Performance of trigger tools in identifying adverse drug events in emergency department patients: a validation study

Karpov, Andrei; Parcero, Catherine; Mok, Catherine Pui Yin; Panditha, Chandima; Yu, Eugenia; Dempster, Linda; Hohl, Corinne M.

doi:10.1111/bcp.13032

Cited by 33 publications

(42 citation statements)

References 47 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Moreover, a recent prospective cohort study found that both of these trigger tools had sensitivities of < 0.20 for identifying ED ADEs. 28 Based on these facts, both models reported in this article identify bleeding ADEs well enough to be potentially useful as a trigger tool for quality improvement and safety surveillance activities. The CART model is particularly promising because of its greater transparency and sensitivity, and also because it could be implemented much the same as typical trigger tools.…”

Section: Discussionmentioning

confidence: 94%

Automated Screening of Emergency Department Notes for Drug-Associated Bleeding Adverse Events Occurring in Older Adults

Boyce¹,

Jao²,

Miller³

et al. 2017

Appl Clin Inform

View full text Add to dashboard Cite

Objective To conduct research to show the value of text mining for automatically identifying suspected bleeding adverse drug events (ADEs) in the emergency department (ED). Methods A corpus of ED admission notes was manually annotated for bleeding ADEs. The notes were taken for patients ≥ 65 years of age who had an ICD-9 code for bleeding, the presence of hemoglobin value ≤ 8 g/dL, or were transfused> 2 units of packed red blood cells. This training corpus was used to develop bleeding ADE algorithms using Random Forest and Classification and Regression Tree (CART). A completely separate set of notes was annotated and used to test the classification performance of the final models using the area under the ROC curve (AUROC). Results The best performing CART resulted in an AUROC on the training set of 0.882. The model's AUROC on the test set was 0.827. At a sensitivity of 0.679, the model had a specificity of 0.908 and a positive predictive value (PPV) of 0.814. It had a relatively simple and intuitive structure consisting of 13 decision nodes and 14 leaf nodes. Decision path probabilities ranged from 0.041 to 1.0. The AUROC for the best performing Random Forest method on the training set was 0.917. On the test set, the model's AUROC was 0.859. At a sensitivity of 0.274, the model had a specificity of 0.986 and a PPV of 0.92. Conclusion Both models accurately identify bleeding ADEs using the presence or absence of certain clinical concepts in ED admission notes for older adult patients. The CART model is particularly noteworthy because it does not require significant technical overhead to implement. Future work should seek to replicate the results on a larger test set pulled from another institution.

show abstract

Section: Discussionmentioning

confidence: 94%

Automated Screening of Emergency Department Notes for Drug-Associated Bleeding Adverse Events Occurring in Older Adults

Boyce¹,

Jao²,

Miller³

et al. 2017

Appl Clin Inform

View full text Add to dashboard Cite

show abstract

“…The conscious work of end-users cannot be replaced by analyzing traces of data left as documentation within medical records that were never specifically intended to capture robust ADE information [28,29]. Instead, we stress the need to rethink the rationale and systems designed for reporting ADEs.…”

Section: Discussionmentioning

confidence: 99%

Designing an Adverse Drug Event Reporting System to Prevent Unintentional Reexposures to Harmful Drugs: Study Protocol for a Multiple Methods Design

Peddie¹,

Small²,

Badke³

et al. 2016

JMIR Res Protoc

View full text Add to dashboard Cite

BackgroundAdverse drug events (ADEs) are unintended and harmful events related to medication use. Up to 30% of serious ADEs recur within six months because culprit drugs are unintentionally represcribed and redispensed. Improving the electronic communication of ADE information between care providers, and across care settings, has the potential to reduce recurrent ADEs.ObjectiveWe aim to describe the methods used to design Action ADE, a novel electronic ADE reporting system that can be leveraged to prevent unintentional reexposures to harmful drugs in British Columbia, Canada.MethodsTo develop the new system, our team will use action research and participatory design, approaches that employ social scientific research methods and practitioner participation to generate insights into work settings and problem resolution. We will develop a systematic search strategy to review existing ADE reporting systems identified in academic and grey literature, and analyze the content of these systems to identify core data fields used to communicate ADE information. We will observe care providers in the emergency departments and on the wards of two urban tertiary hospitals and one urban community hospital, in one rural ambulatory care center, and in three community pharmacies in British Columbia, Canada. We will also conduct participatory workshops with providers to understand their needs and priorities related to communicating ADEs and preventing erroneous represcribing or redispensing of culprit medications. These methods will inform the iterative development of a preliminary paper-based reporting form, which we will then pilot test with providers in a real-world setting.ResultsThis is an ongoing project with results being published as analyses are completed. The systematic review has been completed; field observations, focus groups, and pilot testing of a preliminary paper-based design are ongoing. Results will inform the development of software that will enable clinically useful user-friendly documentation and communication of ADEs.ConclusionsWe take this approach with the recognition that information technology-based solutions in health care often fall short of expectations as a result of designers’ failure to account for organizational and work practice considerations, and the needs of end-users. We describe how integrating qualitative methods into an iterative participatory design process (planned in partnership with end-users) will allow us to address specific clinical needs, conceptualize linkages between systems, integrate the reporting system into clinicians’ workflow, and design the system to optimize its uptake into practice.

show abstract

“…However, a study carried out in Canada has investigated these parameters and observed a low sensitivity, from 2.6 to 15.8%, depending on the trigger, but a high specificity. The authors suggest that more sensitive triggers can be developed using clinical decision-making methods, in which clinical judgment is initially used to define options as predictors 28 .…”

Section: And In This Study No Patients Needed Hospitalizationmentioning

confidence: 99%

Uso de rastreadores para busca de reações adversas a medicamentos como motivo de admissão de idosos em pronto-socorro

Nagai

Takahashi

Pinto

et al. 2018

Ciênc. saúde coletiva

View full text Add to dashboard Cite

Resumo Reações adversas a medicamentos (RAM) podem ser causa de enfermidades, incapacidades ou óbitos, principalmente em idosos. Realizou-se busca ativa de suspeitas de RAM que motivaram a procura de idosos por cuidados em pronto socorro (PS) utilizando rastreadores e discutindo seu uso nessa unidade de atendimento. Tratou-se de um estudo transversal e retrospectivo que utilizou adaptação dos rastreadores do “Institute of Healthcare Improvement”. Foram analisados 287 prontuários, nos quais foram encontrados 38 rastreadores que identificaram 7 suspeitas de RAM. Uma foi encontrada sem uso dos rastreadores. Assim, no total, foram encontradas 8 reações (2,79%). Os medicamentos mais implicados foram os para trato alimentar e metabolismo e sistema cardiovascular. Dos rastreadores testados, alguns são essenciais para uso em PS, como os que indicam problemas com anticoagulantes, hipoglicemiantes e anti-hipertensivos, medicamentos bastante utilizados por idosos. Outros são adequados a estudos prospectivos, por necessitarem avaliação detalhada para confirmar da suspeita de RAM. Os rastreadores devem ser adaptados para cada instituição, mas se mostraram úteis para a detecção das suspeitas de RAM, identificando problemas ocorridos fora do ambiente hospitalar e sinalizando medicamentos que envolvem maior risco ao paciente idoso.

show abstract

Performance of trigger tools in identifying adverse drug events in emergency department patients: a validation study

Cited by 33 publications

References 47 publications

Automated Screening of Emergency Department Notes for Drug-Associated Bleeding Adverse Events Occurring in Older Adults

Automated Screening of Emergency Department Notes for Drug-Associated Bleeding Adverse Events Occurring in Older Adults

Designing an Adverse Drug Event Reporting System to Prevent Unintentional Reexposures to Harmful Drugs: Study Protocol for a Multiple Methods Design

Uso de rastreadores para busca de reações adversas a medicamentos como motivo de admissão de idosos em pronto-socorro

Contact Info

Product

Resources

About