Dataset Information

Natural language processing diagnosed behavioral disturbance vs confusion assessment method for the intensive care unit: prevalence, patient characteristics, overlap, and association with treatment and outcome.

ABSTRACT:

Purpose

To compare the prevalence, characteristics, drug treatment for delirium, and outcomes of patients with Natural Language Processing (NLP) diagnosed behavioral disturbance (NLP-Dx-BD) vs Confusion Assessment Method for intensive care unit (CAM-ICU) positivity.

Methods

In three combined medical-surgical ICUs, we obtained data on demographics, treatment with antipsychotic medications, and outcomes. We applied NLP to caregiver progress notes to diagnose behavioral disturbance and analyzed simultaneous CAM-ICU.

Results

We assessed 2313 patients with a median lowest Richmond Agitation-Sedation Scale (RASS) score of - 2 (- 4.0 to - 1.0) and median highest RASS score of 1 (0 to 1). Overall, 1246 (53.9%) patients were NLP-Dx-BD positive (NLP-Dx-BD^pos) and 578 (25%) were CAM-ICU positive (CAM-ICU^pos). Among NLP-Dx-BD^pos patients, 539 (43.3%) were also CAM-ICU^pos. In contrast, among CAM-ICU^pos patients, 539 (93.3%) were also NLP-Dx-BD^pos. The use of antipsychotic medications was highest in patients in the CAM-ICU^pos and NLP-Dx-BD^pos group (24.3%) followed by the CAM-ICU^neg and NLP-Dx-BD^pos group (10.5%). In NLP-Dx-BD^neg patients, antipsychotic medication use was lower at 5.1% for CAM-ICU^pos and NLP-Dx-BD^neg patients and 2.3% for CAM-ICU^neg and NLP-Dx-BD^neg patients (overall P < 0.001). Regardless of CAM-ICU status, after adjustment and on time-dependent Cox modelling, NLP-Dx-BD was associated with greater antipsychotic medication use. Finally, regardless of CAM-ICU status, NLP-Dx-BD^pos patients had longer duration of ICU and hospital stay and greater hospital mortality (all P < 0.001).

Conclusion

More patients were NLP-Dx-BD positive than CAM-ICU positive. NLP-Dx-BD and CAM-ICU assessment describe partly overlapping populations. However, NLP-Dx-BD identifies more patients likely to receive antipsychotic medications. In the absence of NLP-Dx-BD, treatment with antipsychotic medications is rare.

SUBMITTER: Young M

PROVIDER: S-EPMC9050783 | biostudies-literature | 2022 May

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Natural language processing diagnosed behavioral disturbance vs confusion assessment method for the intensive care unit: prevalence, patient characteristics, overlap, and association with treatment and outcome.

Young Marcus M Holmes Natasha N Kishore Kartik K Marhoon Nada N Amjad Sobia S Serpa-Neto Ary A Bellomo Rinaldo R

Intensive care medicine 20220323 5

<h4>Purpose</h4>To compare the prevalence, characteristics, drug treatment for delirium, and outcomes of patients with Natural Language Processing (NLP) diagnosed behavioral disturbance (NLP-Dx-BD) vs Confusion Assessment Method for intensive care unit (CAM-ICU) positivity.<h4>Methods</h4>In three combined medical-surgical ICUs, we obtained data on demographics, treatment with antipsychotic medications, and outcomes. We applied NLP to caregiver progress notes to diagnose behavioral disturbance a ...[more]

PMID: 35322288

Similar Datasets

Project description:BackgroundEncephalopathy is a severe co-morbid condition in critically ill patients that includes different clinical constellation of neurological symptoms. However, even for the most recognised form, delirium, this medical condition is rarely recorded in structured fields of electronic health records precluding large and unbiased retrospective studies. We aimed to identify patients with encephalopathy using a machine learning-based approach over clinical notes in electronic health records.MethodsWe used a list of ICD-9 codes and clinical concepts related to encephalopathy to define a cohort of patients from the MIMIC-III dataset. Clinical notes were annotated with MedCAT and vectorized with a bag-of-word approach or word embedding using clinical concepts normalised to standard nomenclatures as features. Machine learning algorithms (support vector machines and random forest) trained with clinical notes from patients who had a diagnosis of encephalopathy (defined by ICD-9 codes) were used to classify patients with clinical concepts related to encephalopathy in their clinical notes but without any ICD-9 relevant code. A random selection of 50 patients were reviewed by a clinical expert for model validation.ResultsAmong 46,520 different patients, 7.5% had encephalopathy related ICD-9 codes in all their admissions (group 1, definite encephalopathy), 45% clinical concepts related to encephalopathy only in their clinical notes (group 2, possible encephalopathy) and 38% did not have encephalopathy related concepts neither in structured nor in clinical notes (group 3, non-encephalopathy). Length of stay, mortality rate or number of co-morbid conditions were higher in groups 1 and 2 compared to group 3. The best model to classify patients from group 2 as patients with encephalopathy (SVM using embeddings) had F1 of 85% and predicted 31% patients from group 2 as having encephalopathy with a probability >90%. Validation on new cases found a precision ranging from 92% to 98% depending on the criteria considered.ConclusionsNatural language processing techniques can leverage relevant clinical information that might help to identify patients with under-recognised clinical disorders such as encephalopathy. In the MIMIC dataset, this approach identifies with high probability thousands of patients that did not have a formal diagnosis in the structured information of the EHR.

Project description:ObjectivesIn the emergency department (ED), health care providers miss delirium approximately 75% of the time, because they do not routinely screen for this syndrome. The Confusion Assessment Method for the Intensive Care Unit (CAM-ICU) is a brief (<1 minute) delirium assessment that may be feasible for use in the ED. The study objective was to determine its validity and reliability in older ED patients.MethodsIn this prospective observational cohort study, patients aged 65 years or older were enrolled at an academic, tertiary care ED from July 2009 to February 2012. An emergency physician (EP) and research assistants (RAs) performed the CAM-ICU. The reference standard for delirium was a comprehensive (~30 minutes) psychiatrist assessment using the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, Text Revision criteria. All assessments were blinded to each other and were conducted within 3 hours. Sensitivities, specificities, and likelihood ratios were calculated for both the EP and the RAs using the psychiatrist's assessment as the reference standard. Kappa values between the EP and RAs were also calculated to measure reliability.ResultsOf 406 patients enrolled, 50 (12.3%) had delirium. The median age was 73.5 years old (interquartile range [IQR] = 69 to 80 years), 202 (49.8%) were female, and 57 (14.0%) were nonwhite. The CAM-ICU's sensitivities were 72.0% (95% confidence interval [CI] = 58.3% to 82.5%) and 68.0% (95% CI = 54.2% to 79.2%) in the EP and RAs, respectively. The CAM-ICU's specificity was 98.6% (95% CI = 96.8% to 99.4%) for both raters. The negative likelihood ratios (LR-) were 0.28 (95% CI = 0.18 to 0.44) and 0.32 (95% CI = 0.22 to 0.49) in the EP and RAs, respectively. The positive likelihood ratios (LR+) were 51.3 (95% CI = 21.1 to 124.5) and 48.4 (95% CI = 19.9 to 118.0), respectively. The kappa between the EP and RAs was 0.92 (95% CI = 0.85 to 0.98), indicating excellent interobserver reliability.ConclusionsIn older ED patients, the CAM-ICU is highly specific, and a positive test is nearly diagnostic for delirium when used by both the EP and RAs. However, the CAM-ICU's sensitivity was modest, and a negative test decreased the likelihood of delirium by a small amount. The consequences of a false-negative CAM-ICU are unknown and deserve further study.

Dataset Information

Natural language processing diagnosed behavioral disturbance vs confusion assessment method for the intensive care unit: prevalence, patient characteristics, overlap, and association with treatment and outcome.

Purpose

Methods

Results

Conclusion

Publications

Natural language processing diagnosed behavioral disturbance vs confusion assessment method for the intensive care unit: prevalence, patient characteristics, overlap, and association with treatment and outcome.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets