Which clinical decisions benefit from automation? A task complexity approach

https://doi.org/10.1016/S1386-5056(03)00040-6Get rights and content

Abstract

Objective: To describe a model for analysing complex medical decision making tasks and for evaluating their suitability for automation. Method: Assessment of a decision task's complexity in terms of the number of elementary information processes (EIPs) and the potential for cognitive effort reduction through EIP minimisation using an automated decision aid. Results: The model consists of five steps: (1) selection of the domain and relevant tasks; (2) evaluation of the knowledge complexity for tasks selected; (3) identification of cognitively demanding tasks; (4) assessment of unaided and aided effort requirements for this task accomplishment; and (5) selection of computational tools to achieve this complexity reduction. The model is applied to the task of antibiotic prescribing in critical care and the most complex components of the task identified. Decision aids to support these components can provide a significant reduction of cognitive effort suggesting this is a decision task worth automating. Conclusion: We view the role of decision support for complex decision to be one of task complexity reduction, and the model described allows for task automation without lowering decision quality and can assist decision support systems developers.

Introduction

There is a need to improve clinical decision-making in order to reduce practice variation, preventable medical errors and support the delivery of evidence-based medicine [1], [2]. Information technology has been recognised as a key enabler in the improvement of clinical decision quality [3]. However, the decision to use technology in support of clinical decision making is itself often empirical rather than guided by sound theoretical principles. It thus remains unclear when technology should be brought in to support decision-making for complex clinical problems. Not surprisingly, information-technology initiatives have provided mixed results and the uptake of decision aids by clinicians has been slow [3], [4], [5].

The decision to select one task over another for computational support should be based on some principled methods [6]. The effectiveness of instructional aids and decision support systems (DSS) is predicated on their usefulness in addressing the specific problems that lead to sub-optimal decisions [7]. The objective of our study is to design a framework for the rational selection of clinical tasks for automation using a cognitive task complexity approach and to investigate its potential benefits.

Section snippets

Theoretical framework

Task complexity affects information use and has been shown to be one of the important determinants of decision-making efficiency [8]. Based on a cognitive engineering approach, Woods postulated that all tasks contain three essential components such as products, required acts, and information cues [9]. He has argued that complexity describes the relationships between acts and information cues as task inputs and has used these components to derive three dimensions of task complexity: component

Choice of the domain

Prescribing is an activity central to clinical practice [11]. The prescribing task in critical care is accomplished under time pressure and with limited diagnostic information. Infection is a common presenting problem and antimicrobials are among the most frequently prescribed drugs. Therefore antibiotic use has long been considered an important target for analysis of decision-making [11]. Several studies demonstrated significant level of antibiotic misuse ranging at university medical centres

Prescribing effort and quality

We have constructed a prescribing decision tree that incorporates the two main management strategies (‘treat first’ and ‘test first and treat after’) for suspected VAP (Fig. 3). It is structured around the prescribing sub-goals: (1) control of potentially treatable infection; (2) prevention/delay of antibiotic resistance; (3) reduction of drug-related adverse effects; (4) prevention of super-infection with multi-resistant microorganisms [15]. This decision tree allows the calculation of the

Discussion and conclusions

We argue that prescribing is a complex clinical task because it involves: (a) integration of complex information from a variety of sources; (b) incomplete or imperfect information; (c) the presence of uncertainty and time pressure; (d) a complex interaction between the clinician and the patient with different utilities and values to the alternatives in the decision.

Such high complexity is a general risk factor in clinical decision making. It has, for example, been suggested that clinicians

References (21)

There are more references available in the full text version of this article.

Cited by (48)

  • Basic complexity criteria and their impact on manual assembly quality in actual production

    2017, International Journal of Industrial Ergonomics
    Citation Excerpt :

    Many task complexity models conform to this view point, e.g. by Bonner (1994) and Ham et al. (2012). In other viewpoints task complexity seems to be synonymous with task load or task demand (e.g. Sintchenko and Coiera, 2003; Bedny et al., 2012). The huge amount of variants and build options in automotive assembly offers a challenge in production planning and for the operator who is supposed to manage many different tasks in paced assembly lines.

  • Task complexity as a performance shaping factor: A review and recommendations in Standardized Plant Analysis Risk-Human Reliability Analysis (SPAR-H) adaption

    2015, Safety Science
    Citation Excerpt :

    Size complexity closely resembles “task scope”, which is a function of the subtasks, products, and information processing requirements (Harvey and Koubek, 2000). This category refers to the number of actions or steps that are required by the operator to complete the task (Lazzara et al., 2010; Sintchenko and Coiera, 2003). A higher number of continuous actions or steps will make a task more complex (Park et al., 2001; Wood, 1986).

  • Task complexity: A review and conceptualization framework

    2012, International Journal of Industrial Ergonomics
    Citation Excerpt :

    It is hard to say which perspective is more appropriate. In reality, resource requirement is usually used as a measure of task complexity (see, cognitive efforts: Chu and Spires, 2000; Sintchenko and Coiera, 2003; Bedny et al., 2012). In this viewpoint, task complexity seems to be synonymous with task load or task demand.

View all citing articles on Scopus

This paper was presented at the MIE2002.

View full text