REAL-WORLD DECISION-MAKING

In telephone triage, nurses must repeatedly make decisions in a matter of minutes based on limited information. Information may be partial or inaccurate and circumstances may involve life-and-death decisions—a high-stakes activity. One of the pitfalls of telephone triage is time pressure, which can lead nurses to perform cursory assessments and jump to conclusions, resulting in suboptimal outcomes.

Decision-making is made more complex by sensory deprivation, conflicting goals (such as call quotas vs. quality interactions), “noise” (irrelevant data, long-winded histories), interruptions, and multitasking (thinking, listening, talking, questioning, reading, writing, synthesizing information, pattern recognition). The nurse must focus on the meaningful bits of information (salient information), recognize patterns, estimate symptom urgency, and formulate an impression or working diagnosis.

How do clinicians make decisions under such conditions of uncertainty and urgency? How does decision-making on the phone differ from critical thinking at the bedside?

Experience and Decision-Making

Certain qualities and skills have been studied and found to be essential for nurses working in the field of telephone triage. Nurses with more years of clinical experience have more confidence in making decisions regarding patient assessment and disposition. Clinical judgment, experience, and use of critical-thinking skills all contribute to decision-making accuracy when working with patients in the practice of telephone triage. For this reason, nurses who enter this field most often need to have at least three years of nursing experience or more to be candidates for telephone triage nursing positions (Manetti, 2018).

Nurses often use intuition based on this previous experience to help guide decision-making. Klein (2003, 2010) advises learning to detect problems through emotional cues—a “gut feeling” when something is not right. He recommends developing an active stance, so that if something does not make sense, it acts as an alarm that is not to be dismissed. He also suggests becoming conscious of organizational barriers such as rigid procedures or institutionalized inertia. Finally, he suggests reframing the situation and consulting with colleagues to review with fresh eyes.

Good decision-making relies on this sort of balance of conscious and instinctive thinking. Reducing complex problems to their simplest elements aids in decision-making. It is also essential for nurses to continuously improve critical-thinking skills by avoiding behaviors such as stereotyping others, resisting change, and seeking conformity (Manetti, 2018).

Heuristics, or Rules of Thumb

Many experts recommend using heuristics, or “rules of thumb,” as practical decision-making tools. Rules of thumb are defined as “general principles regarded as roughly correct but not intended to be scientifically accurate” (Merriam-Webster, 2020). They are considered easily applied procedures for approximation—an educated guess, intuitive judgment, or common sense. Rules of thumb may be used to expedite decision-making, guide decisions, and reduce error.

Medical researchers have explored how skilled workers develop mental short cuts/heuristics, finding that rules of thumb are useful and even necessary to guide decisions under conditions of urgency and uncertainty (Manetti, 2018).

A study of groups of ICU nurses, firefighters, and others who make decisions under pressure found that those professionals do not logically and systematically compare all available options. In real life, that methodology is too slow. Instead, they quickly size up situations and act, drawing on experience and intuition. Rules of thumb often underlie their expertise and intuition (Manetti, 2018). Likewise, both emergency medical dispatchers and nurses employ these mental shortcuts and intuition to simplify relevant information and lessen required mental strain (Clawson & Democoeur, 2020; Wheeler, 2017a).

Examples of several types of rules are described below. Many of these rules of thumb represent expert nurses’ collective “pearls of wisdom” in telephone triage. They can be used by novice practitioners to improve their decision-making proficiency and efficiency. These examples are only a partial list.

CARDINAL RULES

Cardinal rules include associated strategies to avoid root causes of error (i.e., errors of communication, assessment, continuity of care, informed consent, human error). Following are example of cardinal rules, with the types of errors they are intended to avoid:

• Always err on the side of caution; when in doubt, bring the patient in sooner rather than later. (Continuity)
• “When in doubt, send ’em out.” (Continuity)
• Always speak directly to the patient; if “too sick to talk,” bring them in. (Assessment, communication)
• Always obtain the age of the patient. (Assessment)
• All frequent calls (within hours or days) are a red flag. Always ask how many calls the patient has made regarding this problem. (Communication, continuity)
• Treatment delayed is treatment denied. (Continuity)
• Trust but verify. Always update, correct, and confirm the back-story (patient history). (Assessment, communication)
• Always remain suspicious of the “nondiagnostic diagnosis” (patient’s self-diagnosis). (Assessment, communication)
• Always remain suspicious of the previous diagnosis (at recent ED or office visit); it may be wrong or complications may have arisen. (Assessment, communication)
• Always remain on the line with callers in crisis. (Continuity)
• Always treat flu-like symptoms with suspicion. (Assessment)
• Beware the middle-of-the-night call; it may be a red flag. (Assessment, communication)
• Speed does not equal competence; avoid premature closure. (Continuity)
• Time is tissue, time is muscle (with MI or CVA symptoms). (Continuity)
• To err is human, to delay is deadly. (Continuity)
• Beware of “failure to improve” on current prescription (antibiotic, antipsychotic, pain medications). (Assessment, human error)
• If a symptom (or symptoms) is unlike any you have experienced before, make the call, get a “reality check.” (Assessment, human error)
  (Clawson & Democoeur, 2020; Wheeler, 2009, 2017a)

AGE-BASED RULES

Pediatric age-based rules include:

• “Kids get sicker quicker.”
• Always err on the side of caution with children, especially with infants and toddlers.
• Infants under 3 months with fever of 38 °C or 100.4 °F should be seen immediately.
• Pediatric populations are at greater risk for hypothermia and hyperthermia; the younger the patient, the greater the risk.
• Assess all sick children for possible dehydration and sepsis.
• Assume any symptom of sexually transmitted infection (discharge, lesions) in a child to be sexual abuse until proven otherwise.
• All parents have the potential to physically abuse their children at some time.
• All sudden confusion in children is considered emergent.
• Always elicit an immunization history; lack of or inadequate immunizations place a child at risk. Appointment is required for serious delays in completing the schedule of immunizations.
• Under 4 years of age: symptoms tend to be very generalized; over 4 years of age: symptoms tend to be more specific.
• Depressed teenagers are at risk for suicide.
  (Wheeler, 2009, 2017b)

Geriatric age-based rules include:

• The older the patient, the greater the risk of hypo- or hyperthermia.
• Assess all sick elderly for possible dehydration and sepsis.
• All sudden confusion in a frail older adult is considered emergent.
• Greatest suicide risk is in white males, over 65 years of age, widower, retired, or jobless.
• Incontinence in the elderly may be related to urinary tract infection.
• Developmentally disabled populations typically age prematurely.

(Wheeler, 2009, 2017a)

SYMPTOM-BASED RULES

General symptom-based rules include:

• All severe pain should be seen within eight hours or less.
• All first-time seizures must be seen.
• All rashes are considered contagious until proven otherwise.
• Once an ectopic (pregnancy), always an ectopic.
• Any bleeding in pregnancy is an ectopic until proven otherwise.
• Presentations may be atypical, silent, or novel/unique, especially with children and older adults.
• Remain suspicious of “flu” symptoms, which might be symptoms of MI, sepsis, or other serious conditions.
• The vaguer the symptoms, the greater the need for good data collection.
• Beware of pain that awakens the patient or prevents sleep at night.
• Beware of afebrile pelvic inflammatory disease symptoms (possible ectopic pregnancy or ovarian cyst).
• Beware of shoulder pain with or without abdominal pain in women of childbearing age (possible ectopic pregnancy).
• Epigastric pain in males over 35 years and females over 45 years is considered an MI until proven otherwise.
• Symptoms should improve after 24 to 48 hours on antibiotics.
• Rectal temperature of 100 °F in an infant of 3 months or less must be seen immediately.
• Extremes of outside temperature often trigger medical symptoms.
  (Clawson & Democoeur, 2020; Wheeler, 2009, 2017a)

Chest pain-related symptom-based rules include:

• The first symptom of an MI is often denial.
• Smokers who have chest pain are more likely to die and die suddenly (within the hour) of MI.
• Chest pain in men over 35 years or women over 45 years is suspicious.
• Time is heart muscle (possible MI); patients treated within the first hour have a substantially improved outcome.
• A little chest pain may be as bad as a lot.
• Any chest pain in a high-risk caller should be treated as MI until proven otherwise.
• Beware atypical or novel presentations.
• MI in women, diabetics, and older adults may present as vague, silent, or atypical symptoms.
• Patients over 70 years typically do not experience chest pain.
• Some minority and low-literacy callers may fail to recognize acute symptoms, report fewer symptoms, or attribute them to other causes, due to a variety of reasons (cultural, health literacy, and financial).
• All chest trauma is considered urgent until proven differently.
  (Clawson & Democoeur, 2020; US DHHS, 2015; Wheeler, 2009, 2017a)

Distractors, or “Red Herrings”

A major task in decision-making is to determine which data are relevant and which are not. Data must be collected, considered, weighed, and even ignored in order to perform pattern recognition and arrive at a proper disposition. For example, key pieces of contextual information—age, gender, and previous medical history—are always salient; key symptoms may be salient; and other more general and nonspecific symptoms may be given less weight.

Some information is irrelevant and must at times be consciously ignored in order to come to safe decisions. Such data is called a “red herring” and diverts the nurse from more significant data. Red herrings can originate from many sources. Both patients and nurses may misinterpret symptoms or miscommunicate. Red herrings may cause the nurse to jump to conclusions, rely on stereotypes, or end the call prematurely.

Examples of common sources of red herrings are described below:

• Age and gender. With chest pain, nurses may stereotype and discount symptoms of MI because the patient is “too young” or the “wrong sex.” With common sexually transmitted infection symptoms, nurses might stereotype the patient as “too old” or “too young” to have an active sex life.
• “Nondiagnostic diagnosis.” A patient’s interpretation of symptoms and symptom acuity may not be accurate. For example, a patient who calls in with “the flu” may actually be experiencing symptoms of sepsis or MI.
• Previous or recent medical diagnoses. A patient who is immediately postoperative may call with symptoms of severe nausea and vomiting. The nurse may prematurely conclude that the symptoms are due to effects of anesthesia and fail to explore or assess the patient further for possible complications, leading to a possible delay in care. Or a patient who was “seen recently in the ED” may now be experiencing new, unrelated symptoms, which must be re-evaluated onsite.
• Existing treatment plan appears to be incorrect or ineffective. The nurse or patient may not recognize that new symptoms might be due to complications. Either of them might ascribe it to a failure of treatment or medication, i.e., inadequate pain medication in the case of uncontrolled post-op pain. A patient who is “on an antibiotic” should not lull the nurse into thinking that it is effective (as with MRSA).
• Denial, downplaying, explaining away. For example, the first symptom of an MI is denial.
• Nonacute initial presentations. The rule of thumb to beware of the developing disease applies particularly in cases of abdominal pain, respiratory problems, diarrhea, nausea and vomiting, fever, or marked change in activities of daily living (ADLs). For example, what starts out as vague abdominal pain with low-grade fever may quickly develop into the classic picture of appendicitis.

Cognitive Biases

Being aware of and avoiding one’s cognitive biases is another important aspect to effective decision-making. Common cognitive biases include:

• Confirmation bias, or selective search for evidence. Tending to gather facts that support certain conclusions while disregarding other facts that support different conclusions
• Premature termination of search for evidence. Accepting the first alternative that looks like it might work (i.e., jumping to a conclusion)
• Recency. Placing more attention on more recent information and either ignoring or forgetting more distant information
• Selective perception. Actively screening out information that one does not think is important (e.g., stereotyping of patient or symptoms)
• Inertia. Being unwilling to change old thought patterns in the face of new circumstances
• Wishful thinking or optimism bias. Wanting to see things in a positive light, which can distort one’s perception and thinking (over-reassurance)
• Anchoring and adjustment. Allowing initial information to shape and unduly influence one’s view of subsequent information (closed-mindedness)
• Source credibility bias. Rejecting something if one has a bias against the person or group to which the person belongs; accepting something if one likes the person (prejudice)
  (Adapted from Plous, 1993)