Skip to main content

Table 3 Prototype components and refinement considerations are the foundation for the design prompts based on observations and interviews

From: Cautiously optimistic: paediatric critical care nurses’ perspectives on data-driven algorithms in low-resource settings—a human-centred design study in Malawi

Prototype component

Design prompts

Source

1. Risk representation

Refinement considerations:

Ensure swift data comprehension

The presentation should be easily understandable and, preferably, intuitive

Users should readily grasp whether the score indicates a high or low risk

We selected four prompts:

i. Colours: Employing commonly recognised warning indicator colours (red, yellow and green)

ii. Number: Providing a numerical score (0–10), aligning with standard health parameters

iii. Percentage: Utilising a universal percentage scale (0–100%)

iv. Text: Integrating descriptive text

Interviews and literature [25, 26]. The risk was determined to be a key aspect in the HDU workflow and for determining the chance of deterioration

2. Representation of score input

Refinement considerations:

Propose various solutions to achieve a balance between comprehensive information and simplicity

Five prompts were designed to illustrate different levels of explainability:

i. Score without background data: providing the score without additional contextual information

ii. Score with an iconic representation of vitals with alarm indication: incorporating icons to symbolise the status of vital signs and alarm indicators

iii. Score with numeric vital signs and alarm indication: displaying numeric values of vital signs alongside alarm indicators

iv. Score with numeric vital signs of only the alarming vital: highlighting numeric values for vital signs that contribute to the score that triggers an alarm

v. Score with incongruent vital signs values/colour coding: displaying vital signs or colour codes that do not match the score presented

Interviews and literature [24, 27]. Participants expressed a desire for an understanding of the factors influencing the data-driven algorithm, deeming explanatory factors for risk necessary

3. Representation of changes in the score

Refinement considerations:

Use various change indicators to enhance the understanding of different cues

Five prompts were designed to represent patient deterioration:

i. An upward arrow: indicating an increase in the score (presented to assess intuitive interpretation)

ii. A downward arrow: indicating a decline in health (presented to assess intuitive interpretation)

iii. Previous and new scores: displaying two scores simultaneously

iv. A score accompanied by a numerical representation: showcasing the score change

v. A visual bar: illustrating the current and previous score points

Observations revealed that the current way of working does not yet involve examining the patient’s history to inform decisions, although nurses do acknowledge the significance of changes in health status