Connecting the Dots: The Influence of Risk Perception on Patient Safety*
Risk perception, including the way we think (cognition), feel (emotion), and behave (behavior) in response to risk, is highly relevant to considerations of patient safety. Unfortunately, there is limited research on the factors that affect such perceptions in the healthcare environment, including intensive care settings, and what has been reported is largely based on survey data rather than any in-depth analysis into the underlying cognitive processes (5, 6). D’Lima et al (4) use semistructured interviews to elicit attitudes of intensive care professionals in an effort to identify cognitive processes underlying their perceptions of ICU staffing risks. This is a step forward in elucidating perceptions of risk, and some underlying patterns of thinking by the health professionals, on suitable staffing levels for intensive care environments.
To capture the underlying cognitive processes for perception of risks, robust methodologies are required. One well-known technique for modeling problem-solving and decision-making processes is to use “think-aloud” approaches, which involve having participants describe what they are thinking as they are performing a set of specified tasks (7, 8). In medicine, these tasks can include processes such as interviewing a patient or walking through a clinical problem-solving scenario (e.g., reaching a diagnosis or devising a management or therapeutic plan) (9). This approach is particularly useful in capturing error progression and risk assessment in complex real-time environments, such as the ICU, where the clinical workflow process does not proceed in regimented steps but is nonlinear, with associated varying boundaries of risks over time (10, 11). Using think-aloud techniques during task performance, the process of making decisions under different risk conditions, which is usually implicit, is made more explicit and is available for further inspection and analysis.
A variation of this method is cognitive interviewing, the technique used by D’Lima et al (4). In their work, the subject is asked predetermined (structured) survey questions, followed by having the interviewer probe for more details (through partially open-ended questions). The responses are recorded and analyzed qualitatively, using a theoretically driven method of data coding (12), which helps the researchers to identify specific strategies that participants use to perceive, and to interpret, the questions.