What place for intelligent automation and artificial intelligence to preserve and strengthen vigilance expertise in the face of increasing declarations?

In 2018, the "Ateliers de Giens" (Giens Workshops) devoted a workshop to artificial intelligence (AI) and led its experts to confirm the potential contribution and theoretical benefit of AI in clinical research, pharmacovigilance, and in improving the efficiency of care. The 2022 workshop is a continuation of this reflection on AI and intelligent automation (IA) by focusing on its contribution to pharmacovigilance and the applications and tasks could be optimized to preserve and strengthen medical and pharmacological expertise in pharmacovigilance. The evolution of pharmacovigilance work is characterized by many tasks with low added value, a growing volume of pharmacovigilance reporting of suspected side effects, and a scarcity of medical staff with expertise in clinical pharmacology and pharmacovigilance and human resources to support this growing need. Together, these parameters contribute to an embolization of the pharmacovigilance system at risk of missing its primary mission: to identify and characterize a risk or even a health alert on a drug. The participants of the workshop (representatives of the Regional Pharmacovigilance Centres (CRPV), the French National Agency for Safety of Medicinal Products (ANSM), patients, the pharmaceutical industry, or start-ups working in the development of AI in the field of medicine) shared their experiences, their pilot projects and their expectations on the expected potential, theoretical or proven, AI and IA. This work has made it possible to identify the needs and challenges that AI or IA represent, in the current or future modes of organization of pharmacovigilance activities. This approach led to the development of a SWOT matrix (strengths, weaknesses, opportunities, threats), a basis for reflection to identify critical points and consider four main recommendations: (1) preserve and develop business expertise in pharmacovigilance (including research and development in methods) with the integration of new technologies; (2) improve the quality of pharmacovigilance reports; (3) adapt technical and regulatory means; (4) implement a development strategy for AI and IA tools at the service of expertise.

Population Pharmacokinetic Study of the Suitability of Standard Dosing Regimens of Amikacin in Critically Ill Patients with Open-Abdomen and Negative-Pressure Wound Therapy.

The aim was to assess the appropriateness of recommended regimens for empirical MIC coverage in critically ill patients with open-abdomen and negative-pressure therapy (OA/NPT). Over a 5-year period, every critically ill patient who received amikacin and who underwent therapeutic drug monitoring (TDM) while being treated by OA/NPT was retrospectively included. A population pharmacokinetic (PK) modeling was performed considering the effect of 10 covariates (age, sex, total body weight [TBW], adapted body weight [ABW], body surface area [BSA], modified sepsis-related organ failure assessment [SOFA] score, vasopressor use, creatinine clearance [CLCR], fluid balance, and amount of fluids collected by the NPT over the sampling day) in patients who underwent continuous renal replacement therapy (CRRT) or did not receive CRRT. Monte Carlo simulations were employed to determine the fractional target attainment (FTA) for the PK/pharmacodynamic [PD] targets (maximum concentration of drug [Cmax]/MIC ratio of ≥8 and a ratio of the area under the concentration-time curve from 0 to 24 h [AUC0-24]/MIC of ≥75). Seventy critically ill patients treated by OA/NPT (contributing 179 concentration values) were included. Amikacin PK concentrations were best described by a two-compartment model with linear elimination and proportional residual error, with CLCR and ABW as significant covariates for volume of distribution (V) and CLCR for CL. The reported V) in non-CRRT and CRRT patients was 35.8 and 40.2 liters, respectively. In Monte Carlo simulations, ABW-adjusted doses between 25 and 35 mg/kg were needed to reach an FTA of >85% for various renal functions. Despite an increased V and a wide interindividual variability, desirable PK/PD targets may be achieved using an ABW-based loading dose of 25 to 30 mg/kg. When less susceptible pathogens are targeted, higher dosing regimens are probably needed in patients with augmented renal clearance (ARC). Further studies are needed to assess the effect of OA/NPT on the PK parameters of antimicrobial agents.

Study protocol for the assessment of nurses internal contamination by antineoplastic drugs in hospital centres: a cross-sectional multicentre descriptive study.

INTRODUCTION: Antineoplastic drugs (AD) are potentially carcinogenic and/or reprotoxic molecules. Healthcare professionals are increasingly exposed to these drugs and can be potentially contaminated by them. Internal contamination of professionals is a key concern for occupational physicians in the assessment and management of occupational risks in healthcare settings. Objectives of this study are to report AD internal contamination rate in nursing staff and to identify factors associated with internal contamination. METHODS AND ANALYSIS: This trial will be conducted in two French hospital centres: University Hospital of Bordeaux and IUCT-Oncopole of Toulouse. The target population is nurses practicing in one of the fifteen selected care departments where at least one of the five studied AD is handled (5-fluorouracil, cyclophosphamide, doxorubicin, ifosfamide, methotrexate). The trial will be conducted with the following steps: (1) development of analytical methods to quantify AD urine biomarkers, (2) study of the workplace and organization around AD in each care department (transport and handling, professional practices, personal and collective protection equipments available) (3) development of a self-questionnaire detailing professional activities during the day of inclusion, (4) nurses inclusion (urine samples and self-questionnaire collection), (5) urine assays, (6) data analysis. ETHICS AND DISSEMINATION: The study protocol has been approved by the French Advisory Committee on the Treatment of Information in Health Research (CCTIRS) and by the French Data Protection Authority (CNIL). Following the opinion of the Regional Committee for the Protection of Persons, this study is outside the scope of the provisions governing biomedical research and routine care (n°2014/87). The results will be submitted to peer-reviewed journals and reported at suitable national and international meetings. TRIAL REGISTRATION NUMBER: NCT03137641.

National trends in use and costs of oral anticancer drugs in France: An 8-year population-based study.

PURPOSE: During the last decade, many oral anticancer drugs (OAcDs) have been marketed, providing interesting but potentially costly pharmaceutical alternatives to intravenous treatments. This study aims to provide updated information on their use and costs. METHODS: A cross-sectional yearly repeated study was conducted from 2006 to 2014 using the representative sample of the French national health care insurance system claims database (EGB). OAcD use was described for each year, among prevalent (ie, patients with at least 1 OAcD reimbursement) and incident users (ie, patients with no OAcD reimbursement within the prior year) and according to their pharmacological classes (Hormone Therapy [HT], Cytotoxic Therapy [CT], Targeted Therapy [TT], and others). Demographic characteristics were described for both users; comorbidities and direct medical costs were described for incident users only. RESULTS: The yearly prevalence and incidence of OAcD use, mainly represented by HT, remained stable from 2006 to 2014 (1.2%; 0.4%). Compared with users of other OAcD classes, the proportion of TT users substantially increased over the 8-year study period (+9.3%), and TT incident users had more severe comorbidities at treatment initiation. The health expenditures were the most important in TT users with median monthly medical direct costs varying from 2995€ to 4968€ per patient between 2006 and 2014. CONCLUSION: With the development of new OAcDs, the TTs use reaches a wider population of patients but is responsible for increasing health expenditures.

Chronic obstructive pulmonary disease exacerbation and inhaler device handling: real-life assessment of 2935 patients.

Acute exacerbations of chronic obstructive pulmonary disease (COPD) can be prevented by inhaled treatment. Errors in inhaler handling, not taken into account in clinical trials, could impact drug delivery and minimise treatment benefit. We aimed to assess real-life inhaler device handling in COPD patients and its association with COPD exacerbations.To this end, 212 general practitioners and 50 pulmonologists assessed the handling of 3393 devices used for continuous treatment of COPD in 2935 patients. Handling errors were observed in over 50% of handlings, regardless of the device used. Critical errors compromising drug delivery were respectively made in 15.4%, 21.2%, 29.3%, 43.8%, 46.9% and 32.1% of inhalation assessment tests with Breezhaler® (n=876), Diskus® (n=452), Handihaler® (n=598), pressurised metered-dose inhaler (pMDI) (n=422), Respimat® (n=625) and Turbuhaler® (n=420).The proportion of patients requiring hospitalisation or emergency room visits in the past 3 months for severe COPD exacerbation was 3.3% (95% CI 2.0-4.5) in the absence of error and 6.9% (95% CI 5.3-8.5) in the presence of critical error (OR 1.86, 95% CI 1.14-3.04, p<0.05).Handling errors of inhaler devices are underestimated in real life and are associated with an increased rate of severe COPD exacerbation. Training in inhaler use is an integral part of COPD management.

Cost Effectiveness of the Long-Acting ß2-Adrenergic Agonist (LABA)/Long-Acting Muscarinic Antagonist Dual Bronchodilator Indacaterol/Glycopyrronium Versus the LABA/Inhaled Corticosteroid Combination Salmeterol/Fluticasone in Patients with Chronic Obstructive Pulmonary Disease: Analyses Conducted for Canada, France, Italy, and Portugal.

OBJECTIVE: The objective of this study was to assess the cost effectiveness of the dual bronchodilator indacaterol/glycopyrronium (IND/GLY) compared with salmeterol/fluticasone combination (SFC) in patients with moderate-to-severe chronic obstructive pulmonary disease (COPD) who had a history of one or no exacerbations in the previous year, in Canada, France, Italy, and Portugal. METHODS: A patient-level simulation was developed to compare the costs and outcomes of IND/GLY versus SFC based on data from the LANTERN trial (NCT01709903). Monte-Carlo simulation methods were employed to follow individual patients over various time horizons. Population and efficacy inputs were derived from the LANTERN trial. Considering the payers' perspective, only direct costs were included. Costs and health outcomes were discounted annually at 3.0 % for all countries. Unit costs were taken from publically available sources with all costs converted to euros (€). The cost base year was 2015. Deterministic and probabilistic sensitivity analyses were undertaken to test the robustness of the model results. RESULTS: IND/GLY was found to be the dominant (more effective and less costly) treatment option compared with SFC in all four countries. The use of IND/GLY was associated with mean total cost savings per patient over a lifetime of €6202, €1974, €1611, and €220 in Canada, France, Italy, and Portugal, respectively. Sensitivity analysis showed that exacerbation rates had the largest impact on incremental costs and quality-adjusted life-years (QALYs). The probability of IND/GLY being cost effective was estimated to be >95 % for thresholds above €5000/QALY. CONCLUSION: In patients with moderate to severe COPD, IND/GLY is likely to be a cost-effective treatment alternative compared with SFC.

Information and communication on risks related to medications and proper use of medications for healthcare professionals and the general public: precautionary principle, risk management, communication during and in the absence of crisis situations.

Recent drug crises have highlighted the complexity, benefits and risks of medication communication. The difficulty of this communication is due to the diversity of the sources of information and the target audience, the credibility of spokespersons, the difficulty to communicate on scientific uncertainties and the precautionary principle, which is influenced by variable perceptions and tolerances of the risk. Globally, there is a lack of training in risk management with a tendency of modern society to refuse even the slightest risk. Communication on medications is subject to regulatory or legal requirements, often uses tools and messages that are not adapted to the target audience and is often based on a poor knowledge of communication techniques. In order to improve this situation, the available information must be coordinated by reinforcing the unique medication information website and by coordinating communication between authorities by means of a single spokesperson. A particular effort must be made in the field of training in the proper use and risk of medications for both the general population and patients but also for healthcare professionals, by setting up a unified academic on-line teaching platform for continuing medical education on medications and their proper use.

Switching from branded to generic inhaled medications: potential impact on asthma and COPD.

Pressure on healthcare budgets is increasing, while at the same time patent protection for many branded inhaled medications has expired, leading to the development and growing availability of generic inhaled medicines. Generic inhaled drugs are therapeutically equivalent to original branded options but may differ in their formulation and inhalation device. This new situation raises questions about the potential impact of switching from branded to generic drug/inhaler combination products in patients with asthma or COPD, with or without their consent, in countries where this is permitted. Inhalation devices, particularly dry powder inhalers, vary markedly in their design, method of operation and drug delivery to the lungs. Current guidelines stress the importance of training patients how to use their inhalers but offer little or no guidance on how this should be achieved. Non-adherence to therapy and incorrect inhaler usage are recognised as major factors in poorly or uncontrolled asthma and COPD and switching patients to a different inhaler device may exacerbate these problems, particularly in patients who disagree to switch. Where switching is permitted or mandatory, adequate patient instruction and follow-up monitoring should be provided routinely.

Use of principal component analysis in the evaluation of adherence to statin treatment: a method to determine a potential target population for public health intervention.

The prevalence of statin use is high but adherence low. For public health intervention to be rational, subpopulations of nonadherent subjects must be defined. To categorise statin users with respect to patterns of reimbursement, this study was performed using the main French health reimbursement database for the Aquitaine region of south-western France. The cohort included subjects who submitted a reimbursement for at least one delivery of a statin (index) during the inclusion period (1st of September 2004-31st of December 2004). Indicators of adherence from reimbursement data were considered for principal component analysis. The 119,570 subjects included and analysed had a sex ratio of 1.1, mean (SD) age of 65.9 (11.9), and 13% were considered incident statin users. Principal component analysis found three dimensions that explained 67% of the variance. Using a K-means classification combined with a hierarchical ascendant classification, six groups were characterised. One group was considered nonadherent (10% of study population) and one group least adherent (1%). This novel application of principal component analysis identified groups that may be potential targets for intervention. The least adherent group appears to be one of the most appropriate because of both its relatively small size for case review with prescribing physicians and its very poor adherence.

Assessing cancer drugs for reimbursement: methodology, relationship between effect size and medical need.

Reimbursement is assessed by the Transparency Commission from the Health Authority (HAS) using a medical benefit (SMR) score that gives access to reimbursement, an "improvement of medical service rendered" (ASMR) that determines the added therapeutic value, and the target population. Assessing cancer drugs for reimbursement raises the same issues as other therapeutic classes, with some key differences. Overall survival (OS) is considered by the Transparency Commission as the endpoint for assessing clinical benefit, and yet it is not an applicable primary endpoint in all types of cancer. Later lines of treatment, particularly during the development process, may make it difficult to interpret OS as the primary endpoint. Therefore, progression-free survival (PFS) for metastatic situations and disease-free survival (DFS) in adjuvant situations are wholly relevant endpoints for decisions on the reimbursement of a new cancer drug. Effect size is assessed using actuarial survival curves of the product versus the comparator, and it is difficult to summarise them into one single parameter. Results are generally interpreted based on median survival, which is fragmented because it only measures one point of the curve. The hazard ratio measures the effect of treatment throughout the duration of survival and is therefore more comprehensive in quantifying clinical benefit. Determining an effect size threshold for granting reimbursement is difficult given the diversity of cancer settings and the level of medical need, which influences assessment of the clinical relevance of the observed difference. Rapid progress in comparators (700 molecules in development) and the identification of predictive factors of efficacy (biomarkers, histology, etc.) during development may lead to different ASMR scores per population, or to the restriction of the target population to a subgroup of the marketing authorisation (MA) population in which the expected effect size is greater. To address these issues, the roundtable recommends the possibility of early scientific opinions by the office of the Transparency Commission in order to discuss comparators and the relevance of responder subgroups. It also recommends the possibility of granting a temporary ASMR, on condition of subsequent confirmation by production of data, when reimbursement appears justified in a subpopulation of the MA for which only subgroup analysis is available.

How reimbursement databases can be used to support drug utilisation studies: example using the main French national health insurance system database.

INTRODUCTION: Reimbursement databases are potentially invaluable tools to develop and conduct pharmacoepidemiological studies on drug use. However, two types of factors that may influence the performance of a database can be distinguished: firstly, factors related to the constitution of the database, and secondly, factors related to the data. For the latter, we think that two are important: the presence of the drug in the database and the capacity to capture real-life use, both of which are influenced by the marketing status (e.g., OTC) and whether reimbursement is possible. OBJECTIVES: To illustrate and discuss to what extent reimbursement databases are relevant tools to conduct drug utilization studies with regard to the data on drugs. METHODS: In order to illustrate the reliability of data in reimbursement databases, data from the main French national health insurance database (55 million individuals) were compared to national drug sales in France (units) during the same year. RESULTS: Depending on the ATC class, the capture in the database of drugs actually sold ranged from 32 to 81%. DISCUSSION: Capture of classes of drugs in the database may be explained by the specific characteristics of the French health insurance system (reimbursable drugs, OTC market share). These characteristics influence the studies that can be performed both in terms of the topic but also the methodology. This problem probably exists for the other reimbursement databases used worldwide. CONCLUSION: Studies should be designed according to the strengths and weaknesses of reimbursement databases that were not originally developed for pharmacoepidemiology.

Imatinib assay by HPLC with photodiode-array UV detection in plasma from patients with chronic myeloid leukemia: Comparison with LC-MS/MS.

BACKGROUND: Imatinib, a competitive inhibitor of BCR-ABL tyrosine kinase, is now the first-line treatment for chronic myelogenous leukemia (CML). Therapeutic drug monitoring targeting trough plasma levels of about 1000ng/mL may help to optimize the therapeutic effect. METHODS: We developed a high-performance liquid chromatography (HPLC) method with UV/Diode Array Detection (DAD) for trough imatinib concentration determination in human plasma. Imatinib trough levels were measured in plasma from 65 CML patients using our method and LC-MS/MS as the reference method. Results with these two methods were compared using Deming regression, chi-square test, and sign test. RESULTS: The calibration curve was prepared in blank human plasma. HPLC-UV/DAD calibration curves were linear from 80 to 4000ng/mL, and the limit of quantification was set at 80ng/mL. The between-day variation was 6.1% with greater than 96% recovery after direct plasma deproteinization and greater than 98% recovery from the column. No significant differences in imatinib plasma levels were found between HPLC-UV/DAD and LC-MS/MS. CONCLUSIONS: This HPLC-UV/DAD method was sufficiently specific and sensitive for imatinib TDM, with no evidence of interference. Our rapid inexpensive HPLC-UV/DAD method that requires only widely available equipment performs well for plasma imatinib assays.

Medical and non-medical direct costs of chronic low back pain in patients consulting primary care physicians in France.

A retrospective, observational, cohort study in primary care. To determine the total direct medical and non-medical cost of chronic low back pain (LBP) in France and its associated factors. Chronic LBP affects 5-10% of the population its burden in France is unknown. Ninety-eight randomly selected general practitioners included 796 adult patients with chronic LBP between October 2001 and December 2002. Direct costs due to physician visits, investigations, medications, hospitalizations, and other medical and non-medical resource use were collected for the 6 months prior to study visit. Costs both reimbursed and not by the French health insurance system were considered. Quality of life (QoL) and disease severity were measured using Short Form (SF)-8 and Roland-Morris disability questionnaire (RMDQ), respectively. Costs were updated to represent 2007 prices. Men represented 50.6% of the 796 patients, mean age was 53 +/- 11.3 years, and the duration of LBP was more than 1 year in 80.9% of patients. The total mean cost per patient over six months was 715.6 euro (95% CI: 644.2-797.8). Of these costs, 22.9% related to care provided by physiotherapists and allied specialists, 19.5% to medications, 17.4% to hospitalizations, 9.6% to investigations, and 12.5% to physician fees. In multivariate analysis, the factors associated with the cost of chronic LBP were disease severity (RMDQ score) and age of the patients. LBP is a disease that is both common and costly.

Adherence to anti-inflammatory treatment for asthma in clinical practice in France.

OBJECTIVES: This study evaluated adherence to anti-inflammatory controller medication for asthma in a French population. METHODS: This was an observational cohort study that employed data from the health insurance database for the Aquitaine region of France. Eligible subjects were aged between 15 and 45 years and had > or = 1 reimbursement claim for anti-inflammatory controller medication between December 1, 2003, and December 31, 2003. Data were collected from June 1, 2003, to May 28, 2005. New users were defined as those having no reimbursement claim for any asthma controller medication in the 6 months preceding the index date; all others were considered previous users. All subjects were followed for 17 months. Continuous multiple-interval measures of medication availability and medication gaps, treatment persistence, and time to first renewal were calculated for users of inhaled corticosteroids (ICS) alone, users of ICS combined with a long-acting beta2-agonist (LABA) in a single inhaler (ICS/LABA) or in separate inhalers (ICS + LABA), and users of oral leukotriene antagonists. RESULTS: The study population contained 12,502 new and previous users of anti-inflammatory controller medication for asthma. Their mean (SD) age was 32.0 (8.9) years and 58.3% were female. Previous users had better adherence to treatment than new users with respect to all measures. Adherence was best in previous users of leukotriene antagonists. However, only 40.2% of previous users of leukotriene antagonists had sufficient medication (95% CI, 37.6-42.9), compared with 56.8% of previous users of ICS + LABA (95% CI, 50.7-62.8), 65.5% of previous users of ICS only (95% CI, 63.1-67.9), and 69.8% of previous users of ICS/LABA (95% CI, 68.2-71.4). The rate of persistence over a 1-year period was 44.3% for previous users of leukotriene antagonists (95% CI, 41.7-46.8), compared with 27.1% for previous users of ICS only (95% CI, 25.1-29.1), 32.1% for previous users of ICS/LABA (95% CI, 30.5-33.6), and 44.1% for previous users of ICS + LABA (95% CI, 38.2-50.0). CONCLUSION: Adherence to treatment was poor (< or = 44%) in these users of anti-inflammatory controller medications for asthma.

Inter-expert agreement of seven criteria in causality assessment of adverse drug reactions.

AIMS: To evaluate agreement between five senior experts when assessing seven causality criteria and the probability of drug causation. METHODS: A sample of 31 adverse event-drug pairs was constituted. For each pair, five experts separately assessed (i) the probability of drug causation, which was secondarily divided into seven causality levels: ruled out (0-0.05), unlikely (0.06-0.25), doubtful (0.26-0.45), indeterminate (0.46-0.55), plausible (0.56-0.75), likely (0.76-0.95), and certain (0.96-1); and (ii) seven causality criteria. To test discrepancies between experts, the kappa index was used. RESULTS: The agreement of the five experts was very poor (kappa = 0.05) for the probability of drug causation. Among the seven levels of causality, only 'doubtful' showed a significant rate of agreement (kappa = 0.32, P < 0.001). For all criteria, the kappa index was significant except for the item 'risk(s) factor(s)' (kappa = 0.09). Agreement between experts was good (0.64, P < 0.001) only for the criterion 'reaction at site of application or toxic plasma concentration of the drug or validated test'. However, the rate of agreement with kappa indices of the causality criteria ranged from 0.12 to 0.38. CONCLUSIONS: This study confirms that in the absence of an operational procedure, agreement between experts is low. This should be considered when designing a causality assessment method. In particular, criteria inducing a low level of agreement should have their weight reduced.

Agreement of expert judgment in causality assessment of adverse drug reactions.

BACKGROUND: Global introspection is, with operational algorithms and Bayes' theorem, one of the three main approaches used to assess the causal relationship between a drug treatment and the occurrence of an adverse event. OBJECTIVE: To analyze and compare the judgments of five senior experts using global introspection about drug causation on a random set of putative adverse drug reactions. METHODS: A random sample of 150 drug-effect pairs was constituted. For each pair, five senior experts had to independently assess the probability of drug causation from 0 to 1 by using a 100 mm visual analog scale (VAS). For analysis, those probabilities were secondarily split into seven levels of causality: excluded (0-0.05); unlikely (0.06-0.25); doubtful (0.26-0.45); unassessable/unclassifiable (0.46-0.55); plausible (0.56-0.75); likely (0.75-0.95); and certain (0.95-1). Agreement among the five experts was assessed using kappa coefficients (kappa). RESULTS: The overall agreement between experts was poor (kappa=0.20), although significantly different from chance, and varied according to the level of causality. It was lower for the unlikely, doubtful, unassessable/unclassifiable, and plausible categories (kappa=0.03, 0.03, -0.01, and 0.13, respectively) than for VAS extremes: excluded, likely, and certain (kappa=0.40, 0.32, and 0.30, respectively). CONCLUSION: This study confirms that experts express marked disagreements when assessing drug causality independently. The agreement rate was lower for intermediate levels of causality, especially when strong evidence was lacking for confirming or ruling out drug causality. Therefore, in a decision-making context, a step-by-step consensual approach such as the Delphi method seems necessary to make the assessment of such cases more reliable.