Search

Introduction: Clinical notes, biomarkers, and neuroimaging have proven valuable in dementia prediction models. Whether commonly available structured clinical data can predict dementia is an emerging area of research. We aimed to predict gold-standard, research-based diagnoses of dementia including Alzheimer's disease (AD) and/or Alzheimer's disease related dementias (ADRD), in addition to ICD-based AD and/or ADRD diagnoses, in a well-phenotyped, population-based cohort using a machine learning approach., Methods: Administrative healthcare data (k = 163 diagnostic features), in addition to census/vital record sociodemographic data (k = 6 features), were linked to the Cache County Study (CCS, 1995-2008)., Results: Among successfully linked UPDB-CCS participants (n = 4206), 522 (12.4%) had incident dementia (AD alone, AD comorbid with ADRD, or ADRD alone) as per the CCS "gold standard" assessments. Random Forest models, with a 1-year prediction window, achieved the best performance with an Area Under the Curve (AUC) of 0.67. Accuracy declined for dementia subtypes: AD/ADRD (AUC = 0.65); ADRD (AUC = 0.49). Accuracy improved when using ICD-based dementia diagnoses (AUC = 0.77)., Discussion: Commonly available structured clinical data (without labs, notes, or prescription information) demonstrate modest ability to predict "gold-standard" research-based AD/ADRD diagnoses, corroborated by prior research. Using ICD diagnostic codes to identify dementia as done in the majority of machine learning dementia prediction models, as compared to "gold-standard" dementia diagnoses, can result in higher accuracy, but whether these models are predicting true dementia warrants further research., (© 2024. The Author(s).)

Background and Objectives: Isolated REM sleep behavior disorder (iRBD) carries increased risk of neurodegenerative parkinsonian disorder or dementia (NPD) but is difficult to accurately screen for in the community. Health care data offer the opportunity to identify large numbers of iRBD cases among outpatients. We aimed to determine the positive predictive value (PPV) of an RBD International Classification of Disorders (ICD) code for actual iRBD based on manual review of the electronic health record (EHR), examine risk of NPD diagnosis, and explore whether a statistical model developed using selected EHR data can identify individuals with the RBD ICD code who have high probability for actual iRBD., Methods: In this retrospective cohort study, a search of the EHR at a single health care system was conducted to identify outpatients who received the ICD9 or ICD10 RBD code in 2011-2021. The EHR for each case was manually reviewed. Secondary RBD cases were excluded. Remaining cases were classified as no iRBD or actual iRBD (possible, probable, or definite). Incident cases of NPD were identified. PPV of presence of the RBD ICD code for actual iRBD was calculated. Cumulative incidence of NPD with death as a competing event was compared in those with vs without iRBD. Least absolute shrinkage and selection operator (LASSO) regression was used to build a prediction model for iRBD, and the model was validated in an independent data set., Results: Among 1,130 cases with the RBD ICD code, 499 had secondary causes of RBD. For the remaining 628 cases, EHR review indicated no iRBD in 168 (26.8%). PPV of the RBD ICD code was 73.25%. Over a median follow-up of 4.7 years, compared with the no iRBD group, the iRBD group had a higher risk of NPD (subdistribution hazard ratio = 10.4 [95% CI 2.5-43.1]). The LASSO prediction model for iRBD had an area under the receiver operating characteristic curve of 0.844 (95% CI 0.806-0.880)., Discussion: PPV of an RBD ICD code is moderate. In the real-world setting, patients with iRBD had a high risk of incident diagnosis of NPD over 4.7 years. Results indicate feasibility of using statistical models developed using EHR data to accurately predict iRBD., Competing Interests: The authors report no relevant disclosures. Full disclosure form information provided by the authors is available with the full text of this article at Neurology.org/cp., (© 2024 American Academy of Neurology.)

Clinical decision support systems (CDSS) are routinely employed in clinical settings to improve quality of care, ensure patient safety, and deliver consistent medical care. However, rule-based CDSS, currently available, do not feature reusable rules. In this study, we present CDSS with reusable rules. Our solution includes a common CDSS module, electronic medical record (EMR) specific adapters, CDSS rules written in the clinical quality language (CQL) (derived from CDC immunization recommendations), and patient records in fast healthcare interoperability resources (FHIR) format. The proposed CDSS is entirely browser-based and reachable within the user's EMR interface at the client-side. This helps to avoid the transmission of patient data and privacy breaches. Additionally, we propose to provide means of managing and maintaining CDSS rules to allow the end users to modify them independently. Successful implementation and deployment were achieved in OpenMRS and OpenEMR during initial testing.

Importance: β-lactam (BL) allergies are the most common drug allergy worldwide, but most are reported in error. BL allergies are also well-established risk factors for adverse drug events and antibiotic-resistant infections during inpatient health care encounters, but the understanding of the long-term outcomes of patients with BL allergies remains limited., Objective: To evaluate the long-term clinical outcomes of patients with BL allergies., Design, Setting, and Participants: This longitudinal retrospective cohort study was conducted at a single regional health care system in western Pennsylvania. Electronic health records were analyzed for patients who had an index encounter with a diagnosis of sepsis, pneumonia, or urinary tract infection between 2007 and 2008. Patients were followed-up until death or the end of 2018. Data analysis was performed from January 2022 to January 2024., Exposure: The presence of any BL class antibiotic in the allergy section of a patient's electronic health record, evaluated at the earliest occurring observed health care encounter., Main Outcomes and Measures: The primary outcome was all-cause mortality, derived from the Social Security Death Index. Secondary outcomes were defined using laboratory and microbiology results and included infection with methicillin-resistant Staphylococcus aureus (MRSA), Clostridium difficile, or vancomycin-resistant Enterococcus (VRE) and severity and occurrence of acute kidney injury (AKI). Generalized estimating equations with a patient-level panel variable and time exposure offset were used to evaluate the odds of occurrence of each outcome between allergy groups., Results: A total of 20 092 patients (mean [SD] age, 62.9 [19.7] years; 12 231 female [60.9%]), of whom 4211 (21.0%) had BL documented allergy and 15 881 (79.0%) did not, met the inclusion criteria. A total of 3513 patients (17.5%) were Black, 15 358 (76.4%) were White, and 1221 (6.0%) were another race. Using generalized estimating equations, documented BL allergies were not significantly associated with the odds of mortality (odds ratio [OR], 1.02; 95% CI, 0.96-1.09). BL allergies were associated with increased odds of MRSA infection (OR, 1.44; 95% CI, 1.36-1.53), VRE infection (OR, 1.18; 95% CI, 1.05-1.32), and the pooled rate of the 3 evaluated antibiotic-resistant infections (OR, 1.33; 95% CI, 1.30-1.36) but were not associated with C difficile infection (OR, 1.04; 95% CI, 0.94-1.16), stage 2 and 3 AKI (OR, 1.02; 95% CI, 0.96-1.10), or stage 3 AKI (OR, 1.06; 95% CI, 0.98-1.14)., Conclusions and Relevance: Documented BL allergies were not associated with the long-term odds of mortality but were associated with antibiotic-resistant infections. Health systems should emphasize accurate allergy documentation and reduce unnecessary BL avoidance.

Translational research requires data at multiple scales of biological organization. Advancements in sequencing and multi-omics technologies have increased the availability of these data, but researchers face significant integration challenges. Knowledge graphs (KGs) are used to model complex phenomena, and methods exist to construct them automatically. However, tackling complex biomedical integration problems requires flexibility in the way knowledge is modeled. Moreover, existing KG construction methods provide robust tooling at the cost of fixed or limited choices among knowledge representation models. PheKnowLator (Phenotype Knowledge Translator) is a semantic ecosystem for automating the FAIR (Findable, Accessible, Interoperable, and Reusable) construction of ontologically grounded KGs with fully customizable knowledge representation. The ecosystem includes KG construction resources (e.g., data preparation APIs), analysis tools (e.g., SPARQL endpoint resources and abstraction algorithms), and benchmarks (e.g., prebuilt KGs). We evaluated the ecosystem by systematically comparing it to existing open-source KG construction methods and by analyzing its computational performance when used to construct 12 different large-scale KGs. With flexible knowledge representation, PheKnowLator enables fully customizable KGs without compromising performance or usability., (© 2024. The Author(s).)

Competing Interests: Declaration of Conflicting InterestsThe authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Increased sales of natural products (NPs) in the US and growing safety concerns highlight the need for NP pharmacovigilance. A challenge for NP pharmacovigilance is ambiguity when referring to NPs in spontaneous reporting systems. We used a combination of fuzzy string-matching and a neural network to reduce this ambiguity. Our aim is to increase the capture of reports involving NPs in the US Food and Drug Administration Adverse Event Reporting System (FAERS). For this, we utilized Gestalt pattern-matching (GPM) and Siamese neural network (SM) to identify potential mentions of NPs of interest in 389,386 FAERS reports with unmapped drug names. A team of health professionals refined the candidates identified in the previous step through manual review and annotation. After candidate adjudication, GPM identified 595 unique NP names and SM 504. There was little overlap between candidates identified by each (Non-overlapping: GPM 347, SM 248). We identified a total of 686 novel NP names from FAERS reports. Including these names in the FAERS collection yielded 3,486 additional reports mentioning NPs., (© 2024. The Author(s).)

Background: There is a lack of understanding of the barriers reported by healthcare providers when evaluating beta-lactam allergies, but knowledge of these barriers is required for practical and effective implementation interventions., Methods: Twenty-five healthcare providers, consisting of physicians, nurses and pharmacists practicing in the areas of intensive care, emergency medicine, infectious disease and general hospital practice, were interviewed between September 2021 and July 2023. Twenty-three of these providers were practising in the USA. A semi-structured interview guide grounded in the Theoretical Domain Framework was used for the interviews. Deductive and inductive analysis was performed on the interview transcripts, and translated into intervention recommendations using the Behaviour Change Wheel., Results: Widely held beliefs included a lack of clear policy for the evaluation of allergies, confusing or missing documentation of allergy information, confidence in their own and their colleagues' ability to evaluate allergies when information is available, and pharmacists as the provider most equipped to evaluate beta-lactam allergies., Conclusions: Health systems should adopt and disseminate policies for the evaluation of beta-lactam allergies, and promote the use of pharmacists in the evaluation of drug allergies when possible. Allergy sections of electronic health records should be reworked to encourage unambiguous documentation of allergy reactions and support using previously tolerated beta-lactam antibiotics., (Copyright © 2023 The Healthcare Infection Society. Published by Elsevier Ltd. All rights reserved.)

Background: Colchicine has a narrow therapeutic index. Its toxicity can be increased due to concomitant exposure to drugs inhibiting its metabolic pathway; these are cytochrome P450 3A4 (CYP3A4) and P-glycoprotein (P-gp)., Objective: To examine clinical outcomes associated with colchicine drug interactions using the spontaneous reports of the US Food and Drug Administration (FDA) Adverse Event Reporting System (FAERS)., Methods: We conducted a disproportionality analysis using FAERS data from January 2004 through June 2020. The reporting odds ratio (ROR) and observed-to-expected ratio (O/E) with shrinkage for adverse events related to colchicine's toxicity (ie, rhabdomyolysis/myopathy, agranulocytosis, hemorrhage, acute renal failure, hepatic failure, arrhythmias, torsade de pointes/QT prolongation, and cardiac failure) were compared between FAERS reports., Results: A total of 787 reports included the combined mention of colchicine, an inhibitor of both CYP3A4 and P-gp drug, and an adverse event of interest. Among reports that indicated the severity, 61% mentioned hospitalization and 24% death. A total of 37 ROR and 34 O/E safety signals involving colchicine and a CYP3A4/P-gp inhibitor were identified. The strongest ROR signal was for colchicine + atazanavir and rhabdomyolysis/myopathy (ROR = 35.4, 95% CI: 12.8-97.6), and the strongest O/E signal was for colchicine + atazanavir and agranulocytosis (O/E = 3.79, 95% credibility interval: 3.44-4.03)., Conclusion and Relevance: This study identifies numerous safety signals for colchicine and CYP3A4/P-gp inhibitor drugs. Avoiding the interaction or monitoring for toxicity in patients when co-prescribing colchicine and these agents is highly recommended.

Introduction: This study aims to explore machine learning (ML) methods for early prediction of Alzheimer's disease (AD) and related dementias (ADRD) using the real-world electronic health records (EHRs)., Methods: A total of 23,835 ADRD and 1,038,643 control patients were identified from the OneFlorida+ Research Consortium. Two ML methods were used to develop the prediction models. Both knowledge-driven and data-driven approaches were explored. Four computable phenotyping algorithms were tested., Results: The gradient boosting tree (GBT) models trained with the data-driven approach achieved the best area under the curve (AUC) scores of 0.939, 0.906, 0.884, and 0.854 for early prediction of ADRD 0, 1, 3, or 5 years before diagnosis, respectively. A number of important clinical and sociodemographic factors were identified., Discussion: We tested various settings and showed the predictive ability of using ML approaches for early prediction of ADRD with EHRs. The models can help identify high-risk individuals for early informed preventive or prognostic clinical decisions., (© 2023 The Authors. Alzheimer's & Dementia published by Wiley Periodicals LLC on behalf of Alzheimer's Association.)

Objective: Despite the benefits of the tailored drug-drug interaction (DDI) alerts and the broad dissemination strategy, the uptake of our tailored DDI alert algorithms that are enhanced with patient-specific and context-specific factors has been limited. The goal of the study was to examine barriers and health care system dynamics related to implementing tailored DDI alerts and identify the factors that would drive optimization and improvement of DDI alerts., Methods: We employed a qualitative research approach, conducting interviews with a participant interview guide framed based on Proctor's taxonomy of implementation outcomes and informed by the Theoretical Domains Framework. Participants included pharmacists with informatics roles within hospitals, chief medical informatics officers, and associate medical informatics directors/officers. Our data analysis was informed by the technique used in grounded theory analysis, and the reporting of open coding results was based on a modified version of the Safety-Related Electronic Health Record Research Reporting Framework., Results: Our analysis generated 15 barriers, and we mapped the interconnections of these barriers, which clustered around three entities (i.e., users, organizations, and technical stakeholders). Our findings revealed that misaligned interests regarding DDI alert performance and misaligned expectations regarding DDI alert optimizations among these entities within health care organizations could result in system inertia in implementing tailored DDI alerts., Conclusion: Health care organizations primarily determine the implementation and optimization of DDI alerts, and it is essential to identify and demonstrate value metrics that health care organizations prioritize to enable tailored DDI alert implementation. This could be achieved via a multifaceted approach, such as partnering with health care organizations that have the capacity to adopt tailored DDI alerts and identifying specialists who know users' needs, liaise with organizations and vendors, and facilitate technical stakeholders' work. In the future, researchers can adopt the systematic approach to study tailored DDI implementation problems from other system perspectives (e.g., the vendors' system)., Competing Interests: None declared., (Thieme. All rights reserved.)

Objective: Between 2012 and 2017, the U.S. Food and Drug Administration (FDA) approved 10 antidiabetic indicated therapies. Due to the limited literature on voluntarily reported safety outcomes for recently approved antidiabetic drugs, this study investigated adverse drug reactions (ADRs) reported in the FDA Adverse Event Reporting System (FAERS)., Research Design and Methods: A disproportionality analysis of spontaneously reported ADRs was conducted. FAERS reports from January 1, 2012 to March 31, 2022 were compiled, allowing a 5-year buffer following drug approval in 2017. Reporting odds ratios were calculated for the top 10 ADRs, comparing new diabetic agents to the other approved drugs in their therapeutic class., Results: 127,525 reports were identified for newly approved antidiabetic medications listed as the primary suspect (PS). For sodium-glucose co-transporter-2 (SGLT-2) inhibitors, the odds of blood glucose increased, nausea, and dizziness being reported was greater for empagliflozin. Dapagliflozin was associated with greater reports of weight decreased. Canagliflozin was found to have a disproportionally higher number of reports for diabetic ketoacidosis, toe amputation, acute kidney injury, fungal infections, and osteomyelitis. Assessing glucagon-like peptide-1 (GLP-1) receptor agonists, dulaglutide and semaglutide were associated with greater reports of gastrointestinal adverse drug reactions. Exenatide was disproportionally associated with injection site reactions and pancreatic carcinoma reports., Conclusion: Pharmacovigilance studies utilizing a large publicly available dataset allow an essential opportunity to evaluate the safety profile of antidiabetic drugs utilized in clinical practice. Additional research is needed to evaluate these reported safety concerns for recently approved antidiabetic medications to determine causality., Competing Interests: The authors declare that there is no conflict of interest., (© The Author(s), 2023.)

Kratom is a widely used Asian botanical that has gained popularity in the United States due to a perception that it can treat pain, anxiety, and opioid withdrawal symptoms. The American Kratom Association estimates 10-16 million people use kratom. Kratom-associated adverse drug reactions (ADRs) continue to be reported and raise concerns about the safety profile of kratom. However, studies are lacking that describe the overall pattern of kratom-associated adverse events and quantify the association between kratom and adverse events. ADRs reported to the US Food and Drug Administration Adverse Event Reporting System from January 2004 through September 2021 were used to address these knowledge gaps. Descriptive analysis was conducted to analyze kratom-related adverse reactions. Conservative pharmacovigilance signals based on observed-to-expected ratios with shrinkage were estimated by comparing kratom to all other natural products and drugs. Based on 489 deduplicated kratom-related ADR reports, users were young (mean age 35.5 years), and more often male (67.5%) than female patients (23.5%). Cases were predominantly reported since 2018 (94.2%). Fifty-two disproportionate reporting signals in 17 system-organ-class categories were generated. The observed/reported number of kratom-related accidental death reports was 63-fold greater than expected. There were eight strong signals related to addiction or drug withdrawal. An excess proportion of ADR reports were about kratom-related drug complaints, toxicity to various agents, and seizures. Although further research is needed to assess the safety of kratom, clinicians and consumers should be aware that real-world evidence points to potential safety threats., (© 2023 The Authors. Clinical and Translational Science published by Wiley Periodicals LLC on behalf of American Society for Clinical Pharmacology and Therapeutics.)

Background: Causal feature selection is essential for estimating effects from observational data. Identifying confounders is a crucial step in this process. Traditionally, researchers employ content-matter expertise and literature review to identify confounders. Uncontrolled confounding from unidentified confounders threatens validity, conditioning on intermediate variables (mediators) weakens estimates, and conditioning on common effects (colliders) induces bias. Additionally, without special treatment, erroneous conditioning on variables combining roles introduces bias. However, the vast literature is growing exponentially, making it infeasible to assimilate this knowledge. To address these challenges, we introduce a novel knowledge graph (KG) application enabling causal feature selection by combining computable literature-derived knowledge with biomedical ontologies. We present a use case of our approach specifying a causal model for estimating the total causal effect of depression on the risk of developing Alzheimer's disease (AD) from observational data., Methods: We extracted computable knowledge from a literature corpus using three machine reading systems and inferred missing knowledge using logical closure operations. Using a KG framework, we mapped the output to target terminologies and combined it with ontology-grounded resources. We translated epidemiological definitions of confounder, collider, and mediator into queries for searching the KG and summarized the roles played by the identified variables. We compared the results with output from a complementary method and published observational studies and examined a selection of confounding and combined role variables in-depth., Results: Our search identified 128 confounders, including 58 phenotypes, 47 drugs, 35 genes, 23 collider, and 16 mediator phenotypes. However, only 31 of the 58 confounder phenotypes were found to behave exclusively as confounders, while the remaining 27 phenotypes played other roles. Obstructive sleep apnea emerged as a potential novel confounder for depression and AD. Anemia exemplified a variable playing combined roles., Conclusion: Our findings suggest combining machine reading and KG could augment human expertise for causal feature selection. However, the complexity of causal feature selection for depression with AD highlights the need for standardized field-specific databases of causal variables. Further work is needed to optimize KG search and transform the output for human consumption., Competing Interests: Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2023. Published by Elsevier Inc.)

Common data models solve many challenges of standardizing electronic health record (EHR) data but are unable to semantically integrate all of the resources needed for deep phenotyping. Open Biological and Biomedical Ontology (OBO) Foundry ontologies provide computable representations of biological knowledge and enable the integration of heterogeneous data. However, mapping EHR data to OBO ontologies requires significant manual curation and domain expertise. We introduce OMOP2OBO, an algorithm for mapping Observational Medical Outcomes Partnership (OMOP) vocabularies to OBO ontologies. Using OMOP2OBO, we produced mappings for 92,367 conditions, 8611 drug ingredients, and 10,673 measurement results, which covered 68-99% of concepts used in clinical practice when examined across 24 hospitals. When used to phenotype rare disease patients, the mappings helped systematically identify undiagnosed patients who might benefit from genetic testing. By aligning OMOP vocabularies to OBO ontologies our algorithm presents new opportunities to advance EHR-based deep phenotyping., (© 2023. The Author(s).)

Background: Pharmacokinetic natural product-drug interactions (NPDIs) occur when botanical or other natural products are co-consumed with pharmaceutical drugs. With the growing use of natural products, the risk for potential NPDIs and consequent adverse events has increased. Understanding mechanisms of NPDIs is key to preventing or minimizing adverse events. Although biomedical knowledge graphs (KGs) have been widely used for drug-drug interaction applications, computational investigation of NPDIs is novel. We constructed NP-KG as a first step toward computational discovery of plausible mechanistic explanations for pharmacokinetic NPDIs that can be used to guide scientific research., Methods: We developed a large-scale, heterogeneous KG with biomedical ontologies, linked data, and full texts of the scientific literature. To construct the KG, biomedical ontologies and drug databases were integrated with the Phenotype Knowledge Translator framework. The semantic relation extraction systems, SemRep and Integrated Network and Dynamic Reasoning Assembler, were used to extract semantic predications (subject-relation-object triples) from full texts of the scientific literature related to the exemplar natural products green tea and kratom. A literature-based graph constructed from the predications was integrated into the ontology-grounded KG to create NP-KG. NP-KG was evaluated with case studies of pharmacokinetic green tea- and kratom-drug interactions through KG path searches and meta-path discovery to determine congruent and contradictory information in NP-KG compared to ground truth data. We also conducted an error analysis to identify knowledge gaps and incorrect predications in the KG., Results: The fully integrated NP-KG consisted of 745,512 nodes and 7,249,576 edges. Evaluation of NP-KG resulted in congruent (38.98% for green tea, 50% for kratom), contradictory (15.25% for green tea, 21.43% for kratom), and both congruent and contradictory (15.25% for green tea, 21.43% for kratom) information compared to ground truth data. Potential pharmacokinetic mechanisms for several purported NPDIs, including the green tea-raloxifene, green tea-nadolol, kratom-midazolam, kratom-quetiapine, and kratom-venlafaxine interactions were congruent with the published literature., Conclusion: NP-KG is the first KG to integrate biomedical ontologies with full texts of the scientific literature focused on natural products. We demonstrate the application of NP-KG to identify known pharmacokinetic interactions between natural products and pharmaceutical drugs mediated by drug metabolizing enzymes and transporters. Future work will incorporate context, contradiction analysis, and embedding-based methods to enrich NP-KG. NP-KG is publicly available at https://doi.org/10.5281/zenodo.6814507. The code for relation extraction, KG construction, and hypothesis generation is available at https://github.com/sanyabt/np-kg., Competing Interests: Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2023 The Author(s). Published by Elsevier Inc. All rights reserved.)

Colchicine is useful for the prevention and treatment of gout and a variety of other disorders. It is a substrate for CYP3A4 and P-glycoprotein (P-gp), and concomitant administration with CYP3A4/P-gp inhibitors can cause life-threatening drug-drug interactions (DDIs) such as pancytopenia, multiorgan failure, and cardiac arrhythmias. Colchicine can also cause myotoxicity, and coadministration with other myotoxic drugs may increase the risk of myopathy and rhabdomyolysis. Many sources of DDI information including journal publications, product labels, and online sources have errors or misleading statements regarding which drugs interact with colchicine, as well as suboptimal recommendations for managing the DDIs to minimize patient harm. Furthermore, assessment of the clinical importance of specific colchicine DDIs can vary dramatically from one source to another. In this paper we provide an evidence-based evaluation of which drugs can be expected to interact with colchicine, and which drugs have been stated to interact with colchicine but are unlikely to do so. Based on these evaluations we suggest management options for reducing the risk of potentially severe adverse outcomes from colchicine DDIs. The common recommendation to reduce the dose of colchicine when given with CYP3A4/P-gp inhibitors is likely to result in colchicine toxicity in some patients and therapeutic failure in others. A comprehensive evaluation of the almost 100 reported cases of colchicine DDIs is included in table form in the electronic supplementary material. Colchicine is a valuable drug, but improvements in the information about colchicine DDIs are needed in order to minimize the risk of serious adverse outcomes., (© 2022. The Author(s), under exclusive licence to Springer Nature Switzerland AG.)

Objective: The goal of this work is to evaluate if there is an increase in the risk of thromboembolic events (TEEs) due to concomitant exposure to dexamethasone and apixaban or rivaroxaban. Direct oral anticoagulants (DOACs), as well as corticosteroid dexamethasone, are commonly used to treat individuals hospitalised with COVID-19. Dexamethasone induces cytochrome P450-3A4 enzyme that also metabolises DOACs apixaban and rivaroxaban. This raises a concern about possible interaction between dexamethasone and DOACs that may reduce the efficacy of the DOACs and result in an increased risk of TEE., Design: We used nested case-control study design., Setting: This study was conducted in the National COVID Cohort Collaborative (N3C), the largest electronic health records repository for COVID-19 in the USA., Participants: Study participants were adults over 18 years who were exposed to a DOAC for 10 or more consecutive days. Exposure to dexamethasone was at least 5 or more consecutive days., Primary and Secondary Outcome Measures: Our primary exposure variable was concomitant exposure to dexamethasone for 5 or more days after exposure to either rivaroxaban or apixaban for 5 or more consecutive days. We used McNemar's Χ 2 test and adjusted logistic regression to evaluate association between concomitant use of dexamethasone with either apixaban or rivaroxaban., Results: McNemar's Χ 2 test did not find a discernible association of TEE in patients concomitantly exposed to dexamethasone and a DOAC (χ 2 =0.5, df=1, p=0.48). In addition, a conditional logistic regression model did not find an increase in the risk of TEE (adjusted OR 1.15, 95% CI 0.32 to 4.18)., Conclusion: This nested case-control study did not find evidence of an association between concomitant exposure to dexamethasone and a DOAC with an increase in risk of TEE. Due to small sample size, an association cannot be completely ruled out., Competing Interests: Competing interests: After the completion of this work, KMA became a full-time employee of Pfizer. GCA is past Chair and a current member of Food and Drug Administration's(FDA) Peripheral and Central Nervous System Advisory Committee; is a co-founding Principal and equity holder in Monument Analytics, a healthcare consultancy whose clients include the life sciences industry as well as plaintiffs in opioid litigation; and is a past member of OptumRx’s National P&T Committee. This arrangement has been reviewed and approved by Johns Hopkins University in accordance with its conflict of interest policies. CEL is an Executive Committee Member of the University of Pennsylvania’s Center for Pharmacoepidemiology Research and Training. The centre receives funds from Pfizer and Sanofi to support pharmacoepidemiology education. CEL recently received honoraria from the American College of Clinical Pharmacy Foundation, the University of Florida, the University of Massachusetts, and the Scientific and Data Coordinating Center for the NIDDK-funded Chronic Renal Insufficiency Cohort Study. CEL is a Special Government Employee of the US FDA and consults for their Reagan-Udall Foundation. CEL receives travel support from John Wiley & Sons. CEL’s spouse is an employee of Merck. Neither CEL nor his spouse owns stock in the company. There are no other disclosures to report., (© Author(s) (or their employer(s)) 2022. Re-use permitted under CC BY-NC. No commercial re-use. See rights and permissions. Published by BMJ.)

Introduction: Hydroxychloroquine can induce QT/QTc interval prolongation for some patients; however, little is known about its interactions with other QT-prolonging drugs., Objective: The purpose of this retrospective electronic health records study was to evaluate changes in the QTc interval in patients taking hydroxychloroquine with or without concomitant QT-prolonging medications., Methods: De-identified health records were obtained from the Cerner Health Facts ® database. Variables of interest included demographics, diagnoses, clinical procedures, laboratory tests, and medications. Patients were categorized into six cohorts based on exposure to hydroxychloroquine, methotrexate, or sulfasalazine alone, or the combination of any those drugs with any concomitant drug known to prolong the QT interval. Tisdale QTc risk score was calculated for each patient cohort. Two-sample paired t-tests were used to test differences between the mean before and after QTc measurements within each group and ANOVA was used to test for significant differences across the cohort means., Results: A statistically significant increase in QTc interval from the last measurement prior to concomitant exposure of 18.0 ms (95% CI 3.5-32.5; p < 0.05) was found in the hydroxychloroquine monotherapy cohort. QTc changes varied considerably across cohorts, with standard deviations ranging from 40.9 (hydroxychloroquine monotherapy) to 57.8 (hydroxychloroquine + sulfasalazine). There was no difference in QTc measurements among cohorts. The hydroxychloroquine + QTc-prolonging agent cohort had the highest average Tisdale Risk Score compared with those without concomitant exposure (p < 0.05)., Conclusion: Our analysis of retrospective electronic health records found hydroxychloroquine to be associated with a moderate increase in the QTc interval compared with sulfasalazine or methotrexate. However, the QTc was not significantly increased with concomitant exposure to other drugs known to increase QTc interval., (© 2022. The Author(s).)

Objective: The purpose of the study was to develop and validate a model to predict the risk of experiencing a fall for nursing home residents utilizing data that are electronically available at the more than 15 000 facilities in the United States., Materials and Methods: The fall prediction model was built and tested using 2 extracts of data (2011 through 2013 and 2016 through 2018) from the Long-term Care Minimum Dataset (MDS) combined with drug data from 5 skilled nursing facilities. The model was created using a hybrid Classification and Regression Tree (CART)-logistic approach., Results: The combined dataset consisted of 3985 residents with mean age of 77 years and 64% female. The model's area under the ROC curve was 0.668 (95% confidence interval: 0.643-0.693) on the validation subsample of the merged data., Discussion: Inspection of the model showed that antidepressant medications have a significant protective association where the resident has a fall history prior to admission, requires assistance to balance while walking, and some functional range of motion impairment in the lower body; even if the patient exhibits behavioral issues, unstable behaviors, and/or are exposed to multiple psychotropic drugs., Conclusion: The novel hybrid CART-logit algorithm is an advance over the 22 fall risk assessment tools previously evaluated in the nursing home setting because it has a better performance characteristic for the fall prediction window of ≤90 days and it is the only model designed to use features that are easily obtainable at nearly every facility in the United States., (© The Author(s) 2022. Published by Oxford University Press on behalf of the American Medical Informatics Association. All rights reserved. For permissions, please email: journals.permissions@oup.com.)

Introduction: Tizanidine is primarily metabolized via cytochrome P450 (CYP) 1A2 and therefore medications that inhibit the enzyme will affect the clearance of tizanidine, leading to increased plasma concentrations of tizanidine and potentially serious adverse events., Objectives: Our aim was to study the occurrence of adverse events reported in the FDA Adverse Event Reporting System (FAERS) involving the combination of tizanidine and drugs that inhibit the metabolic activity of CYP1A2., Methods: A disproportionality analysis of FAERS reports from 2004 quarter 1 through 2020 quarter 3 was conducted to calculate the reporting odds ratio (ROR) of reports mentioning tizanidine in a suspect or interacting role or having any role, a CYP1A2 inhibitor, and the following adverse events: hypotension, bradycardia, syncope, shock, cardiorespiratory arrest, and fall or fracture., Results: A total of 89 reports were identified mentioning tizanidine, at least one CYP1A2 inhibitor, and one of the adverse events of interest. More than half of the reports identified tizanidine as having a suspect or interacting role (n = 59, 66.3%), and the reports more frequently involved women (n = 58, 65.1%). The median age was 56.1 years (standard deviation 17.1). Some of the important safety signals included interactions between tizanidine in a suspect or interacting role and ciprofloxacin (ROR for hypotension 28.1, 95% confidence interval [CI] 19.2-41.2) or fluvoxamine (ROR for hypotension 36.9, 95% CI 13.1-103.4), and also when reported in "any role" with ciprofloxacin (ROR for hypotension 6.3, 95% CI 4.7-8.5), fluvoxamine (ROR for hypotension 11.4, 95% CI 4.5-28.8), and zafirlukast (ROR for falls 16.0, 95% CI 6.1-42.1)., Conclusions: Reports involving tizanidine and a CYP1A2 inhibitor have higher odds of reporting hypotension. This study suggests that concurrent use of tizanidine with CYP1A2 inhibitors may lead to serious health consequences associated with low blood pressure such as falls and fractures., (© 2022. The Author(s), under exclusive licence to Springer Nature Switzerland AG.)

Ineffective computerized alerts for potential Drug-Drug Interactions (DDI) is a longstanding informatics issue. Prescribing clinicians often ignore or override such alerts due to lack of context and clinical relevance, among various other reasons. In this study, we reveiwed published data on the rate of DDI alert overrides and medications involved in the overrides. We identified 34 eligible studies from sites across Asia, Europe, the United States, and the United Kingdom. The override rate of DDI alerts ranged from 55% to 98%, with more than half of the studies reporting the most common drug pairs or medications involved in acceptance or overriding of alerts. The high prevalance of alert overrides highlights the need for decision support systems that take user, drug, and institutional factors into consideration, as well as actionable metrics to better characterize harm associated with overrides.

We created an online calculator using machine learning (ML) algorithms to impute the partial pressure of oxygen (PaO 2 )/fraction of delivered oxygen (FiO 2 ) ratio using the non-invasive peripheral saturation of oxygen (SpO 2 ) and compared the accuracy of the ML models we developed to published equations. We generated three ML algorithms (neural network, regression, and kernel-based methods) using seven clinical variable features (N = 9900 ICU events) and subsequently three features (N = 20,198 ICU events) as input into the models. Data from mechanically ventilated ICU patients were obtained from the publicly available Medical Information Mart for Intensive Care (MIMIC III) database and used for analysis. Compared to seven features, three features (SpO 2 , FiO 2 and PEEP) were sufficient to impute PaO 2 from the SpO 2 . Any of the ML models enabled imputation of PaO 2 from the SpO 2 with lower error and showed greater accuracy in predicting PaO 2 /FiO 2  ≤ 150 compared to the previously published log-linear and non-linear equations. To address potential hidden hypoxemia that occurs more frequently in Black patients, we conducted sensitivity analysis and show ML models outperformed published equations in both Black and White patients. Imputation using data from an independent validation cohort of ICU patients (N = 133) showed greater accuracy with ML models., (© 2022. The Author(s).)

Purpose: Between 2012 and 2017, the FDA approved 29 therapies for a cardiovascular disease (CVD) indication. Due to the limited literature on patient safety outcomes for recently approved CVD medications, this study investigated adverse drug reports (ADRs) reported in the FDA Adverse Event Reporting System (FAERS)., Methods: A disproportionality analysis of spontaneously reported ADR was conducted. Reports in FAERS from Quarter 1, 2012, through Quarter 1, 2019, were compiled, allowing a 2-year buffer following drug approval in 2017. Top 10 reported ADRs and reporting odds ratios (ROR; confidence interval (CI)), a measure of disproportionality, were analyzed and compared to drugs available prior to 2012 as appropriate., Results: Of 7,952,147 ADR reports, 95,016 (1.19%) consisted of reports for newly approved CVD medications. For oral anticoagulants, apixaban had significantly lower reports for anemia and renal failure compared to dabigatran and rivaroxaban but greater reports for neurological signs/symptoms, and arrhythmias. Evaluating heart failure drugs, sacubitril/valsartan had greater reports for acute kidney injury, coughing, potassium imbalances, and renal impairment but notably, lower for angioedema compared to lisinopril. Assessing familial hypercholesterolemia drugs, alirocumab had greater reports for joint-related-signs/symptoms compared to other agents in this category. A newer pulmonary arterial hypertension treatment, selexipag, had greater reports of reporting for bone/joint-related-signs/symptoms but riociguat had greater reports for hemorrhages and vascular hypotension., Conclusion: Pharmacovigilance studies allow an essential opportunity to evaluate the safety profile of CVD medications in clinical practice. Additional research is needed to evaluate these reported safety concerns for recently approved CVD medications., (© 2021. Springer Science+Business Media, LLC, part of Springer Nature.)

Background: Despite advances in interoperability standards, it remains challenging and often costly to share clinical decision support (CDS) across healthcare organizations. This is due in part to limited coordination among CDS components. To improve coordination of CDS components, Health Level 7 (HL7) has developed a suite of interoperability standards with Fast Health Interoperability Resources (FHIR) specification as a common information model. Evidence is needed to determine the feasibility of implementing these CDS components; therefore, the objective of this study was to investigate the coordination of emerging HL7 standards with modular CDS architecture components., Methods: We used a modular, standards-based architecture consisting of four components: data, logic, services, and applications. The implementation use-case was an application to support shared decision making in the context of drug-drug interactions (DDInteract)., Results: DDInteract uses FHIR as the data representation model, Clinical Quality Language for logic representation, CDS Hooks for the services layer, and Substitutable Medical Apps Reusable Technologies for application integration. DDInteract was first implemented in a sandbox environment and then in an electronic health record (Epic®) test environment. DDInteract can be integrated in clinical workflows through on-demand access from a menu or through CDS Hooks upon opening a patient's record or placing a medication order., Conclusion: In the context of drug interactions, DDInteract is the first application to leverage a full stack of emerging interoperability standards for each component of modular CDS architecture. The demonstrated feasibility of interoperable components can be generalized to other modular CDS applications., (Copyright © 2022 Elsevier B.V. All rights reserved.)

The University of Pittsburgh Medical Center Hillman Cancer Center Academy (Hillman Academy) has the primary goal of reaching high school students from underrepresented and disadvantaged backgrounds and guiding them through a cutting-edge research and professional development experience that positions them for success in STEM. With this focus, the Hillman Academy has provided nearly 300 authentic mentored research internship opportunities to 239 students from diverse backgrounds over the past 13 years most of whom matriculated into STEM majors in higher education. These efforts have helped shape a more diverse generation of future scientists and clinicians, who will enrich these fields with their unique perspectives and lived experiences. In this paper, we describe our program and the strategies that led to its growth into a National Institutes of Health Youth Enjoy Science-funded program including our unique multi-site structure, tiered mentoring platform, multifaceted recruitment approach, professional and academic development activities, and a special highlight of a set of projects with Deaf and Hard of Hearing students. We also share student survey data from the past six years that indicate satisfaction with the program, self-perceived gains in key areas of scientific development, awareness of careers in STEM, and an increased desire to pursue advanced degrees in STEM.

Background: Between 2012 and 2017, 25 new medications or combination products were approved by the Food and Drug Administration (FDA) for use in treatment of chronic lower respiratory diseases (CLRDs). With limited data on post-marketing patient exposure to these drugs, their safety profiles remain unknown. This study aims to provide post-marketing surveillance of these medications., Methods: A list of new CLRD medications approved between 2012 and 2017 was generated through searches on Drugs.com (https://www.drugs.com), FDA.gov (https://www.fda.gov), and IBM Micromedex (https://www.micromedexsolutions.com/home/dispatch/ssl/true). Data describing adverse drug reactions (ADRs) were collected from the FDA Adverse Event Reporting System for analysis. Of the 25 identified medications, we selected 4 medications indicated for asthma or COPD with at least 500 reports. Only ADRs catalogued with these medications as the primary suspect were analyzed. Reporting odds ratios were calculated for the top 10 ADRs of each CLRD medication., Results: A total of 61,682 ADR reports were collected for newly approved CLRD medications ( n = 27,190 older adults; n = 30,502 male). Reports of COPD medications (umeclidinium and umeclidinium/vilanterol) indicate that umeclidinium/vilanterol yielded a higher reporting odds ratio than umeclidinium alone for reports of pain. Fluticasone furoate/vilanterol had higher reporting odds ratios for cough, pain, and dizziness than budesonide/formoterol and fluticasone propionate/salmeterol., Conclusions: Our findings suggest that the incidence of different adverse events experienced by patients in post-marketing reports resembles the incidence reported in pre-marketing clinical trials for COPD medications, except for fluticasone furoate/vilanterol, which has several differences., Competing Interests: The authors have disclosed no conflicts of interest., (Copyright © 2021 by Daedalus Enterprises.)

Introduction: Drug safety research asks causal questions but relies on observational data. Confounding bias threatens the reliability of studies using such data. The successful control of confounding requires knowledge of variables called confounders affecting both the exposure and outcome of interest. However, causal knowledge of dynamic biological systems is complex and challenging. Fortunately, computable knowledge mined from the literature may hold clues about confounders. In this paper, we tested the hypothesis that incorporating literature-derived confounders can improve causal inference from observational data., Methods: We introduce two methods (semantic vector-based and string-based confounder search) that query literature-derived information for confounder candidates to control, using SemMedDB, a database of computable knowledge mined from the biomedical literature. These methods search SemMedDB for confounders by applying semantic constraint search for indications treated by the drug (exposure) and that are also known to cause the adverse event (outcome). We then include the literature-derived confounder candidates in statistical and causal models derived from free-text clinical notes. For evaluation, we use a reference dataset widely used in drug safety containing labeled pairwise relationships between drugs and adverse events and attempt to rediscover these relationships from a corpus of 2.2 M NLP-processed free-text clinical notes. We employ standard adjustment and causal inference procedures to predict and estimate causal effects by informing the models with varying numbers of literature-derived confounders and instantiating the exposure, outcome, and confounder variables in the models with dichotomous EHR-derived data. Finally, we compare the results from applying these procedures with naive measures of association (χ 2 and reporting odds ratio) and with each other., Results and Conclusions: We found semantic vector-based search to be superior to string-based search at reducing confounding bias. However, the effect of including more rather than fewer literature-derived confounders was inconclusive. We recommend using targeted learning estimation methods that can address treatment-confounder feedback, where confounders also behave as intermediate variables, and engaging subject-matter experts to adjudicate the handling of problematic covariates., (Published by Elsevier Inc.)

The botanical natural product goldenseal can precipitate clinical drug interactions by inhibiting cytochrome P450 (CYP) 3A and CYP2D6. Besides P-glycoprotein, effects of goldenseal on other clinically relevant transporters remain unknown. Established transporter-expressing cell systems were used to determine the inhibitory effects of a goldenseal extract, standardized to the major alkaloid berberine, on transporter activity. Using recommended basic models, the extract was predicted to inhibit the efflux transporter BCRP and uptake transporters OATP1B1/3. Using a cocktail approach, effects of the goldenseal product on BCRP, OATP1B1/3, OATs, OCTs, MATEs, and CYP3A were next evaluated in 16 healthy volunteers. As expected, goldenseal increased the area under the plasma concentration-time curve (AUC 0-inf ) of midazolam (CYP3A; positive control), with a geometric mean ratio (GMR) (90% confidence interval (CI)) of 1.43 (1.35-1.53). However, goldenseal had no effects on the pharmacokinetics of rosuvastatin (BCRP and OATP1B1/3) and furosemide (OAT1/3); decreased metformin (OCT1/2, MATE1/2-K) AUC 0-inf (GMR, 0.77 (0.71-0.83)); and had no effect on metformin half-life and renal clearance. Results indicated that goldenseal altered intestinal permeability, transport, and/or other processes involved in metformin absorption, which may have unfavorable effects on glucose control. Inconsistencies between model predictions and pharmacokinetic outcomes prompt further refinement of current basic models to include differential transporter expression in relevant organs and intestinal degradation/metabolism of the precipitant(s). Such refinement should improve in vitro-in vivo prediction accuracy, contributing to a standard approach for studying transporter-mediated natural product-drug interactions., (© 2020 The Authors. Clinical Pharmacology & Therapeutics © 2020 American Society for Clinical Pharmacology and Therapeutics.)

Objective: Alert fatigue is a common issue with off-the-shelf clinical decision support. Most warnings for drug-drug interactions (DDIs) are overridden or ignored, likely because they lack relevance to the patient's clinical situation. Existing alerting systems for DDIs are often simplistic in nature or do not take the specific patient context into consideration, leading to overly sensitive alerts. The objective of this study is to develop, validate, and test DDI alert algorithms that take advantage of patient context available in electronic health records (EHRs) data., Methods: Data on the rate at which DDI alerts were triggered but for which no action was taken over a 3-month period (override rates) from a single tertiary care facility were used to identify DDIs that were considered a high-priority for contextualized alerting. A panel of DDI experts developed algorithms that incorporate drug and patient characteristics that affect the relevance of such warnings. The algorithms were then implemented as computable artifacts, validated using a synthetic health records data, and tested over retrospective data from a single urban hospital., Results: Algorithms and computable knowledge artifacts were developed and validated for a total of 8 high priority DDIs. Testing on retrospective real-world data showed the potential for the algorithms to reduce alerts that interrupt clinician workflow by more than 50%. Two algorithms (citalopram/QT interval prolonging agents, and fluconazole/opioid) showed potential to filter nearly all interruptive alerts for these combinations., Conclusion: The 8 DDI algorithms are a step toward addressing a critical need for DDI alerts that are more specific to patient context than current commercial alerting systems. Data commonly available in EHRs can improve DDI alert specificity., (© The Author(s) 2021. Published by Oxford University Press on behalf of the American Medical Informatics Association.)

Despite the significant health impacts of adverse events associated with drug-drug interactions, no standard models exist for managing and sharing evidence describing potential interactions between medications. Minimal information models have been used in other communities to establish community consensus around simple models capable of communicating useful information. This paper reports on a new minimal information model for describing potential drug-drug interactions. A task force of the Semantic Web in Health Care and Life Sciences Community Group of the World-Wide Web consortium engaged informaticians and drug-drug interaction experts in in-depth examination of recent literature and specific potential interactions. A consensus set of information items was identified, along with example descriptions of selected potential drug-drug interactions (PDDIs). User profiles and use cases were developed to demonstrate the applicability of the model. Ten core information items were identified: drugs involved, clinical consequences, seriousness, operational classification statement, recommended action, mechanism of interaction, contextual information/modifying factors, evidence about a suspected drug-drug interaction, frequency of exposure, and frequency of harm to exposed persons. Eight best practice recommendations suggest how PDDI knowledge artifact creators can best use the 10 information items when synthesizing drug interaction evidence into artifacts intended to aid clinicians. This model has been included in a proposed implementation guide developed by the HL7 Clinical Decision Support Workgroup and in PDDIs published in the CDS Connect repository. The complete description of the model can be found at https://w3id.org/hclscg/pddi., Competing Interests: Author EG was employed by the company Pharmacy Consulting International. Author OH was employed by IBM. Author BH was employed by Genelex. The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest., (Copyright © 2021 Hochheiser, Jing, Garcia, Ayvaz, Sahay, Dumontier, Banda, Beyan, Brochhausen, Draper, Habiel, Hassanzadeh, Herrero-Zazo, Hocum, Horn, LeBaron, Malone, Nytrø, Reese, Romagnoli, Schneider, Zhang and Boyce.)

Background/objectives: Federally-mandated consultant pharmacist-conducted retrospective medication regimen reviews (MRRs) are designed to improve medication safety in nursing homes (NH). However, MRRs are potentially ineffective. A new model of care that improves access to and efficiency of consultant pharmacists is needed. The objective of this study was to determine the impact of pharmacist-led telemedicine services on reducing high-risk medication adverse drug events (ADEs) for NH residents using medication reconciliation and prospective MRR on admission plus ongoing clinical decision support alerts throughout the residents' stay., Design: Quality improvement study using a stepped-wedge design comparing the novel service to usual care in a one-year evaluation from November 2016 to October 2017., Setting: Four NHs (two urban, two suburban) in Southwestern Pennsylvania., Participants: All residents in the four NHs were screened. There were 2,127 residents admitted having 652 alerts in the active period., Intervention: Upon admission, pharmacists conducted medication reconciliation and prospective MRR for residents and also used telemedicine for communication with cognitively-intact residents. Post-admission, pharmacists received clinical decision support alerts to conduct targeted concurrent MRRs and telemedicine., Measurement: Main outcome was incidence of high-risk medication, alert-specific ADEs. Secondary outcomes included all-cause hospitalization, 30-day readmission rates, and consultant pharmacists' recommendations., Results: Consultant pharmacists provided 769 recommendations. The intervention group had a 92% lower incidence of alert-specific ADEs than usual care (9 vs 31; 0.14 vs 0.61/1,000-resident-days; adjusted incident rate ratio (AIRR) = 0.08 (95% confidence interval (CI) = 0.01-0.40]; P = .002). All-cause hospitalization was similar between groups (149 vs 138; 2.33 vs 2.70/1,000-resident-days; AIRR = 1.06 (95% CI = 0.72-1.58); P = .75), as were 30-day readmissions (110 vs 102; 1.72 vs 2.00/1,000-resident-days; AIRR = 1.21 (95% CI = 0.76-1.93); P = .42)., Conclusions: This is the first evaluation of the impact of pharmacist-led patient-centered telemedicine services to manage high-risk medications during transitional care and throughout the resident's NH stay, supporting a new model of patient care., (© 2020 The American Geriatrics Society.)

A longstanding issue with knowledge bases that discuss drug-drug interactions (DDIs) is that they are inconsistent with one another. Computerized support might help experts be more objective in assessing DDI evidence. A requirement for such systems is accurate automatic classification of evidence types. In this pilot study, we developed a hierarchical classifier to classify clinical DDI studies into formally defined evidence types. The area under the ROC curve for sub-classifiers in the ensemble ranged from 0.78 to 0.87. The entire system achieved an F1 of 0.83 and 0.63 on two held-out datasets, the latter consisting focused on completely novel drugs from what the system was trained on. The results suggest that it is feasible to accurately automate the classification of a sub-set of DDI evidence types and that the hierarchical approach shows promise. Future work will test more advanced feature engineering techniques while expanding the system to classify a more complex set of evidence types., (©2020 AMIA - All rights reserved.)

Background and Objectives: Medication exposure is a potential risk factor for falls and subsequent death and functional decline among older adults. However, controversy remains on the best way to assess medication exposure and which approach best predicts falls. The objective of the current study was to examine the association between different measures of medication exposure and falls risk among community-dwelling older adults., Research Design and Methods: This retrospective cohort study was conducted using Falls Free PA program data and a linked prescription claims data from Pennsylvania's Pharmaceutical Assistance Contract for the Elderly program. Participants were community-dwelling older adults living in Pennsylvania, United States. Three measures of medication exposure were assessed: (a) total number of regular medications (polypharmacy); (b) counts of potentially inappropriate medications derived from current prescription guidance tools (Fall Risk-Increasing Drugs [FRIDs], Beers Criteria); and (c) medication burden indices based on pharmacologic mechanisms (Anticholinergic Cognitive Burden, Drug Burden Index) all derived from claims data. The associations between the different medication risk measures and self-reported falls incidence were examined with univariate and multivariable negative binomial regression models to estimate incidence rate ratios (IRRs)., Results: Overall 343 older adults were included and there were 236 months with falls during 2,316 activity-adjusted person-months (10.2 falls per 100 activity-adjusted person-months). Of the 6 measures of medication risk assessed in multivariate models, only the use of 2 or more FRIDs (adjusted IRR 1.67 [95% CI: 1.04-2.68]) independently predicted falls risk. Among the 13 FRID drug classes, the only FRID class associated with an increased fall risk was antidepressants., Discussion and Implications: The presence of multiple FRIDs in a prescription is an independent risk factor for falls, even in older adults with few medications. Further investigation is required to examine whether deprescribing focused on FRIDs effectively prevents falls among this population., (© The Author(s) 2021. Published by Oxford University Press on behalf of The Gerontological Society of America.)

There are many gaps in scientific knowledge about the clinical significance of pharmacokinetic natural product-drug interactions (NPDIs) in which the natural product (NP) is the precipitant and a conventional drug is the object. The National Center for Complimentary and Integrative Health created the Center of Excellence for NPDI Research (NaPDI Center) (www.napdi.org) to provide leadership and guidance on the study of pharmacokinetic NPDIs. A key contribution of the Center is the first user-friendly online repository that stores and links pharmacokinetic NPDI data across chemical characterization, metabolomics analyses, and pharmacokinetic in vitro and clinical experiments (repo.napdi.org). The design is expected to help researchers more easily arrive at a complete understanding of pharmacokinetic NPDI research on a particular NP. The repository will also facilitate multidisciplinary collaborations, as the repository links all of the experimental data for a given NP across the study types. The current work describes the design of the repository, standard operating procedures used to enter data, and pharmacokinetic NPDI data that have been entered to date. To illustrate the usefulness of the NaPDI Center repository, more details on two high-priority NPs, cannabis and kratom, are provided as case studies. SIGNIFICANCE STATEMENT: The data and knowledge resulting from natural product-drug interaction (NPDI) studies is distributed across a variety of information sources, rendering difficulties to find, access, and reuse. The Center of Excellence for NPDI Research addressed these difficulties by developing the first user-friendly online repository that stores data from in vitro and clinical pharmacokinetic NPDI experiments and links them with study data from chemical characterization and metabolomics analyses of natural products that are also stored in the repository., (Copyright © 2020 by The Author(s).)

Background: The implementation of clinical decision support systems (CDSSs) as an intervention to foster clinical practice change is affected by many factors. Key factors include those associated with behavioral change and those associated with technology acceptance. However, the literature regarding these subjects is fragmented and originates from two traditionally separate disciplines: implementation science and technology acceptance., Objective: Our objective is to propose an integrated framework that bridges the gap between the behavioral change and technology acceptance aspects of the implementation of CDSSs., Methods: We employed an iterative process to map constructs from four contributing frameworks-the Theoretical Domains Framework (TDF); the Consolidated Framework for Implementation Research (CFIR); the Human, Organization, and Technology-fit framework (HOT-fit); and the Unified Theory of Acceptance and Use of Technology (UTAUT)-and the findings of 10 literature reviews, identified through a systematic review of reviews approach., Results: The resulting framework comprises 22 domains: agreement with the decision algorithm; attitudes; behavioral regulation; beliefs about capabilities; beliefs about consequences; contingencies; demographic characteristics; effort expectancy; emotions; environmental context and resources; goals; intentions; intervention characteristics; knowledge; memory, attention, and decision processes; patient-health professional relationship; patient's preferences; performance expectancy; role and identity; skills, ability, and competence; social influences; and system quality. We demonstrate the use of the framework providing examples from two research projects., Conclusions: We proposed BEAR (BEhavior and Acceptance fRamework), an integrated framework that bridges the gap between behavioral change and technology acceptance, thereby widening the view established by current models., (©Jhon Camacho, Manuela Zanoletti-Mannello, Zach Landis-Lewis, Sandra L Kane-Gill, Richard D Boyce. Originally published in the Journal of Medical Internet Research (http://www.jmir.org), 06.08.2020.)

Introduction: Colchicine is currently approved for the treatment of gout and familial Mediterranean fever, among other conditions. Clarithromycin, a strong inhibitor of CYP3A4 and P-glycoprotein, dramatically increases colchicine's half-life, augmenting the risk of a life-threatening adverse reaction when used inadvertently with colchicine., Objectives: The aim of this study was to examine the evidence and clinical implications of concomitant use of colchicine and clarithromycin., Methods: Case reports of colchicine-clarithromycin co-administration were searched using the FDA's Adverse Event Reporting System (FAERS) database. PubMed, EMBASE, and Web of Science electronic databases were also searched from January 2005 through November 2019 for articles reporting colchicine-clarithromycin concomitant use. Individual reports were reviewed to identify consequences of coadministration, dose, days to onset of interaction, symptoms, evidence of renal disease, time to resolution of symptoms, and Drug Interaction Probability Scale (DIPS) rating., Results: The FAERS search identified 58 reported cases, nearly 53% of which were from patients aged between 65 and 85 years. Of 30 reported deaths, 11 occurred in males, and 19 in females. Other frequent complications reported in FAERS included diarrhea (31%), pancytopenia (22%), bone marrow failure (14%), and vomiting (14%). From published literature, we identified 20 case reports of concomitant exposure, 19 of which were rated 'probable' and one 'possible' according to DIPS rating. Of these cases, four 'probable' patients expired. The documented onset of colchicine toxicity occurred within 5 days of starting clarithromycin, and death within 2 weeks of concomitant exposure., Conclusion: Clinical manifestations of colchicine-clarithromycin interaction may resemble other systemic diseases and may be life threatening. Understanding this clinically meaningful interaction can help clinicians avoid unsafe medication combinations.

Background:  Warfarin use can trigger the occurrence of bleeding independently or as a result of a drug-drug interaction when used in combination with nonsteroidal anti-inflammatory drugs (NSAIDs)., Objectives:  This article examines the risk of bleeding in individuals exposed to concomitant warfarin and NSAID compared with those taking warfarin alone (Prospero Registry ID 145237)., Methods:  PubMed, EMBASE, Scopus, and Web of Science were searched. The primary outcome of interest was gastrointestinal bleeding and general bleeding. Summary effects were calculated to estimate average treatment effect using random effects models. Heterogeneity was assessed using Cochran's Q and I 2 . Risk of bias was also assessed using the Agency for Healthcare Research and Quality bias assessment tool., Results:  A total of 651 studies were identified, of which 11 studies met inclusion criteria for meta-analysis. The odds ratio (OR) for gastrointestinal bleeding when exposed to warfarin and an NSAID was 1.98 (95% confidence interval [CI]: 1.55-2.53). The risk of gastrointestinal bleeding was also significantly elevated with exposure to a COX-2 inhibitor and warfarin relative to warfarin alone (OR = 1.90, 95% CI: 1.46-2.46). There was an increased risk of general bleeding with the combination of warfarin with NSAIDs (OR = 1.58, 95% CI: 1.18-2.12) or COX-2 inhibitors (OR = 1.54, 95% CI: 0.86-2.78) compared with warfarin alone., Conclusion:  Risk of bleeding is significantly increased among persons taking warfarin and a NSAID or COX-2 inhibitor together as compared with taking warfarin alone. It is important to caution patients about taking these medications in combination., Competing Interests: None declared., (Georg Thieme Verlag KG Stuttgart · New York.)

This paper introduces a database derived from Structured Product Labels (SPLs). SPLs are legally mandated snapshots containing information on all drugs released to market in the United States. Since publication is not required for pre-trial findings, we hypothesize that SPLs may contain knowledge absent in the literature, and hence "novel." SemMedDB is an existing database of computable knowledge derived from the literature. If SPL content could be similarly transformed, novel clinically relevant assertions in the SPLs could be identified through comparison with SemMedDB. After we derive a database (containing 4,297,481 assertions), we compare the extracted content with SemMedDB for recent FDA drug approvals. We find that novelty between the SPLs and the literature is nuanced, due to the redundancy of SPLs. Highlighting areas for improvement and future work, we conclude that SPLs contain a wealth of novel knowledge relevant to research and complementary to the literature., (©2020 AMIA - All rights reserved.)

Low concordance between drug-drug interaction (DDI) knowledge bases is a well-documented concern. One potential cause of inconsistency is variability between drug experts in approach to assessing evidence about potential DDIs. In this study, we examined the face validity and inter-rater reliability of a novel DDI evidence evaluation instrument designed to be simple and easy to use., Methods: A convenience sample of participants with professional experience evaluating DDI evidence was recruited. Participants independently evaluated pre-selected evidence items for 5 drug pairs using the new instrument. For each drug pair, participants labeled each evidence item as sufficient or insufficient to establish the existence of a DDI based on the evidence categories provided by the instrument. Participants also decided if the overall body of evidence supported a DDI involving the drug pair. Agreement was computed both at the evidence item and drug pair levels. A cut-off of ≥ 70% was chosen as the agreement threshold for percent agreement, while a coefficient > 0.6 was used as the cut-off for chance-corrected agreement. Open ended comments were collected and coded to identify themes related to the participants' experience using the novel approach., Results: The face validity of the new instrument was established by two rounds of evaluation involving a total of 6 experts. Fifteen experts agreed to participate in the reliability assessment, and 14 completed the study. Participant agreement on the sufficiency of 22 of the 34 evidence items (65%) did not exceed the a priori agreement threshold. Similarly, agreement on the sufficiency of evidence for 3 of the 5 drug pairs (60%) was poor. Chance-corrected agreement at the drug pair level further confirmed the poor interrater reliability of the instrument (Gwet's AC 1  = 0.24, Conger's Kappa = 0.24). Participant comments suggested several possible reasons for the disagreements including unaddressed subjectivity in assessing an evidence item's type and study design, an infeasible separation of evidence evaluation from the consideration of clinical relevance, and potential issues related to the evaluation of DDI case reports., Conclusions: Even though the key findings were negative, the study's results shed light on how experts approach DDI evidence assessment, including the importance situating evidence assessment within the context of consideration of clinical relevance. Analysis of participant comments within the context of the negative findings identified several promising future research directions including: novel computer-based support for evidence assessment; formal evaluation of a more comprehensive evidence assessment approach that requires consideration of specific, explicitly stated, clinical consequences; and more formal investigation of DDI case report assessment instruments., (Copyright © 2019 Elsevier Inc. All rights reserved.)

Potential drug-drug interactions (PDDI) rules are currently represented without any common standard making them difficult to update, maintain, and exchange. The PDDI minimum information model developed by the Semantic Web in the Healthcare and Life Sciences Community Group describes PDDI knowledge in an actionable format. In this paper, we report implementation and evaluation of CDS Services which represent PDDI knowledge with Clinical Quality Language (CQL). The suggested solution is based on emerging standards including CDS Hooks, FHIR, and CQL. Two use cases are selected, implemented with CQL rules and tested at the Connectathon held at the 32nd Annual Plenary & Working Group Meeting of HL7.

Background: Preventing drug interactions is an important goal to maximize patient benefit from medications. Summarizing potential drug-drug interactions (PDDIs) for clinical decision support is challenging, and there is no single repository for PDDI evidence. Additionally, inconsistencies across compendia and other sources have been well documented. Standard search strategies for complete and current evidence about PDDIs have not heretofore been developed or validated., Objective: This study aimed to identify common methods for conducting PDDI literature searches used by experts who routinely evaluate such evidence., Methods: We invited a convenience sample of 70 drug information experts, including compendia editors, knowledge-base vendors, and clinicians, via emails to complete a survey on identifying PDDI evidence. We created a Web-based survey that included questions regarding the (1) development and conduct of searches; (2) resources used, for example, databases, compendia, search engines, etc; (3) types of keywords used to search for the specific PDDI information; (4) study types included and excluded in searches; and (5) search terms used. Search strategy questions focused on 6 topics of the PDDI information-(1) that a PDDI exists; (2) seriousness; (3) clinical consequences; (4) management options; (5) mechanism; and (6) health outcomes., Results: Twenty participants (response rate, 20/70, 29%) completed the survey. The majority (17/20, 85%) were drug information specialists, drug interaction researchers, compendia editors, or clinical pharmacists, with 60% (12/20) having >10 years' experience. Over half (11/20, 55%) worked for clinical solutions vendors or knowledge-base vendors. Most participants developed (18/20, 90%) and conducted (19/20, 95%) search strategies without librarian assistance. PubMed (20/20, 100%) and Google Scholar (11/20, 55%) were most commonly searched for papers, followed by Google Web Search (7/20, 35%) and EMBASE (3/20, 15%). No respondents reported using Scopus. A variety of subscription and open-access databases were used, most commonly Lexicomp (9/20, 45%), Micromedex (8/20, 40%), Drugs@FDA (17/20, 85%), and DailyMed (13/20, 65%). Facts and Comparisons was the most commonly used compendia (8/20, 40%). Across the 6 attributes of interest, generic drug name was the most common keyword used. Respondents reported using more types of keywords when searching to identify the existence of PDDIs and determine their mechanism than when searching for the other 4 attributes (seriousness, consequences, management, and health outcomes). Regarding the types of evidence useful for evaluating a PDDI, clinical trials, case reports, and systematic reviews were considered relevant, while animal and in vitro data studies were not., Conclusions: This study suggests that drug interaction experts use various keyword strategies and various database and Web resources depending on the PDDI evidence they are seeking. Greater automation and standardization across search strategies could improve one's ability to identify PDDI evidence. Hence, future research focused on enhancing the existing search tools and designing recommended standards is needed., (©Amy J Grizzle, John Horn, Carol Collins, Jodi Schneider, Daniel C Malone, Britney Stottlemyer, Richard David Boyce. Originally published in the Journal of Medical Internet Research (http://www.jmir.org), 04.01.2019.)