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.

Impact of single and combined rare diseases on adult inpatient outcomes: a retrospective, cross-sectional study of a large inpatient population.

BACKGROUND: Little is known about the impact of rare diseases on inpatient outcomes. OBJECTIVE: To compare outcomes of inpatients with 0, 1, or > 1 rare disease. A catalogue of 628 ICD-10 coded rare diseases was applied to count rare diseases. DESIGN: Retrospective, cross-sectional study. SUBJECTS: 165,908 inpatients, Swiss teaching hospital. MAIN MEASURES: Primary outcome: in-hospital mortality. SECONDARY OUTCOMES: length of stay (LOS), intensive care unit (ICU) admissions, ICU LOS, and 30-day readmissions. Associations with single and combined rare diseases were analyzed by multivariable regression. KEY RESULTS: Patients with 1 rare disease were at increased risk of in-hospital death (odds ratio [OR]: 1.80; 95% confidence interval [CI]: 1.67, 1.95), combinations of rare diseases showed stronger associations (OR 2.78; 95% CI 2.39, 3.23). Females with 1 rare disease had an OR of 1.69 (95% CI 1.50, 1.91) for in-hospital death, an OR of 2.99 (95% CI 2.36, 3.79) if they had a combination of rare diseases. Males had an OR of 1.85 (95% CI 1.68, 2.04) and 2.61 (95% CI 2.15, 3.16), respectively. Rare diseases were associated with longer LOS (for 1 and > 1 rare diseases: increase by 28 and 49%), ICU admissions (for 1 and > 1: OR 1.64 [95% CI 1.57, 1.71] and 2.23 [95% CI 2.01, 2.48]), longer ICU LOS (for 1 and > 1 rare diseases: increase by 14 and 40%), and 30-day readmissions (for 1 and > 1: OR 1.57 [95% CI 1.47, 1.68] and 1.64 [95% CI 1.37, 1.96]). CONCLUSIONS: Rare diseases are independently associated with worse inpatient outcomes. This might be the first study suggesting even stronger associations of combined rare diseases with in-hospital deaths, increased LOS, ICU admissions, increased ICU LOS, and 30-day readmissions.

Towards The Automated, Empirical Filtering of Drug-Drug Interaction Alerts in Clinical Decision Support Systems: Historical Cohort Study of Vitamin K Antagonists.

BACKGROUND: Drug-drug interactions (DDIs) involving vitamin K antagonists (VKAs) constitute an important cause of in-hospital morbidity and mortality. However, the list of potential DDIs is long; the implementation of all these interactions in a clinical decision support system (CDSS) results in over-alerting and alert fatigue, limiting the benefits provided by the CDSS. OBJECTIVE: To estimate the probability of occurrence of international normalized ratio (INR) changes for each DDI rule, via the reuse of electronic health records. METHODS: An 8-year, exhaustive, population-based, historical cohort study including a French community hospital, a group of Danish community hospitals, and a Bulgarian hospital. The study database included 156,893 stays. After filtering against two criteria (at least one VKA administration and at least one INR laboratory result), the final analysis covered 4047 stays. Exposure to any of the 145 drugs known to interact with VKA was tracked and analyzed if at least 3 patients were concerned. The main outcomes are VKA potentiation (defined as an INR≥5) and VKA inhibition (defined as an INR≤1.5). Groups were compared using the Fisher exact test and logistic regression, and the results were expressed as an odds ratio (95% confidence limits). RESULTS: The drugs known to interact with VKAs either did not have a statistically significant association regarding the outcome (47 drug administrations and 14 discontinuations) or were associated with significant reduction in risk of its occurrence (odds ratio<1 for 18 administrations and 21 discontinuations). CONCLUSIONS: The probabilities of outcomes obtained were not those expected on the basis of our current body of pharmacological knowledge. The results do not cast doubt on our current pharmacological knowledge per se but do challenge the commonly accepted idea whereby this knowledge alone should be used to define when a DDI alert should be displayed. Real-life probabilities should also be considered during the filtration of DDI alerts by CDSSs, as proposed in SPC-CDSS (statistically prioritized and contextualized CDSS). However, these probabilities may differ from one hospital to another and so should probably be calculated locally.

Developing strategies for predicting hyperkalemia in potassium-increasing drug-drug interactions.

OBJECTIVE: To compare different strategies predicting hyperkalemia (serum potassium level ≥5.5 mEq/l) in hospitalized patients for whom medications triggering potassium-increasing drug-drug interactions (DDIs) were ordered. MATERIALS AND METHODS: We investigated 5 strategies that combined prediction triggered at onset of DDI versus continuous monitoring and taking into account an increasing number of patient parameters. The considered patient parameters were identified using generalized additive models, and the thresholds of the prediction strategies were calculated by applying Youden's J statistic to receiver operation characteristic curves. Half of the data served as the calibration set, half as the validation set. RESULTS: We identified 132 incidences of hyperkalemia induced by 8413 potentially severe potassium-increasing DDIs among 76 467 patients. The positive predictive value (PPV) of those strategies predicting hyperkalemia at the onset of DDI ranged from 1.79% (undifferentiated anticipation of hyperkalemia due to the DDI) to 3.02% (additionally considering the baseline serum potassium) and 3.10% (including further patient parameters). Continuous monitoring significantly increased the PPV to 8.25% (considering the current serum potassium) and 9.34% (additional patient parameters). CONCLUSION: Continuous monitoring of the risk for hyperkalemia based on current potassium level shows a better predictive power than predictions triggered at the onset of DDI. This contrasts with efforts to improve DDI alerts by taking into account more patient parameters at the time of ordering.

Clinical decision support systems.

Clinical decision support (CDS) systems link patient data with an electronic knowledge base in order to improve decision-making and computerised physician order entry (CPOE) is a requirement to set up electronic CDS. The medical informatics literature suggests categorising CDS tools into medication dosing support, order facilitators, point-of-care alerts and reminders, relevant information display, expert systems and workflow support. To date, CDS has particularly been recognised for improving processes. CDS successfully fostered prevention of deep-vein thrombosis, improved adherence to guidelines, increased the use of vaccinations, and decreased the rate of serious medication errors. However, CDS may introduce errors, and therefore the term "e-iatrogenesis" has been proposed to address unintended consequences. At least two studies reported severe treatment delays due to CPOE and CDS. In addition, the phenomenon of "alert fatigue" - arising from a high number of CDS alerts of low clinical significance - may facilitate overriding of potentially critical notifications. The implementation of CDS needs to be carefully planned, CDS interventions should be thoroughly examined in pilot wards only, and then stepwise introduced. A crucial feature of CPOE in combination with CDS is speed, since time consumption has been found to be a major factor determining failure. In the near future, the specificity of alerts will be improved, notifications will be prioritised and offer detailed advice, customisation of CDS will play an increasing role, and finally, CDS is heading for patient-centred decision support. The most important research question remains whether CDS is able to improve patient outcomes beyond processes.

Use of an on-demand drug-drug interaction checker by prescribers and consultants: a retrospective analysis in a Swiss teaching hospital.

BACKGROUND: Offering a drug-drug interaction (DDI) checker on-demand instead of computer-triggered alerts is a strategy to avoid alert fatigue. OBJECTIVE: The purpose was to determine the use of such an on-demand tool, implemented in the clinical information system for inpatients. METHODS: The study was conducted at the University Hospital Zurich, an 850-bed teaching hospital. The hospital-wide use of the on-demand DDI checker was measured for prescribers and consulting pharmacologists. The number of DDIs identified on-demand was compared to the number that would have resulted by computer-triggering and this was compared to patient-specific recommendations by a consulting pharmacist. RESULTS: The on-demand use was analyzed during treatment of 64,259 inpatients with 1,316,884 prescriptions. The DDI checker was popular with nine consulting pharmacologists (648 checks/consultant). A total of 644 prescribing physicians used it infrequently (eight checks/prescriber). Among prescribers, internists used the tool most frequently and obtained higher numbers of DDIs per check (1.7) compared to surgeons (0.4). A total of 16,553 DDIs were identified on-demand, i.e., <10 % of the number the computer would have triggered (169,192). A pharmacist visiting 922 patients on a medical ward recommended 128 adjustments to prevent DDIs (0.14 recommendations/patient), and 76 % of them were applied by prescribers. In contrast, computer-triggering the DDI checker would have resulted in 45 times more alerts on this ward (6.3 alerts/patient). CONCLUSIONS: The on-demand DDI checker was popular with the consultants only. However, prescribers accepted 76 % of patient-specific recommendations by a pharmacist. The prescribers' limited on-demand use indicates the necessity for developing improved safety concepts, tailored to suit these consumers. Thus, different approaches have to satisfy different target groups.

Sustained impact of electronic alerts on rate of prophylaxis against venous thromboembolism.

Advanced electronic alerts (eAlerts) and computerised physician order entry (CPOE) increase adequate thromboprophylaxis orders among hospitalised medical patients. It remains unclear whether eAlerts maintain their efficacy over time, after withdrawal of continuing medical education (CME) on eAlerts and on thromboprophylaxis indications from the study staff. We analysed 5,317 hospital cases from the University Hospital Zurich during 2006-2009: 1,854 cases from a medical ward with eAlerts (interventiongroup) and 3,463 cases from a surgical ward without eAlerts (controlgroup). In the intervention group, an eAlert with hospital-specific venous thromboembolism (VTE) prevention guidelines was issued in the electronic patient chart 6 hours after admission if no pharmacological or mechanical thromboprophylaxis had been ordered. Data were analysed for three phases: pre-implementation (phase 1), eAlert implementation with CME (phase 2), and post-implementation without CME (phase3). The rates of thromboprophylaxis in the intervention group were 43.4% in phase 1 and 66.7% in phase 2 (p<0.001), and increased further to 73.6% in phase3 (p=0.011). Early thromboprophylaxis orders within 12 hours after admission were more often placed in phase 2 and 3 as compared to phase 1 (67.1% vs. 52.1%, p<0.001). In the surgical control group, the thromboprophylaxis rates in the three phases were 88.6%, 90.7%, 90.6% (p=0.16). Advanced eAlerts may provide sustained efficacy over time, with stable rates of thromboprophylaxis orders among hospitalised medical patients.