Re: Development and validation of PRE-DELIRIC (PREdiction of DELIRium in ICu patients) delirium prediction model for intensive care patients: observational multicentre study
CCBYNCOpen access
Rapid response to:
Research
Development and validation of PRE-DELIRIC (PREdiction of DELIRium in ICu patients) delirium prediction model for intensive care patients: observational multicentre study
Re: Development and validation of PRE-DELIRIC (PREdiction of DELIRium in ICu patients) delirium prediction model for intensive care patients: observational multicentre study
Dear Editor,
van den Boogard et al propose the PRE-DELIRIC model that has a high predictive value, for early detection of delirium in intensive care patients. (1) The authors excluded “use of lorazepam” as a risk factor because the prevalence rate in the sample was less than 10%. However, they have included “use of sedatives” as a predictor in the model. It is not clear, how the term “sedatives” was defined, and why lorazepam was excluded from this definition. Furthermore, on examining the automatic version of the model, downloaded from the author’s website, “Use of sedative” is defined as “Use of propofol, midazolam, lorazepam or combination”. It is not clear, why lorazepam was used in this definition, while “use of lorazepam” was excluded from the risk factor list in the analysis. It is also not clear why the term “sedative” was restricted to the use of propofol, midazolam, lorazepam or a combination of these.
Exclusion of alcohol misuse and dementia, due to low prevalence in the sample, question the generalisability and utility of the PRE-DELIRIC model on its own to predict delirium, especially in a population where 4 to 5% of all hospital admissions in a year is for an alcohol related medical condition. (2) The authors acknowledge in the discussion that use of this model alone may miss out on people with alcohol / other substance misuse, who may undergo withdrawal, and those with an underlying dementia (sometimes subclinical , and may not have been clinically diagnosed).
Reference
1. Boogaard M, Pickkers P, Slooter AJ, Kuiper MA, Spronk PE, Voort PH, Hoeven JG, Donders R, Achterberg T, Schoonhoven L. Development and validation of PRE-DELIRIC (PREdiction of DELIRium in ICu patients) delirium prediction model for intensive care patients: observational multicentre study. BMJ. 2012 Feb 9;344:e420. doi: 10.1136/bmj.e420.
Rapid Response:
Re: Development and validation of PRE-DELIRIC (PREdiction of DELIRium in ICu patients) delirium prediction model for intensive care patients: observational multicentre study
Dear Editor,
van den Boogard et al propose the PRE-DELIRIC model that has a high predictive value, for early detection of delirium in intensive care patients. (1) The authors excluded “use of lorazepam” as a risk factor because the prevalence rate in the sample was less than 10%. However, they have included “use of sedatives” as a predictor in the model. It is not clear, how the term “sedatives” was defined, and why lorazepam was excluded from this definition. Furthermore, on examining the automatic version of the model, downloaded from the author’s website, “Use of sedative” is defined as “Use of propofol, midazolam, lorazepam or combination”. It is not clear, why lorazepam was used in this definition, while “use of lorazepam” was excluded from the risk factor list in the analysis. It is also not clear why the term “sedative” was restricted to the use of propofol, midazolam, lorazepam or a combination of these.
Exclusion of alcohol misuse and dementia, due to low prevalence in the sample, question the generalisability and utility of the PRE-DELIRIC model on its own to predict delirium, especially in a population where 4 to 5% of all hospital admissions in a year is for an alcohol related medical condition. (2) The authors acknowledge in the discussion that use of this model alone may miss out on people with alcohol / other substance misuse, who may undergo withdrawal, and those with an underlying dementia (sometimes subclinical , and may not have been clinically diagnosed).
Reference
1. Boogaard M, Pickkers P, Slooter AJ, Kuiper MA, Spronk PE, Voort PH, Hoeven JG, Donders R, Achterberg T, Schoonhoven L. Development and validation of PRE-DELIRIC (PREdiction of DELIRium in ICu patients) delirium prediction model for intensive care patients: observational multicentre study. BMJ. 2012 Feb 9;344:e420. doi: 10.1136/bmj.e420.
2. Alcohol statistics Scotland 2011. ISD Scotland Publications.
Competing interests: No competing interests