrong (failure causes).
  5. Identify the consequences of potential failures (failure effects).
  6. Prioritize the hypothetical failures by their probability of occurring and the severity of their effects.
  7. Implement corrective measures proactively.
• Example
  1. Component: medication administration
  2. Process mapping: Outline the steps involved in the medication administration process (e.g., medication ordering, preparation, and dispensing).
  3. Failure mode: dispensing error (e.g., wrong medication given to a patient)
  4. Failure causes
     • A nurse confused one medication with another due to similar appearance or name.
     • The packaging contained the wrong medication.
     • The medication accidentally got swapped during dispensing.
  5. Failure effects: ranging from no effect to severe patient harm/death
  6. Prioritization
     1. A nurse confused one medication with another due to similar appearance or name.
     2. The medication accidentally got swapped during dispensing.
     3. The packaging contained the wrong medication.
  7. Implementation of corrective measures
     1. Introduce measures to reduce the probability of one medication getting confused with another, e.g., changes in storage, tall man lettering.
     2. Implement measures to prevent medication getting swapped during dispensing, e.g., physical barriers between the single medications, not dispensing the medications for all patients on the ward in one run.
     3. Educate staff to double-check packaging before removing medication from the wrapper.

Morbidity and mortality review (M&M) ^[129]

• Definition: a retrospective analysis of patient care errors and complications performed in a clinical peer review activity format
• Goals
  • Identify medical errors.
  • Determine areas for improvement.
  • Increase situational awareness among team members.
• Procedure
  • Meetings are held at regular intervals to confidentially present, review, and discuss selected cases among peers.
  • Traditionally, the meetings were held within a single department and focused on individual actions that led to an adverse event.
  • Today, meetings more often include health care providers from different departments and focus on systemic causes for medical errors rather than the errors of individuals. ^[130]
  • The meeting's proceedings, findings, and recommendations (as well as any documentation thereof) are protected from legal discovery and inadmissible in malpractice lawsuits in most (though not all) states. ^[131][132]