The decision to extubate intensive care patients is critical to the long-term recovery of patients. Extubate too early and the patient may still have difficulties breathing and may need to be re-intubated which leads to delayed recovery and may even increase mortality. Delaying extubation exposes patients to many risks e.g. long term damage to the patient’s lungs and airways and infections.
Clinicians generally rely on a subjective assessment of observations during a patient’s treatment and recovery in order to decide whether or not to extubate, but would benefit from a tool to help them make that decision.
By recording data from patients over several years, together with the extubation outcome, it is possible to identify the most important predictive factors for successful extubation. These are related to the amount of secretions in the patient’s airways and lungs and also the patient’s ability to cough since, by coughing, the patient is able to maintain a clear and patent airway to facilitate ease of breathing. These factors can be combined within a statistical model to predict the chance of successful extubation for new patients, given the observations available for that patient. The model can also be used to suggest a cut-off probability beyond which a clinician would be encouraged to extubate (more details on this model are available here).
By analysing historical data it’s possible to provide clinicians with advice to help guide their decision as to whether or not to extubate their patient. This improves the objectivity of the decision and, in particular, helps clinicians with less experience make the right decision for the patients in their care.