ICD-10 codes used to identify adverse drug events in administrative data: a systematic review

Adverse Drug Events in Ambulatory Care: A Cross-Sectional Study

BACKGROUND

Adverse drug events (ADEs) are understudied in the ambulatory care setting. We aim to estimate the prevalence and characteristics of ADEs in outpatient care using electronic health records (EHRs).

METHODS

This cross-sectional study included EHR data for patients who had an outpatient encounter at an academic medical center from 1 October 2018 through 31 December 2019. We developed a stratified sampling strategy based on a comprehensive set of 994 ADE-related International Classification of Disease (ICD-10) codes to identify clinical encounters and notes likely to contain ADEs. Within each ICD-10 likelihood group, clinical notes were randomly sampled and annotated for present or possible ADE-drug relationships and severity. The overall estimated population prevalence of ADEs presenting in the outpatient setting was calculated. The generalizability of the findings was assessed by comparing ICD-10 code frequencies against a large commercial database.

RESULTS

The study included 3126 notes (unique patient encounters) from 2882 unique patients. Of these, 1383 patient encounters (44.2%) had a present or possible ADE documented (6308 mentions). Of the 6038 ADEs mentioned, 14.1% were hypersensitivity reactions, 1.1% were life-threatening, 22.4% were serious, and 60.4% were significant. Main causal agents included anti-infectives (19.3%), central nervous system agents (12.8%), and cardiovascular agents (11.5%). The overall prevalence of present ADEs mentioned in the clinical notes was estimated to be 1.97 per 100 patient encounters (or 2.52 per 100 patient encounters when possible ADEs are included).

CONCLUSIONS

This study identified the overall population prevalence per encounter of ADEs in the outpatient population by leveraging ICD-10 codes and investigating ADEs documented in clinical notes. Understanding the ADE characteristics in a large corpus of outpatient documentation advances pharmacovigilance knowledge, enhancing the detection, monitoring, and prevention of ADEs in ambulatory care.

ICD-10 Codes to Identify Adverse Drug Events Associated with Antibiotics in Administrative Data

Antibiotics are among the most used therapeutics in primary care, and while their benefits are clear, the potential harms related to adverse drug events (ADEs) cannot be ignored. We outline the creation of a comprehensive list of diagnostic codes describing antibiotic-associated ADEs resulting in presentations to acute care hospitals. Methods: Previously published ADE codes were used to link BC hospitalizations to prior outpatient antibiotic prescriptions and were restricted based on whether patients received an antibiotic within a month prior to the ADE-related hospitalization. The code list was reviewed by two clinical experts independently for the likelihood of being antibiotic-associated. The inter-rater reliability was calculated using Kappa scores with 95% confidence intervals (CIs). Results: Of the 695 ICD-10 ADE codes with evidence of recent antibiotic administration, 72, 68, and 555 codes were considered likely, possibly, and unlikely antibiotic-associated, respectively. Conclusions: We outline a methodology for developing an ICD-10 code list for antibiotic-associated ADEs severe enough to warrant hospital admission. This will help to improve the use of administrative data to capture antibiotic-associated ADEs.

Impact of in-hospital medication changes on clinical outcomes in older inpatients: the journey and destination

BACKGROUND

Medication review is integral in the pharmacological management of older inpatients.

OBJECTIVE

To assess the association of in-hospital medication changes with 28-day postdischarge clinical outcomes.

METHODS

Retrospective cohort of 2000 inpatients aged ≥75 years. Medication changes included the number of increases (medications started or dose-increased) and decreases (medications stopped or dose-decreased) for (i) all medications, (ii) Drug Burden Index (DBI)-contributing medications and (iii) Beers Criteria 2015 medications (potentially inappropriate medications, PIMs). Changes also included differences in (i) the number of medications, (ii) the number of PIMs and (iii) DBI score, at discharge versus admission. Associations with clinical outcomes (28-day ED visit, readmission and mortality) were ascertained using logistic regression, adjusted for age, gender and principal diagnosis. For mortality, sensitivity analysis excluded end-of-life patients due to higher death risk. Patients were stratified into : (i) ≤4, (ii) 5-9 and (iii) ≥10 discharge medications.

RESULTS

The mean age was 86 years (SD = 5.8), with 59.1% female. Medication changes reduced ED visits and readmission risk for patients prescribed five to nine discharge medications, with no associations in patients prescribed ≤4 and ≥ 10 medications. In the five to nine medications group, decreasing PIMs reduced risks of ED visit (adjusted odds ratio, aOR 0.55, 95% CI 0.34-0.91, P = .02) and readmission (aOR 0.62, 95% CI 0.38-0.99, P = .04). Decreasing DBI-contributing medications reduced readmission risk (aOR 0.71, 95% CI 0.51-0.99, P = .04). Differences in PIMs reduced ED visit risk (aOR 0.65, 95% CI 0.43-0.99, P = .04). There were no associations with mortality in sensitivity analyses in all groups.

DISCUSSION

Medication changes were associated with reduced ED visits and readmission for patients prescribed five to nine discharge medications.

Drug-related hospitalizations - insights from the Czech Republic

Drug-related hospitalizations - insights from the Czech Republic Background and objective: Drug-related hospitalizations represent a significant burden on healthcare. The aim of the study was to determine the prevalence of drug-related hospitalizations and identify medications and clinical manifestations associated with drug-related hospitalizations in patients admitted to hospital through the emergency department.

METHODS

This cross-sectional study examined unplanned hospitalizations at the University Hospital Hradec Kralove through the Department of Emergency Medicine between August and November 2018. Data were obtained from electronic health records. The methodology for identifying drug-related hospitalizations was based on the guideline of the European project OPERAM. This article focuses on a subgroup of drug-related problems related to the medication safety.

RESULTS

Of the total 1252 hospitalizations analyzed, 145 cases were identified as drug-related. The prevalence of drug-related hospitalizations was 12% (95% confidence interval 10-13). In 62% of cases, medications only contributed to the cause of hospitalization. Antithrombotics, cytostatics, diuretics, and systemic corticosteroids were the most common medication classes leading to drug-related hospitalizations. Gastrointestinal bleeding was the most common cause of drug-related hospitalizations. The potential preventability of drug-related hospitalizations was 34%.

CONCLUSION

Drug-related hospitalizations remain relatively common, while some of them could be potentially prevented. Pharmacists can contribute to enhancing patient safety by detecting drug-related problems and proposing measures to minimize risks.

Graph neural network-based subgraph analysis for predicting adverse drug events

PURPOSE

Adverse drug events (ADEs) are a significant global public health concern, and they have resulted in high rates of hospital admissions, morbidity, and mortality. Prior to the use of machine learning and deep learning methods, ADEs may not become well recognized until long after a drug has been approved and is widely used, which poses a significant challenge for ensuring patient safety. Consequently, there is a need to develop computational approaches for earlier identification of ADEs not detected during pre-registration clinical trials.

METHODS

This paper presents a state-of-the-art network-based approach that models patients as subgraphs composed of nodes of International Classification of Diseases (ICD) codes and directed edges illustrating disease progression. Four Graph Neural Network (GNN) variants were employed to make sub-graph level predictions that answer three Research Questions (RQ): 1) whether ADE(s) would occur given a patient's prior diagnoses history, 2) when an ADE would occur, and 3) which ADE would occur. The first and second RQs were addressed using a binary classification approach. The third RQ was addressed using a multi-label classification model.

RESULTS

The proposed network-based approach demonstrated superior performance in predicting ADEs, with the GraphSage model exhibiting the highest accuracy for both RQ 1 (0.8863) and RQ 3 (0.9367), while the Graph Attention Networks (GAT) model was found to perform best for RQ 2 (0.8769). Furthermore, an analysis segmented by ADE classification revealed that while RQs 1 and 3 exhibited minimal variance across different ADE categories, a distinct advantage was observed for categories B, C, and E in the context of RQ 2 when applying this sub-graph method.

CONCLUSION

The network-based approach demonstrates the potential of GNNs in supporting the early detection and prevention of ADEs. Accurately predicting ADEs could enable healthcare professionals to make informed clinical decisions, take preventive measures and adjust medication regimens before serious adverse events occur. The proposed prediction method could also lead to optimized usage of healthcare resources by preventing hospital admissions and reducing the overall burden of adverse drug events on the healthcare systems.

Effectiveness and Safety of an Emergency Department Code Sepsis Protocol: A Pragmatic Clinical Trial

Rationale: Sepsis care delivery-including the initiation of prompt, appropriate antimicrobials-remains suboptimal. Objectives: This study was conducted to determine direct and off-target effects of emergency department (ED) sepsis care reorganization. Methods: This pragmatic pilot trial enrolled adult patients who presented from November 2019 to February 2021 to an ED in Utah before and after implementation of a multimodal, team-based "Code Sepsis" protocol. Patients who presented to two other EDs where usual care was continued served as contemporaneous control subjects. The primary outcome was door-to-antimicrobial time among patients meeting Sepsis-3 criteria before ED departure. Secondary and safety outcomes included all-cause 30-day mortality, antimicrobial utilization and overtreatment, and antimicrobial-associated adverse events. Multivariable regression analyses used difference-in-differences methods to account for trends in outcomes unrelated to the studied intervention. Results: Code Sepsis protocol activation (N = 307) exhibited 8.5% sensitivity and 66% positive predictive value for patients meeting sepsis criteria before ED departure. Among 10,151 patients who met sepsis criteria during the study, adjusted difference-in-differences analysis demonstrated a 13-minute (95% confidence interval = 7-19) decrease in door-to-antimicrobial time associated with Code Sepsis implementation (P < 0.001). Mortality and clinical safety outcomes were unchanged, but Code Sepsis implementation was associated with increased false-positive presumptive infection diagnoses among patients who met sepsis criteria in the ED and increased antimicrobial utilization. Conclusions: Implementation of a team-based protocol for rapid sepsis evaluation and treatment during the coronavirus disease (COVID-19) pandemic's first year was associated with decreased ED door-to-antimicrobial time but also increased antimicrobial utilization. Measurement of both patient-centered and off-target effects of sepsis care improvement interventions is essential to comprehensive assessment of their value. Clinical trial registered with www.clinicaltrials.gov (NCT04148989).

Risk of emergency hospital admission related to adverse events after antibiotic treatment in adults with a common infection: impact of COVID-19 and derivation and validation of risk prediction models

BACKGROUND

With the global challenge of antimicrobial resistance intensified during the COVID-19 pandemic, evaluating adverse events (AEs) post-antibiotic treatment for common infections is crucial. This study aims to examines the changes in incidence rates of AEs during the COVID-19 pandemic and predict AE risk following antibiotic prescriptions for common infections, considering their previous antibiotic exposure and other long-term clinical conditions.

METHODS

With the approval of NHS England, we used OpenSAFELY platform and analysed electronic health records from patients aged 18-110, prescribed antibiotics for urinary tract infection (UTI), lower respiratory tract infections (LRTI), upper respiratory tract infections (URTI), sinusitis, otitis externa, and otitis media between January 2019 and June 2023. We evaluated the temporal trends in the incidence rate of AEs for each infection, analysing monthly changes over time. The survival probability of emergency AE hospitalisation was estimated in each COVID-19 period (period 1: 1 January 2019 to 25 March 2020, period 2: 26 March 2020 to 8 March 2021, period 3: 9 March 2021 to 30 June 2023) using the Kaplan-Meier approach. Prognostic models, using Cox proportional hazards regression, were developed and validated to predict AE risk within 30 days post-prescription using the records in Period 1.

RESULTS

Out of 9.4 million patients who received antibiotics, 0.6% of UTI, 0.3% of URTI, and 0.5% of LRTI patients experienced AEs. UTI and LRTI patients demonstrated a higher risk of AEs, with a noted increase in AE incidence during the COVID-19 pandemic. Higher comorbidity and recent antibiotic use emerged as significant AE predictors. The developed models exhibited good calibration and discrimination, especially for UTIs and LRTIs, with a C-statistic above 0.70.

CONCLUSIONS

The study reveals a variable incidence of AEs post-antibiotic treatment for common infections, with UTI and LRTI patients facing higher risks. AE risks varied between infections and COVID-19 periods. These findings underscore the necessity for cautious antibiotic prescribing and call for further exploration into the intricate dynamics between antibiotic use, AEs, and the pandemic.

Accuracy assessment of patient safety incident (PSI) codes and present-on-admission (POA) indicators: a cross-sectional analysis using the Patient Safety Incidents Inquiry (PSII) in Korea

BACKGROUND

Among the various methods used, administrative data collected for claims and billing purposes, such as diagnosis codes and present-on-admission (POA) indicators, can easily be employed to assess patient safety status. However, it is crucial that administrative data be accurate to generate valid estimates of adverse event (AE) occurrence. Thus, we aimed to evaluate the accuracy of diagnosis codes and POA indicators in patients with confirmed AEs in the hospital admission setting.

METHODS

We analysed the diagnosis codes of 1,032 confirmed AE cases and 6,754 non-AE cases from the 2019 Patient Safety Incidents Inquiry, which was designed as a cross-sectional study, to determine their alignment with the Korean Patient Safety Incidents (PSIs) Code Classification System. The unit of analysis was the individual case rather than the patient, because two or more AEs may occur in one patient. We examined whether the primary and secondary diagnostic codes had PSIs codes matching the AE type and checked each PSI code for whether the POA indicator had an 'N' tag. We reviewed the presence of PSI codes in patients without identified AEs and calculated the correlation between the AE incidence rate and PSI code and POA indicator accuracy across 15 hospitals.

RESULTS

Ninety (8.7%) of the AE cases had PSI codes with an 'N' tag on the POA indicator compared to 294 (4.4%) of the non-AE cases. Infection- (20.4%) and surgery/procedure-related AEs (13.6%) had relatively higher instances of correctly tagged PSI codes. We did not identify any PSI codes for diagnosis-related incidents. While we noted significant differences in AE incidence rates, PSI code accuracy, and POA indicator accuracy among the hospitals, the correlations between these variables were not statistically significant.

CONCLUSION

Currently, PSI codes and POA indicators in South Korea appear to have low validity. To use administrative data in medical quality improvement activities such as monitoring patient safety levels, improving the accuracy of administrative data should be a priority. Possible strategies include targeted education on PSI codes and POA indicators and introduction of new evaluation indicators regarding the accuracy of administrative data.

Prescribing Patterns of Nonrecommended Medications for Children With Acute COVID-19

OBJECTIVE

Repurposed medications for acute coronavirus disease 2019 (COVID-19) continued to be prescribed after results from rigorous studies and national guidelines discouraged use. We aimed to describe prescribing rates of nonrecommended medications for acute COVID-19 in children, associations with demographic factors, and provider type and specialty.

METHODS

In this retrospective cohort of children <18 years in a large United States all-payer claims database, we identified prescriptions within 2 weeks of an acute COVID-19 diagnosis. We calculated prescription rate, performed multivariable logistic regression to identify risk factors, and described prescriber type and specialty during nonrecommended periods defined by national guidelines.

RESULTS

We identified 3 082 626 COVID-19 diagnoses in 2 949 118 children between March 7, 2020 and December 31, 2022. Hydroxychloroquine (HCQ) and ivermectin were prescribed in 0.03% and 0.14% of COVID-19 cases, respectively, during nonrecommended periods (after September 12, 2020 for HCQ and February 5, 2021 for ivermectin) with considerable variation by state. Prescription rates were 4 times the national average in Arkansas (HCQ) and Oklahoma (ivermectin). Older age, nonpublic insurance, and emergency department or urgent care visit were associated with increased risk of either prescription. Additionally, residence in nonurban and low-income areas was associated with ivermectin prescription. General practitioners had the highest rates of prescribing.

CONCLUSIONS

Although nonrecommended medication prescription rates were low, the overall COVID-19 burden translated into high numbers of ineffective and potentially harmful prescriptions. Understanding overuse patterns can help mitigate downstream consequences of misinformation. Reaching providers and parents with clear evidence-based recommendations is crucial to children's health.

Assessing the impact of climatic variability on acute respiratory diseases across diverse climatic zones in South Africa

Acute respiratory diseases are a significant public health concern in South Africa, with climatic variables such as temperature and rainfall being key influencers. This study investigates the associations between these variables and the prevalence of acute respiratory diseases in Johannesburg, Cape Town, and Gqeberha (Port Elizabeth), representing distinct climatic zones. Spearman's correlation analyses showed negative correlations in Johannesburg for respiratory disease claims with maximum temperature (r = -0.12, p < 0.0001) and mean temperature (r = -0.13, p < 0.0001), and a negative correlation with daily rainfall (r = -0.12, p < 0.0001). Cape Town demonstrated a negative correlation with maximum temperature (r = -0.18, p < 0.0001) and a positive correlation with rainfall (r = 0.08, p < 0.0001). Utilizing Distributed Lag Non-linear Models (DLNM), the study revealed that in Johannesburg, the relative risk (RR) of respiratory claims increases notably at temperatures below 12 °C, and again at a Tmax between 16 and 23 °C. The risk escalates further at >30 °C, although with a considerable error margin. For Cape Town, a stable level of moderate RR is seen from Tmax 15-24 °C, with a significant increase in RR and error margin above 30 °C. In Gqeberha, the DLNM results are less definitive, reflecting the city's moderate climate and year-round rainfall. The RR of acute respiratory diseases did not show clear patterns with temperature changes, with increasing error margins outside the 22 °C threshold. These findings emphasize the imperative for region-specific public health strategies that account for the complex, non-linear influences of climate on respiratory health. This detailed understanding of the climate-health nexus provides a robust basis for enhancing public health interventions and future research directed at reducing the impacts of climate factors.

International Trends in Adverse Drug Event-Related Mortality from 2001 to 2019: An Analysis of the World Health Organization Mortality Database from 54 Countries

BACKGROUND AND OBJECTIVE

Adverse drug events (ADEs) are becoming a significant public health issue. However, reports on ADE-related mortality are limited to national-level evaluations. Therefore, we aimed to reveal overall trends in ADE-related mortality across the 21st century on an international level.

METHODS

This observational study analysed long-term trends in ADE-related mortality rates from 2001 to 2019 using the World Health Organization Mortality Database. The rates were analysed according to sex, age and region. North America, Latin America and the Caribbean, Western Europe, Eastern Europe and Western Pacific regions were assessed. Fifty-four countries were included with four-character International Statistical Classification of Disease and Related Health Problems, Tenth Revision codes in the database, population data in the World Population Prospects 2019 report, mortality data in more than half of the study period, and high-quality or medium-quality death registration data. A locally weighted regression curve was used to show international trends in age-standardised rates.

RESULTS

The global ADE-related mortality rate per 100,000 population increased from 2.05 (95% confidence interval 0.92-3.18) in 2001 to 6.86 (95% confidence interval 5.76-7.95) in 2019. Mortality rates were higher among men than among women, especially in those aged 20-50 years. The population aged ≥ 75 years had higher ADE-related mortality rates than the younger population. North America had the highest mortality rate among the five regions. The global ADE-related mortality rate increased by approximately 3.3-fold from 2001 to 2019.

CONCLUSIONS

The burden of ADEs has increased internationally with rising mortality rates. Establishing pharmacovigilance systems can facilitate efforts to reduce ADE-related mortality rates globally.

Clarification of adverse drug reactions by a pharmacovigilance team results in increased antibiotic re-prescribing at a freestanding United States children's hospital

Documentation of adverse drug reactions (ADRs) is a key factor in guiding future prescribing. However, incomplete documentation is common and often fails to distinguish implicated drugs as true allergies. This in turn leads to unnecessary avoidance of implicated drug classes and may result in sub-optimal prescribing. Pharmacovigilance (PV) programs utilize a systematic approach to clarify ADR documentation and are known to improve patient safety. Yet it remains unclear if PV alters prescribing. Or, if the existence of the ADR documentation itself continues to prompt avoidance of implicated drugs. To address this, our work presents a retrospective cohort study assessing if clarification of antibiotic ADRs by a hospital-wide PV team was associated with future, safe, re-prescribing at a freestanding pediatric hospital in the midwestern United States. First, we compared the likelihood of future prescribing in an antibiotic class with an active ADR, as compared to alternative drug classes, between PV-clarified and non-clarified patients. Second, we assessed differences in adverse event rates 30-days after future prescribing based on PV clarification status. For robustness, analyses were performed on patients with ADRs in four antibiotic classes: penicillin-based beta-lactams (n = 45,642), sulfonamides/trimethoprim (n = 5,329), macrolides (n = 3,959), and glycopeptides (n = 622). Results illustrate that clarification of an ADR by PV was associated with an increased odds of future prescribing in the same drug class (Odds Ratio [95%-CI]): penicillin-based beta-lactams (1.59 [1.36-1.89]), sulfonamides/trimethoprim (2.29 [0.89-4.91]), macrolides (0.77 [0.33-1.61]), and glycopeptide (1.85 [1.12-3.20]). Notably, patients clarified by PV experienced no increase in the rate of adverse events within 30-days following the prescribing of antibiotics in the same class as an active ADR. Overall, this study provides strong evidence that PV reviews safely increase the rate of re-prescribing antibiotics even in the presence of an existing implicated drug ADR.

Enhancing antidepressant safety surveillance: comparative analysis of adverse drug reaction signals in spontaneous reporting and healthcare claims databases

Background/Objective: Spontaneous reporting systems (SRS) such as the Korea Adverse Event Reporting System (KAERS) are limited in their ability to detect adverse drug reaction (ADR) signals due to their limited data on drug use. Conversely, the national health insurance claim (NHIC) data include drug use information for all qualifying residents. This study aimed to compare ADR signal profiles for antidepressants between KAERS and NHIC, evaluating the extent to which detected signals belong to common ADRs and labeling information. Materials and Methods: ADR signal detection in KAERS and NHIC databases, spanning January to December 2017, employed disproportionality analysis. Signal classes were determined based on System Organ Class (SOC) of the Medical Dictionary for Regulatory Activities (MedDRA). Also, Common ADR Coverage (CAC), the proportion of detected signals deemed common ADRs, and labeling information coverage (LIC) represented by mean average precision (mAP) were calculated. Additionally, protopathic bias and relative risk (RR) evaluation were performed to check for signal robustness. Results: Signal detection revealed 51 and 62 signals in KAERS and NHIC databases, respectively. Both systems predominantly captured signals related to nervous system disorders, comprising 33.3% (N = 17) in KAERS and 50.8% (N = 31) in NHIC. Regarding the type of antidepressants, KAERS predominantly reported signals associated with tricyclic antidepressants (TCAs) (N = 21, 41.2%), while NHIC produced most signals linked to selective serotonin reuptake inhibitors (SSRIs) (N = 22, 35.5%). KAERS exhibited higher CAC (68.63% vs. 29.03%) than NHIC. LIC was also higher in KAERS than in NHIC (mAP for EB05: 1.00 vs. 0.983); i.e., NHIC identified 5 signals not documented in drug labeling information, while KAERS found none. Among the unlabeled signals, one (Duloxetine-Myelopathy) was from protopathic bias, and two (duloxetine-myelopathy and tianeptine-osteomalacia) were statistically significant in RR. Conclusion: NHIC exhibited greater capability in detecting ADR signals associated with antidepressant use, encompassing unlabeled ADR signals, compared to KAERS. NHIC also demonstrated greater potential for identifying less common ADRs. Further investigation is needed for signals detected exclusively in NHIC but not covered by labeling information. This study underscores the value of integrating different sources of data, offering substantial regulatory insights and enriching the scope of pharmacovigilance.

Patterns of Mental Health Service Use During the Transition to Adulthood Among Autistic Adolescents and Young Adults

Background

The time of transition into adulthood, especially when leaving school, is a time when many autistic adolescents and young adults (AYA) may stop receiving mental health services that they have relied on, leading to worse mental health outcomes. The purpose of this study was to describe patterns of mental health service use during transition to adulthood among autistic AYAs.

Methods

We performed a cross-sectional study using electronic health records from years 2015 to 2019 from one large university health care system. We included autistic individuals ages 11-27 with at least one clinical encounter annually in the cohort. Outcomes included psychotropic medications and psychotherapy received, psychotropic polypharmacy, psychiatric emergency department (ED) visits, and adverse drug events.

Results

Almost half of the 529 patients in the cohort received polypharmacy. The most common treatment was medication only (56.9%), followed by no treatment (22.7%), medication plus psychotherapy (18.7%), and psychotherapy only (data masked). The 17-21 age group had the highest odds of a psychiatric ED visit, whereas the 22-27 age group had the highest odds of receiving psychotropic medications and polypharmacy. Black AYA were more likely to receive psychotherapy and less likely to receive psychotropic polypharmacy than non-Hispanic Whites.

Conclusion

Autistic individuals may benefit from more support from the health care system for their transition into adulthood to maintain use of beneficial mental health services as they leave school and to reduce the frequency of adverse outcomes. Access to providers experienced treating the complex needs of autistic individuals is important to reduce disparities.

Sharing Adverse Drug Event Reports Between Hospitals and Community Pharmacists to Inform Re-dispensing: An Analysis of Reports and Process Outcomes

INTRODUCTION

Adverse drug events (ADEs) are a leading cause of unplanned hospital visits. We designed ActionADE, an online ADE reporting platform, and integrated it with PharmaNet, British Columbia's (BC's) provincial medication dispensing system, to overcome identified barriers in ADE reporting and communicate ADEs to community pharmacies. Our objectives were to characterise ADEs reported in ActionADE, explore associations between patients' age, sex and ADE characteristics, and estimate the re-dispensation rate of culprit medications in community pharmacies.

METHODS

We conducted a prospective observational study of ADE reporting in four BC hospitals between April 1, 2020 and October 31, 2022. We described the characteristics of ADEs reported into ActionADE, used logistic regression modelling to examine associations between age and sex and ADE characteristics, and calculated rates of avoided culprit drug re-dispensations using community pharmacists' responses to ActionADE alerts.

RESULTS

In total, 3591 ADE reports were initiated by hospital clinicians, 3174 of which were included in this analysis. Serious or life-threatening ADEs resulting in permanent disability, hospitalisation, extended hospitalisation, and/or death accounted for 28.5% (906/3174; 95% CI 27.0-30.1%) of reports. Males were more likely to have non-adherence reported compared to females and experienced life threatening ADEs at a younger age than females. Of 592 patients who had ≥ 1 adverse drug reaction or allergy report (a subset of ADEs) transmitted to community pharmacies, 200 subsequently attempted to re-fill the culprit or a same class drug. Community pharmacists responded to preventative alerts by avoiding re-dispensation in 33.0% (66/200; 95% CI 26.5-39.5%).

INTERPRETATION

ActionADE is the first interoperable system that communicates ADEs via a central medication database to community pharmacies. Every 10th ADE reported in ActionADE and shared to PharmaNet resulted in community pharmacists' avoiding one culprit or same class drug re-exposure. Further research is needed to understand ActionADE's impact on patient and health system outcomes.

Real-world evidence of incidence and outcomes of aplastic anaemia following administration of immune checkpoint inhibitors

Aplastic anaemia (AA) is a rare immune-related adverse events (irAEs) after immune checkpoint inhibitors (ICIs) administration with poorly understood incidence and outcomes. We analysed an electronic health record database of 52 303 ICI-treated patients and found 77 (0.15%) cases of AA, with a median onset of 126 days (interquartile range, 58-363 days). The most used treatment for AA was systemic glucocorticoids 60 (77.9%) and 32 (41.6%) patients were able to resume ICI within 1 year. Patients diagnosed with AA had a steep decline in overall survival (OS) within the first 120 days; when compared to propensity score-matched patients without AA, they had a significantly worse OS (hazard ratio 1.72, 95% confidence interval 1.19-2.50; p = 0.003).

Hospitalisations Related to Adverse Drug Reactions in Switzerland in 2012-2019: Characteristics, In-Hospital Mortality, and Spontaneous Reporting Rate

INTRODUCTION

Adverse drug reactions (ADRs) contribute to morbidity, and serious ADRs may cause hospitalisation and death. This study characterises and quantifies ADR-related hospitalisations and subsequent in-hospital deaths, and estimates the spontaneous reporting rate to regulatory authorities in Switzerland, where healthcare professionals are legally obliged to report ADRs.

METHODS

This retrospective cohort study from 2012 to 2019 analysed nationwide data from the Federal Statistical Office. ICD-10 coding rules identified ADR-related hospitalisations. To estimate the reporting rate, individual case safety reports (ICSRs) collected in the Swiss spontaneous reporting system during the same period were considered.

RESULTS

Among 11,240,562 inpatients, 256,550 (2.3%) were admitted for ADRs, 132,320 (51.6%) were female, 120,405 (46.9%) were aged ≥ 65 (median of three comorbidities, interquartile range [IQR] 2-4), and 16,754 (6.5%) were children/teenagers (0 comorbidities, IQR 0-1). Frequent comorbidities were hypertension (89,938 [35.1%]), fluid/electrolyte disorders (54,447 [21.2%]), renal failure (45,866 [17.9%]), cardiac arrhythmias (37,906 [14.8%]), and depression (35,759 [13.9%]). Physicians initiated 113,028 (44.1%) of hospital referrals, and patients/relatives 73,494 (28.6%). Frequently ADR-affected were the digestive system (48,219 [18.8%], e.g. noninfective gastroenteritis and colitis), the genitourinary system (39,727 [15.5%], e.g. acute renal failure), and the mental/behavioural state (39,578 [15.4%], e.g. opioid dependence). In-hospital mortality was 2.2% (5669). Since ICSRs indicated 14,109 hospitalisations and 700 in-hospital deaths, estimated reporting rates were 5% and 12%, respectively.

CONCLUSIONS

This 8-year observation in Switzerland revealed that 2.3%, or roughly 32,000 admissions per year, were caused by ADRs. The majority of ADR-related admissions were not reported to the regulatory authorities, despite legal obligations.

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

PURPOSE

To investigate drug-related causes attributed to acute kidney injury (DAKI) and their documentation in patients admitted to the Intensive Care Unit (ICU).

METHODS

This study was conducted in an academic hospital in the Netherlands by reusing electronic health record (EHR) data of adult ICU admissions between November 2015 to January 2020. First, ICU admissions with acute kidney injury (AKI) stage 2 or 3 were identified. Subsequently, three modes of DAKI documentation in EHR were examined: diagnosis codes (structured data), allergy module (semi-structured data), and clinical notes (unstructured data).

RESULTS

n total 8124 ICU admissions were included, with 542 (6.7%) ICU admissions experiencing AKI stage 2 or 3. The ICU physicians deemed 102 of these AKI cases (18.8%) to be drug-related. These DAKI cases were all documented in the clinical notes (100%), one in allergy module (1%) and none via diagnosis codes. The clinical notes required the highest time investment to analyze.

CONCLUSIONS

Drug-related causes comprise a substantial part of AKI in the ICU patients. However, current unstructured DAKI documentation practice via clinical notes hampers our ability to gain better insights about DAKI occurrence. Therefore, both automating DAKI identification from the clinical notes and increasing structured DAKI documentation should be encouraged.

Combinations of medicines in patients with polypharmacy aged 65-100 in primary care: Large variability in risks of adverse drug related and emergency hospital admissions

BACKGROUND

Polypharmacy can be a consequence of overprescribing that is prevalent in older adults with multimorbidity. Polypharmacy can cause adverse reactions and result in hospital admission. This study predicted risks of adverse drug reaction (ADR)-related and emergency hospital admissions by medicine classes.

METHODS

We used electronic health record data from general practices of Clinical Practice Research Datalink (CPRD GOLD) and Aurum. Older patients who received at least five medicines were included. Medicines were classified using the British National Formulary sections. Hospital admission cases were propensity-matched to controls by age, sex, and propensity for specific diseases. The matched data were used to develop and validate random forest (RF) models to predict the risk of ADR-related and emergency hospital admissions. Shapley Additive eXplanation (SHAP) values were calculated to explain the predictions.

RESULTS

In total, 89,235 cases with polypharmacy and hospitalised with an ADR-related admission were matched to 443,497 controls. There were over 112,000 different combinations of the 50 medicine classes most implicated in ADR-related hospital admission in the RF models, with the most important medicine classes being loop diuretics, domperidone and/or metoclopramide, medicines for iron-deficiency anaemias and for hypoplastic/haemolytic/renal anaemias, and sulfonamides and/or trimethoprim. The RF models strongly predicted risks of ADR-related and emergency hospital admission. The observed Odds Ratio in the highest RF decile was 7.16 (95% CI 6.65-7.72) in the validation dataset. The C-statistics for ADR-related hospital admissions were 0.58 for age and sex and 0.66 for RF probabilities.

CONCLUSIONS

Polypharmacy involves a very large number of different combinations of medicines, with substantial differences in risks of ADR-related and emergency hospital admissions. Although the medicines may not be causally related to increased risks, RF model predictions may be useful in prioritising medication reviews. Simple tools based on few medicine classes may not be effective in identifying high risk patients.

Clinical Relevance of Drug-Drug Interactions With Antibiotics as Listed in a National Medication Formulary: Results From Two Large Population-Based Case-Control Studies in Patients Aged 65-100 Years Using Linked English Primary Care and Hospital Data

This study evaluated drug-drug interactions (DDIs) between antibiotic and nonantibiotic drugs listed with warnings of severe outcomes in the British National Formulary based on adverse drug reaction (ADR) detectable with routine International Classification of Diseases, Tenth Revision coding. Data sources were Clinical Practice Research Databank GOLD and Aurum anonymized electronic health records from English general practices linked to hospital admission records. In propensity-matched case-control study, outcomes were ADR or emergency admissions. Analyzed were 121,546 ADR-related admission cases matched to 638,238 controls. For most antibiotics, adjusted odds ratios (aORs) for ADR-related hospital admission were large (aOR for trimethoprim 4.13; 95% confidence interval (CI), 3.97-4.30). Of the 51 DDIs evaluated for ADR-related admissions, 38 DDIs (74.5%) had statistically increased aORs of concomitant exposure compared with nonexposure (mean aOR 3.96; range 1.59-11.42); for the 89 DDIs for emergency hospital admission, the results were 75 (84.3%) and mean aOR 2.40; range 1.43-4.17. Changing reference group to single antibiotic exposure reduced aORs for concomitant exposure by 76.5% and 83.0%, respectively. Medicines listed to cause nephrotoxicity substantially increased risks that were related to number of medicines (aOR was 2.55 (95% CI, 2.46-2.64) for current use of 1 and 10.44 (95% CI, 7.36-14.81) for 3 or more medicines). In conclusion, no evidence of substantial risk was found for multiple DDIs with antibiotics despite warnings of severe outcomes in a national formulary and flagging in electronic health record software. It is proposed that the evidence base for inclusion of DDIs in national formularies be strengthened and made publicly accessible and indiscriminate flagging, which compounds alert fatigue, be reduced.

Comparing risk prediction models aimed at predicting hospitalizations for adverse drug events in community dwelling older adults: a protocol paper

Background

The objective of this paper is to describe the creation, validation, and comparison of two risk prediction modeling approaches for community-dwelling older adults to identify individuals at highest risk for adverse drug event-related hospitalizations. One approach will use traditional statistical methods, the second will use a machine learning approach.

Methods

We will construct medication, clinical, health care utilization, and other variables known to be associated with adverse drug event-related hospitalizations. To create the cohort, we will include older adults (≥ 65 years of age) empaneled to a primary care physician within the Cedars-Sinai Health System primary care clinics with polypharmacy (≥ 5 medications) or at least 1 medication commonly implicated in ADEs (certain oral hypoglycemics, anti-coagulants, anti-platelets, and insulins). We will use a Fine-Gray Cox proportional hazards model for one risk modeling approach and DataRobot, a data science and analytics platform, to run and compare several widely used supervised machine learning algorithms, including Random Forest, Support Vector Machine, Extreme Gradient Boosting (XGBoost), Decision Tree, Naïve Bayes, and K-Nearest Neighbors. We will use a variety of metrics to compare model performance and to assess the risk of algorithmic bias.

Discussion

In conclusion, we hope to develop a pragmatic model that can be implemented in the primary care setting to risk stratify older adults to further optimize medication management.

Pharmacological data science perspective on fatal incidents of morphine treatment

Morphine prescribed for analgesia has caused drug-related deaths at an estimated incidence of 0.3% to 4%. Morphine has pharmacological properties that make it particularly difficult to assess the causality of morphine administration with a patient's death, such as its slow transfer between plasma and central nervous sites of action and the existence of the active metabolite morphine-6-glucuronide with opioid agonistic effects, Furthermore, there is no well-defined toxic dose or plasma/blood concentration for morphine. Dosing is often adjusted for adequate pain relief. Here, we summarize reported deaths associated with morphine therapy, including associated morphine exposure and modulating patient factors such as pharmacogenetics, concomitant medications, or comorbidities. In addition, we systematically analyzed published numerical information on the stability of concentrations of morphine and its relevant metabolites in biological samples collected postmortem. A medicolegal case is presented in which the causality of morphine administration with death was in dispute and pharmacokinetic modeling was applied to infer the administered dose. The results of this analytical review suggest that (i) inference from postmortem blood concentrations to the morphine dose administered has low validity and (ii) causality between a patient's death and the morphine dose administered remains a highly context-dependent and collaborative assessment among experts from different medical specialties.

Development of the Korean Patient Safety Incidents Code Classification System

OBJECTIVES

Attempts to understand patient safety using administrative data in Korea have been rare. This study develops a Korean patient safety incident code classification system and identifies its characteristics to boost diagnosis code usage for assessing patient safety.

METHODS

Based on existing literature, we selected Korean Standard Classification of Diseases 7 codes for characterizing patient safety incidents using diagnosis codes. We conducted 2 rounds of review to evaluate the codes applicability to different patient safety incidents using the Delphi method. The verified diagnosis codes were then classified by incident type.

RESULTS

Of the 54,259 Korean Standard Classification of Diseases 7 codes, 4509 were applicable for Korean patients, which were divided into 2435 code groups and 2074 candidate groups. The codes were classified into 6 categories (diagnosis, medication, patient care, operation or procedure, infection related, and other) and then further classified into 35 subcategories. The major categories of patient safety incidents, in the order of frequency, involved medication, fluid and blood related (1719, 38.1%), operation and procedure related (1339, 29.7%), and patient care related (991, 22.0%). Meanwhile, there were only 2 codes related to diagnosis.

CONCLUSIONS

Our study provides a basis for estimating patient safety incidents using diagnosis codes. We suggest that gradually increasing the utilization and accuracy of the patient safety incident codes will help develop effective patient safety indicators in Korea similar to other countries. Moreover, clinicians are also needed to be aware of using the developed code classification system.

Developing a deep learning natural language processing algorithm for automated reporting of adverse drug reactions

The detection of adverse drug reactions (ADRs) is critical to our understanding of the safety and risk-benefit profile of medications. With an incidence that has not changed over the last 30 years, ADRs are a significant source of patient morbidity, responsible for 5%-10% of acute care hospital admissions worldwide. Spontaneous reporting of ADRs has long been the standard method of reporting, however this approach is known to have high rates of under-reporting, a problem that limits pharmacovigilance efforts. Automated ADR reporting presents an alternative pathway to increase reporting rates, although this may be limited by over-reporting of other drug-related adverse events. We developed a deep learning natural language processing algorithm to identify ADRs in discharge summaries at a single academic hospital centre. Our model was developed in two stages: first, a pre-trained model (DeBERTa) was further pre-trained on 1.1 million unlabelled clinical documents; secondly, this model was fine-tuned to detect ADR mentions in a corpus of 861 annotated discharge summaries. This model was compared to a version without the pre-training step, and a previously published RoBERTa model pretrained on MIMIC III, which has demonstrated strong performance on other pharmacovigilance tasks. To ensure that our algorithm could differentiate ADRs from other drug-related adverse events, the annotated corpus was enriched for both validated ADR reports and confounding drug-related adverse events using. The final model demonstrated good performance with a ROC-AUC of 0.955 (95% CI 0.933 - 0.978) for the task of identifying discharge summaries containing ADR mentions, significantly outperforming the two comparator models.

Exploring the prevalence and characteristics of adverse drug events among older adults in South Korea using a national health insurance database

Background: Adverse drug events (ADEs) in the elderly frequently occur because of their multiple chronic diseases and complexity of drug therapy. To better understand adverse drug events, the prevalence and characteristics of adverse drug events in elderly South Korean patients were assessed.

Methods: The National Health Insurance databases for 2015 and 2016 were used for the analysis. We included patients aged ≥65 years that had at least one claim with the diagnosis codes ‘drug-induced,’ ‘poisoning by drug,’ and ‘vaccine-associated’ each year for the base-case analysis. To minimize the underestimation of adverse drug event prevalence, we also used an extended definition analysis by adding the ‘adverse drug event very likely’ codes. We estimated the prevalence of adverse drug events by sex, age group, and type of insurance and examined the frequent types of adverse drug events in 2015 and 2016.

Results: In the base-case analysis, adverse drug event prevalence in individuals aged 65 years and older was 2.75% in 2015 and 2.77% in 2016. With advanced age, the prevalence of adverse drug event tended to increase, peaking in the age group of 75–79 years. In addition, the adverse drug event prevalence was higher in females and Medical Aid enrollees. The most frequently occurring adverse drug event was ‘allergy, unspecified,’ followed by ‘other drug-induced secondary parkinsonism,’ and ‘generalized skin eruption due to drugs and medicaments.’ When we examined the extended definition analysis, the prevalence of adverse drug events was 4.47% in 2015 and 4.52% in 2016, which significantly increased from those estimated in the base-case analysis.

Conclusion: Among the older adults, the prevalence of adverse drug event was higher in advanced age, females, and Medical Aid enrollees. In particular, allergy and drug-induced secondary parkinsonism frequently occurred. This study provides evidence that health policies addressing the prevention and management of adverse drug events should be a priority for the most vulnerable elderly patients.

Automatic Detection of Adverse Drug Events in Geriatric Care: Study Proposal

BACKGROUND

One-third of older inpatients experience adverse drug events (ADEs), which increase their mortality, morbidity, and health care use and costs. In particular, antithrombotic drugs are among the most at-risk medications for this population. Reporting systems have been implemented at the national, regional, and provider levels to monitor ADEs and design prevention strategies. Owing to their well-known limitations, automated detection technologies based on electronic medical records (EMRs) are being developed to routinely detect or predict ADEs.

OBJECTIVE

This study aims to develop and validate an automated detection tool for monitoring antithrombotic-related ADEs using EMRs from 4 large Swiss hospitals. We aim to assess cumulative incidences of hemorrhages and thromboses in older inpatients associated with the prescription of antithrombotic drugs, identify triggering factors, and propose improvements for clinical practice.

METHODS

This project is a multicenter, cross-sectional study based on 2015 to 2016 EMR data from 4 large hospitals in Switzerland: Lausanne, Geneva, and Zürich university hospitals, and Baden Cantonal Hospital. We have included inpatients aged ≥65 years who stayed at 1 of the 4 hospitals during 2015 or 2016, received at least one antithrombotic drug during their stay, and signed or were not opposed to a general consent for participation in research. First, clinical experts selected a list of relevant antithrombotic drugs along with their side effects, risks, and confounding factors. Second, administrative, clinical, prescription, and laboratory data available in the form of free text and structured data were extracted from study participants' EMRs. Third, several automated rule-based and machine learning-based algorithms are being developed, allowing for the identification of hemorrhage and thromboembolic events and their triggering factors from the extracted information. Finally, we plan to validate the developed detection tools (one per ADE type) through manual medical record review. Performance metrics for assessing internal validity will comprise the area under the receiver operating characteristic curve, F1-score, sensitivity, specificity, and positive and negative predictive values.

RESULTS

After accounting for the inclusion and exclusion criteria, we will include 34,522 residents aged ≥65 years. The data will be analyzed in 2022, and the research project will run until the end of 2022 to mid-2023.

CONCLUSIONS

This project will allow for the introduction of measures to improve safety in prescribing antithrombotic drugs, which today remain among the drugs most involved in ADEs. The findings will be implemented in clinical practice using indicators of adverse events for risk management and training for health care professionals; the tools and methodologies developed will be disseminated for new research in this field. The increased performance of natural language processing as an important complement to structured data will bring existing tools to another level of efficiency in the detection of ADEs. Currently, such systems are unavailable in Switzerland.

INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID)

DERR1-10.2196/40456.

Validating ICD-10 codes for adverse drug events in hospitalised older adults: protocol for a cross-sectional study

INTRODUCTION

Adverse drug events (ADEs) among hospitalised older adults are common yet often preventable. Efforts to recognise ADEs using pharmacist review and electronic health record adaptations have had mixed results. Our health system developed and implemented a geriatric prescribing context designed to offer age-friendly dose and frequency defaults for hospitalised patients 75 years and older. The impact of this context on ADEs remains unknown. To measure its impact, our team created a list of ADE-related International Classification of Diseases (ICD) codes specific to 10 commonly used medications at our institution. This protocol paper presents the process of designing a screening tool for ADEs, validating the tool with manual chart reviews and measuring the impact of the context on ADEs.

METHODS AND ANALYSIS

This retrospective cross-sectional study will assess our list of ICD-10 codes against manual chart review to determine its accuracy. An electronic health record report for patients aged 75 years and older admitted to the hospital for a minimum of two nights was generated to identify 100 test positives and 100 test negatives. Test positives need at least one code from each level of our ICD-10 code list. The first level of codes identifies any possible ADEs while the second level is more symptom based. Test negatives must not have any code from the list. Two physicians blinded to test status will complete a structured chart review to determine if a patient had an ADE during their hospitalisation. Acceptable inter-rater reliability will need to be met before proceeding with independent chart review. Positive predictive value and negative predictive value will be calculated once all the chart reviews are completed.

ETHICS AND DISSEMINATION

The Oregon Health & Science University Institutional Review Board approved this study (#21385). The results of the study will be disseminated in peer-reviewed journals and conference presentations.

Evaluation of Factors Associated with Adverse Drug Events in South Korea Using a Population-Based Database

This retrospective study aims to investigate the factors associated with the occurrence of ADEs using nationally representative claims data. All patients with at least one claim with diagnosis codes denoting potential ADE between 1 July 2015 and 31 December 2015 were included. Potential ADE was defined as ADE identified in the claims data, because it was not verified. The index date was defined as the date of the first claim with potential ADEs. Demographic data were collected at the index date, while data on comorbidities and number of medications used were collected six months before the index date. Multivariate logistic regression was used to explore the association between potential ADEs and several factors, including sex, age group, insurance type, comorbidities, and number of prescribed medications. Patients with potential ADEs were older, had more chronic diseases, and used more medications than those without potential ADEs. In the multivariate analysis, occurrence of potential ADEs was associated with age (≥65 years, odds ratio [OR] 1.15, 95% confidence interval [CI] 1.08–1.21), Medical Aid program (OR 1.37, 95% CI 1.27–1.47), Charlson Comorbidity Index scores (≥5, OR 2.87, 95% CI 2.56–3.20), and use of six or more medications (6–10 medications, OR 1.89, 95% CI 1.79–1.99). Age, Medical Aid program, comorbidities, and number of medications were associated with occurrence of potential ADEs.

[Deaths from adverse drug events in Brazil: Mortality Information System as a source of information]

Adverse drug events (ADEs) are harmful events caused by medication, and some of which can lead to death. Death records are an important source of information when using codes from the 10th revision of the International Classification of Diseases (ICD-10) suggestive of ADE. This study aimed to identify the ADEs registered in Brazililian Mortality Information System (SIM), analyzing data distribution by year, age group, and type of event. This is an ecological study with retrospective data collection, identifying ADEs in the SIM, using the ICD-10 codes. The study included deaths that occurred in Brazil from 2008 to 2016. An increase in the number of deaths associated with ADE was observed from 2008 to 2016, with a mortality rate per 1 million inhabitants ranging from 8.70 to 14.40 in the period. Most events corresponded to mental and behavioral disorders due to the use of psychotropic drugs. Most deaths (12,311) related to ADE codes were identified in several chapters of the ICD-10. Chapter XX, about adverse events, allowed the identification of a smaller number of deaths (4,893). Higher event rates were observed among individuals aged 60 years and over (39.8/1 million) and children younger than one year (22.0/1 million). The identification of ADE-related deaths on the SIM is an important strategy for addressing undesirable drug-related events. Deaths related to the use of psychotropic drugs were the most frequent ADE-related deaths and the elderly were the age group most affected by ADEs.

Analysis of pharmacovigilance databases for spontaneous reports of adverse drug reactions related to substandard and falsified medical products: A descriptive study

Introduction: The public health threat of substandard and falsified medicines has been well known in the last two decades, and several studies focusing on the identification of products affected and preventing consumption have been published. However, the number of these products reaching patients and causing health consequences and adverse drug reactions is not a well-researched area.

Objectives: Our aim was to identify and describe the characteristics of cases that are related to adverse drug reactions potentially originating from counterfeit medication using publicly available pharmacovigilance data.

Methods: A descriptive study was performed based on pharmacovigilance data retrieved from Individual Case Safety Reports (ICSRs) identified in the European Medicines Agency’s EudraVigilance and FDA Adverse Event Reporting System (FAERS) databases in April 2022 using selected MedDRA preferred terms: counterfeit product administered, product counterfeit, product label counterfeit, product packaging counterfeit, suspected counterfeit product, adulterated product, product tampering, and suspected product tampering. ICSRs were analyzed by age and gender, by year of reporting, region of origin, reporter’s profession, and severity of the outcome. The disproportionality method was used to calculate pharmacovigilance signal measures.

Results: A total of 5,253 cases in the FAERS and 1,049 cases in the EudraVigilance database were identified, generally affecting middle-aged men with a mean age of 51.055 (±19.62) in the FAERS and 64.18% of the cases between 18 and 65 years, while the male to female ratios were 1.18 and 1.5. In the FAERS database, we identified 138 signals with 95% confidence interval including sildenafil (n = 314; PRR, 12.99; ROR, 13.04; RRR, 11.97), tadalafil (n = 200; PRR, 11.51; ROR, 11.55; RRR, 10.94), and oxycodone (n = 190; PRR, 2.47; ROR, 2.14; RRR, 2.47). While in the EV data 31, led by vardenafil (n = 16, PRR = 167.19; 101.71–274.84; 95% CI, RRR = 164.66; 100.17–270.66; 95% CI, ROR = 169.47; 103.09–278.60; 95% CI, p &lt; 0.001), entecavir (n = 46, PRR = 161.26, RRR = 154.24, ROR = 163.32, p &lt; 0.001), and tenofovir (n = 20, PRR = 142.10, RRR = 139.42, ROR = 143.74, p &lt; 0.001).

Conclusion: The application of pharmacovigilance datasets to identify potential counterfeit medicine ADRs can be a valuable tool in recognition of potential risk groups of consumers and the affected active pharmaceutical ingredients and products. However, the further development and standardization of ADR reporting, pharmacovigilance database analysis, and prospective and real-time collection of potential patients with health consequences are warranted in the future.

Clinical and Cost-Saving Effects of the Drug Utilization Review Modernization Project in Inpatient and Outpatient Settings in Korea

OBJECTIVES

Korea's national health insurance authority introduced a drug utilization review modernization pilot project in which health professionals provided follow-up services to monitor adverse drug events. We aimed to evaluate the effects of the project on clinical and economic outcomes.

METHODS

We conducted difference-in-differences analysis using National Health Insurance claims data from the Health Insurance Review and Assessment Service. We calculated the number of adverse drug events and allergic reactions as a clinical indicator and medical costs incurred to manage these events as an economic indicator. Absolute difference in each outcome measure was defined as the value after the project minus the value before the project. Difference-in-differences was defined as a difference in absolute differences between the intervention group and the control group.

RESULTS

Overall, difference-in-differences were -43 and -826 for the number of drug-related adverse events and allergic reactions and -$198,700 and $53,318 for medical costs in the inpatient and outpatient settings, respectively. For outpatients, the monthly number of adverse drug events and allergic reactions has grown higher for the control group than for the intervention group after implementation of the pilot project.

CONCLUSIONS

Implementation of the pilot project lowered the number of adverse drug events and allergic reactions in the inpatient and outpatient setting. The project also lowered medical costs incurred to manage these events in the inpatient setting only. Based on our findings, we recommend that the pilot project be expanded on a nationwide level at least in the inpatient setting.

Marked Increase of Gamma-Glutamyltransferase as an Indicator of Drug-Induced Liver Injury in Patients without Conventional Diagnostic Criteria of Acute Liver Injury

INTRODUCTION

Clinically significant drug-induced liver injury (DILI) is defined by elevations of alanine aminotransferase (ALT) ≥5 times the upper limit of normal (ULN), alkaline phosphatase (ALP) ≥2 × ULN, or ALT ≥3 × ULN and total bilirubin TBIL >2 × ULN. However, DILI might also occur in patients who do not reach those thresholds and still may benefit from discontinuation of medication.

METHODS

Fifteen patients recruited for our prospective study on potentially hepatotoxic drugs were included. DILI diagnosis was based on RUCAM (Roussel Uclaf Causality Assessment Method) score and expert opinion and was supported by an in vitro test using monocyte-derived hepatocyte-like (MH) cells.

RESULTS

Median RUCAM score was 6 (range 4-8), indicating that DILI was possible or probable in all cases. The predominant types of liver injury were mixed (60%) and cholestatic (40%). While no elevation above 2 × ULN of ALP and TBIL was observed, gamma-glutamyltransferase (GGT) above 2 × ULN was identified in 8 of the patients. Six of the 15 patients did not achieve full remission and showed persistent elevation of GGT, which was significantly associated with peak GGT elevation above 2 × ULN (p = 0.005).

CONCLUSION

Here we present a case series of patients with liver enzyme elevation below the conventional thresholds who developed DILI with a predominant GGT elevation leading to drug withdrawal and/or chronic elevation of liver parameters, in particular of GGT. Thus, we propose that DILI should be considered in particular in cases with marked increase of GGT even if conventional DILI threshold levels are not reached, resulting in discontinuation of the causative drug and/or close monitoring of the patients.

Opportunities for penicillin allergy evaluation in dental clinics

OBJECTIVE

To evaluate opportunities for assessing penicillin allergies among patients presenting to dental clinics.

DESIGN

Retrospective cross-sectional study.

SETTING

VA dental clinics.

PATIENTS

Adult patients with a documented penicillin allergy who received an antibiotic from a dentist between January 1, 2015, and December 31, 2018, were included.

METHODS

Chart reviews were completed on random samples of 100 patients who received a noncephalosporin antibiotic and 200 patients who received a cephalosporin. Each allergy was categorized by severity. These categories were used to determine patient eligibility for 3 testing groups based on peer-reviewed algorithms: (1) no testing, (2) skin testing, and (3) oral test-dose challenge. Descriptive and bivariate statistics were used to compare facility and patient demographics first between true penicillin allergy, pseudo penicillin allergy, and missing allergy documentation, and between those who received a cephalosporin and those who did not at the dental visit.

RESULTS

Overall, 19% lacked documentation of the nature of allergic reaction, 53% were eligible for skin testing, 27% were eligible for an oral test-dose challenge, and 1% were contraindicated from testing. Male patients and African American patients were less likely to receive a cephalosporin.

CONCLUSIONS

Most penicillin-allergic patients in the VA receiving an antibiotic from a dentist are eligible for penicillin skin testing or an oral penicillin challenge. Further research is needed to understand the role of dentists and dental clinics in assessing penicillin allergies.

Evaluating the Impact of Medication Risk Mitigation Services in Medically Complex Older Adults

Adverse drug events (ADEs) represent an expensive societal burden that disproportionally affects older adults. Therefore, value-based organizations that provide care to older adults—such as the Program of All-Inclusive Care for the Elderly (PACE)—should be highly motivated to identify actual or potential ADEs to mitigate risks and avoid downstream costs. We sought to determine whether PACE participants receiving medication risk mitigation (MRM) services exhibit improvements in total healthcare costs and other outcomes compared to participants not receiving structured MRM. Data from 2545 PACE participants from 19 centers were obtained for the years 2018 and 2019. We compared the year-over-year changes in outcomes between patients not receiving (control) or receiving structured MRM services. Data were adjusted based on participant multimorbidity and geographic location. Our analyses demonstrate that costs in the MRM cohort exhibited a significantly smaller year-to-year increase compared to the control (MRM: USD 4386/participant/year [95% CI, USD 3040−5732] vs. no MRM: USD 9410/participant/year [95% CI, USD 7737−11,084]). Therefore, receipt of structured MRM services reduced total healthcare costs (p < 0.001) by USD 5024 per participant from 2018 to 2019. The large majority (75.8%) of the reduction involved facility-related expenditures (e.g., hospital admission, emergency department visits, skilled nursing). In sum, our findings suggest that structured MRM services can curb growing year-over-year healthcare costs for PACE participants.

The detection and prevention of adverse drug events in nursing home and home care patients: Study protocol of a quasi-experimental study

AIM

To estimate the cost-effectiveness of an intervention facilitating the early detection of adverse drug events through the means of health professional training and the application of a digital screening tool.

DESIGN

Multi-centred non-randomized controlled trial from August 2018 to March 2020 including 65 nursing homes or home care providers.

METHODS

We aim to estimate the effect of the intervention on the rate of adverse drug events as primary outcome through a quasi-experimental empirical study design. As secondary outcomes, we use hospital admissions and falls. All outcomes will be measured on patient-month level. Once the causal effect of the intervention is estimated, cost-effectiveness will be calculated. For cost-effectiveness, we include all patient costs observed by the German statutory health insurance.

RESULTS

The results of this study will inform about the cost-effectiveness of the optimized drug supply intervention and provide evidence for potential reimbursement within the German statutory health insurance system.

Adverse Drug Events Detected by Clinical Pharmacists in an Emergency Department: A Prospective Monocentric Observational Study

OBJECTIVES

Adverse drug events (ADEs) are a major public health issue in hospitals. They are difficult to detect because of incomplete or unavailable medication history. In this study, we aimed to assess the rate and characteristics of ADEs identified by pharmacists in an emergency department (ED) to identify factors associated with ADEs.

METHODS

In this prospective observational study, we included consecutive adult patients presenting to the ED of a French 2600-bed tertiary care university hospital from November 2011 to April 2015. Clinical pharmacists conducted structured interviews and collected the medication history to detect ADEs (i.e., injuries resulting directly or indirectly from adverse drug reactions and noncompliance to medication prescriptions). Unsure ADE cases were reviewed by an expert committee. Relations between patient characteristics, type of ED visit, and ADE risk were analyzed using logistic regression.

RESULTS

Among the 8275 included patients, 1299 (15.7%) presented to the ED with an ADE. The major ADE symptoms were bleeding, endocrine problems, and neurologic disorders. Moreover, ADEs led to the ED visit, hospitalization, and death in 87%, 49.3%, and 2.2% of cases, respectively. Adverse drug event risk was independently associated with male sex, ED visit for neurological symptoms, visit to the ED critical care unit, or ED short stay hospitalization unit, use of blood, anti-infective, antineoplastic, and immunomodulating drugs.

CONCLUSIONS

This study improves the knowledge about ADE characteristics and on the patients at risk of ADE. This could help ED teams to better identify and manage ADEs and to improve treatment quality and safety.

Hospitalizations and deaths related to adverse drug events worldwide: Systematic review of studies with national coverage

PURPOSE

Adverse drug events are related to negative outcomes in healthcare, including hospitalization, increased duration of hospital stay and death. The aim of this study was to conduct a systematic review to evaluate hospitalizations and deaths related to adverse drug events worldwide, reported in studies with national coverage.

METHODS

The protocol was registered in PROSPERO (CRD42020157008). We performed a systematic search on Medline, Embase, CINAHL, LILACS, and the Cochrane Library (until March 2020) using pre-specified terms. We included published studies that reported data on hospitalizations and/or deaths related to adverse drug events from a national perspective and the use of secondary data as a source of information. Two reviewers independently extracted and synthesized data. The quality of the studies was assessed using an adapted version of the Joanna Briggs Institute critical appraisal checklist for prevalence studies. Narrative summaries of findings were undertaken.

RESULTS

Among 59,336 citations, 62 studies were included for data extraction and synthesis. Among these studies, 41 studies included the outcome of hospitalization, 16 included the death outcome, and five included both outcomes. Administrative databases regarding discharges and registries of vital statistics were the most common sources of information. The relative frequency of hospitalizations ranged from 0.03% to 7.3%, and from 9.7 to 383.0/100,000 population, whereas mortality rate ranged from 0.1 to 7.88/100,000 population.

CONCLUSION

Our study highlights information about adverse drug events using large administrative databases in a national scenario and provides an overview of databases and methods implemented to detect adverse drug events.

Use of a hospital administrative database to identify and characterize community-acquired, hospital-acquired and drug-induced acute kidney injury

BACKGROUND

Acute kidney injury (AKI) has serious short- and long-term consequences. The objective of the present study of a cohort of hospitalized patients with AKI was to (i) evaluate the proportion of patients with hospital-acquired (HA) AKI and community-acquired (CA) AKI, the characteristics of these patients and the AKIs, and the short-term outcomes, and (ii) determine the performance of several ICD-10 codes for identifying AKI (both CA and HA) and drug-induced AKI.

METHODS

A cohort of hospitalized patients with AKI was constituted by screening hospital's electronic medical records (EMRs) for cases of AKI. We distinguished between and compared CA-AKI and HA-AKI and evaluated the proportion of AKIs that were drug-induced. The EMR data were merged with hospital billing codes (according to the International Classification of Diseases, 10th Edition (ICD-10)) for each hospital stay. The ability of ICD-10 codes to identify AKIs (depending on the type of injury) was determined by calculating the sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV). Lastly, we sought to validate specific ICD-10 codes for drug-induced AKI.

RESULTS

Of the 2473 patients included, 1557 experienced an AKI (HA-AKI: 59.3%; CA-AKI: 40.7%). Patients with CA-AKI had a better short-term outcome and a lower death rate (7.6%, vs. 20% for HA-AKI). One AKI in three was drug-induced. The combination of AKI codes had a very high specificity (94.8%), a high PPV (94.9%), a moderate NPV (56.7%) and moderate sensitivity (57.4%). The sensitivity was higher for CA-AKI (72.2%, vs. 47.2% for HA-AKI), for more severe AKI (82.8% for grade 3 AKI vs. 43.7% for grade 1 AKI), and for patients with CKD. Use of a specific ICD-10 code for drug-induced AKI (N14x) alone gave a very low sensitivity (1.8%), whereas combining codes for adverse drug reactions with AKI-specific codes increased the sensitivity.

CONCLUSION

Our results show that the combination of an EMR-based analysis with ICD-10-based hospital billing codes gives a comprehensive "real-life" picture of AKI in hospital settings. We expect that this approach will enable researchers to study AKI in more depth.

Comparative assessment of manual chart review and ICD claims data in evaluating immunotherapy-related adverse events

BACKGROUND

The aim of this retrospective study was to demonstrate that irAEs, specifically gastrointestinal and pulmonary, examined through International Classification of Disease (ICD) data leads to underrepresentation of true irAEs and overrepresentation of false irAEs, thereby concluding that ICD claims data are a poor approach to electronic health record (EHR) data mining for irAEs in immunotherapy clinical research.

METHODS

This retrospective analysis was conducted in 1,063 cancer patients who received ICIs between 2011 and 2017. We identified irAEs by manual review of medical records to determine the incidence of each of our endpoints, namely colitis, hepatitis, pneumonitis, other irAE, or no irAE. We then performed a secondary analysis utilizing ICD claims data alone using a broad range of symptom and disease-specific ICD codes representative of irAEs.

RESULTS

16% (n = 174/1,063) of the total study population was initially found to have either pneumonitis 3% (n = 37), colitis 7% (n = 81) or hepatitis 5% (n = 56) on manual review. Of these patients, 46% (n = 80/174) did not have ICD code evidence in the EHR reflecting their irAE. Of the total patients not found to have any irAEs during manual review, 61% (n = 459/748) of patients had ICD codes suggestive of possible irAE, yet were not identified as having an irAE during manual review.

DISCUSSION

Examining gastrointestinal and pulmonary irAEs through the International Classification of Disease (ICD) data leads to underrepresentation of true irAEs and overrepresentation of false irAEs.

Clinical factors associated with increased length of stay and readmission in patients with medication-related hospital admissions: a retrospective study

BACKGROUND

Adverse drug events (ADEs) remain a key contributor to hospitalisations, resulting in long hospital stays and readmissions. Information pertaining to the specific medications and clinical factors associated with these outcomes is limited. Hence, a better understanding of these factors and their relationship to ADEs is required.

OBJECTIVES

To investigate medications involved, clinical manifestations of ADE-related hospitalisations, and their association with length of stay and readmission.

METHODS

A retrospective medical record review of patients admitted to a major, tertiary referral hospital in NSW, Australia, from January 2019 to August 2020 was conducted. ADEs were identified using Australian Refined Diagnosis Related Group (AR-DRG) codes: X40, X61, X62 and X64. Medications were classified per the Anatomical Therapeutic Chemical (ATC) classification system and clinical symptoms were classified per the International Classification of Disease (ICD) 9-CM. Logistic regression was performed to assess the relationship between medication and presentation classes with length of stay (≥2 days vs <2 days) and readmission.

RESULTS

There were 125 patients who met inclusion criteria (median age = 64 [interquartile range, 45-75] years; 53.6% male). Anti-thrombotic agents, opioids, antidepressants, antipsychotics, insulins and NSAIDs were the most implicated pharmacological classes. Neurological medications and falls were associated with a length of stay ≥2 days (adjusted odds ratio [aOR] 3.92, 95% confidence interval [CI] 1.48-10.33 and aOR 3.24, 95% CI 1.05-10.06, respectively). Neurological medications and neurological and cognitive disorders were associated with an increased likelihood of 90-day readmission (aOR 2.63, 95% CI 1.05-6.57 and aOR 3.20, 95% CI 1.17-8.75, respectively).

CONCLUSION

This study identified neurological medications as high-risk for increased length of stay and readmission in those hospitalised due to ADEs. This highlights the need for judicious prescribing and monitoring of these medications.

Development and Validation of a Trigger Tool for Identifying Drug-Related Emergency Department Visits

There are various trigger tools for detecting adverse drug events (ADEs), however, a drug-related emergency department (ED) visit trigger tool (DrEDTT) has not yet been developed. We aimed to develop and validate a DrEDTT with a multi-center cohort. In this cross-sectional study, we developed the DrEDTT consisting of 28 triggers through a comprehensive literature review and three phase expert group discussion. Next, we evaluated the performance of the DrEDTT by applying it to relevant medical records retrieved from four hospitals from January 2016 to June 2016. Two experts performed an in-depth chart review of a 25% of random sample of trigger flagged and unflagged ED visits and a true ADE was determined through causality assessment. Among 66,564 patients who visited the ED for reasons other than traffic accident and trauma during the study period, at least one trigger was found in 21,268 (32.0%) patients. A total of 959 true ADE cases (5.8%) were identified from a randomly selected 25% of ED visit cases. The overall positive predictive value was 14.0% (range: 8.3-66.7%). Sensitivity and specificity of DrEDTT were 77.7% and 70.4%, respectively. In conclusion, this newly developed trigger tool might be helpful to detect ADE-related ED visits.

Detection of adverse drug events in e-prescribing and administrative health data: a validation study

BACKGROUND

Administrative health data are increasingly used to detect adverse drug events (ADEs). However, the few studies evaluating diagnostic codes for ADE detection demonstrated low sensitivity, likely due to narrow code sets, physician under-recognition of ADEs, and underreporting in administrative data. The objective of this study was to determine if combining an expanded ICD code set in administrative data with e-prescribing data improves ADE detection.

METHODS

We conducted a prospective cohort study among patients newly prescribed antidepressant or antihypertensive medication in primary care and followed for 2 months. Gold standard ADEs were defined as patient-reported symptoms adjudicated as medication-related by a clinical expert. Potential ADEs in administrative data were defined as physician, ED, or hospital visits during follow-up for known adverse effects of the study medication, as identified by ICD codes. Potential ADEs in e-prescribing data were defined as study drug discontinuations or dose changes made during follow-up for safety or effectiveness reasons.

RESULTS

Of 688 study participants, 445 (64.7%) were female and mean age was 64.2 (SD 13.9). The study drug for 386 (56.1%) patients was an antihypertensive, and for 302 (43.9%) an antidepressant. Using the gold standard definition, 114 (16.6%) patients experienced an ADE, with 40 (10.4%) among antihypertensive users and 74 (24.5%) among antidepressant users. The sensitivity of the expanded ICD code set was 7.0%, of e-prescribing data 9.7%, and of the two combined 14.0%. Specificities were high (86.0-95.0%). The sensitivity of the combined approach increased to 25.8% when analysis was restricted to the 27% of patients who indicated having reported symptoms to a physician.

CONCLUSION

Combining an expanded diagnostic code set with e-prescribing data improves ADE detection. As few patients report symptoms to their physician, higher detection rates may be achieved by collecting patient-reported outcomes via emerging digital technologies such as patient portals and mHealth applications.

Population-Based Observational Study of Adverse Drug Event-Related Mortality in the Super-Aged Society of Japan

INTRODUCTION

Adverse drug events (ADEs) are a major cause of mortality.

OBJECTIVE

We examined long-term trends for ADE-related deaths in Japan.

METHODS

This observational study was conducted using the Japanese Vital Statistics from 1999 to 2016. Data for all ADE-related deaths were extracted using International Classification of Diseases, Tenth Revision codes. We analysed ADE-related deaths by age and sex and calculated crude and age-standardised mortality rates (ASMR) per 100,000 people. We used Joinpoint regression analysis to identify significant changing points in mortality trends and to estimate annual percentage change (APC).

RESULTS

In total, 16,417 ADE-related deaths were identified. The crude mortality rate for individuals aged ≥ 65 years was higher than that of young individuals. The ASMR per 100,000 people increased from 0.44 in 1999 to 0.64 in 2016. The crude mortality rate increased from 0.44 in 1999 to 1.01 in 2016. The APC of ASMR increased at a rate of 2.8% (95% confidence interval [CI] 1.4-4.2) throughout the study period. In addition, crude mortality increased at a rate of 5.7% (95% CI 4.2-7.3) annually from 1999 to 2016. The ADE-related mortality rate was higher for men than for women during the study period.

CONCLUSIONS

The number of and trend in ADE-related deaths increased in Japan from 1999 to 2016, particularly in the older population.

Utility of a Computerized ICD-10 Algorithm to Identify Idiosyncratic Drug-Induced Liver Injury Cases in the Electronic Medical Record

INTRODUCTION

Idiosyncratic drug-induced liver injury (DILI) is an important cause of liver injury that is difficult to diagnose and identify in the electronic medical record (EMR).

OBJECTIVE

Our objective was to develop a computerized algorithm that can reliably identify DILI cases from the EMR.

METHODS

The EMR was searched for all encounters with an International Classification of Diseases, Tenth Revision (ICD-10) T code for drug toxicity and a K-71 code for toxic liver injury between 1 October 2015 and 30 September 2018. Clinically significant liver injury was defined using predetermined laboratory values. An expert opinion causality score (1-3 = probable DILI, 4/5 = non-DILI), Roussel Uclaf Causality Assessment Method (RUCAM) score, and severity score was assigned to each case.

RESULTS

Among the 1,211,787 encounters searched, 517 had both an ICD-10 T code and a K-71 code, with 257 patients meeting the laboratory criteria. After excluding 75 cases of acetaminophen hepatotoxicity, the final study sample included 182 cases of potential DILI, with antineoplastics and antibiotics being the most frequently implicated agents. Causality assessment identified probable DILI in 121 patients (66.5%), whereas 61 (33.5%) had an alternative cause of liver injury. Although age, sex, race, and suspect drugs were similar, the probable DILI cases were more likely to present with a hepatocellular injury profile and have more severe liver injury than the non-DILI cases (p < 0.05).

CONCLUSION

A computerized algorithm based on a combination of ICD-10 codes identified 182 potential DILI cases with 121 true positives, 61 false positives, and a positive predictive value of 66.5%. Future studies incorporating natural language processing may further improve the utility of this algorithm in identifying high-causality idiosyncratic DILI cases.

Validated methods for identifying individuals with obesity in health care administrative databases: A systematic review

BACKGROUND

Health care administrative databases are increasingly used for health studies and public health surveillance. Cases of individuals with obesity are selected using case-identification methods. However, the validity of these methods is fragmentary and particularly challenging for obesity case identification.

OBJECTIVE

The objectives of this systematic review are to (1) determine the case-identification methods used to identify individuals with obesity in health care administrative databases and (2) to summarize the validity of these case-identification methods when compared with a reference standard.

METHODS

A systematic literature search was conducted in six bibliographic databases for the period January 1980 to June 2019 for all studies evaluating obesity case-identification methods compared with a reference standard.

RESULTS

Seventeen articles met the inclusion criteria. International Classification of Diseases (ICD) codes were the only case-identification method utilized in selected articles. The performance of obesity-identification methods varied widely across studies, with positive predictive value ranging from 19% to 100% while sensitivity ranged from 3% to 92%. The sensitivity of these methods was usually low while the specificity was higher.

CONCLUSION

When obesity is reported in health care administrative databases, it is usually correctly reported; however, obesity tends to be highly underreported in databases. Therefore, case-identification methods to monitor the prevalence and incidence of obesity within health care administrative databases are not reliable. In contrast, the use of these methods remains relevant for the selection of individuals with obesity for cohort studies, particularly when identifying cohorts of individuals with severe obesity or cohorts where obesity is associated with comorbidities.

Natural language processing with deep learning for medical adverse event detection from free-text medical narratives: A case study of detecting total hip replacement dislocation

BACKGROUND

Accurate and timely detection of medical adverse events (AEs) from free-text medical narratives can be challenging. Natural language processing (NLP) with deep learning has already shown great potential for analyzing free-text data, but its application for medical AE detection has been limited.

METHOD

In this study, we developed deep learning based NLP (DL-NLP) models for efficient and accurate hip dislocation AE detection following primary total hip replacement from standard (radiology notes) and non-standard (follow-up telephone notes) free-text medical narratives. We benchmarked these proposed models with traditional machine learning based NLP (ML-NLP) models, and also assessed the accuracy of International Classification of Diseases (ICD) and Current Procedural Terminology (CPT) codes in capturing these hip dislocation AEs in a multi-center orthopaedic registry.

RESULTS

All DL-NLP models outperformed all of the ML-NLP models, with a convolutional neural network (CNN) model achieving the best overall performance (Kappa = 0.97 for radiology notes, and Kappa = 1.00 for follow-up telephone notes). On the other hand, the ICD/CPT codes of the patients who sustained a hip dislocation AE were only 75.24% accurate.

CONCLUSIONS

We demonstrated that a DL-NLP model can be used in largescale orthopaedic registries for accurate and efficient detection of hip dislocation AEs. The NLP model in this study was developed with data from the most frequently used electronic medical record (EMR) system in the U.S., Epic. This NLP model could potentially be implemented in other Epic-based EMR systems to improve AE detection, and consequently, quality of care and patient outcomes.

Evaluation of the nature, severity, likelihood and preventability of medication-related hospital-acquired complications

Objective Pricing for safety and quality was introduced into Australian hospitals using a defined list of hospital-acquired complications (HACs). Medication-related HACs include drug-related respiratory complications (DRRC), haemorrhagic disorder due to circulating anticoagulants (HDDCA) and hypoglycaemia. The aim of this study was to determine the probability, severity and preventability of medication-related HACs, common contributory medications and themes, and whether medication-related HACs are a suitable data source to inform risk associated with medicines use. Methods Medical notes were reviewed retrospectively for all patients discharged from a tertiary referral metropolitan hospital between 1 July and 31 December 2018 who were flagged as experiencing a medication-related HAC. Naranjo, Hartwig's and Schumock and Thornton tools were used to assess the probability, severity and preventability of medication-related HACs. Results Over the 6-month period, 88 patients experienced a medication-related HAC. An HAC was not identified in five (5.7%) patient charts. The most common HAC was hypoglycaemia (n=59; 67%), followed by HDDCA (n=23; 26%) and DRRC (n=6; 7%). Fifteen patients (17%) flagged with a hypoglycaemia HAC were not on a medicine associated with hypoglycaemia. Overall, 6% (n=4) of HACs were severe, 72% (n=49) were moderate and 22% (n=15) were mild. Where the HAC and causal medication(s) were identified (n=68), over half were probable (51.5%, n=35) and 44.1% (n=30) were possible causes of the adverse drug reaction; only two (2.9%) were definite causes. None of the DRRC HACs was preventable. Over half the HDDCA HACs (52.2%; n=12) and almost half the hypoglycaemia HACs (46.2%; n=18) were not preventable. Common themes included appropriate anticoagulant agent, dose and monitoring, as well as periprocedural hypoglycaemic management, which considers oral intake and comorbidities. Conclusion Not all patients who experience medication-related HACs were on causative medications. Of those who were, medications were probable causal agents in over 50% of cases. Only a small number of HACs were severe and under half of medication-related HACs were preventable. What is known about the topic? The relationship between pricing for safety and quality and improvements in patient outcomes has shown mixed results. Medication-related harm is a problem within Australia and system-wide changes should be considered to improve patient care. What does this paper add? This paper adds evidence to the use of medication-related HACs as a source of data to inform risk associated with medicines use and provides details on the preventability and severity of medication-related HACs and the likelihood that medicines contribute to these complications. What are the implications for practitioners? This paper provides clinicians and policy makers details on the utility of using medication-related HACs as a measure of risk associated with medicines use. It discusses merit in using HACs as a source for quality improvement, but recommends that definitions may need to be reviewed to enhance utility.

Association Between Adverse Events and Discontinuation of Antiepileptic Drugs Among Drug-Naïve Adults with Epilepsy

Background

Adherence to antiepileptic drugs (AEDs) remains the primary management tool to prevent recurrent seizures in patients with epilepsy. Adverse events associated with AEDs could have an impact on adherence and result in treatment failures.

Objective

The goal of this study was to assess the association between adverse events and discontinuation of AEDs for AED-naïve patients with epilepsy. Our second objective was to estimate the economic burden of AED discontinuation.

Methods

We retrospectively analyzed IBM MarketScan administrative data from 2014 to 2017. The cohort consisted of new users of AEDs with an epilepsy diagnosis and with two or more subsequent AED claims. Outpatient and inpatient cohorts were analyzed separately. Adverse events were identified by injury codes (E-CODES) or by International Classification of Diseases, Ninth/Tenth Edition (ICD-9/10) codes for disease manifestations reported in the literature or product inserts (LADE). Discontinuation of AEDs was defined as a gap of ≥ 60 days without a refill. All cost comparisons were based on 1:1 propensity-score matching. Associations between adverse events and discontinuation were estimated using logistic regression, adjusting for predefined covariates such as age, sex, Charlson Comorbidity Index, insurance type, and AED type.

Results

The overall discontinuation rate was 9% (E-CODES rate was 0.1% and LADE rate was 27%). The discontinued group was older (56.1 vs. 52.8 years; p < 0.0001). Adults aged ≥ 65 years had the highest discontinuation rate (11%). Patients who discontinued had fewer AED claims (6.8 vs. 9.2; p < 0.0001), more outpatient claims (19.3 vs. 17.8; p < 0.0001), and longer hospital stays (6.6 vs. 5.3 days; p < 0.0001). Differences in daily outpatient costs between patients with and without adverse events were statistically significant (E-CODES $US213 vs. 105; p = 0.001; LADE $US188 vs. 161; p < 0.0001). Additionally, total cost of AEDs in the outpatient cohort was higher for patients with adverse events (E-CODES and LADE). There was no association between E-CODES and AED discontinuation; however, there was a positive association between LADE and discontinuation in the outpatient cohort but a negative association in the inpatient cohort.

Conclusion

We found that total costs of prescriptions claimed and total costs of outpatient visits among the outpatient cohort were higher for those with adverse drug events than for those without. An association between adverse events and discontinuation was inconclusive because it depended on the target population and how the adverse events were identified.

Comparison of laboratory threshold criteria in drug-induced liver injury detection algorithms for use in pharmacovigilance

PURPOSE

For the purpose of pharmacovigilance, we sought to determine the best performing laboratory threshold criteria to detect drug-induced liver injury (DILI) in the electronic medical records (EMR).

METHODS

We compared three commonly used liver chemistry criteria from the DILI expert working group (DEWG), DILI network (DILIN), and Council for International Organizations of Medical Sciences (CIOMS), based on hospital EMR for years 2010 and 2011 (42 176 admissions), using independent medical record review. The performance characteristics were compared in terms of sensitivity, specificity, positive predictive value (PPV), negative predictive value, accuracy, F-measure, and area under the receiver operating characteristic curve (AUROC).

RESULTS

DEWG had the highest PPV (5.5%, 95% CI: 4.1%-7.2%), specificity (97.0%, 95% CI: 96.8%-97.2%), accuracy (96.8%, 95% CI: 96.6%-97.0%) and F-measure (0.099). CIOMS had the highest sensitivity (74.0%, 95% CI: 64.3%-82.3%) and AUROC (85.2%, 95% CI: 80.8%-89.7%). Besides the laboratory criteria, including additional keywords in the classification algorithm improved the PPV and F-measure to a maximum of 29.0% (95% CI: 22.3%-36.5%) and 0.379, respectively.

CONCLUSIONS

More stringent criteria (DEWG and DILIN) performed better in terms of PPV, specificity, accuracy and F-measure. CIOMS performed better in terms of sensitivity. An algorithm with high sensitivity is useful in pharmacovigilance for detecting rare events and to avoid missing cases. Requiring at least two abnormal liver chemistries during hospitalization and text-word searching in the discharge summaries decreased false positives without loss in sensitivity.