2021
DOI: 10.1371/journal.pone.0253515
|View full text |Cite
|
Sign up to set email alerts
|

Risk of adverse events in gastrointestinal endoscopy: Zero-inflated Poisson regression mixture model for count data and multinomial logit model for the type of event

Abstract: Background and aims We analyze the possible predictive variables for Adverse Events (AEs) during sedation for gastrointestinal (GI) endoscopy. Methods We consider 23,788 GI endoscopies under sedation on adults between 2012 and 2019. A Zero-Inflated Poisson Regression Mixture (ZIPRM) model for count data with concomitant variables is applied, accounting for unobserved heterogeneity and evaluating the risks of multi-drug sedation. A multinomial logit model is also estimated to evaluate cardiovascular, respirat… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
4
1

Citation Types

0
8
0
2

Year Published

2022
2022
2024
2024

Publication Types

Select...
7
1
1

Relationship

0
9

Authors

Journals

citations
Cited by 13 publications
(10 citation statements)
references
References 37 publications
0
8
0
2
Order By: Relevance
“…We found that the OvO-PC method had comparative performance to MLR across all scenarios in the simulation, but it has been shown to have better performance than MLR in a clinical example when variable selection was implemented on each sub-model. 4 It is therefore possible OvO-PC could provide improved performance over the commonly implemented MLR, [18][19][20][21][22][23][24][25][26] in scenarios where variable selection is required. Variable selection for MLR is beyond the scope of this work but should be considered in the future.…”
Section: Discussionmentioning
confidence: 99%
See 1 more Smart Citation
“…We found that the OvO-PC method had comparative performance to MLR across all scenarios in the simulation, but it has been shown to have better performance than MLR in a clinical example when variable selection was implemented on each sub-model. 4 It is therefore possible OvO-PC could provide improved performance over the commonly implemented MLR, [18][19][20][21][22][23][24][25][26] in scenarios where variable selection is required. Variable selection for MLR is beyond the scope of this work but should be considered in the future.…”
Section: Discussionmentioning
confidence: 99%
“…[11][12][13][14][15][16][17] However, an outstanding question is which of these methods are most suitable for developing prediction models for nominal polytomous outcomes. In practice, MLR is the most commonly used method to model nominal outcomes, [18][19][20][21][22][23][24][25][26] however a comprehensive evaluation of the most appropriate method in this setting does not exist. A recent simulation has compared the calibration of different models for the prediction of an ordinal outcome, 27 but no such simulation exists for nominal outcomes.…”
Section: Introductionmentioning
confidence: 99%
“…In gastroenterology, AI has been found widely applied in various aspects, including disease detection, diagnosis, treatment decision-making, long-term prognosis, and the optimization of medical environments, 12 with a particular emphasis on image recognition in endoscopy. While several prediction models have been developed for various digestive diseases, they often focus on a single specific outcome of interest; for instance, prediction models for cancers, upper GI bleeding, [13][14][15] endoscopic complications, 16 or adverse effects, 17 assessment for irritable bowel syndrome, [18][19][20] and liver disease. 21 Many of these models also necessitate laboratory tests, which introduce limitations in terms of usability and result accuracy.…”
Section: Discussionmentioning
confidence: 99%
“…In this work, the analysis of the vital parameters from the dataset of patients who underwent an endoscopy with sedoanalgesia showed that there were no clinically significant differences between the average values of BP, HR and SpO 2 at the beginning of the procedure and those recorded at the end of the procedure, denoting the safety of sedoanalgesia in this setting. As for anesthesia-related complications, these occurred in 0.8% (4 cases out of 529), while the literature reports adverse events during sedoanalgesia with benzodiazepine and opioids (hypotension, respiratory depression) in percentages ranging from 0.47% to 17% [ 31 ]. In this study, all the cases of anesthesia-related complication required an IV bolus infusion of atropine due to persistent bradycardia (HR < 50 bpm) and hypotension (mean pre-sedation level BP value < 50 mmhg).…”
Section: Discussionmentioning
confidence: 99%