Drug-related causes attributed to acute kidney injury and their documentation in intensive care patients

Leveraging Natural Language Processing and Machine Learning Methods for Adverse Drug Event Detection in Electronic Health/Medical Records: A Scoping Review

BACKGROUND

Natural language processing (NLP) and machine learning (ML) techniques may help harness unstructured free-text electronic health record (EHR) data to detect adverse drug events (ADEs) and thus improve pharmacovigilance. However, evidence of their real-world effectiveness remains unclear.

OBJECTIVE

To summarise the evidence on the effectiveness of NLP/ML in detecting ADEs from unstructured EHR data and ultimately improve pharmacovigilance in comparison to other data sources.

METHODS

A scoping review was conducted by searching six databases in July 2023. Studies leveraging NLP/ML to identify ADEs from EHR were included. Titles/abstracts were screened by two independent researchers as were full-text articles. Data extraction was conducted by one researcher and checked by another. A narrative synthesis summarises the research techniques, ADEs analysed, model performance and pharmacovigilance impacts.

RESULTS

Seven studies met the inclusion criteria covering a wide range of ADEs and medications. The utilisation of rule-based NLP, statistical models, and deep learning approaches was observed. Natural language processing/ML techniques with unstructured data improved the detection of under-reported adverse events and safety signals. However, substantial variability was noted in the techniques and evaluation methods employed across the different studies and limitations exist in integrating the findings into practice.

CONCLUSIONS

Natural language processing (NLP) and machine learning (ML) have promising possibilities in extracting valuable insights with regard to pharmacovigilance from unstructured EHR data. These approaches have demonstrated proficiency in identifying specific adverse events and uncovering previously unknown safety signals that would not have been apparent through structured data alone. Nevertheless, challenges such as the absence of standardised methodologies and validation criteria obstruct the widespread adoption of NLP/ML for pharmacovigilance leveraging of unstructured EHR data.

Artificial intelligence: revolutionizing cardiology with large language models

Natural language processing techniques are having an increasing impact on clinical care from patient, clinician, administrator, and research perspective. Among others are automated generation of clinical notes and discharge letters, medical term coding for billing, medical chatbots both for patients and clinicians, data enrichment in the identification of disease symptoms or diagnosis, cohort selection for clinical trial, and auditing purposes. In the review, an overview of the history in natural language processing techniques developed with brief technical background is presented. Subsequently, the review will discuss implementation strategies of natural language processing tools, thereby specifically focusing on large language models, and conclude with future opportunities in the application of such techniques in the field of cardiology.

Nephrotoxic drug burden and predictors of exposure among patients with renal impairment in Ethiopia: A multi-center study

Background

Nephrotoxic drugs may hasten the decline in kidney function and worsen the progression of renal impairment as a result; these drugs should be avoided or used with caution in patients with pre-existing renal insufficiency. The purpose of this study was to assess the burden of nephrotoxic medication use and its predictors among patients with underlying renal impairment.

Methods

A multicenter, institution-based, cross-sectional study was conducted from May 30, 2021 to July 30, 2021, at medical wards. Renal impaired patients admitted during the data collection period who took at least one medication were enrolled in the study. A simple random sampling technique was used to select the study participants. Data was collected through an interview and a medical card review. Both bivariable and multivariable binary logistic regression analyses were fitted to identify factors associated with nephrotoxic drug use.

Results

Among the 422 participants, more than half of them (53.6 %) were male. The mean patient's age was 47.5 (±16.7) years. A total of 1310 drugs were prescribed for 422 patients with renal impairment, of which 80.15 % were nephrotoxic. Nephrotoxic drugs were prescribed for 66.4 % of patients. The burden of nephrotoxic medication prescription was significantly associated with variables like the presence of comorbidity (AOR = 6.31, 95 % CI: 2.01-19.79), the number of medications prescribed (AOR = 1.43, 95 % CI: 1.05-1.93), and the age of participants (AOR = 1.12, 95 % CI: 1.07-1.17).

Conclusion

The present study demonstrated that two-third of the patients with renal impairment were exposed to nephrotoxic medications. Furosemide, Enalapril, and vancomycin were the most frequently prescribed nephrotoxic medications. The study suggests that prescribers need to give special attention to older patients who have underlying renal insufficiency, a comorbid condition, and polypharmacy regarding exposure to contraindicated nephrotoxic medication.

Moving toward a contemporary classification of drug-induced kidney disease

Drug-induced kidney disease (DIKD) accounts for about one-fourth of all cases of acute kidney injury (AKI) in hospitalized patients, especially in critically ill setting. There is no standard definition or classification system of DIKD. To address this, a phenotype definition of DIKD using expert consensus was introduced in 2015. Recently, a novel framework for DIKD classification was proposed that incorporated functional change and tissue damage biomarkers. Medications were stratified into four categories, including "dysfunction without damage," "damage without dysfunction," "both dysfunction and damage," and "neither dysfunction nor damage" using this novel framework along with predominant mechanism(s) of nephrotoxicity for drugs and drug classes. Here, we briefly describe mechanisms and provide examples of drugs/drug classes related to the categories in the proposed framework. In addition, the possible movement of a patient's kidney disease between certain categories in specific conditions is considered. Finally, opportunities and barriers to adoption of this framework for DIKD classification in real clinical practice are discussed. This new classification system allows congruencies for DIKD with the proposed categorization of AKI, offering clarity as well as consistency for clinicians and researchers.