The clinical pharmacist’s role in enhancing the relevance of a clinical decision support system. Cuvelier, E., Robert, L., Musy, E., Rousselière, C., Marcilly, R., Gautier, S., Odou, P., Beuscart, J. -., & Décaudin, B. International Journal of Medical Informatics, 155:104568, November, 2021.
The clinical pharmacist’s role in enhancing the relevance of a clinical decision support system [link]Paper  doi  abstract   bibtex   
Background Clinical decision support systems (CDSSs) can improve the quality of patient care by helping physicians to review their prescriptions and thus to optimize drug treatments. Nevertheless, the “alert fatigue” brought on by a large number of irrelevant alerts can decrease a CDSS’s effectiveness and thus clinical value. Involving a clinical pharmacist in the development and management of a CDSS can reduce the number of irrelevant alerts presented to physicians. Clinical pharmacists screen alerts and suggest PIs for physicians, corresponding to any proposed therapeutic change about health products, only for relevant alerts could improve the relevance and the acceptance of the information given to physicians about the risks faced by their patients. Objective To assess the value of involving clinical pharmacists in the development and maintenance of decision support rules for generating alerts and pharmaceutical interventions (PIs) and to describe the level of acceptance of these PIs by the physicians. Method In a retrospective, single-centre study, we evaluated the number of PIs accepted from alerts generated by the CDSS when a clinical pharmacist had developed and managed this tool. During the first 7 months of development of the CDSS, a clinical pharmacist analyzed alerts triggered by the CDSS according to its technical validity and pharmaceutical relevance. Lastly, for alerts that led to a PI, the level of acceptance by physicians was documented. Results During the study, 1430 alerts were analysed: 186 (13%) were considered to be technically invalid – mainly due to the characteristics of the interface. Of the 1244 (87.0%) technically valid alerts, 353 (24.6%) were pharmaceutically relevant and led to a PI. The three main causes of pharmaceutical irrelevance were a lack of specificity in the CDSS (70.8%), lack of relevance with regard to the ward’s habits (15.6%), and the pharmacist’s decision to recommend monitoring for the patient rather than sending a PI immediately (10.8%). 64.6% of the submitted PIs were accepted by the physicians. Conclusion The standardized analysis of alerts by a clinical pharmacist appears to be a good way of improving the development of CDSS by limiting the generation of irrelevant alerts and the latter’s transmission to physicians. The involvement of a clinical pharmacist in the development and implementation of a CDSS appears to be novel and may help to optimize drug treatment.
@article{cuvelier_clinical_2021,
	title = {The clinical pharmacist’s role in enhancing the relevance of a clinical decision support system},
	volume = {155},
	issn = {1386-5056},
	url = {https://www.sciencedirect.com/science/article/pii/S1386505621001945},
	doi = {10.1016/j.ijmedinf.2021.104568},
	abstract = {Background
Clinical decision support systems (CDSSs) can improve the quality of patient care by helping physicians to review their prescriptions and thus to optimize drug treatments. Nevertheless, the “alert fatigue” brought on by a large number of irrelevant alerts can decrease a CDSS’s effectiveness and thus clinical value. Involving a clinical pharmacist in the development and management of a CDSS can reduce the number of irrelevant alerts presented to physicians. Clinical pharmacists screen alerts and suggest PIs for physicians, corresponding to any proposed therapeutic change about health products, only for relevant alerts could improve the relevance and the acceptance of the information given to physicians about the risks faced by their patients.
Objective
To assess the value of involving clinical pharmacists in the development and maintenance of decision support rules for generating alerts and pharmaceutical interventions (PIs) and to describe the level of acceptance of these PIs by the physicians.
Method
In a retrospective, single-centre study, we evaluated the number of PIs accepted from alerts generated by the CDSS when a clinical pharmacist had developed and managed this tool. During the first 7 months of development of the CDSS, a clinical pharmacist analyzed alerts triggered by the CDSS according to its technical validity and pharmaceutical relevance. Lastly, for alerts that led to a PI, the level of acceptance by physicians was documented.
Results
During the study, 1430 alerts were analysed: 186 (13\%) were considered to be technically invalid – mainly due to the characteristics of the interface. Of the 1244 (87.0\%) technically valid alerts, 353 (24.6\%) were pharmaceutically relevant and led to a PI. The three main causes of pharmaceutical irrelevance were a lack of specificity in the CDSS (70.8\%), lack of relevance with regard to the ward’s habits (15.6\%), and the pharmacist’s decision to recommend monitoring for the patient rather than sending a PI immediately (10.8\%). 64.6\% of the submitted PIs were accepted by the physicians.
Conclusion
The standardized analysis of alerts by a clinical pharmacist appears to be a good way of improving the development of CDSS by limiting the generation of irrelevant alerts and the latter’s transmission to physicians. The involvement of a clinical pharmacist in the development and implementation of a CDSS appears to be novel and may help to optimize drug treatment.},
	language = {en},
	urldate = {2021-09-17},
	journal = {International Journal of Medical Informatics},
	author = {Cuvelier, E. and Robert, L. and Musy, E. and Rousselière, C. and Marcilly, R. and Gautier, S. and Odou, P. and Beuscart, J. -B. and Décaudin, B.},
	month = nov,
	year = {2021},
	keywords = {ADE, Alert fatigue, Clinical Pharmacist, Clinical decision support system, Evidence-based pharmacy practice, Prescription drug monitoring program},
	pages = {104568},
}

Downloads: 0