Detection of adverse drug events in e-prescribing and administrative health data: a validation study.
Ontology highlight
ABSTRACT: 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.
SUBMITTER: Habib B
PROVIDER: S-EPMC8063436 | biostudies-literature |
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:BackgroundAdverse drug events are a frequent cause of emergency department presentations. Administrative data could be used to identify patients presenting with adverse drug events for post-market surveillance, and to conduct research in patient safety and in drug safety and effectiveness. However, such data sources have not been evaluated for their completeness with regard to adverse drug event reporting. Our objective was to determine the proportion of adverse drug events to outpatient medications diagnosed at the point-of-care in emergency departments that were documented in administrative data.MethodsWe linked the records of patients enrolled in a prospective observational cohort study on adverse drug events conducted in two Canadian tertiary care emergency departments to their administrative data. We compared the number of adverse drug events diagnosed and recorded at the point-of-care in the prospective study with the number of adverse drug events recorded in the administrative data.ResultsAmong 1574 emergency department visits, 221 were identified as adverse drug event-related in the prospective database. We found 15 adverse drug events documented in administrative records with ICD-10 codes clearly indicating an adverse drug event, indicating a sensitivity of 6.8% (95% CI 4.0-11.2%) of this code set. When the ICD-10 code categories were broadened to include codes indicating a very likely, likely or possible adverse event to a medication, 62 of 221 events were identifiable in administrative data, corresponding to a sensitivity of 28.1% (95% CI 22.3-34.6%).ConclusionsAdverse drug events to outpatient medications were underreported in emergency department administrative data compared to the number of adverse drug events diagnosed and recorded at the point-of-care.
| S-EPMC3842633 | biostudies-literature
Project description:Inappropriate prescribing (IP) in older patients is highly prevalent and is associated with an increased risk of adverse drug events (ADEs), morbidity, mortality and healthcare utilisation. Consequently, IP is a major safety concern and with changing population demographics, it is likely to become even more prevalent in the future. IP can be detected using explicit or implicit prescribing indicators. Theoretically, the routine clinical application of these IP criteria could represent an inexpensive and time efficient method to optimise prescribing practice. However, IP criteria must be sensitive, specific, have good inter-rater reliability and incorporate those medications most commonly associated with ADEs in older people. To be clinically relevant, use of prescribing appropriateness tools must translate into positive patient outcomes, such as reduced rates of ADEs. To accurately measure these outcomes, a reliable method of assessing the relationship between the administration of a drug and an adverse clinical event is required. The Naranjo criteria are the most widely used tool for assessing ADE causality, however, they are often difficult to interpret in the context of older patients. ADE causality criteria that allow for the multiple co-morbidities and prescribed medications in older people are required. Ultimately, the current high prevalence of IP and ADEs is unacceptable. IP screening criteria need to be tested as an intervention to assess their impact on the incidence of ADEs in vulnerable older patients. There is a role for IP screening tools in everyday clinical practice. These should enhance, not replace good clinical judgement, which in turn should be based on sound pharmacogeriatric training.
| S-EPMC2642820 | biostudies-other
Project description:BackgroundAdverse drug events, the unintended and harmful effects of medications, are important outcome measures in health services research. Yet no universally accepted set of International Classification of Diseases (ICD) revision 10 codes or coding algorithms exists to ensure their consistent identification in administrative data. Our objective was to synthesize a comprehensive set of ICD-10 codes used to identify adverse drug events.MethodsWe developed a systematic search strategy and applied it to five electronic reference databases. We searched relevant medical journals, conference proceedings, electronic grey literature and bibliographies of relevant studies, and contacted content experts for unpublished studies. One author reviewed the titles and abstracts for inclusion and exclusion criteria. Two authors reviewed eligible full-text articles and abstracted data in duplicate. Data were synthesized in a qualitative manner.ResultsOf 4241 titles identified, 41 were included. We found a total of 827 ICD-10 codes that have been used in the medical literature to identify adverse drug events. The median number of codes used to search for adverse drug events was 190 (IQR 156-289) with a large degree of variability between studies in the numbers and types of codes used. Authors commonly used external injury (Y40.0-59.9) and disease manifestation codes. Only two papers reported on the sensitivity of their code set.ConclusionsSubstantial variability exists in the methods used to identify adverse drug events in administrative data. Our work may serve as a point of reference for future research and consensus building in this area.
| S-EPMC3994866 | biostudies-literature
Project description:BackgroundMost prevalence estimates of chronic pain are derived from surveys and vary widely, both globally (2%-54%) and in Canada (6.5%-44%). Health administrative data are increasingly used for chronic disease surveillance, but their validity as a source to ascertain chronic pain cases is understudied.AimThe aim of this study was to derive and validate an algorithm to identify cases of chronic pain as a single chronic disease using provincial health administrative data.MethodsA reference standard was developed and applied to the electronic medical records data of a Newfoundland and Labrador general population sample participating in the Canadian Primary Care Sentinel Surveillance Network. Chronic pain algorithms were created from the administrative data of patient populations with chronic pain, and their classification performance was compared to that of the reference standard via statistical tests of selection accuracy.ResultsThe most performant algorithm for chronic pain case ascertainment from the Medical Care Plan Fee-for-Service Physicians Claims File was one anesthesiology encounter ever recording a chronic pain clinic procedure code OR five physician encounter dates recording any pain-related diagnostic code in 5 years with more than 183 days separating at least two encounters. The algorithm demonstrated 0.703 (95% confidence interval [CI], 0.685-0.722) sensitivity, 0.668 (95% CI, 0.657-0.678) specificity, and 0.408 (95% CI, 0.393-0.423) positive predictive value. The chronic pain algorithm selected 37.6% of a Newfoundland and Labrador provincial cohort.ConclusionsA health administrative data algorithm was derived and validated to identify chronic pain cases and estimate disease burden in residents attending fee-for-service physician encounters in Newfoundland and Labrador.
| S-EPMC7967902 | biostudies-literature
Project description:Background More than 600 000 coronary stents are implanted during percutaneous coronary interventions (PCIs) annually in the United States. Because no real-world surveillance system exists to monitor their long-term safety, claims data are often used for this purpose. The extent to which adverse events identified with claims data can be reasonably attributed to a specific medical device is uncertain. Methods and Results We used deterministic matching to link the NCDR (National Cardiovascular Data Registry) CathPCI Registry to Medicare fee-for-service claims for patients aged ≥65 years who underwent PCI with drug-eluting stents (DESs) between July 1, 2009 and December 31, 2013. We identified subsequent PCIs within 1 year of the index procedure in Medicare claims as potential safety events. We linked these subsequent PCIs back to the NCDR CathPCI Registry to ascertain how often the revascularization could be reasonably attributed to the same coronary artery as the index PCI (ie, target vessel revascularization). Of 415 306 DES placements in 368 194 patients, 33 174 repeat PCIs were identified in Medicare claims within 1 year. Of these, 28 632 (86.3%) could be linked back to the NCDR CathPCI Registry; 16 942 (51.1% of repeat PCIs) were target vessel revascularizations. Of these, 8544 (50.4%) were within a previously placed DES: 7652 for in-stent restenosis and 1341 for stent thrombosis. Of 16 176 patients with a claim for acute myocardial infarction in the follow-up period, 4446 (27.5%) were attributed to the same coronary artery in which the DES was implanted during the index PCI (ie, target vessel myocardial infarction). Of 24 288 patients whose death was identified in claims data, 278 (1.1%) were attributed to the same coronary artery in which the DES was implanted during the index PCI. Conclusions Most repeat PCIs following DES stent implantation identified in longitudinal claims data could be linked to real-world registry data, but only half could be reasonably attributed to the same coronary artery as the index procedure. Attribution among those with acute myocardial infarction or who died was even less frequent. Safety signals identified using claims data alone will require more in-depth examination to accurately assess stent safety.
| S-EPMC7070203 | biostudies-literature
Project description:Adverse drug events are unintended and harmful events related to medications. Adverse drug events are important for patient care, quality improvement, drug safety research, and postmarketing surveillance, but they are vastly underreported.Our objectives were to identify barriers to adverse drug event documentation and factors contributing to underreporting.This qualitative study was conducted in 1 ambulatory center, and the emergency departments and inpatient wards of 3 acute care hospitals in British Columbia between March 2014 and December 2016. We completed workplace observations and focus groups with general practitioners, hospitalists, emergency physicians, and hospital and community pharmacists. We analyzed field notes by coding and iteratively analyzing our data to identify emerging concepts, generate thematic and event summaries, and create workflow diagrams. Clinicians validated emerging concepts by applying them to cases from their clinical practice.We completed 238 hours of observations during which clinicians investigated 65 suspect adverse drug events. The observed events were often complex and diagnosed over time, requiring the input of multiple providers. Providers documented adverse drug events in charts to support continuity of care but never reported them to external agencies. Providers faced time constraints, and reporting would have required duplication of documentation.Existing reporting systems are not suited to capture the complex nature of adverse drug events or adapted to workflow and are simply not used by frontline clinicians. Systems that are integrated into electronic medical records, make use of existing data to avoid duplication of documentation, and generate alerts to improve safety may address the shortcomings of existing systems and generate robust adverse drug event data as a by-product of safer care.
| S-EPMC5849794 | biostudies-other
Project description:BackgroundFrail older people with multiple interacting conditions, polypharmacy, and complex care needs are particularly exposed to health care-related adverse events. Among these, anticoagulant-related thromboembolic and hemorrhagic events are particularly frequent and serious in older inpatients. The growing use of anticoagulants in this population and their substantial risk of toxicity and inefficacy have therefore become an important patient safety and public health concern worldwide. Anticoagulant-related adverse events and the quality of anticoagulation management should thus be routinely assessed to improve patient safety in vulnerable older inpatients.ObjectiveThis project aims to develop and validate a set of outcome and process indicators based on linked administrative health data (ie, insurance claims data linked to hospital discharge data) assessing older inpatient safety related to anticoagulation in both Switzerland and France, and enabling comparisons across time and among hospitals, health territories, and countries. Geriatric patient safety indicators (GPSIs) will assess anticoagulant-related adverse events. Geriatric quality indicators (GQIs) will evaluate the management of anticoagulants for the prevention and treatment of arterial or venous thromboembolism in older inpatients.MethodsGPSIs will measure cumulative incidences of thromboembolic and bleeding adverse events based on hospital discharge data linked to insurance claims data. Using linked administrative health data will improve GPSI risk adjustment on patients' conditions that are present at admission and will capture in-hospital and postdischarge adverse events. GQIs will estimate the proportion of index hospital stays resulting in recommended anticoagulation at discharge and up to various time frames based on the same electronic health data. The GPSI and GQI development and validation process will comprise 6 stages: (1) selection and specification of candidate indicators, (2) definition of administrative data-based algorithms, (3) empirical measurement of indicators using linked administrative health data, (4) validation of indicators, (5) analyses of geographic and temporal variations for reliable and valid indicators, and (6) data visualization.ResultsStudy populations will consist of 166,670 Swiss and 5,902,037 French residents aged 65 years and older admitted to an acute care hospital at least once during the 2012-2014 period and insured for at least 1 year before admission and 1 year after discharge. We will extract Swiss data from the Helsana Group data warehouse and French data from the national health insurance information system (SNIIR-AM). The study has been approved by Swiss and French ethics committees and regulatory organizations for data protection.ConclusionsValidated GPSIs and GQIs should help support and drive quality and safety improvement in older inpatients, inform health care stakeholders, and enable international comparisons. We discuss several limitations relating to the representativeness of study populations, accuracy of administrative health data, methods used for GPSI criterion validity assessment, and potential confounding bias in comparisons based on GQIs, and we address these limitations to strengthen study feasibility and validity.
| S-EPMC5445236 | biostudies-literature
Project description:IntroductionMeasure and monitor adverse events (AEs) following hip arthroplasty is challenging. The aim of this study was to create a model for measuring AEs after hip arthroplasty using administrative data, such as length of stay and readmissions, with equal or better precision than an ICD-code based model.Materials and methodsThis study included 1 998 patients operated with an acute or elective hip arthroplasty in a national multi-centre study. We collected AEs within 90 days following surgery with retrospective record review. Additional data came from the Swedish Hip Arthroplasty Register, the Swedish National Patient Register and the Swedish National Board of Health and Welfare. We made a 2:1 split of the data into a training and a holdout set. We used the training set to train different machine learning models to predict if a patient had sustained an AE or not. After training and cross-validation we tested the best performing model on the holdout-set. We compared the results with an established ICD-code based measure for AEs.ResultsThe best performing model was a logistic regression model with four natural age splines. The variables included in the model were as follows: length of stay at the orthopaedic department, discharge to acute care, age, number of readmissions and ED visits. The sensitivity and specificity for the new model was 23 and 90% for AE within 30 days, compared with 5 and 94% for the ICD-code based model. For AEs within 90 days the sensitivity and specificity were 31% and 89% compared with 16% and 92% for the ICD-code based model.ConclusionWe conclude that a prediction model for AEs following hip arthroplasty surgery, relying on administrative data without ICD-codes is more accurate than a model based on ICD-codes.
| S-EPMC7644076 | biostudies-literature
Project description:RationaleDespite the high prevalence of obstructive sleep apnea (OSA) and concurrent use of opioid therapy, no large-scale population studies have investigated whether opioid use and pre-existing OSA may interact synergistically to increase the risk of adverse health consequences. To address this knowledge gap, we conducted a retrospective cohort study using provincial health administrative data to evaluate whether the combined presence of opioid use and OSA increases the risk of adverse health consequences, such as mortality, hospitalizations, and emergency department (ED) visits; and if it does, whether this co-occurrence has synergistic clinical relevance.MethodsWe included all adults who underwent a diagnostic sleep study in Ontario, Canada, between 2013 and 2016. Individuals were considered exposed to opioids if they filled a prescription that overlapped with the date of their sleep study (Opioid+). Individuals with at least a 50% probability of having a diagnosis of moderate to severe OSA (OSA+) were identified using a previously externally validated case-ascertainment model. The primary outcome was all-cause mortality; secondary outcomes were all-cause or ischemic heart disease hospitalizations, all-cause ED visits, and motor vehicle collisions (MVC) requiring hospital or ED visit. We used multivariable Cox regression models to compare hazards between four mutually exclusive groups: (1) Opioid+ OSA+; (2) Opioid+ OSA-; (3) Opioid- OSA+, and (4) OSA- Opioid- (reference for comparison). Relative excess risks due to interaction (RERI) were calculated to test for additive interaction.ResultsOf 300,663 adults who underwent a sleep study, 15,713 (5.2%) were considered as Opioid+ and 128,351 (42.7%) as OSA+. Over a median of two years, 6,223 (2.1%) died from any cause. Regardless of OSA status, opioid use at the date of the sleep study was associated with an increased hazard for all-cause mortality with the greatest hazard associated with Opioid+ OSA- (adjusted hazard ratio [aHR]: 1.75, 95% CI 1.57-1.94), but not Opioid+ OSA+ (aHR: 1.14, 95% CI 1.02-1.27) as hypothesized. Regardless of OSA status, opioid use at the date of the sleep study was associated with an increased hazard for all secondary outcomes. Opioid+ OSA+ was associated with the greatest hazards of all-cause hospitalizations (aHR 1.55, 95% CI 1.49-1.61) and MVC (aHR of 1.39; 95% CI 1.09-1.77); however, no statistically significant synergistic effects were observed.ConclusionsAdults referred for sleep disorder assessment who used opioids had a significantly increased hazard of adverse health outcomes than those who did not, regardless of whether they had a high probability of moderate to severe OSA. The use of opioids and OSA was associated with the greatest hazard of all-cause hospitalizations and MVC requiring hospital or ED visit. The interaction of opioids and OSA did not confer a synergistic risk for poor outcomes.
| S-EPMC9239451 | biostudies-literature
Project description:MotivationAdverse drug reactions (ADRs) are a central consideration during drug development. Here we present a machine learning classifier to prioritize ADRs for approved drugs and pre-clinical small-molecule compounds by combining chemical structure (CS) and gene expression (GE) features. The GE data is from the Library of Integrated Network-based Cellular Signatures (LINCS) L1000 dataset that measured changes in GE before and after treatment of human cells with over 20 000 small-molecule compounds including most of the FDA-approved drugs. Using various benchmarking methods, we show that the integration of GE data with the CS of the drugs can significantly improve the predictability of ADRs. Moreover, transforming GE features to enrichment vectors of biological terms further improves the predictive capability of the classifiers. The most predictive biological-term features can assist in understanding the drug mechanisms of action. Finally, we applied the classifier to all >20 000 small-molecules profiled, and developed a web portal for browsing and searching predictive small-molecule/ADR connections.Availability and implementationThe interface for the adverse event predictions for the >20 000 LINCS compounds is available at http://maayanlab.net/SEP-L1000/ CONTACT: avi.maayan@mssm.eduSupplementary informationSupplementary data are available at Bioinformatics online.
| S-EPMC4965635 | biostudies-literature