expertise, including correlational measures, rely in part or in all on consensus.
C-CWS provided performance assessments at roughly 1 to 2 min intervals. Unexpectedly, drops in C-CWS values were discovered to be predictive of some (but not all) errors before they occurred. That is, a rapid decline in C-CWS was often followed by
an operational error 1 to 3 min later. C-CWS may be sensitive to deterioration prior to an overt error.
It appears that performance
C-CWS proved successful in assessing development of expertise in a complex, dynamically evolving task. Other approaches to assessing expertise were less useful. These findings have
implications for the evaluation of experts
selection, training, and In addition, this line of
research has consequences for rule learning, development of mental models, influence of context on decision making, and the role of individual differences on decision behavior.
The discovery that falloffs in C-CWS scores may be predictive of subsequent errors has important ramifications. This may provide a performance-based approach to identifying and preventing errors before they occur. We are now exploring the generality of this finding in other contexts.
For more information on CWS, see our website www.ksu.edu/psych/cws
Diagnosing Chronic Heart Failure in Primary Health Care
Karolinska Institute, Stockholm, Sweden E-mail: firstname.lastname@example.org
The aim of my project is to describe General
Practitioners’ (GPs’) diagnostic strategies regarding patients with
failure. I Judgement
have done two studies with Clinical Analysis technique (CJA) and I am now
supplementing them with a study in which I use Think Aloud technique (TA). In the two CJA studies I have used case vignettes based on authentic patients. In the first study, the patients were collected from two health centres. In order to get patients with more valid diagnoses (heart-failure or not), the patients in the second study were collected among patients referred from GPs to a cardiology out-patient clinic. GPs, cardiologists and medical students were compared as regards diagnostic accomplishment and diagnostic strategies, and I found them to be similar on the group level, but very different at the individual
The most important cues for the participants cardiac enlargement and pulmonary stasis.
Strategies, in which cardiac enlargement was the predominating cue, led to a higher diagnostic accomplishment; a third of the participants used such strategies. The cues given in the vignettes could have been utilized more efficiently; cardiac enlargement seems to be more important while “classical” symptoms are less important for predicting heart failure than the participants
process, I will now use TA technique. GPs will be presented with six case descriptions, based on authentic patients from primary health care (a sample of patients utilized in the second study, representing different levels of difficulty). Each case description will be presented on consecutive computer screens (history, symptoms and signs;
echocardiography), and the think aloud session will be taped, transcribed, coded, and analysed. Our central questions concern how information about the patient is collected and integrated and how decision rules and knowledge are applied in the decison.
General Practitioners' decision making and the role of clinical guidelines.
Liz Smith University of Aberdeen Email: email@example.com
I am finishing off work on
decisions in depression and the role of the clinical guideline within these. The goal of the project was to discover the factors that influence their prescribing and to investigate how guideline use could be increased so as to promote clinical effectiveness. A lens model study found that GPs
tend to over prescribe compared with guidelines and place much more emphasis thoughts of suicide and sleep disturbance than
the on the
higher LME scores guidelines) than those
(when compared with in medium or smaller
practices. Another lens model out comparing GPs in England
study was carried and Scotland and
again differences were found.
England those in
quantitative studies were used questions for the final study which
from these to generate used in-depth
Newsletter 2002 page 24 of 28