Predicting adverse drug events in older inpatients: a machine learning study

Hu, Qiaozhi; Wu, Bin; Wu, Jian; Tian, Fangyuan

doi:10.1007/s11096-022-01468-7

Cited by 8 publications

(27 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These studies were conducted in China [5,26,30], the United States [16], India 28, and Sweden [27,29,[31][32][33]. The included population of these studies included older [5], pediatric [26], and all population [16, [27][28][29][30][31][32].…”

Section: Characteristics Of the Included Studiesmentioning

confidence: 99%

“…Among 10 included studies, seven studies were with high risk of bias and three were with high risk of applicability concerns. Only 3 studies had both low risk of bias and applicability concerns [5,16,26]. The evaluation result detail and diagram of PROBAST are showed in Supplement Table 3 and Supplement Fig.…”

Section: Evaluating the Quality Of Studiesmentioning

confidence: 99%

“…Eight studies reported the types of ADRs/ADEs found, which number ranged 14 to 102 [5,16,26,27,29,[31][32][33]. The detail can be found in Supplement Table 5 The drug allergy was mentioned in eight studies, including angioneurotic oedema, Steven-Johnson syndrome, anaphylactic shock, and contact dermatitis.…”

Section: Adr/adementioning

confidence: 99%

“…Convulsions, hyperglycemia, respiratory depression, bronchospasm, and dyspnea were only mentioned in the study for children [26], while bradycardia was only in the study for older patients [5]. The incidences of above ADRs/ADEs were low, all below 1% [5,26].…”

Section: Adr/adementioning

confidence: 99%

“…ML can handle complex nonlinear relationships between variables and outcomes, exhibiting stronger generalization capabilities and improved accuracy [17]. With the advancement of computer technology and the development of medical databases, ML has become a new research hotspot in medical practice, which has shown great capability in capability the elds of disease diagnosis [18], prescription analysis [19], complication monitoring [20], and disease prediction [5,16]. However, no detailed overview and critical appraisal for the developed prediction model were provided.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Machine learning to predict adverse drug reaction or event based on electronic health records: a systematic review and meta-analysis

Hu,

Li,

Zou

et al. 2024

Preprint

Self Cite

View full text Add to dashboard Cite

Introduction Machine learning (ML) for adverse drug reaction or event (ADR/ADE) prediction is emerging as the promising method for improving medical quality. We aimed to conduct a systematic review to comprehensively summarize ML prediction for ADR/ADE based on electronic health record (EHR). Materials and methods We systematically searched the PubMed, Web of Science, Embase, and IEEE Xplore databases from database inception to 21 Nov. 2023, to identify eligible studies. Any study that developed ML model to predict multiple ADR/ADE based on EHR was included in the final review. The pooled sensitivity, specificity and their 95% CI were calculated. Binary accuracy data were extracted for meta-analysis to derive area under the curve (AUC). Results 5704 studies were identified, of which 10 studies met the inclusion criteria. Among the 20 ML methods reported in the including studies, Random Forest (RF) was reported the most (n=9), followed by Adaboost (n=4), eXtreme Gradient Boosting (n=3) and support vector machine (n=3). The mean AUC for ML prediction was 75.71% (26.00-94.57). RF combined with resampling based approaches might get high AUC, the mean was 82.92% (94.48-94.57). The length of stay, number of drugs, admission type, age, and high-risk drug used, such as antiviral agents, rifamycin, were the common risk factors for ADR/ADE prediction. The pooled estimated AUC of summary receiver operator characteristics was 72.00% (68.00-75.00). Conclusions Acceptable prediction performance of ADR/ADE with ML algorithmwas highlighted. More rigorous reporting standards and the new ML methods that take into account the unique challenges of ML research could improve future studies and help the application of ML models in clinical practice.

show abstract

Section: Characteristics Of the Included Studiesmentioning

confidence: 99%

Section: Evaluating the Quality Of Studiesmentioning

confidence: 99%

Section: Adr/adementioning

confidence: 99%

Section: Adr/adementioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Machine learning to predict adverse drug reaction or event based on electronic health records: a systematic review and meta-analysis

Hu,

Li,

Zou

et al. 2024

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

A model for identifying potentially inappropriate medication used in older people with dementia: a machine learning study

Hu,

Zhao,

Teng

et al. 2024

Int J Clin Pharm

View full text Add to dashboard Cite

Background Older adults with dementia often face the risk of potentially inappropriate medication (PIM) use. The quality of PIM evaluation is hindered by researchers' unfamiliarity with evaluation criteria for inappropriate drug use. While traditional machine learning algorithms can enhance evaluation quality, they struggle with the multilabel nature of prescription data. Aim This study aimed to combine six machine learning algorithms and three multilabel classification models to identify correlations in prescription information and develop an optimal model to identify PIMs in older adults with dementia. Method This study was conducted from January 1, 2020, to December 31, 2020. We used cluster sampling to obtain prescription data from patients 65 years and older with dementia. We assessed PIMs using the 2019 Beers criteria, the most authoritative and widely recognized standard for PIM detection. Our modeling process used three problem transformation methods (binary relevance, label powerset, and classifier chain) and six classification algorithms. Results We identified 18,338 older dementia patients and 36 PIMs types. The classifier chain + categorical boosting (CatBoost) model demonstrated superior performance, with the highest accuracy (97.93%), precision (95.39%), recall (94.07%), F1 score (95.69%), and subset accuracy values (97.41%), along with the lowest Hamming loss value (0.0011) and an acceptable duration of the operation (371s). Conclusion This research introduces a pioneering CC + CatBoost warning model for PIMs in older dementia patients, utilizing machine-learning techniques. This model enables a quick and precise identification of PIMs, simplifying the manual evaluation process.

show abstract

Using Machine Learning or Deep Learning Models in a Hospital Setting to Detect Inappropriate Prescriptions: A Systematic Review

Johns

Godet

Alkanj

et al. 2023

Preprint

View full text Add to dashboard Cite

Objectives: The emergence of artificial intelligence (AI) is catching the interest of hospitals pharmacists. Massive collection of pharmaceutical data is now available to train AI models and hold the promise of disrupting codes and practices. The objective of this systematic review was to examine the state of the art of machine learning or deep learning models that detect inappropriate hospital medication orders. Methods: A systematic review was conducted according to the PRISMA statement. PubMed and Cochrane database were searched from inception to May 2023. Studies were included if they reported and described an AI model intended for use by clinical pharmacists in hospitals. Results: After reviewing, thirteen articles were selected. Eleven studies were published between 2020 and 2023; eight were conducted in North America and Asia. Six analyzed orders and detected inappropriate prescriptions according to patient profiles and medication orders, seven detected specific inappropriate prescriptions. Various AI models were used, mainly supervised learning techniques. Conclusions: This systematic review points out that, to date, few original research studies report AI tools based on machine or deep learning in the field of hospital clinical pharmacy. However, these original articles, while preliminary, highlighted the potential value of integrating AI into clinical hospital pharmacy practice.

show abstract

Predicting adverse drug events in older inpatients: a machine learning study

Cited by 8 publications

References 35 publications

Machine learning to predict adverse drug reaction or event based on electronic health records: a systematic review and meta-analysis

Machine learning to predict adverse drug reaction or event based on electronic health records: a systematic review and meta-analysis

A model for identifying potentially inappropriate medication used in older people with dementia: a machine learning study

Using Machine Learning or Deep Learning Models in a Hospital Setting to Detect Inappropriate Prescriptions: A Systematic Review

Contact Info

Product

Resources

About