Have a personal or library account? Click to login
Management reasoning scripts: Qualitative exploration using simulated physician-patient encounters Cover

Management reasoning scripts: Qualitative exploration using simulated physician-patient encounters

Open Access
|Jun 2022

Figures & Tables

Table 1

Differences between illness scripts and management scripts

Characteristic

Illness script

Management script

End product

End product is a diagnosis

End product is a management plan

Cognitive representation of end product

Concrete, objective, single correct solution

Abstract, conceptual, multiple correct solutions

Temporal relationship

Retrospective (story [temporal evolution] of patient’s illness up to point of clinician’s involvement)

Prospective (guides temporal evolution of clinician’s future actions)

Script goal

The script is often synonymous with the diagnosis

The script is the pathway that leads to a management plan

Script activation, selection, and instantiation

Multiple candidate diagnoses are activated.

The most likely diagnosis (“general clinical picture of disease”) is selected and instantiated (populated with typical and atypical features of individual patient’s story)

Multiple candidate pathways are activated.

The best pathway is selected and instantiated (populated with known and assumed features of individual patient’s problem, comorbidities, preferences, etc.). Occasionally, a highly-developed script may lead by default to a specific management plan

Dual process thinking

Often remains largely Type 1 (non-analytical) thinking

Nearly always involves Type 2 (analytical) thinking

Key features

Predisposing conditions;

faults; consequences

Problem to be solved; management options; preferences, values, constraints; education needs; interactions; encounter flow (see Tab. 2)

How script is developed/built: Exposure to numerous varied cases enables …

Recognition of atypical presentations of an illness

Tailoring of the pathway (and thereby the subsequent plan) to the unique needs of the individual patient and context

Rate of development

Faster to develop because there are fewer permutations of a single diagnosis

Slower to develop because there are numerous permutations for each problem (varying comorbidities, preferences, interactions, etc.)

Table 2

Key features of the management script and example (hypothetical) instantiation

Feature

Potential script elementsa

Example of an instantiated scriptb

Problem to be solved

– Diagnosis

– Patient medical history/context (medical facts [not preferences]): comorbidities, allergies, disease severity, etc.

– Level of specificity can vary (“chest pain” or “myocardial ischemia” or “occlusion of the left anterior descending artery”)

– 44-year-old man with newly diagnosed HTN

– Comorbidities include obesity and impaired fasting glucose

– Allergy to sulfa

– Patient [does | does not] have edema. Potassium levels are [normal | high | low]

Management options

– Drugs

– Non-drug treatments

– Diagnostic tests

– Consultations

– Benefits

– Costs

– Side effects

– Monitoring and follow-up

– Hydrochlorothiazide, lisinopril, and amlodipine are top drug options [(unless edema or abnormal potassium is present)]

– Lifestyle measures including dietary change, exercise, and weight loss are essential

– Current lifestyle approach [is optimal | suggests minor improvements possible | suggests major change required]

Preferences, values, constraints

– … Of patient

– … Of providers

– … Of system

– … Confirmed, to be confirmed, assumed, or imposed

– Confirmed: Patient understands that HTN is important to treat. Nurse can provide education on diet and blood pressure monitoring

– To be confirmed: Patient [does | does not] want to try lifestyle measures a little longer before starting drug treatment

– Assumed (not typically confirmed): Patient wants to be treated with drug, is willing to take daily drug, can afford drug that costs $ 10 per month, and can return for periodic follow-up

– Imposed: Clinician is running behind schedule and feels time pressure that may limit capacity for education and shared decision-making

Education needs

– Before decision-making (what is going on, implications, prognosis)

– During decision-making (options)

– After decision-making (next steps)

– Before decision-making: what is HTN, what are long-term effects of untreated HTN, what are benefits and costs of prolonged treatment of HTN?

– During decision-making: what benefits, costs, and side effects are likely for this particular patient? What can he do to exercise and lose weight?

– After decision-making: does patient understand the illness and management plan (i.e., confirmation of understanding), when will he return for follow-up, what questions remain in his mind, does he know how to check his own blood pressure?

Interactions

– Human-human (communication, negotiation, shared decision-making; with patient, nurse)

– Human-computer (EHR, knowledge resource)

– Human-system (care pathway, insurance preapproval)

– Check EHR to confirm potassium is normal

– Invite patient to join in decision making

– Patient [does | does not] want to use HTN treatment decision aid

– Plan to pause and assess understanding at the end

– Use computer to send prescription to pharmacy

Encounter flow

– Timing and sequence of events

– Start with teaching about initial explanation of diagnosis, health impact, benefits of treatment

– Pause and confirm understanding

– Next teach about lifestyle measures

– Next describe drug options

– Next [use | skip] decision aid

– Come to agreement on drug (will probably be hydrochlorothiazide)

– Pause and confirm understanding

– Arrange follow-up with nurse in 2 weeks and with clinician in 2 months

– All of this will need to be a bit rushed

HTN hypertension, EHR electronic health record

aList of potential elements is illustrative, not intended to be complete

bUnderlined text indicates “instantiation” using features of this particular patient and context; text in brackets and italics are empty “slots” that are not yet fully instantiated. The remaining text is relatively generic for all patients who match this (hypothetical) management script. This script reflects the approach of a fairly experienced clinician who sees patients similar to this one every week, and thus can anticipate many of the issues that need to be addressed; less experienced clinicians or clinicians who see this problem infrequently would have less-well-developed scripts. Actions within a given feature (table row) are in approximate sequential order, but the features themselves do not follow exclusively in the order presented herein

Fig. 1

Model of management reasoning script. Management scripts are “precompiled conceptual knowledge structures that represent and connect management options and clinician tasks in a temporal or logical sequence to facilitate development of a rational management plan” [14]. Although initial script activation, selection, and instantiation are likely (usually) predominantly Type 1 thinking, with each iteration the script and associated management tasks (usually) employ more Type 2 thinking. The list of management tasks is illustrative, not comprehensive. EHR electronic health record

40037_2022_714_Fig1_HTML.png

Language: English
Submitted on: Feb 17, 2022
|
Accepted on: Apr 26, 2022
|
Published on: Jun 2, 2022
In partnership with: Paradigm Publishing Services
Publication frequency: 1 issue per year

© 2022 David A. Cook, Christopher R. Stephenson, Larry D. Gruppen, Steven J. Durning, published by Bohn Stafleu van Loghum
This work is licensed under the Creative Commons Attribution 4.0 License.