Rapid diagnosis has been demonstrated as being essential in improving outcomes for patients displaying potential symptoms of cardiopulmonary problems including pneumonia, acute coronary syndrome, pulmonary embolism, heart failure, pneumothorax, mediastinal disease and aortic dissection.
Visual cues of a humorous cartoon, a close-up of a surprised face and a picture of someone crying were shown to patients in an emergency department.
The most popular tool among physicians for making this initial decision on whether invasive and expensive diagnostic tests are required is known as "the gestalt method." The gestalt method is a rapid visual assessment that is more reliant on human thought processes than references, and so is unique in its application to each patient.
Although the subjectivity of the gestalt method may sound worrying, a 2011 meta-analysis found that it has a similar diagnostic accuracy compared with objective methods, and that the addition of gestalt assessment to other decision rules improves diagnostic performance.
Other evidence suggests, though, that the accuracy of gestalt reasoning improves in accordance with clinical experience. This variability in the accuracy of the diagnostic from physician to physician - as well as a lack of transparency in how it is applied - therefore makes some clinicians uneasy with its use.
By breaking down the gestalt method into identifiable elements, the researchers behind the new study reasoned these elements can be individually assessed for accuracy and taught as part of a more standardized diagnostic process.
Facial expression as indicators of diseaseThe researchers chose to investigate facial expressions because clinicians take into account "emotional tension" in the face of their patients as part of gestalt reasoning, and also because previous research has shown a relationship between facial muscles and cardiopulmonary health.
The patients' responses to the pictures were evaluated using the "Facial Action Coding System (FACS)," which scores changes in facial muscle activity.
Tension in the face, for instance, has been linked to patients with reactive airway disease, and patients with myocardial ischemia have been found to have "significantly more facial expressions of anger, non-enjoyment smiles and brow lowering" than people without this condition.
The team tested the diagnostic accuracy of reduced facial expression range in 50 adults with shortness of breath and chest pain who presented to an emergency care department, and who were scanned for serious heart or lung disease and monitored for 2 weeks.
The patients were each shown three visual cues that were designed to evoke an emotional response, while their facial responses were recorded using a webcam.
The visual cues were a humorous cartoon, a close-up of a surprised face and a picture of someone crying. The patients' responses to these pictures were evaluated using the "Facial Action Coding System (FACS)," which evaluates changes in facial muscle activity.
The researchers analyzed the webcam recordings and compared the FACS scores with the objective medical diagnoses the patients received while in the hospital.
They found that patients with chest pain and shortness of breath who had a potentially serious condition affecting their heart or lungs had a significantly reduced range of facial expressions in response to the visual cues than people who did not have these conditions.
The authors write:
"We believe that due to the gravity of their illness, [these] patients may not have been able to process and respond to an emotional stimulus in the way that would be expected of most people under normal conditions.
The ultimate goal of this work is to provide clinicians with a new physical finding that can be associated with a healthy state to avoid unnecessary [computed tomography] scanning."
The authors believe their findings emphasize the importance of a doctor being able to accurately "read" a patient's face, suggesting this may become even more important as consultations over Skype become more common.
Written by David McNamee