Grégoire Ficheur

Data Mining to Generate Adverse Drug Events Detection Rules

Adverse drug events (ADEs) are a public health is sue. Their detection usually relies on voluntary reporting or medical chart reviews. The objective of this paper is to automatically detect cases of ADEs by data mining. 115 447 complete past hospital stays are extracted from six French, Danish, and Bulgarian hospitals using a common data model including diagnoses, drug administrations, laboratory results, and free-text records. Different kinds of outcomes are traced, and supervised rule induction methods (decision trees and association rules) are used to discover ADE detection rules, with respect to time constraints. The rules are then filtered, validated, and reorganized by a committee of experts. The rules are described in a rule repository, and several statistics are automatically computed in every medical department, such as the confidence, relative risk, and median delay of outcome appearance. 236 validated ADE-detection rules are discovered; they enable to detect 27 different kinds of outcomes. The rules use a various number of conditions related to laboratory results, diseases, drug administration, and demographics. Some rules involve innovative conditions, such as drug discontinuations.

Proposal and evaluation of FASDIM, a Fast And Simple De-Identification Method for unstructured free-text clinical records.

PURPOSE Medical free-text records enable to get rich information about the patients, but often need to be de-identified by removing the Protected Health Information (PHI), each time the identification of the patient is not mandatory. Pattern matching techniques require pre-defined dictionaries, and machine learning techniques require an extensive training set. Methods exist in French, but either bring weak results or are not freely available. The objective is to define and evaluate FASDIM, a Fast And Simple De-Identification Method for French medical free-text records. METHODS FASDIM consists in removing all the words that are not present in the authorized word list, and in removing all the numbers except those that match a list of protection patterns. The corresponding lists are incremented in the course of the iterations of the method. For the evaluation, the workload is estimated in the course of records de-identification. The efficiency of the de-identification is assessed by independent medical experts on 508 discharge letters that are randomly selected and de-identified by FASDIM. Finally, the letters are encoded after and before de-identification according to 3 terminologies (ATC, ICD10, CCAM) and the codes are compared. RESULTS The construction of the list of authorized words is progressive: 12h for the first 7000 letters, 16 additional hours for 20,000 additional letters. The Recall (proportion of removed Protected Health Information, PHI) is 98.1%, the Precision (proportion of PHI within the removed token) is 79.6% and the F-measure (harmonic mean) is 87.9%. In average 30.6 terminology codes are encoded per letter, and 99.02% of those codes are preserved despite the de-identification. CONCLUSION FASDIM gets good results in French and is freely available. It is easy to implement and does not require any predefined dictionary.

Fetal Heart-Rate Variability: Validation of a New Continuous, Noninvasive Computerized Analysis

Objective: We developed a computerized heart-rate variability index related to the fetal parasympathetic activity: the Fetal Stress Index (FSI). The objective was to determine whether the FSI is related to the visual analysis of the fetal heart rate (FHR). Methods: Thirty tracings recorded at a labor ward were classified according to the NICHD categories: (I) normal FHR tracing, (II) intermediate risk of acidosis, and (III) high risk. FSI was calculated as minimum, maximum, and mean, and was evaluated before the onset of the FHR pattern, during the 10 min following, and between 10 and 20 min after that. Results: The FSI for categories II and III was similar to that of category I before the onset of the FHR pattern. FSI min was lower just after the onset of the abnormal FHR in category III, compared with that of category I (33 vs. 43, p < 0.001). Between 10 and 20 min after the onset of the abnormal FHR, we observed a significant reduction in FSI min in categories II and III (44 vs. 39 vs. 29.7, p < 0.0001). Conclusion: Although further studies are necessary for the sake of clinical validation, FSI could constitute an interesting method for the evaluation of fetal well-being.

Adverse drug events with hyperkalaemia during inpatient stays: evaluation of an automated method for retrospective detection in hospital databases

BackgroundAdverse drug reactions and adverse drug events (ADEs) are major public health issues. Many different prospective tools for the automated detection of ADEs in hospital databases have been developed and evaluated. The objective of the present study was to evaluate an automated method for the retrospective detection of ADEs with hyperkalaemia during inpatient stays.MethodsWe used a set of complex detection rules to take account of the patient’s clinical and biological context and the chronological relationship between the causes and the expected outcome. The dataset consisted of 3,444 inpatient stays in a French general hospital. An automated review was performed for all data and the results were compared with those of an expert chart review. The complex detection rules’ analytical quality was evaluated for ADEs.ResultsIn terms of recall, 89.5% of ADEs with hyperkalaemia “with or without an abnormal symptom” were automatically identified (including all three serious ADEs). In terms of precision, 63.7% of the automatically identified ADEs with hyperkalaemia were true ADEs.ConclusionsThe use of context-sensitive rules appears to improve the automated detection of ADEs with hyperkalaemia. This type of tool may have an important role in pharmacoepidemiology via the routine analysis of large inter-hospital databases.

Randomized controlled trial on promoting influenza vaccination in general practice waiting rooms

Background Most of general practitioners (GPs) use advertising in their waiting rooms for patient’s education purposes. Patients vaccinated against seasonal influenza have been gradually lessening. The objective of this trial was to assess the effect of an advertising campaign for influenza vaccination using posters and pamphlets in GPs’ waiting rooms. Methods and findings Registry based 2/1 cluster randomized controlled trial, a cluster gathering the enlisted patients of 75 GPs aged over 16 years. The trial, run during the 2014–2015 influenza vaccination campaign, compared patient’s awareness from being in 50 GPs’ standard waiting rooms (control group) versus that of waiting in 25 rooms from GPs who had received and exposed pamphlets and one poster on influenza vaccine (intervention group), in addition to standard mandatory information. The main outcome was the number of vaccination units delivered in pharmacies. Data were extracted from the SIAM-ERASME claim database of the Health Insurance Fund of Lille-Douai (France). The association between the intervention (yes/no) and the main outcome was assessed through a generalized estimating equation. Seventy-five GPs enrolled 10,597 patients over 65 years or suffering from long lasting diseases (intervention/control as of 3781/6816 patients) from October 15, 2014 to February 28, 2015. No difference was found regarding the number of influenza vaccination units delivered (Relative Risk (RR) = 1.01; 95% Confidence interval: 0.97 to 1.05; p = 0.561). Conclusion Effects of the monothematic campaign promoting vaccination against influenza using a poster and pamphlets exposed in GPs’ waiting rooms could not be demonstrated.

Case-crossover study to examine the change in postpartum risk of pulmonary embolism over time

BackgroundAlthough the current guidelines recommend anticoagulation up until 6 weeks after delivery in women at high risk of venous thromboembolism (VTE), the risk of VTE may extend beyond 6 weeks. Our objective was to estimate the risk of a pulmonary embolism in successive 2-week intervals during the postpartum period.MethodsIn a population-based, case-crossover study, we analyzed the French national inpatient database from 2007 to 2013 (n = 5,517,680 singleton deliveries). Using ICD-10 codes, we identified women who were diagnosed with a postpartum pulmonary embolism between July 1st, 2008, and December 31st, 2013. Deliveries were identified during a case “period” immediately before the pulmonary embolism, and five different control periods one year before the pulmonary embolism. Using conditional logistic regression, Odds ratios (ORs) and 95% confidential intervals (CIs) were estimated for ten successive 2-week intervals that preceded the diagnosis of pulmonary embolism.ResultsWe identified 167,103 cases with a pulmonary embolism during the inclusion period. After delivery, the risk of pulmonary embolism declined progressively over time, with an OR [95%CI] of 17.2 [14.0–21.3] in postpartum weeks 1 to 2 and 1.9 [1.4–2.7] in postpartum weeks 11 to 12. The OR [95%CI] in postpartum weeks 13 to 14 was 1.4 [0.9–2.0], and the OR did not fall significantly after postpartum week 14.ConclusionsOur findings indicate that women are at risk of a pulmonary embolism up to 12 weeks after delivery. The shape of the risk curve suggests that the risk decreases exponentially over time. Future research is needed to establish whether the duration of postpartum anticoagulation should be extended beyond 6 weeks.

Inappropriate anticholinergic drugs prescriptions in older patients: analysing a hospital database

Background Although many anticholinergics are inappropriate in older patients, the prescription of these drugs in a hospital setting has not been extensively studied. Objective To describe prescriptions of anticholinergic drugs in terms of frequency, at risk situations and constipation in hospitalized, older adults. Setting Using a database from a French general hospital (period 2009–2013), we extracted information on 14,090 hospital stays by patients aged 75 and over. Methods Anticholinergic drug prescriptions were automatically detected, with a focus on prescriptions in three well-known at-risk situations: falls, dementia, and benign prostatic hyperplasia. Cases of constipation that might have been causally related to the administration of anticholinergic drugs were screened for and reviewed. Main outcome measure Prescriptions with a high associated risk of anticholinergic related adverse reactions. Results Administration of an anticholinergic drug was detected in 1412 (10.0%) of the hospital stays by older patients. At-risk situations were identified in 413 (36.5%) of these stays: 137 (9.7%) for falls, 243 (17.2%) for dementia, and 114 (8.1%) for benign prostatic hyperplasia; 78 (18.9%) of these 413 stays featured a combination of two or three at-risk situations. Cases of constipation induced by anticholinergic drug administration were identified in 188 (13.3%) patient stays by using validated adjudication rules for adverse drug reactions: 85 and 103 cases were respectively evaluated as “possible” or “probable” adverse drug reactions. Conclusions Anticholinergic drugs prescription was found in 10.0% of hospitalized, older patients. More than one third of these prescriptions occurred in at-risk situations and more than one in ten prescriptions induced constipation.

Hospital‐acquired hyperkalemia events in older patients are mostly due to avoidable, multifactorial, adverse drug reactions

Drug‐induced hyperkalemia is a frequent and severe complication in the hospital setting. Other risk factors may also induce hyperkalemia but the combination of drugs and precipitating factors has not been extensively studied. The aim was to identify drug‐induced hyperkalemia events in hospitalized older patients and to describe their combinations with precipitating factors. Two experts independently analyzed retrospective data of patients aged 75 years or more. Experts identified 471 hyperkalemia events and concluded that 379 (80.5%) were induced by drugs. The cause was multifactorial (i.e., at least one drug with a precipitating factor) in 300 (79.2%) of the 379 drug‐induced hyperkalemia. Most of the drug‐induced hyperkalemia events were avoidable (79.9%)—mainly because of the multifactorial cause (e.g., dosage adaptation during acute kidney injury). Drug‐induced hyperkalemia events are frequently combined with precipitating factors in hospitalized older patients and their prevention should focus on these combinations.

How to Compare the Length of Stay of Two Samples of Inpatients? A Simulation Study to Compare Type I and Type II Errors of 12 Statistical Tests☆

BACKGROUND Although many researchers in the field of health economics and quality of care compare the length of stay (LOS) in two inpatient samples, they often fail to check whether the sample meets the assumptions made by their chosen statistical test. In fact, LOS data show a highly right-skewed, discrete distribution in which most of the observations are tied; this violates the assumptions of most statistical tests. OBJECTIVES To estimate the type I and type II errors associated with the application of 12 different statistical tests to a series of LOS samples. METHODS The LOS distribution was extracted from an exhaustive French national database of inpatient stays. The type I error was estimated using 19 sample sizes and 1,000,000 simulations per sample. The type II error was estimated in three alternative scenarios. For each test, the type I and type II errors were plotted as a function of the sample size. RESULTS Gamma regression with log link, the log rank test, median regression, Poisson regression, and Weibull survival analysis presented an unacceptably high type I error. In contrast, the Student standard t test, linear regression with log link, and the Cox models had an acceptable type I error but low power. CONCLUSIONS When comparing the LOS for two balanced inpatient samples, the Student t test with logarithmic or rank transformation, the Wilcoxon test, and the Kruskal-Wallis test are the only methods with an acceptable type I error and high power.

The risks of pulmonary embolism and upper gastrointestinal bleeding beyond 35 days after total hip replacement for coxarthrosis among middle-aged patients: A cross-over cohort

Prophylactic anticoagulation is recommended up to 35days after total hip replacement (THR). Although several observational studies have assessed the incidence of thrombotic events or bleeding events after THR, the corresponding measures of association have never been studied concomitantly. Here, we evaluated the duration of the elevated risks (relative to the baseline risk) of both venous thromboembolic events and bleeding events after THR for coxarthrosis among middle-aged patients. This was a population-based, cross-over cohort study of data extracted from the French national inpatient database between 2007 and 2013. We included middle-aged patients (aged 45 to 69) having undergone THR for coxarthrosis. We compared the numbers of pulmonary embolisms (PEs) (respectively upper gastrointestinal bleedings (UGIBs)) following the THR with the numbers occurring during three unexposed periods one year later. This enabled us to estimate the odds ratio (OR) [95% confidence interval (CI)] for each of six successive 35-day intervals. The study included 108,099 patients. The ORs for PE were respectively 12.4 (95% CI, 8.6-17.8) (absolute risk difference rate per 100,000 (ARD/100,000)=130) and 5.0 (95% CI, 3.4-7.4) (ARD/100,000=52) for the first two 35-day intervals, and the risk was close to 1 thereafter. The risk of UGIB fell quickly, with an OR of 6.5 (95% CI, 4.6-9.1) (ARD/100,000=83) and 0.8 (95% CI, 0.4-1.6) for the first two 35-day intervals, respectively. The majority of UGIBs occurred during the inpatient stay for THR. Among middle-aged patients, the risk of a PE remains elevated beyond 35days after THR for coxarthrosis, whereas the risk of a UGIB remains elevated for the first 35days only.