Modeling the economic value of Ceribell Rapid Response EEG in the inpatient hospital setting

Figures & data

Figure 1. Rapid Response EEG System. Rapid-EEG is a new EEG device developed by Ceribell, Inc. (Mountain View, CA) and cleared by US Food and Drug Administration (FDA) since 2018. It consists of a headband with 10 electrodes (5 left and 5 right), which enables eight-channel EEG acquisition with a pocket-size battery-operated EEG recording device. The EEG can be set up by anyone without needing specialized EEG techs. The acquired data is accessible at the bedside as well as on a cloud portal that can be accessed remotely in real time.

Table 1. DECIDE trial summary.

Trial characteristic/finding	Result
n (final)	164 patients
Age (mean + SD)	58.6 ± 18.7 year
Percent female	45%
Time to rapid EEG (median, IQR)	5 min (4–10 min)
Time to conventional EEG (median, IQR)	239 min (134–471 min)
Sensitivity-clinical suspicion (95%CI)	77.8% (40.0%, 97.2%)
Sensitivity-rapid EEG (95%CI)	100% (66.4%, 100%)
Specificity-clinical suspicion (95%CI)	63.9% (55.8%, 71.4%)
Specificity-rapid EEG (95%CI)	89% (83.0%, 93.5%)

Patient characteristics and findings from the DECIDE trial¹¹.

Figure 2. Decision tree model for cost–benefit analysis. Two-armed decision model with intervention (Rapid-EEG vs clinical suspicion only) at the point of suspected seizure and subsequent event nodes corresponding with diagnostic categories of Rapid-EEG interpretation (true positive, false positive, true negative, false negative). Length of stay data at terminal nodes correspond to Nationwide Inpatient Sample 2017 cross-sectional cohorts for coma (ICD10 codes R40.2*), seizure (Clinical Classification Software diagnostic grouper CCS-83), and mechanical ventilation (ICD10 procedure = 5A1935Z, 5A1945Z, 5A1955Z). Abbreviations. EEG, electroencephalography; FN, false negative; FP, false positive; NCSz, non-convulsive seizure; NIS, National Inpatient Sample; TN, true negative; TP, true positive.

Table 2. Model input parameters.

Parameter	Reference Case	SD or (±)	Distribution	Source	Notes
Probabilities
Non-convulsive Seizure Rate	0.21	0.01	β	^15–18	Pooled 4 trials, 199 NCSz in 937 Coma Patients
Rapid EEG Sensitivity	1	0.17	Truncated Normal	^7	Rapid EEG Sensitivity 95% CI 0.66–1.0, mean 1.0, assume sensitivity cannot exceed 100%
Rapid EEG Specificity	0.89	0.03	β	^7
Clinical Suspicion Sensitivity	0.78	0.19	β	^7
Clinical Suspicion Specificity	0.64	0.04	β	^7
Early Seizure Termination	0.35	0.06	β	^19	23/60 NCSz terminated with benzodiazepine
Length of stay (days)
LOS FP	10.2	1.1^a	Lognormal	^20	Coma + no seizure, +mechanical ventilator^b
LOS TP|LOS FN-Seizure	11.5	1.0^a	Lognormal	^20	Coma + seizure + mechanical ventilator^b
LOS TN	5.1	0.8^a	Lognormal	^20	Coma + no seizure, no mechanical ventilator^b
LOS TP|Early Seizure Termination	6	0.8^a	Lognormal	^20	Coma + seizure, no mechanical ventilator^b
Proportion ICU Stay	0.3	^c	Lognormal^c	^21	Proportion of ICU: hospitalization duration
Reduction ICU Stay|Early Seizure Termination	0.5	^c	Lognormal^c	^22	Proportion of ICU: hospital duration given early seizure termination protocol success
Costs (2019 US$)
Daily cost mechanical ventilation	$2,625	$2,625	γ	^21	Additional cost of mechanical ventilation in the ICU, SD = mean, assumed
Daily Cost Hospitalization	$1,957	±$391	Uniform	^23	2019 MS-DRG 81 (atraumatic coma) national standard reimbursement/mean LOS
Cost Intubation	$610	±$123	Uniform	^24	ET Tube, paralytic, sedative, analgesic, oxygen, intubation, disposables + CPT for intubation
Cost Parenteral ASM	$90	$90	γ	^25	Non-benzodiazepine ASMs for status epilepticus

Abbreviations. NIS 2017, Nationwide Inpatient Sample, 2017; CMS, Center for Medicare and Medicaid Services; EEG, electroencephalogram; FP, false positive; TP, true positive; TN, true negative; FN, false negative; LOS, length of stay; β, beta distribution; γ, gamma distribution.

^aStandard deviation (SD) log estimates; ^b2017 NIS data, coma patients (ICD10 R40.2* ± Clinical Classification Software for Seizure (CCS83) ± ICD10 Mechanical Ventilation (5A1935Z 5A1945Z 5A1955Z). ^cRatio of ICU Stay to LOS, each distributed as log normal, with random draws from each distribution to provide ratio in probabilistic model.

Table 3. Results of reference case and probabilistic sensitivity analysis (PSA).

	Reference case				PSA MCS (95% PI)
	Total hospital cost (2019 US$)	LOS (days)	ICU LOS (days)	Intubation +  ASM initiated (%)*	Total hospital cost (2019 US$)	LOS (days)	ICU LOS (days)	Intubation +  ASM initiated (%)^a
Clinical suspicion (control)	$21,261	7.4	1.9	45%	$24,694 ($22,999, $26,389)	7.6 (7.1, 8.0)	2.7 (2.5, 2.9)	45% (44%, 46%)
Rapid-EEG (intervention)	$17,290	6.2	1.5	22%	$19,061 ($17958, $20,164)	6.5 (6.0, 6.8)	2.2 (2.0, 2.4)	21% (21%, 22%)
Difference (net savings)	$3,971	1.2	0.4	23%	$5,633 ($4,649, $6,617	1.1 (0.9, 1.4)	0.5 (0.4, 0.6)	24% (23%, 24%)

Includes 95% prediction intervals (PI) from PSA results from MCS with 1,000 replications.

Abbreviations. US$, US dollars; EEG, electroencephalogram; ICU, intensive care unit; LOS, length of stay; ASM, antiseizure medications; PSA, probabilistic sensitivity analysis; MCS, Monte Carlo simulation.

^a Assumes initiation of intubation and administration of ASM due to suspected seizure.

Figure 3. Monte Carlo simulation. This figure depicts results of Monte Carlo simulation with 1,000 replications of difference in costs between Rapid-EEG and clinical suspicion.

Figure 4. ANCOVA sum of squares analysis. Depicts contribution of model input parameters to variation in the difference in total cost between decisions based on Rapid-EEG vs. clinical suspicion.

Figure 5. One way factor analysis: tornado plot depicting the effect of varying model inputs along a range of values on the difference in costs between Rapid EEG and clinical suspicion for NCSE.

Vespa PM, Olson DM, John S, et al. Evaluating the clinical impact of rapid response electroencephalography: the DECIDE multicenter prospective observational clinical study. Crit Care Med. 2020;48(9):1249–1257.

 PubMed Web of Science ®Google Scholar

Gururangan K, Razavi B, Parvizi J. Diagnostic utility of eight-channel EEG for detecting generalized or hemispheric seizures and rhythmic periodic patterns. Clin Neurophysiol Pract. 2018;3:65–73.

 PubMedGoogle Scholar

Spatola M, Alvarez V, Rossetti AO. Benzodiazepine overtreatment in status epilepticus is related to higher need of intubation and longer hospitalization. Epilepsia. 2013;54(8):e99–e102.

 PubMed Web of Science ®Google Scholar

Sánchez Fernández I, Gaínza-Lein M, Lamb N, et al. Meta-analysis and cost-effectiveness of second-line antiepileptic drugs for status epilepticus. Neurology. 2019;92(20):e2339–e2348.

 PubMed Web of Science ®Google Scholar

DRG data files, files for FY 2019 final rule and correction notice [Internet]. Rockville, MD: Centers for Medicare and Medicaid Services; 2019 [cited 2020 May 1]. Available from: http://www.cms.gov

 Google Scholar

Schmitt SE. Utility of clinical features for the diagnosis of seizures in the intensive care unit. J Clin Neurophysiol. 2017;34(2):158–161.

 PubMed Web of Science ®Google Scholar

National inpatient sample: Healthcare Cost and Utilization Project (HCUP) [Internet]. Rockville (MD): Agency for Healthcare Research and Quality; 2017 [cited 2020 May 1]. Available from: http://www.hcup-us.ahrq.gov/nisoverview.jsp

 Google Scholar

Mihaylova B, Briggs A, O’Hagan A, et al. Review of statistical methods for analysing healthcare resources and costs. Health Econ. 2011;20(8):897–916.

 PubMed Web of Science ®Google Scholar

Physician fee schedule, by HCPCS code [Internet]. Baltimore (MD): Center for Medicare and Medicaid Services; 2020 [cited 2020 May 1]. Available from: www.cms.gov

 Google Scholar