French Health Insurance Database Research Articles

Évaluation des médicaments chez la femme enceinte : à propos de la base de données française EFEMERIS

The present article describes the first French database of drugs prescribed and dispensed during pregnancy and the outcome of these pregnancies, named EFEMERIS (Évaluation chez la Femme Enceinte des MÉdicaments et de leurs RISques). At the present time, EFEMERIS contains anonymous data concerning around 78,000 pregnant women who gave birth to a baby between 1 July 2004 to 31 December 2012 in Haute-Garonne (South West France) and who are registered in the French health insurance service. Data sources include 1- the French health insurance database (drugs prescribed during pregnancy), 2- the mother and child protection centre database (newborn health) 3- the antenatal diagnostic centre database (medical pregnancy interruptions) and 4- medical data from the hospital (PMSI). EFEMERIS provides exact data on period of exposure to drugs, pregnancy terminations, and follow up of the babies 9 months and 2 years after birth. It is incremented each year with around 10000 new pregnancies. Analysis of data until December 2011 shows a prevalence rate of congenital anomalies of 2.4%. Pregnant women were prescribed around 10 different reimbursed drugs during their pregnancy, the potential risk of a lot of these medications having not been evaluated during pregnancy. EFEMERIS can be used to monitor the prescription of reimbursed drugs to French pregnant women as well as to identify adverse pregnancy outcomes such as congenital malformation or effects on psychomotor development of the child. Examples of some studies already performed are given.

Healthcare cost impact of biological drugs compared with traditional systemic treatments in psoriasis: a cohort analysis in the French insurance database

Biological drugs have dramatically improved the management of moderate to severe psoriasis. Little is known about their economic impact in daily clinical practice. The aim of this study was to estimate the costs of biological drugs, in term of health resources consumption, and to compare it with costs induced by traditional systemic treatments. This cohort study was built from the French health insurance database in the Midi Pyrénées area (2.8million inhabitants, South West of France). We compared health care costs between 'exposed' patients treated with biological drugs (adalimumab, etanercept, infliximab or ustekinumab) and 'unexposed' patients defined as patients who received traditional systemic treatments (phototherapy, acitretin, methotrexate or cyclosporin) during a 6-month period. A total of 1924 patients met the inclusion criteria. Sixty-nine patients were 'exposed', whereas 1855 patients were 'unexposed'. 'Exposed' patients had a mean total healthcare cost of 8107€ vs. 1678€ (P<0.001) for 'unexposed' patients. They had higher costs concerning inpatient admission, medication and consultations including dermatology consultations, laboratory, non-medical care and transportation. Biological drug prescription was associated with an increase in the use of anti-infective drugs and with a reduction in the use of psychoactive drugs. The mean total health care expenditure in patients treated with biological drugs was five times higher as compared with patients treated with traditional systemic treatments. The limitation of the study is the short duration of follow-up comprising a loading dose period for some biological drugs. This may have contributed to an overestimation of drug-related costs.

Différences de taux de substitution des médicaments génériques en Midi-Pyrénées en fonction des classes pharmaco-thérapeutiques

BackgroundGeneric substitution has been permitted for several years in France and is promoted in order to reduce health expenditures. However, reluctance concerning use of generic drugs exists for different reasons: suspicions about their efficacy and/or safety, differences in content (excipients) and discussions about bioequivalency. The aim of our study was to determine whether or not the substitution ratio differs according to pharmacological classes used in primary care. MethodsWe conducted a descriptive study in the French Health Insurance Database using reimbursement data on drugs prescribed and delivered in the Midi-Pyrénées administrative district between March 2010 and March 2011. We selected different pharmacological classes largely used in primary care. For each class, a substitution ratio was calculated (DDD of generics delivered/DDD of brand name plus generics). A Chi2 test was used in order to detect a difference between these substitution ratios. ResultsThe global substitution ratio was 72.32%. Values varied from 28.36% for thyroid hormones to 90.39% for antibiotics, with significant statistical difference (P<0.001). The substitution ratio was less than 50% for thyroid hormones (28.36%) and antiepileptics (45.28%). Higher substitution ratios were observed for protein pump inhibitors (88.81%), statins (87.81%), antidepressors (87.37%) and diuretics (86.1%). ConclusionThis study highlights major differences in terms of the generic substitution ratio between different pharmacological classes. This difference can be explained in part by published guidelines. Further studies are needed to ascertain the precise point of view of patients, general practitioners and pharmacists concerning this issue.

Use of health insurance claim patterns to identify patients using nonsteroidal anti‐inflammatory drugs for rheumatoid arthritis

To determine healthcare claim patterns associated using nonsteroidal anti-inflammatory drugs (NSAIDs) for rheumatoid arthritis (RA). The CADEUS study randomly identified NSAID users within the French health insurance database. One-year claims data were extracted, and NSAID indication was obtained from prescribers. Logistic regression was used in a development sample to identify claim patterns predictive of RA and models applied to a validation sample. Analyses were stratified on the dispensation of immunosuppressive agents or specific antirheumatism treatment, and the area under the receiver operating characteristic curve was used to estimate discriminant power. NSAID indication was provided for 26,259 of the 45,217 patients included in the CADEUS cohort; it was RA for 956 patients. Two models were constructed using the development sample (n = 13,143), stratifying on the dispensation of an immunosuppressive agent or specific antirheumatism treatment. Discriminant power was high for both models (AUC > 0.80) and was not statistically different from that found when applied to the validation sample (n = 13,116). The models derived from this study may help to identify patients prescribed NSAIDs who are likely to have RA in claims databases without medical data such as treatment indication.

Prescription of drugs during pregnancy: a study using EFEMERIS, the new French database

Because of the limited data concerning drug risks in pregnancy, health professionals are often deprived of relevant and sufficient information related to prescribing or dispensing during pregnancy. However, previous studies have emphasised the widespread French prescription of several drugs (sometimes "typically French") which have not been assessed in pregnant women. The aim of the present study was to create the first French database of drugs prescribed and dispensed during pregnancy and the outcome of these pregnancies. This feasibility study concerns pregnant women who gave birth to a baby between 1 July 2004 to 30 June 2005 in Haute-Garonne and who are registered in the French Health Insurance Service. Data sources include (1) the French Health Insurance Database (drugs prescribed during pregnancy), (2) the Mother and Child Protection Centre Database (newborn health at birth and 9 months after) and (3) the Antenatal Diagnostic Centre Database (medical pregnancy interruptions). The database is composed of 10,174 "mother-outcome" pairs. The prevalence rate of congenital anomalies was 2.2%. Pregnant women were prescribed 11.3 +/- 8.2 different drugs. Among the 20 most frequently prescribed drugs, around half of them have not been evaluated in pregnant women. The first results of this study show that implementation of a French database on prescription of drugs and pregnancy outcomes is feasible. Compared with several databases available in other countries, EFEMERIS provides exact data on period of exposure to drugs, pregnancy terminations, and follow up of the baby 9 months after birth. Recording these data would make it possible to assess the risk of malformations due to a greater number of drugs and would contribute to international drug evaluation studies.

Short-term effect of pollen exposure on antiallergic drug consumption

Several studies have investigated the association between pollen exposure and asthma emergency admissions, but only 2 have investigated the effect of airborne allergens on consultations for rhinitis or conjunctivitis and none has used drug consumption as the health indicator. To analyze the short-term association between pollen exposure and antiallergic drug consumption in the urban area of Clermont-Ferrand, France, taking into account the potentially confounding effect of air pollution and meteorological factors. We used the French health insurance database to select all individuals from the Clermont-Ferrand urban area having benefited from reimbursement for antiallergic treatment from January 1, 2000, through December 31, 2001, and from January 1, 2003, through December 31, 2004. An episode of treated allergic rhinitis, rhinosinusitus, or conjunctivitis (ARC) was defined as the association of an oral antihistamine and a local antiallergic drug on the same prescription. The relations between daily changes in pollen concentrations and daily changes in the number of treated ARC cases were analyzed using a Poisson regression model with penalized spline functions. The risk of treated ARC associated with an interquartile increase in pollen concentration increased significantly for Poaceae (5%, P < .001), Fraxinus (7%, P < .001), Betula (7%, P < .001), and Corylus (2%, P < .02). This increase was significant in all age groups for Poaceae and Fraxinus pollen and in people younger than 65 years for Betula pollen. The effect was mainly concentrated on the present day, except for Poaceae pollens, for which the risk remained significantly (P < .001) increased until 3 days lag time. This study showed a significant increase in treated ARC cases related to Poaceae, Fraxinus, and Betula. Specific risks are difficult to evaluate for species that share the same pollination period. Time-series studies based on drug consumption are useful to highlight and to supervise pollen-related diseases requiring ambulatory care.

Digest

Digest

Utilisation des bases de données de l’Assurance Maladie par les Centres d’Evaluation et d’Information sur la Pharmacodépendance : exemples, intérêts et perspectives

In France, the observation and evaluation of drug abuse and dependence associated with pychoactive medications are the responsibility of the National Commission for Narcotics and Psychotropic Drugs. In order to assist this commission, several centres for evaluation and information on pharmacodependence (CEIP) were created throughout in France. Recently, in order to complete their epidemiological tools, several centres have developed another pharmacoepidemiological approach using data for refunded prescriptions obtained from the local and regional French Health Insurance database. This article underlines the potential contribution of the Health Insurance database to improving knowledge of drug use in the real-life conditions based on studies performed by the CEIP. Several examples are given showing the extent of the possibilities (population-based studies, cohort studies, development of misuse indicators). In spite of their limitations (e.g. the difference between consumption and delivery), these examples confirm that these database may be a novel tool for CEIP to assess a potential abuse of a medication.