Value of a national administrative database to guide public decisions: From the système national d’information interrégimes de l’Assurance Maladie (SNIIRAM) to the système national des données de santé (SNDS) in France

Predialysis Care Trajectories of Patients With ESKD Starting Dialysis in Emergency in France

Introduction

Emergency dialysis start (EDS) is frequent for patients with chronic kidney disease (CKD). To improve CKD management, new trajectory-based care policies are currently being introduced both in France and in the United States. This study describes the different types of predialysis care trajectories and factors associated with EDS.

Methods

Adults patients who started dialysis in France in 2015 were included. Individual clinical and health care consumption data were retrieved from the French national end-stage kidney disease (ESKD) registry (Renal Epidemiology and Information Network [REIN]) and the French National Health Data system (SNDS), respectively. Hierarchical Clustering on Principal Component was used to identify groups of patients with the same health care consumption profile during the 2 years before dialysis start. Logistic regression analysis was used to identify factors associated with EDS.

Results

Among the 8856 patients included in the analysis, 2681 (30.3%) had EDS. The Hierarchical Clustering on Principal Component identified six types of predialysis care trajectories in which EDS rate ranged from 13.8% to 61.8%. After adjustment for the patients’ characteristics, less frequent or lack of follow-up with a nephrologist was associated with higher risk of EDS (odds ratio [OR]: 1.32; 95% confidence interval [CI]: 1.17–1.50 and OR: 1.83; 95% CI: 1.58–2.12), but not follow-up with a general practitioner.

Conclusions

The care trajectories during the 2 years before dialysis start were heterogeneous and patients with a lesser or lack of follow-up with a nephrologist were more likely to start dialysis in emergency, regardless of the frequency of follow-up by a general practitioner (GP). New CKD policies should include actions to strengthen CKD screening and referral to nephrologists.

Nivolumab treatment in advanced non-small cell lung cancer: real-world long-term outcomes within overall and special populations (the UNIVOC study)

OBJECTIVE

To describe long-term outcomes of patients treated with nivolumab for advanced non-small cell lung cancer (aNSCLC) in everyday clinical practice in France, with a focus on patients aged ⩾80 years, patients with renal impairment and patients with brain metastases.

METHODS

The study included all patients with aNSCLC recorded in the French national hospital database, starting nivolumab in 2015-2016 and followed until December 2018. Patients were stratified by age, the presence of renal impairment and brain metastasis, as documented in the hospital discharge summaries. Information was retrieved on demographics, comorbidities and treatment history at baseline. Time to discontinuation of nivolumab treatment and overall survival were estimated using Kaplan-Meier survival analysis.

RESULTS

Overall, 10,452 patients were included, of whom 514 were octogenarians, 479 had renal impairment and 1800 had brain metastases at baseline. Median duration of nivolumab treatment was 2.8 months in the overall population and in both the octogenarian and renally impaired subgroups, and 2.3 months in patients with brain metastases. Median overall survival in these patient groups was 11.7 months (95% confidence interval: 11.3-12.2), 11.7 months (11.3-12.1), 11.7 months (11.3-12.2) and 9.9 months (9.0-10.9) respectively. Three-year overall survival rates were 19.1% (18.1-20.2) in the overall population, 16.5% (11.6-23.4) in octogenarians, 15.9% (11.8-21.4) in patients with renal impairment and 21.7% (19.4-24.2) in those with brain metastases.

CONCLUSION

This large nationwide retrospective real-life cohort provided narrow estimates of long-term overall survival, which reached 19% at 3 years, consistent with data from phase III trials of nivolumab. Survival rates were comparable in the three special populations of interest and the overall population.

Risk of early neurodevelopmental disorders associated with in utero exposure to valproate and other antiepileptic drugs: a nationwide cohort study in France

Information available on the risks of neurodevelopmental disorders (NDs) associated with in utero exposure to valproate (VPA) and to other antiepileptic drugs (AEDs) is limited. A nationwide population-based cohort study was conducted based on comprehensive data of the French National Health Data System (SNDS). Liveborn infants without brain malformation, born between January 2011 and December 2014, were followed from birth up to December 2016. NDs were identified based on diagnoses of mental or behavioural disorders and utilization of speech therapy, orthoptic or psychiatric services. The risk of NDs was compared between children exposed in utero to AED monotherapy and unexposed children, using Cox proportional hazard models adjusted for maternal and neonatal characteristics. The cohort included 1,721,990 children, 8848 of whom were exposed in utero to AED monotherapy. During a mean follow-up of 3.6 years, 15,458 children had a diagnosis of mental or behavioural disorder. In utero exposure to VPA was associated with an increased risk of NDs overall (aHR: 3.7; 95% CI 2.8–4.9) and among children born to a mother without mental illness (aHR 5.1; 95% CI 3.6–7.3). A dose–response relationship was demonstrated and the risk of NDs was more particularly increased for an exposure to VPA during the second or third trimesters of pregnancy. Among the other AEDs, only pregabalin was consistently associated with an increased risk of NDs (aHR: 1.5; 95% CI 1.0–2.1). This study confirms a four to fivefold increased risk of early NDs associated with exposure to VPA during pregnancy. The risk associated with other AEDs appears much lower.

Cause-Specific Survival After Meningioma Surgery: A Nationwide Population-Based Competing Risk Study

BACKGROUND

Survival after meningioma surgery often is reported with inadequate allowance for competing causes of death.

METHODS

We processed the French Système National des Données de Santé database using an algorithm combining the type of surgical procedure and the International Classification of Diseases to retrieve appropriate cases of meningiomas. The cumulative incidence of meningioma-related death was the primary end point. A competing risk analysis was performed to identify factors associated with meningioma-specific death of patients who underwent meningioma surgery.

RESULTS

The risk of meningioma-related death at 1, 2, and 3 years respectively was 2.4%, 95% confidence interval [CI] 2-2.7; 3%, 95% CI 2.6-3.4; and 3.1%, 95% CI 2.7-3.6. In the adjusted Fine-Gray competing risk regression for meningioma cause-specific survival, age at surgery (subdistribution hazard ratio [SHR] 1.07, 95% CI 1.05-1.09, P < 0.001), mortality-related morbidity index (SHR 1.68, 95% CI 1.07-2.63, P = 0.025), expenditure-related morbidity index (SHR 1.06, 95% CI 1.03-1.09, P < 0.001), spinal location (SHR 0.2, 95% CI 0.08-0.47, P < 0.001), cerebrospinal fluid shunt (SHR 3.13, 95% CI 1.9-5.16, P < 0.001), grade (SHR 1.88, 95% CI 1.13-3.14, P = 0.015) redo surgery for recurrence (SHR 1.6, 95% CI 1.01-2.51, P = 0.043), and progressing meningioma (SHR 2.87, 95% CI 1.23-6.68, P = 0.015) were established as independent prognostic factors of meningioma-related death.

CONCLUSIONS

Cause-specific survival after meningioma surgery is greater in younger, low-comorbidity adults with spinal and benign meningioma. Those with an intracranial, progressing malignant tumor requiring cerebrospinal fluid shunting and having a severe global health-state have a significant increased risk of meningioma-related death. Redo surgery failed to improve the outcome. We recommend the use of competing risk model in meningioma studies in which unrelated mortality may be substantial, as this approach results in more accurate estimates of disease risk and associated predictors.

French hepatitis C care cascade: substantial impact of direct-acting antivirals, but the road to elimination is still long

BACKGROUND

Hepatitis C virus (HCV) elimination by 2030, as targeted by the World Health Organization (WHO), requires that 90% of people with chronic infection be diagnosed and 80% treated. We estimated the cascade of care (CoC) for chronic HCV infection in mainland France in 2011 and 2016, before and after the introduction of direct-acting antivirals (DAAs).

METHODS

The numbers of people (1) with chronic HCV infection, (2) aware of their infection, (3) receiving care for HCV and (4) on antiviral treatment, were estimated for 2011 and 2016. Estimates for 1) and 2) were based on modelling studies for 2011 and on a virological sub-study nested in a national cross-sectional survey among the general population for 2016. Estimates for 3) and 4) were made using the National Health Data System.

RESULTS

Between 2011 and 2016, the number of people with chronic HCV infection decreased by 31%, from 192,700 (95% Credibility interval: 150,900-246,100) to 133,500 (95% Confidence interval: 56,900-312,600). The proportion of people aware of their infection rose from 57.7 to 80.6%. The number of people receiving care for HCV increased by 22.5% (representing 25.7% of those infected in 2016), while the number of people on treatment increased by 24.6% (representing 12.1% of those infected in 2016).

CONCLUSIONS

This study suggests that DAAs substantially impact CoC. However, access to care and treatment for infected people remained insufficient in 2016. Updating CoC estimates will help to assess the impact of new measures implemented since 2016 as part of the goal to eliminate HCV.

Impact of pregnancy on antidepressant treatment course: a population-based comparative cohort study in France

The prevention of relapses and the treatment of depression during pregnancy are difficult challenges. The maintenance of antidepressants in pregnancy with its concomitant risks to mother and child needs to be weighed against those associated with not treating the disease. This study aimed at quantifying the impact of the occurrence of pregnancy on the course of antidepressant treatment among newly treated women (< 6 months). We performed a comparative observational cohort study using the nationwide French reimbursement healthcare system database. Women who conceived in 2014 and initiated an antidepressant at any time in the 6 months before pregnancy were compared with nonpregnant women newly exposed to antidepressants with matching on age, antidepressant exposure, history of psychiatric disorders, and area of residence. The primary outcome was a composite of antidepressant discontinuation, switch to another antidepressant, and concomitant use of antidepressants. The secondary outcome was the resumption of antidepressant during follow-up. We used Cox marginal proportional hazards models to compare time to outcomes between pregnant and nonpregnant women. The pregnant cohort included 6593 women, and the comparison cohort 29,347 nonpregnant women. In the period following the first month of treatment, pregnant women were more likely to experience treatment modification, and especially to stop receiving it, compared with nonpregnant women (adjusted hazard ratio (aHR) 1.58; 95%CI, 1.51-1.62). Pregnant women who discontinued treatment had a 41% decreased incidence of antidepressant resumption compared with nonpregnant women (aHR 0.59; 95%CI, 0.56-0.62). Pregnancy was a determinant of antidepressant treatment modification, and especially of discontinuation.

Teriflunomide-exposed pregnancies in a French cohort of patients with multiple sclerosis

Objective

To describe pregnancies exposed to teriflunomide (TERIF) in women with multiple sclerosis (MS) in France over the period 2014-2016.

Methods

All 15- to 49-year-old women with MS in the national health insurance database were included. Pregnancies that had started between August 2014 and March 2016 were identified from their outcomes. Three groups according to treatment exposure were compared: TERIF, interferons (IFNs) or glatiramer acetate, and no medication.

Results

Among the 44,008 women with MS followed 24.5 months on average, 2,639 pregnancies were identified. There were 1,538 pregnancies (58.3%) that were not exposed to any MS treatment in accordance with the guidelines. A total of 673 pregnancies (25.5%) were exposed to IFN and/or glatiramer acetate, and possible or probable exposure to contra-indicated treatments was observed in 428 pregnancies (16.2%), of whom 47 pregnancies were exposed to TERIF. The annual incidence rate of pregnancies exposed to TERIF was 1.4 per 100 patient-years; i.e., 3 times less than the 2 control groups (5.6 and 4.7, respectively). The median exposure duration to TERIF was 45 days after conception. The outcomes comprised 23 live births, 22 abortions (3 times more than the 2 other groups), and 2 miscarriages. All newborns were healthy at birth.

Conclusions

Despite specific TERIF guidelines for pregnancy-related issues and the availability of alternative therapies, some pregnancies exposed to TERIF were identified. Most of the cases were because of the absence of the recommended accelerated elimination procedure and appeared to be mostly unplanned pregnancies that probably reflect a lack of effective contraception.

Home blood pressure monitoring in France: Device possession rate and associated determinants, the Esteban study

Home blood pressure monitoring (HBPM) is increasingly being promoted in hypertension guidelines to improve hypertension management. Possessing a HBPM device could improve blood pressure (BP) control and prognostic impact. The aims of this study were to estimate the possession rate of HBPM devices in the French population and in hypertensive adults, and to investigate the determinants of possessing such devices at home. Cross-sectional analyses were performed using data from the Esteban survey, which comprised a representative sample of the French population. Among the 2,054 study participants, 673 had hypertension. Of these, 385 were aware they had it. Weighted logistic regressions were performed to investigate the factors (socioeconomic, clinical, drug treatment, and healthcare visits) associated with possessing a HBPM device. 20.9% of the study sample, 42.1% of those with hypertension, and 54% of those aware of their hypertension, possessed a HBPM device. Female gender (OR = 2.03, 95%CI [1.46; 2.60]), smoking (OR = 2.33, 95%CI [1.51; 3.15]), antihypertensive drugs (OR = 1.75, 95%CI [1.06; 2.44]), general practitioner (GP) visits (OR = 3.28, 95%CI [1.84; 4.68]), and diabetes (OR = 0.41 95% CI [0.14; 0.68]) were associated with possessing a HBPM device among those aware of their hypertension. Over 20% of the study population possessed a HBPM device at home. This proportion rose to one in two in those aware or their hypertension. Among the latter, possessing a device was positively associated with female gender, GP visits, and antihypertensive drug use. Increasing possession of HBPM devices in the hypertensive population could foster better management of the condition.

Impact of terminating reimbursement of symptomatic slow-acting drugs in osteoarthritis in France on volume and cost of drug deliveries, assessed with administrative databases

WITH OSTEOARTHRITIS (OA)

As one of the leading causes of disability in adults worldwide, its toll on patients and its economic burden for payers are substantial. The issue of change in OA management with the evolution of reimbursement schemes needs to be addressed.

OBJECTIVE

To assess the impact of terminating the reimbursement of symptomatic slow-acting drugs in OA (SYSADOAs) in France in terms of volume and cost, from a healthcare payer perspective.

PRINCIPAL RESULTS

We obtained costs and volumes from French public national databases. We considered three exposure periods around cutoff dates according to decisions of decreased then terminated SYSADOA reimbursement. The periods included 19 345 (control), 20 066 (secondary), and 16 200 (primary) patients, respectively. Mean ages were 66.2 (±11.8), 65.3 (±11.6) and 64.6 (±11.5) years and about 70% were women. The volume of nonsteroidal anti-inflammatory drug (NSAID) deliveries estimated by defined daily doses (DDDs) decreased during the periods from 40.5 (±76.3) DDDs per patient in 2008 to 29.6 (±66.4) in 2015. The volume of analgesic deliveries increased slowly over the three periods, from 70.2 (±108.9) DDDs in 2008 to 76.9 (±123.1) in 2015 for all patients.

MAJOR CONCLUSIONS

Our results did not show a measurable impact of terminating SYSADOA reimbursement on the delivery of NSAIDs and analgesics or on hospitalizations. However, neither do they allow for concluding that terminating SYSADOA reimbursement did not generate an increase in deliveries of non-reimbursed drugs, with their associated potential risks for public health.

Automatic and Explainable Labeling of Medical Event Logs With Autoencoding

Process mining is a suitable method for knowledge extraction from patient pathways. Structured in event logs, medical events are complex, often described using various medical codes. An efficient labeling of these events before applying process mining analysis is challenging. This paper presents an innovative methodology to handle the complexity of events in medical event logs. Based on autoencoding, accurate labels are created by clustering similar events in latent space. Moreover, the explanation of created labels is provided by the decoding of its corresponding events. Tested on synthetic events, the method is able to find hidden clusters on sparse binary data, as well as accurately explain created labels. A case study on real healthcare data is performed. Results confirm the suitability of the method to extract knowledge from complex event logs representing patient pathways.

Contacts with Health Services During the Year Prior to Suicide Death and Prevalent Conditions A Nationwide Study

BACKGROUND

This study was designed to describe contacts with health services during the year before suicide death in France, and prevalent mental and physical conditions.

METHODS

Data were extracted from the French National Health Data System (SNDS), which comprises comprehensive claims data for inpatient and outpatient care linked to the national causes-of-death registry. Individuals aged ≥15 years who died from suicide in France in 2013-2015 were included. Medical consultations, emergency room visits, and hospitalisations during the year preceding death were collected. Conditions were identified, and standardised prevalence ratios (SPRs) were estimated to compare prevalence rates in suicide decedents with those of the general population.

RESULTS

The study included 19,144 individuals. Overall, 8.5% of suicide decedents consulted a physician or attended an emergency room on the day of death, 34.1% during the week before death, 60.9% during the month before death. Most contacts involved a general practitioner or an emergency room. During the month preceding suicide, 24.4% of individuals were hospitalised at least once. Mental conditions (36.8% of cases) were 7.9-fold more prevalent in suicide decedents than in the general population. The highest SPRs among physical conditions were for liver/pancreatic diseases (SPR=3.3) and epilepsy (SPR=2.7).

LIMITATIONS

The study population was restricted to national health insurance general scheme beneficiaries (76% of the population living in France).

CONCLUSIONS

Suicide decedents have frequent contacts with general practitioners and emergency departments during the last weeks before death. Improving suicide risk identification and prevention in these somatic healthcare settings is needed.

Benzodiazepine exposure in subjects with incident Alzheimer's disease and related syndromes in France: a longitudinal study 2011-2017

Benzodiazepines are commonly used for behavioral and psychiatric symptoms of dementia, despite their numerous adverse effects and the lack of evidence regarding their efficacy in this context. We studied longitudinal benzodiazepines exposure in incident cases of Alzheimer's disease and related syndromes (ADRS) in France. We used a cohort of incident ADRS patients identified in 2012 within the national health data system. Benzodiazepines exposure was measured 1 year before, to 5 years after ADRS identification. Quarterly benzodiazepines prevalence and incidence were computed. We identified factors associated with long half-life benzodiazepines initiation. A total of 106 508 subjects were included. Quarterly benzodiazepines prevalence was stable (around 25%) but we saw an important decrease in long half-life benzodiazepines compensated by an increase in short half-life benzodiazepines. In most benzodiazepine initiations, the first episode lasted less than 3 months. Factors associated with initiating a long half-life benzodiazepine were young age, male gender, no registration with ADRS as a long-term disease, having consulted an ADRS specialist, antipsychotic reimbursement before the index date, no admission in nursing home. Prevalence of benzodiazepines use remains high in subjects with ADRS despite guidelines for their avoidance. However, indicators regarding benzodiazepine initiations (duration, benzodiazepine type) suggest some caution in their use.

ZiMM: A deep learning model for long term and blurry relapses with non-clinical claims data

This paper considers the problems of modeling and predicting a long-term and "blurry" relapse that occurs after a medical act, such as a surgery. We do not consider a short-term complication related to the act itself, but a long-term relapse that clinicians cannot explain easily, since it depends on unknown sets or sequences of past events that occurred before the act. The relapse is observed only indirectly, in a "blurry" fashion, through longitudinal prescriptions of drugs over a long period of time after the medical act. We introduce a new model, called ZiMM (Zero-inflated Mixture of Multinomial distributions) in order to capture long-term and blurry relapses. On top of it, we build an end-to-end deep-learning architecture called ZiMM Encoder-Decoder (ZiMM ED) that can learn from the complex, irregular, highly heterogeneous and sparse patterns of health events that are observed through a claims-only database. ZiMM ED is applied on a "non-clinical" claims database, that contains only timestamped reimbursement codes for drug purchases, medical procedures and hospital diagnoses, the only available clinical feature being the age of the patient. This setting is more challenging than a setting where bedside clinical signals are available. Our motivation for using such a non-clinical claims database is its exhaustivity population-wise, compared to clinical electronic health records coming from a single or a small set of hospitals. Indeed, we consider a dataset containing the claims of almost all French citizens who had surgery for prostatic problems, with a history between 1.5 and 5 years. We consider a long-term (18 months) relapse (urination problems still occur despite surgery), which is blurry since it is observed only through the reimbursement of a specific set of drugs for urination problems. Our experiments show that ZiMM ED improves several baselines, including non-deep learning and deep-learning approaches, and that it allows working on such a dataset with minimal preprocessing work.

Patient stratification for risk of readmission due to heart failure by using nationwide administrative data

BACKGROUND

Identifying patients with heart failure (HF) who are most at risk of readmission permits targeting adapted interventions. The use of administrative data enables regulators to support the implementation of such interventions.

METHODS AND RESULTS

In a French nationwide cohort of patients aged 65 years or older, surviving an index hospitalization for HF in 2015 (N = 70,657), we studied HF readmission predictors available in administrative data, distinguishing HF severity from overall morbidity and taking into account the competing mortality risk, over a 1-year follow-up period. We also computed cumulative incidences and daily rates of HF readmission for patient groups defined according to HF severity and overall morbidity. Of the patients, 31.8% (n = 22,475) were readmitted at least once for HF, and 17.6% (n = 12,416) died without any readmission for HF. HF severity and overall morbidity were the strongest readmission predictors were the strongest readmission predictors (subdistribution hazard ratios 2.66 [95% CI: 2.52-2.81] and 1.37 [1.30-1.45], respectively, when comparing extreme categories). Overall morbidity and age were more strongly associated with the rate of death without HF readmission (cause-specific hazard ratios). The difference in observed HF readmission between patient risk groups was approximately 40% (21.9%, n = 2144/9,786 vs 60.4%, n = 618/1023).

CONCLUSIONS

Segmentation of HF patients into readmission risk groups is possible by using administrative data, and it enables the targeting of preventive interventions.

Economic Evaluations of Anticancer Drugs Based on Medico-Administrative Databases: A Systematic Literature Review

BACKGROUND

Oncology is among the most active therapeutic fields in terms of new drug development projects, with increasingly expensive drugs. The expected clinical benefit and cost effectiveness of these treatments in clinical practice have yet to be fully confirmed. Health medico-administrative databases may be useful for assessing the value of anticancer drugs with real-world data.

OBJECTIVE

The objectives of our systematic literature review (SLR) were to analyse economic evaluations of anticancer drugs based on health medico-administrative databases, to assess the quality of these evaluations, and to identify the inputs from such databases that can be used in economic evaluations of anticancer drugs.

METHODS

We performed an SLR by using PubMed and Web of Science articles published from January 2008 to January 2019. The search strategy focused on anticancer drug cost-effectiveness analyses (CEAs)/cost-utility analyses (CUAs) that were entirely based on medico-administrative databases. The review reported the main choices of economic evaluation methods in the analyses. The quality of the articles was assessed using the Consolidated Health Economic Evaluation Reporting Standards (CHEERS) and risk of bias assessment checklists.

RESULTS

Of the 306 records identified in PubMed, 12 articles were selected, and one additional article was identified through Web of Science. Ten of the 13 articles were CEAs and three were CUAs. Most of the analyses were carried out in North America (n = 11). The economic metric used was the cost per life-year gained (n = 10) or cost per quality-adjusted life-year (n = 3). Reporting of the target analysis population and strategies in the articles was in agreement with the CHEERS guidelines. The structural assumptions underpinning the economic models displayed the poorest reporting quality among the items analysed. Representativeness bias (n = 11) and the issue of censored medical costs (n = 8) were the most frequently analysed risks.

CONCLUSION

A comparison of the economic results was not relevant due to the high heterogeneity of the selected studies. Our SLR highlighted the benefits and pitfalls related to the use of medico-administrative databases in the economic evaluations of anticancer drugs.

Hospitalization costs and risk of mortality in adults with nonalcoholic steatohepatitis: Analysis of a French national hospital database

BACKGROUND

Nonalcoholic fatty liver disease (NAFLD) has reached high prevalence, paralleling the obesity pandemic. The aggressive form of the disease, nonalcoholic steatohepatitis (NASH), is characterized by fatty infiltration and inflammation of the liver, can progress to compensated cirrhosis (CC) and end-stage liver disease (ESLD: decompensated cirrhosis [DCC] and hepatocellular carcinoma [HCC]), and may ultimately require liver transplantation (LT). Real-world data on the burden of NAFLD/NASH are limited. This study aimed to evaluate the clinical and economic burden of NAFLD/NASH to the French hospital system.

METHODS

This retrospective cohort study used data from the French PMSI-MCO database. Adults with NAFLD/NASH diagnosis identified between 2009 and 2015 were categorized into disease severity cohorts (NAFLD/NASH, CC, DCC, HCC, and LT). Demographic and clinical data were assessed at the index (diagnosis) date. Hospitalization resource utilization and costs were assessed in the pre- and post-index periods. Rates of liver disease progression and death were evaluated for each cohort.

FINDINGS

During the median follow-up of 34.8 months, of the 131,656 patients included, 1491 patients developed CC (1.1%), 7846 developed DCC (5.9%), 1144 developed HCC (0.9%), and 52 required LT (0.04%). The diagnosis of NAFLD/NASH was associated with increasing annual costs: €7736 vs €5076 before the diagnosis. Rates of comorbidities, hospitalization resource utilization, and costs increased with disease progression. The rate of death at seven-year follow-up was 7.9% in NAFLD/NASH, CC: 18.0%, DCC: 34.9%, and HCC: 48.8%.

INTERPRETATION

NAFLD/NASH is associated with high economic burden and imparts substantial risk of negative clinical outcomes and mortality at all stages of disease.

Infertility Treatment in France, 2008-2017: A Challenge of Growing Treatment Needs at Older Ages

Objectives. To measure trends in infertility treatment use between 2008 and 2017 in France using data from the national health insurance system.Methods. Between 2008 and 2017, we observed a representative national sample of nearly 1% of all women aged 20 to 49 years who were affiliated with the main health insurance scheme in France (more than 100 000 women observed each year). We exhaustively recorded all health care reimbursed to these women.Results. Among women aged 20 to 49 years, 1.25% were treated for infertility each year. Logistic regression analysis showed a significant interaction between age and year of treatment use (P < .001). Over the decade, infertility treatment use increased by 23.9% among women aged 34 years or older, whereas among women younger than 34 years there was a nonsignificant variation.Conclusions. Women aged 34 years or older were increasingly treated for infertility between 2008 and 2017.Public Health Implications. Treatment efficiency decreases strongly with a woman's age, presenting a challenge for medical infertility care.

Bariatric Surgery in France from 1997 to 2018

BACKGROUND

Bariatric surgery is 1 of the major treatments of obesity.

OBJECTIVES

This study describes the development of the bariatric surgery in France and look at some factors of this evolution.

SETTING

It concerns the activity of all the French hospitals over a 22-year period.

METHODS

Hospitalization databases from 1997 to 2018 have been used to study the evolution of the number of bariatric operations, their types, the characteristics of both patients and hospitals performing this surgery.

RESULTS

The number of operations grew from 2800 in 1997 to 52,500 in 2018, with 2 interruptions in this growth, in 2002 and from 2017 to date. The rate of operations is 4 times higher for women than for men, with a peak in the 35-44 age group. The adjustable gastric banding was the most popular operation until 2010, then replaced by sleeve gastrectomy since 2010. Private for-profit hospitals carried out the majority of these operations, even if the public hospitals activity progressed regularly during the past 2 decades.

CONCLUSIONS

Compared to other countries, the rate of bariatric operations in France is rather high whereas the obesity rate is medium to low. Easy accessibility to bariatric surgery should play a role in the high rate, but specific studies are necessary to evaluate if the operations are delivered adequately to the obese population.

Oral anticoagulants and risk of acute liver injury in patients with nonvalvular atrial fibrillation: a propensity-weighted nationwide cohort study

Insufficient real-world data on acute liver injury (ALI) risk associated with oral anticoagulants (OACs) exist in patients with nonvalvular atrial fibrillation (NVAF). Using the French national healthcare databases, a propensity-weighted nationwide cohort study was performed in NVAF patients initiating OACs from 2011 to 2016, considering separately those (1) with no prior liver disease (PLD) as main population, (2) with PLD, (3) with a history of chronic alcoholism. A Cox proportional hazards model was used to estimate the hazard ratio with 95% confidence interval (HR [95% CI]) of serious ALI (hospitalised ALI or liver transplantation) during the first year of treatment, for each non-vitamin K antagonist (VKA) oral anticoagulant (NOAC: dabigatran, rivaroxaban, apixaban) versus VKA. In patients with no PLD (N = 434,015), only rivaroxaban new users were at increased risk of serious ALI compared to VKA initiation (adjusted HR: 1.41 [1.05–1.91]). In patients with chronic alcoholism history (N = 13,173), only those initiating dabigatran were at increased risk of serious ALI compared to VKA (2.88 [1.74–4.76]) but an ancillary outcome suggested that differential clinical follow-up between groups might partly explain this association. In conclusion, this study does not suggest an increase of the 1-year risk of ALI in NOAC versus VKA patients with AF.

Use of the French healthcare insurance database to estimate the prevalence of exposure to potential drug-drug interactions

PURPOSE

Drug-drug interactions (DDIs) require monitoring in an aging population with increasing polypharmacy exposure. We aimed to estimate the prevalence of exposure to potential DDIs using the French healthcare insurance system database, for six DDIs with various clinical relevance: angiotensin II receptor blockers or angiotensin-converting enzyme inhibitors and nonsteroidal anti-inflammatory drugs (ARBs-ACEIs + NSAIDs), antiplatelet agents and NSAIDs (AAP + NSAIDs), serotonergic drugs and tramadol (SD + T), statins and macrolides (S + M), oral anticoagulant and NSAIDs (OAC + NSAIDs), and colchicine and macrolides (C + M).

METHODS

We used exhaustive healthcare data from a 1/97th random sample of the population covered by the French health insurance system (EGB) between 2006 and 2016. Exposure to a DDI was defined as overlapping exposure to two interacting drugs. The prevalence of exposure was estimated by year.

RESULTS

Prevalence of exposure in 2016 was estimated at 3.7% for ARBs-ACEIs + NSAIDs, 1.5% for AAP + NSAIDs, 0.76% for SD + T, 0.36% for S + M, 0.24% for AOC + NSAIDs, and 0.02% for C + M. In 26% to 58% of episodes of exposure, the two interacting drugs were prescribed by the same physician and dispensed by the same pharmacy the same day. Between 2006 and 2016, the yearly prevalence was increasing for SD + T and for DDIs involving NSAIDs, and it was decreasing for those involving macrolides.

CONCLUSION

Exposures to potential DDIs in France are not uncommon with a high proportion resulting from a co-prescription by the same physician. Monitoring the prevalence of exposure to DDIs is needed to implement prevention measures. Administrative data enable this surveillance in large and representative cohorts.

Ten-year trend of opioid and nonopioid analgesic use in the French adult population

AIMS

Analgesics are the most widely used medicines worldwide. In parallel, opioid abuse has increased and is of major concern. The accessibility of pharmacologically powerful medicines and the addictovigilance signals in France about the risk of opiates addiction call for an overview of analgesic use. The objective of this study was to investigate the use of analgesics reimbursed in France over a 10-year period through its prevalence.

METHODS

A cross-sectional study repeated yearly was conducted by using data from the French reimbursement database from 2006 to 2015. Analgesics were classified according to their pharmacological potency: prevalence of use for each category and sociodemographic characteristics of patients treated were analysed.

RESULTS

The annual prevalence of analgesic use was high and increased during the study period (59.8%, 253 976 users in 2015). In 2015, prevalence was always higher in women and increased with age, except for those older than 84 years. Peripheral analgesics were the most used (55.3%, 234 739 users). The prevalence of weak analgesic use decreased (21.3%, 90 257 users), mainly due to the definitive withdrawal of dextropropoxyphene in France in 2011, which was not offset by an increase in the consumption of other weak analgesics. For strong analgesics (1.2%, 5129 users), morphine was the most widely used, with a dramatic increase in oxycodone use, especially in the elderly.

CONCLUSION

The prevalence of analgesic use is high: approximately 31 million adults had at least 1 analgesic reimbursed in 2015. The most widely used analgesics were peripheral analgesics, far ahead of opioid analgesics.

Association Between Nocturnal Hypoxemia and Cancer Incidence in Patients Investigated for OSA: Data From a Large Multicenter French Cohort

BACKGROUND

Previous studies have yielded inconsistent findings regarding the association between OSA and cancer in humans.

RESEARCH QUESTION

Is there an association between indexes of sleep-disordered breathing severity and cancer incidence in patients investigated for suspected OSA?

STUDY DESIGN AND METHODS

Data from a large multicenter cohort of cancer-free patients investigated for OSA were linked to health administrative data to identify new-onset cancer. Kaplan-Meier survival analysis and Cox proportional hazards models were used to evaluate the association of cancer incidence with OSA severity and nocturnal hypoxemia.

RESULTS

After a median follow-up period of 5.8 years (interquartile range, 3.8-7.8), 718 of 8,748 patients (8.2%) had received a diagnosis of cancer. On unadjusted Kaplan-Meier survival analyses, cancer incidence was associated with increasing severity of OSA (log-rank test, P < .0005) and nocturnal hypoxemia (log-rank test, P < .0001 for both oxygen desaturation index and percent night time with oxygen saturation < 90% [T90]). After adjustment for anthropomorphic data, smoking and alcohol consumption, comorbid cardiac, metabolic, and respiratory diseases, marital status, type of sleep study, and study site, only T90 was associated with cancer incidence (adjusted hazard ratio, 1.33; 95% CI, 1.05-1.68 for T90 ≥ 13% vs < 0.01%; P = .02). On stratified analyses, the association between T90 and cancer appeared stronger in older patients with obesity and no adequate OSA therapy. Among the most frequent cancer sites, nocturnal hypoxemia was associated with lung and breast malignancies.

INTERPRETATION

Nocturnal hypoxemia was associated with all-cancer incidence in patients investigated for OSA. Whether OSA therapy might reduce the risk of cancer needs further evaluation.

Prevalence of prediabetes and undiagnosed type 2 diabetes in France: Results from the national survey ESTEBAN, 2014-2016

AIM

To assess the prevalence of prediabetes, undiagnosed diabetes and diagnosed diabetes in France, among adults between 2014 and 2016 using data from the nationwide Esteban survey.

METHODS

National cross-sectional study conducted between 2014 and 2016 in metropolitan France. Individuals aged 18 to 74 were recruited with a 3-stage geographic sampling. They completed two face-to-face interviews, filled in a self-administered questionnaire and underwent a medical examination with the collection of biological samples. Their data were linked to the National Health Data System to identify anti-diabetic drugs reimbursement. Prediabetes and undiagnosed diabetes were defined as no diagnosis of diabetes and 6.1 mmol/l ≤ Fasting Plasma Glucose (FPG) < 7.0 mmol/l for prediabetes and FPG ≥ 7.0 mmol/l for undiagnosed diabetes. Non-pharmacologically treated diabetes and pharmacologically treated diabetes were defined as self-reported diabetes without or with self-reported or reimbursed antidiabetic medication, respectively. Estimated prevalence were weighted to take into account survey design and non-response.

RESULTS

The ESTEBAN survey recruited 3476 adults, 2270 were included in this analysis. The weighted prevalence was 1.7% [1.1 - 2.4] for undiagnosed diabetes (men: 2.7%, women; 0.9%), 9.9% [8.3 - 11.5] for prediabetes (men: 13.2%, women: 7.0%), 5.7% [4.3 - 7.1] for diagnosed diabetes. Among the diagnosed cases, 79% were pharmacologically treated. Among all diabetes cases, 23% were undiagnosed.

CONCLUSION

The prevalence of undiagnosed diabetes and prediabetes is increasing in France. Our results highlight the need to increase primary prevention and reinforce secondary prevention of diabetes.

Prevalence of mental disorders is higher in patients with multiple sclerosis than in the general population or in patients with rheumatoid arthritis in France

Background

Mental disorders (MDs) in multiple sclerosis (MS) patients decreases treatment adherence and quality of life, and increases the risk of disability progression and care consumption.

Objective

This study was to assess the prevalence of MDs in MS patients compared with healthy controls (HC) of the French general population and rheumatoid arthritis (RA) patients.

Methods

The 2015 prevalence of MDs for MS patients, RA patients and general population was estimated using a random population-based data sample from ‘National Inter-Scheme Information System on Health Insurance’ in the 2011–2015 period. Two control groups (1:5 ratio for the HC and 1:1 for the RA group) were matched to the MS group for year of birth, gender, area of residence and health insurance scheme.

Results

A total of 1145 MS patients were identified (sex ratio of 2.5 (F/M); median age 50 years). The prevalence of MDs was higher in the population of patients with MS (37.3%) than in the French general population (13.6%), and to a lesser extent in the RA group (21.1%) leading to the prevalence ratios of 2.8 (95% confidence intervals (CIs) 2.5–3.0) and 1.9 (95% CI 1.7–2.3), respectively.

Conclusions

This study confirmed that MS patients are at a higher risk of MDs than the French general population or RA patients.

β-Adrenoceptor Drugs and Parkinson's Disease: A Nationwide Nested Case-Control Study

BACKGROUND

Potential relationships between β-adrenergic drugs and α-synuclein synthesis in Parkinson's disease (PD) have been recently suggested.

OBJECTIVE

This study investigated the putative association between β-adrenoceptor drug exposure and PD occurrence.

METHODS

A nested case-control study was performed in the Echantillon Généraliste des Bénéficiaires (EGB) (a 1/97th random sample of affiliates to the French Insurance System). Incident PD patients diagnosed between 01/01/2008 and 31/12/2017 (index date) were matched 1:1 to controls by gender, birth year, and insurance scheme. Exposure to any β-agonist and to any β-antagonist was compared between cases and controls within 1-2 years before the index date, and exposure to salbutamol and to propranolol was individualized. The association between PD and β-adrenoceptor drugs was investigated through conditional logistic regression models adjusted for potential confounding factors. Because of a statistical interaction between β-agonists and diabetes, results were stratified according to the presence of diabetes.

RESULTS

Among the 2225 incident PD patients identified in the EGB (mean age 75.6 ± 10.2 years, sex ratio 1.04), no significant association was found between PD and β-antagonists (adjusted odds ratio [aOR] 1.05 [95% confidence interval 0.91-1.20]), except for propranolol (aOR 2.11 [1.38-3.23]). For β-agonists, a protective association in non-diabetic patients (aOR 0.75 [0.60-0.93]) and an opposite and significant association in diabetic patients (aOR 1.61 [1.02-2.55]) were observed. Similar results were found with salbutamol.

CONCLUSION

This study did not identify an increased risk of PD occurrence after β-antagonist exposure, except for propranolol (potential protopathic bias). The discordant results observed with β-agonists in patients with or without diabetes deserve further exploration of the influence of diabetic comorbidity on PD occurrence and evolution.

Care coordination for severe mental health disorders: an analysis of healthcare provider patient-sharing networks and their association with quality of care in a French region

BACKGROUND

For patients with multiple and complex health needs, such as those suffering from mental health disorders, outcomes are determined by the combined actions of the care providers they visit and their interactions. Care coordination is therefore essential. However, little is known on links between hospitals providing psychiatric care and community-based care providers which could serve as a basis for the creation of formal mental care networks supported by recent policies. In this context, we first aimed to identify and characterize existing types of healthcare provider patient-sharing networks for severe mental health disorders in one French region. Second, we aimed to analyse the association between their characteristics and the quality of the care they provide.

METHODS

Patient flows among healthcare providers involved in treating severe mental health disorders in the Provence-Alpes-Côte-d'Azur region were extracted from the French national health data system, which contains all billing records from the social health insurance. Healthcare provider networks that have developed around public and private non-profit hospitals were identified based on shared patients with other providers (hospitals, community-based psychiatrists, general practitioners and nurses). Hierarchical clustering was conducted to create a typology of the networks. Indicators of quality of care, encompassing multiple complementary dimensions, were calculated across these networks and linked to their characteristics using multivariable methods.

RESULTS

Three main types of existing healthcare provider networks were identified. They were either networks strongly organized around the main hospital providing psychiatric care; scattered networks involving numerous and diverse healthcare providers; or medically-oriented networks involving mainly physician providers. Few significant associations between the structure and composition of healthcare provider networks and indicators of quality of care were found.

CONCLUSIONS

Our findings provide a basis to develop explicit structuring of mental care based on pre-existing working relationships but suggest that healthcare providers' patient-sharing patterns were not the main driver of optimal care provision in the context explored. The shift towards a stronger integration of health and social care in the mental health field might impact these results but is currently not observable in the administrative data available for research purpose which should evolve to include social care.

Comparative Safety and Effectiveness of Oral Anticoagulants in Nonvalvular Atrial Fibrillation: The NAXOS Study

Supplemental Digital Content is available in the text.

Background and Purpose:

The effects of direct oral anticoagulants in nonvalvular atrial fibrillation should be assessed in actual conditions of use. France has near-universal healthcare coverage with a unified healthcare information system, allowing large population-based analyses. NAXOS (Evaluation of Apixaban in Stroke and Systemic Embolism Prevention in Patients With Nonvalvular Atrial Fibrillation) aimed to compare the safety, effectiveness, and mortality of apixaban with vitamin K antagonists (VKAs), rivaroxaban, and dabigatran, in oral anticoagulant-naive patients with nonvalvular atrial fibrillation.

Methods:

This was an observational study using French National Health System claims data and including all adults with nonvalvular atrial fibrillation who initiated oral anticoagulant between 2014 and 2016. Outcomes of interest were major bleeding events leading to hospitalization (safety), stroke and systemic thromboembolic events (effectiveness), and all-cause mortality. Four approaches were used for comparative analyses: matching on propensity score (PS; 1:n); as a sensitivity analysis, matching on high-dimensional PS; adjustment on PS; and adjustment on known confounders. For each outcome, cumulative incidence rates accounting for competing risks of death were estimated.

Results:

Overall, 321 501 patients were analyzed, of whom 35.0%, 27.2%, 31.1%, and 6.6% initiated VKAs, apixaban, rivaroxaban, and dabigatran, respectively. Apixaban was associated with a lower PS–matched risk of major bleeding compared with VKAs (hazard ratio [HR], 0.43 [95% CI, 0.40–0.46]) and rivaroxaban (HR, 0.67 [95% CI, 0.63–0.72]), but not dabigatran (HR, 0.93 [95% CI, 0.81–1.08]). Apixaban was associated with a lower risk of stroke and systemic thromboembolic event compared with VKAs (HR, 0.60 [95% CI, 0.56–0.65]), but not rivaroxaban (HR, 1.05 [95% CI, 0.97–1.15]) or dabigatran (HR, 0.93 [95% CI, 0.78–1.11]). All-cause mortality was lower with apixaban than with VKAs, but not lower than with rivaroxaban or dabigatran.

Conclusions:

Apixaban was associated with superior safety, effectiveness, and lower mortality than VKAs; with superior safety than rivaroxaban and similar safety to dabigatran; and with similar effectiveness when compared with rivaroxaban or dabigatran. These observational data suggest potentially important differences in outcomes between direct oral anticoagulants, which should be explored in randomized trials.

Syndromic Surveillance of Acute Gastroenteritis Using the French Health Insurance Database: Discriminatory Algorithm and Drug Prescription Practices Evaluations

The French national public health agency (Santé publique France) has used data from the national health insurance reimbursement system (SNDS) to identify medicalised acute gastroenteritis (mAGE) for more than 10 years. This paper presents the method developed to evaluate this system: performance and characteristics of the discriminatory algorithm, portability in mainland and overseas French departments, and verification of the mAGE database updating process. Pharmacy surveys with certified mAGE from 2012 to 2015 were used to characterise mAGE and to estimate the sensitivity and predictive positive value (PPV) of the algorithm. Prescription characteristics from these pharmacy surveys and from 2014 SNDS prescriptions in six mainland and overseas departments were compared. The sensitivity (0.90) and PPV (0.82) did not vary according to the age of the population or year. Prescription characteristics were similar within all studied departments. This confirms that the algorithm can be used in all French departments, for both paediatric and adult populations, with stability and durability over time. The algorithm can identify mAGE cases at a municipal level. The validated system has been implemented in a national waterborne disease outbreaks surveillance system since 2019 with the aim of improving the prevention of infectious disease risk attributable to localised tap water systems.

Innovative use of data sources: a cross-sectional study of data linkage and artificial intelligence practices across European countries

Background

The availability of data generated from different sources is increasing with the possibility to link these data sources with each other. However, linked administrative data can be complex to use and may require advanced expertise and skills in statistical analysis. The main objectives of this study were to describe the current use of data linkage at the individual level and artificial intelligence (AI) in routine public health activities, to identify the related estimated health indicators (i.e., outcome and intervention indicators) and health determinants of non-communicable diseases and the obstacles to linking different data sources.

Method

We performed a survey across European countries to explore the current practices applied by national institutes of public health, health information and statistics for innovative use of data sources (i.e., the use of data linkage and/or AI).

Results

The use of data linkage and AI at national institutes of public health, health information and statistics in Europe varies. The majority of European countries use data linkage in routine by applying a deterministic method or a combination of two types of linkages (i.e., deterministic & probabilistic) for public health surveillance and research purposes. The use of AI to estimate health indicators is not frequent at national institutes of public health, health information and statistics. Using linked data, 46 health outcome indicators, 34 health determinants and 23 health intervention indicators were estimated in routine. The complex data regulation laws, lack of human resources, skills and problems with data governance, were reported by European countries as obstacles to routine data linkage for public health surveillance and research.

Conclusions

Our results highlight that the majority of European countries have integrated data linkage in their routine public health activities but only a few use AI. A sustainable national health information system and a robust data governance framework allowing to link different data sources are essential to support evidence-informed health policy development. Building analytical capacity and raising awareness of the added value of data linkage in national institutes is necessary for improving the use of linked data in order to improve the quality of public health surveillance and monitoring activities.

Risk of early neurodevelopmental outcomes associated with prenatal exposure to the antiepileptic drugs most commonly used during pregnancy: a French nationwide population-based cohort study

OBJECTIVES

To assess the association between prenatal exposure to monotherapy with the antiepileptic drugs (AEDs) most commonly used during pregnancy and the risk of various neurodevelopmental outcomes compared with lamotrigine.

DESIGN

Nationwide population-based cohort study.

SETTING

French national healthcare databases.

PARTICIPANTS

Children born alive between 2011 and 2014 and prenatally exposed to AED monotherapy.

PRIMARY AND SECONDARY OUTCOME MEASURES

Outcomes included neurodevelopmental disorders (NDD), defined by International Classification of Diseases, 10th Revision codes F70-F98-pervasive developmental disorders (PDD, F84) and mental retardation (MR, F70-F79) were studied separately-and visits to speech therapists. The reference group comprised children prenatally exposed to lamotrigine. Children were followed until outcome, loss to follow-up, death or 31 December 2016. We performed inverse probability of treatment weighting analyses using the propensity score, which included maternal and infant characteristics. Hazard ratios (HRs) were calculated using Cox models.

RESULTS

The cohort comprised 9034 children, 2916 of which were exposed to lamotrigine, 1627 to pregabalin, 1246 to clonazepam, 991 to valproic acid (VPA), 621 to levetiracetam, 502 to carbamazepine, 477 to topiramate, 378 to gabapentin and 143 to oxcarbazepine. None of these AEDs, except VPA, was associated with an increased risk of any of the four neurodevelopmental outcomes investigated. Exposure to VPA was associated with increased risks of NDDs (HR=2.7, 95% CI (1.8 to 4.0)), PDD (HR=4.4 (2.1 to 9.3)), MR (HR=3.1 (1.5 to 6.2)) and visits to speech therapists (HR=1.5 (1.1 to 1.9)), with a dose-response relationship.

CONCLUSIONS

No increased risk of any of the neurodevelopmental outcomes investigated in this study was observed with prenatal exposure to levetiracetam, pregabalin, oxcarbazepine, topiramate, gabapentin, clonazepam or carbamazepine, compared with lamotrigine. However, this study corroborates the well-known association between maternal use of VPA during pregnancy and the risk of neurodevelopmental disorders in the offspring. Longer follow-up is necessary to confirm these findings.

Impact of persistence with tumour necrosis factor inhibitors on healthcare resource utilization and costs in chronic inflammatory joint diseases

AIM

To assess persistence with subcutaneous (SC) tumour necrosis factor (TNF) inhibitors as well as the impact of persistence on healthcare resource utilization (HCRU) and costs in patients with chronic inflammatory joint diseases.

METHODS

In this cohort study using population-based French claims data (from 2011 to 2014), we measured persistence with SC TNF inhibitors within 12 months (M0-12) following treatment initiation in treatment-naïve and treatment-experienced users (divided into three cohorts: rheumatoid arthritis [RA], ankylosing spondylitis [AS] and psoriatic arthritis [PsA]). Persistent patients were propensity score matched to nonpersistent patients at M12. The impact of persistence status on HCRU and costs was assessed during M12-24.

RESULTS

Of treatment-naïve (n = 3,804) and treatment-experienced (n = 2,279) users, only 56.1% and 46.8% were persistent at M12, respectively. Nonpersistent patients had more outpatient visits, computerized tomography scans, spine or joint magnetic resonance imaging procedures and disease-related hospitalizations, while persistent patients had more rheumatologist visits. Nonpersistent patients had lower drug costs but higher nondrug-related healthcare and hospitalization costs than persistent patients. In AS and PsA, overall healthcare costs were similar in persistent and nonpersistent patients. In RA, overall healthcare costs were lower in persistent patients (15,753€ vs 17,590€ in treatment-naïve and 17,622€ vs 21,177€ in treatment-experienced).

CONCLUSION

Persistence with SC TNF inhibitors within first 12 months following treatment initiation was low in both treatment-naïve and treatment-experienced patients. Differences were observed in distribution of costs between persistent and nonpersistent patients, showing that nonpersistence with SC TNF inhibitors can lead to increased HCRU and higher costs.

SCALPEL3: A scalable open-source library for healthcare claims databases

OBJECTIVE

This article introduces SCALPEL3 (Scalable Pipeline for Health Data), a scalable open-source framework for studies involving Large Observational Databases (LODs). It focuses on scalable medical concept extraction, easy interactive analysis, and helpers for data flow analysis to accelerate studies performed on LODs.

MATERIALS AND METHODS

Inspired from web analytics, SCALPEL3 relies on distributed computing, data denormalization and columnar storage. It was compared to the existing SAS-Oracle SNDS infrastructure by performing several queries on a dataset containing a three years-long history of healthcare claims of 13.7 million patients.

RESULTS AND DISCUSSION

SCALPEL3 horizontal scalability allows handling large tasks quicker than the existing infrastructure while it has comparable performance when using only a few executors. SCALPEL3 provides a sharp interactive control of data processing through legible code, which helps to build studies with full reproducibility, leading to improved maintainability and audit of studies performed on LODs.

CONCLUSION

SCALPEL3 makes studies based on SNDS much easier and more scalable than the existing framework [1]. It is now used at the agency collecting SNDS data, at the French Ministry of Health and soon at the National Health Data Hub in France [2].

Spatial distribution of in- and out-of-hospital mortality one year after acute myocardial infarction in France

OBJECTIVE

To describe the spatial distribution of acute myocardial infarction (AMI) mortality in France in association with the socio-economic characteristics of the patient's place of residence.

METHODS

In this population-based study, we included patients hospitalized for AMI identified according to ICD-10 codes, using data from the national health insurance database from January 1, 2013 to December 31, 2014. In- and out-of-hospital deaths were identified over a period of 1 year following the first hospital stay for AMI.An exploratory analysis was performed to classify area profiles. The spatial analysis of AMI mortality was performed using a principal component analysis followed by an ascending hierarchical classification taking into account socio-economic data, access-time by road to coronary angiography, standardized in-hospital prevalence, and 1 year mortality.

RESULTS

Over the 2 years, 115,418 patients were hospitalized with a diagnosis of AMI. Patients were a mean of 68 ​± ​15 years and most were men (68.5%). The overall mortality rate was 12.2% after 1 year. More than half of patients (65.5%) underwent an early revascularization procedure. The map of standardized 1 year mortality showed a geographic area of high mortality extending diagonally from north-east to south-west France. We identified 6 different area profiles with standardized mortality varying from 15.9 to 54.4 per 100,000 inhabitants. The spatial distribution of higher mortality was associated with lower socioeconomic levels. These findings were not associated with a lower access to coronary angiography.

CONCLUSION

There are considerable geographical differences in the prevalence of AMI and 1 year mortality. The spatial distribution of lower healthcare indicators follows the distribution of social inequalities. This study highlights the importance of focusing national policies on universally accessible prevention programs such as the promotion cardiac rehabilitation and healthy lifestyles.

Outpatient healthcare utilization 30 days before and after hospitalization for heart failure in France: Contribution of the national healthcare database (Systèmenationaldesdonnéesdesanté)

BACKGROUND

Guidelines have been published concerning patient management after hospitalization for heart failure. The French national healthcare database (Systèmenationaldesdonnéesdesanté; SNDS) can be used to compare these guidelines with real-life practice.

AIMS

To study healthcare utilization 30 days before and after hospitalization for heart failure, and the variations induced by the exclusion of institutionalized patients, who are less exposed to outpatient healthcare utilization.

METHODS

We identified the first hospitalization for heart failure in 2015 of adult beneficiaries of the health insurance schemes covering 88% of the French population, who were alive 30 days after hospitalization. Outpatient healthcare utilization rates during the 30 days after hospitalization and the median times to outpatient care, together with their interquartile ranges, were described for all patients, and for a subgroup excluding institutionalized patients.

RESULTS

Among the 104,984 patients included (mean age 79 years; 52% women), 74% were non-institutionalized (mean age 78 years; 47% women). The frequencies of at least one consultation after hospitalization and the median times to consultation were 69% (total sample) vs. 78% (subgroup excluding institutionalized patients) and 8 days (interquartile range 3; 16) vs. 7 days (3; 15) for general practitioners, 20% vs. 21% and 14 days (7; 23) vs. 16 days (9; 24) for cardiologists and 58% vs. 69% and 3 days (1; 9) vs. 2 days (1; 7) for nurses, with reimbursement of diuretics in 77% vs. 86%, angiotensin-converting enzyme inhibitors or angiotensin II receptor blockers in 48% vs. 55% and beta-blockers in 55% vs. 63%. Departmental variations, excluding institutionalized patients, were large: general practice consultations (interquartile range 74%; 83%), cardiology consultations (11%; 23%) and nursing care (68%; 77%).

CONCLUSIONS

Low outpatient healthcare utilization rates, long intervals to first healthcare utilization and departmental variations indicate a mismatch between guidelines and real-life practice, which is accentuated when including institutionalized patients.

Disparities in the Use of General Somatic Care among Individuals Treated for Severe Mental Disorders and the General Population in France

Individuals with severe mental illnesses (SMI) face a striking excess and premature mortality which has been demonstrated in several national contexts. This phenomenon, which constitutes a red-flag indicator of public health inequities, can be hypothesized to result from healthcare access issues which have been insufficiently documented so far. In this context, our objective was to explore patterns of general somatic healthcare use of individuals treated for SMI in comparison to those of the general population in France using national health administrative data and a matched case-control study. Differences in the use of general and specific somatic preventive care services, primary care, routine specialized somatic care and admissions to non-psychiatric hospital departments for somatic causes were described between cases and controls after adjustment on differing clinical needs, socio-economic status, and living environment. Our results show a lower use of general preventive care services and of routine specialized somatic care in the SMI population, despite more frequent comorbidities, and a higher occurrence of avoidable hospitalizations, despite higher contacts with primary care physicians. These findings suggest that the health system fails to address the specific needs of this vulnerable population and support the development of measures aimed at reducing this gap.

Economic burden of coronary artery disease or peripheral artery disease in patients at high risk of ischemic events in the French setting: a claims database analysis

Aims: Estimate the direct costs of high-risk patients presenting with coronary artery disease (CAD) or peripheral artery disease (PAD) in France.Materials and methods: This retrospective cohort study used a representative claims database, the "Echantillon Généraliste de Bénéficiaires" (EGB), to identify patients presenting with CAD or PAD between 2011 and 2016. Among those, patients meeting the COMPASS trial selection criteria were selected, as well as controls matched on age and sex. Direct costs (Euros 2016) were estimated in a societal perspective by comparing case and controls.Results: The adult population presenting with CAD or PAD in the EGB in 2016 was estimated at 29,888 individuals, representing a crude prevalence rate of 5.44%. After using the documented selection criteria of the COMPASS study, this population (COMPASS-like) was estimated at 17,369 individuals (58.1% of the CAD and/or PAD total population). Among them, a proportion of 11.5% presented with CAD + PAD. Compared with the original COMPASS population, patients were older (76.5 vs 68.2 years) and with a lower male predominance (60.0% vs 78.2% males). Compared with controls, the COMPASS-like population was characterized by a higher annual mortality (5.9% vs 3.5%) and the presence of more comorbidities on top of CAD and/or PAD. The annual per capita extra direct cost of the COMPASS-like population was estimated at €4,284, with a main contribution from inpatient care (58.9%). This extra cost was higher in the PAD ± CAD sub-group (€5,552) and the CAD + PAD sub-group (€8,067).Limitations: The EGB had limitations about several clinical features defining high-risk patients that may lead to bias in our estimates.Conclusions: Due to the high prevalence of CAD and/or PAD and the associated high unit costs, this population generates a significant economic burden, which is higher among patients with PAD and in those presenting simultaneously with both conditions.

The impact of health literacy on medico-social follow-up visits among French cancer survivors 5 years after diagnosis: The national VICAN survey

Background

Long‐term medico‐social follow‐up of cancer survivors is a challenge because of frequent subsequent troubles. In particular survivors with lower health literacy (HL) have poorer health and might more often use primary care services. However, the impact of HL on cancer survivors’ medico‐social follow‐up visits is not known. Our aim was to study medico‐social follow‐up and its associated determinants with a focus on HL 5 years after diagnosis.

Methods

VICAN is a national survey of French adult cancer survivors 5 years after a primary cancer. The Single‐Item Literacy Screener was used to define functional HL in this sample. We also asked patients to report the frequency of follow‐up visits with a general practitioner (GP) and/or social worker (SW) regarding their cancer disease.

Results

The 4045 participants were 57.4 ± 12.9 years old at diagnosis (range 20‐82) and 1495 (37%) were classified as having inadequate HL. Most cancer survivors (66.7%) were followed up by a GP regarding their cancer while only 14.5% had contact with a SW. After adjustment for sociodemographic, medical, and psychosocial characteristics, medico‐social follow‐ups (GP and SW visits) were more frequent among survivors with low HL. Furthermore, low income, unemployment, impaired mental health, treatment by chemotherapy, and perception of sequelae and fatigue were also associated with more frequent medico‐social follow‐up. Cancer localization association with medico‐social follow‐up was heterogeneous.

Conclusion

French cancer survivors with limited HL, lower socioeconomic status, and more severe cancer were more likely to use GP care and social services. Raising awareness and training GPs and SWs on medico‐social follow‐up for patients with limited HL seem necessary to support these vulnerable survivors.

Discontinuing statins or not in the elderly? Study protocol for a randomized controlled trial

BACKGROUND

The risk/benefit ratio of using statins for primary prevention of cardiovascular (CV) events in elderly people has not been established. The main objectives of the present study are to assess the cost-effectiveness of statin cessation and to examine the non-inferiority of statin cessation in terms of mortality in patients aged 75 years and over, treated with statins for primary prevention of CV events.

METHODS

The "Statins in the elderly" (SITE) study is an ongoing 3-year follow-up, open-label comparative multi-centre, randomized clinical trial that is being conducted in two parallel groups in outpatient primary care offices. Participants meeting the following criteria are included: people aged 75 years and older being treated with statins as primary prevention for CV events, who provide informed consent. After randomization, patients in the statin-cessation strategy are instructed to withdraw their treatment. In the comparison strategy, patients continue their statin treatment at the usual dosage. The cost-effectiveness of the statin-cessation strategy compared to continuing statins will be estimated through the incremental cost per quality-adjusted life years (QALYs) gained at 36 months, from the perspective of the French healthcare system. Overall mortality will be the primary clinical endpoint. We assumed that the mortality rate at 3 years will be 15%. The sample size was computed to achieve 90% power in showing the non-inferiority of statin cessation, assuming a non-inferiority margin of 5% of the between-group difference in overall mortality. In total, the SITE study will include 2430 individuals.

DISCUSSION

There is some debate on the value of statins in people over 75 years old, especially for primary prevention of CV events, due to a lack of evidence of their efficacy in this population, potential compliance-related events, drug-drug interactions and side effects that could impair quality of life. Data from clinical trials guide the initiation of medication therapy for primary or secondary prevention of CV disease but do not define the timing, safety, or risks of discontinuing the agents. The SITE study is one of the first to examine whether treatment cessation is a cost-effective and a safe strategy in people of 75 years and over, formerly treated with statins.

TRIAL REGISTRATION

ClinicalTrials.gov: NCT02547883. Registered on 11 September 2015.

Incidence of and Risk Factors for Systemic Adverse Events After Screening or Primary Diagnostic Colonoscopy: A Nationwide Cohort Study

OBJECTIVES

To estimate the systemic serious adverse event (SAE) rates after colonoscopy and to identify their risk factors.

METHODS

A nationwide cohort study was conducted using the comprehensive French claims databases SNDS (National Health Data System). Patients aged 30 years and over who underwent a first screening or diagnostic colonoscopy in 2010-2015 were included. The rates of cardiovascular and renal SAEs were estimated within 5 days after colonoscopy. The standardized incidence ratios were calculated to compare these incidence rates with those of the same events in the general population, and the associated risk factors were assessed by multilevel logistic regression.

RESULTS

Among the 4,088,799 included patients (median age, 59 years [interquartile range = 50-67]; 55.2% women; 30.1% with a Charlson index score ≤1), the 5-day SAE incidence rate was 2.8/10,000 procedures for shock, 0.87/10,000 for myocardial infarction, 1.9/10,000 for stroke, 2.9/10,000 for pulmonary embolism, 5.5/10,000 for acute renal failure, and 3.3/10,000 for urolithiasis. These SAEs occurred 3.3 to 15.8 times more often during the first 5 days after colonoscopy than expected in the general population. Thirty-day mortality rates ranged from 2.2/1,000 cases of urolithiasis to 268.1/1,000 cases of shock. Increasing age was associated with an increasing incidence of SAEs. Risks of shock and acute renal failure were associated with a greater number of comorbidities than the other SAEs. Colonoscopies in university hospitals were associated with higher risks, reflecting patient selection processes.

DISCUSSION

The systemic SAEs can be associated with a substantial mortality. They should be taken into account when deciding colonoscopy, in addition to perforation and bleeding, particularly in elderly patients with multiple comorbidities.

The Public Health Burden of Cardiomyopathies: Insights from a Nationwide Inpatient Study

Cardiomyopathies are responsible for heart failure and sudden cardiac death, but epidemiological data are scarce and the public health burden may be underestimated. We studied aggregating data from all public or private hospitals in France. Patients were categorized from relevant ICD-10 codes into dilated, hypertrophic, restrictive, or other cardiomyopathies (DCM, HCM, RCM, or OCM, respectively). Between 2008 and 2015, a total of 326,461 distinct patients had cardiomyopathy-related hospitalizations. The hospital-based prevalence of cardiomyopathy was 809 per million inhabitants (PMI) per year, including 428 PMI for DCM, 101 PMI for HCM, 26 PMI for RCM, and 253 PMI for OCM. Patients with cardiomyopathies accounted for 51% of all heart transplants, 33% of defibrillator implantations, 38% of mechanical circulatory supports, and 11.3% of hospitalizations for heart failure. In patients less than 40 years of age, these figures were 71%, 51%, 63%, and 23%, respectively. Over 2008–2015 and considering all cardiomyopathies, there was a significant increase for heart transplant (average annual percentage change, AAPC: +3.86%, p = 0.0015) and for defibrillator implantation (AAPC: +6.98%, p < 0.0001), and a significant decrease of in-hospital mortality (AAPC: −4.7%, p = 0.0002). This nationwide study shows that cardiomyopathies constitute an important cause of hospitalization, with increasing invasive therapeutic procedures and decreasing mortality.

Hospitalisations with infections related to antimicrobial-resistant bacteria from the French nationwide hospital discharge database, 2016

Massive use of antibiotics has led to increased bacterial resistance to these drugs, making infections more difficult to treat. Few studies have assessed the overall antimicrobial resistance (AMR) burden, and there is a paucity of comprehensive data to inform health policies. This study aims to assess the overall annual incident number of hospitalised patients with AMR infection in France, using the National Hospital Discharge database. All incident hospitalisations with acute infections in 2016 were extracted. Infections which could be linked with an infecting microorganism were first analysed. Then, an extrapolation of bacterial species and resistance status was performed, according to age class, gender and infection site to estimate the total number of AMR cases. Resistant bacteria caused 139 105 (95% CI 127 920-150 289) infections, resulting in a 12.3% (95% CI 11.3-13.2) resistance rate. ESBL-producing Enterobacteriaceae and methicillin-resistant Staphylococcus aureus were the most common resistant bacteria (&gt;50%), causing respectively 49 692 (95% CI 47 223-52 142) and 19 493 (95% CI 15 237-23 747) infections. Although assumptions are needed to provide national estimates, information from PMSI is comprehensive, covering all acute bacterial infections and a wide variety of microorganisms.

Cancer risk in dialyzed patients with and without diabetes

BACKGROUND

The risk of cancer is higher in patients with renal diseases and diabetes compared with the general population. The aim of this study was to assess in dialyzed patients, the association between diabetes and the risk to develop a cancer after dialysis start.

METHODS

All patients who started dialysis in the French region of Poitou-Charentes between 2008 and 2015 were included. Their baseline characteristics were extracted from the French Renal Epidemiology and Information Network and were linked to data relative to cancer occurrence from the Poitou-Charentes General Cancer Registry using a procedure developed by the INSHARE platform. The association between diabetes and the risk of cancer was assessed using the Fine & Gray model that takes into account the competing risk of death.

RESULTS

Among the 1634 patients included, 591 (36.2 %) had diabetes and 91 (5.6 %) patients developed a cancer (n = 24 before or at dialysis start, and n = 67 after dialysis start). The risk to develop a cancer after dialysis initiation was lower in dialyzed patients with diabetes than without diabetes (SHR = 0.54; 95 %CI: 0.32-0.91). Moreover, compared with the general population, the cancer risk was higher in dialyzed patients without diabetes, but not in those with diabetes.

CONCLUSION

The risk of developing a cancer in the region of Poitou-Charentes is higher in dialyzed patients without diabetes than with diabetes.

Untreated alcohol use disorder in people who inject drugs (PWID) in France: a major barrier to HCV treatment uptake (the ANRS-FANTASIO study)

BACKGROUND AND AIMS

Although people who inject drugs (PWID) are the core at-risk population in the hepatitis C virus (HCV) epidemic in industrialized countries, few initiate treatment. Alcohol use disorder (AUD), common within this population, has been identified as a barrier to HCV treatment uptake in the general population. We investigated whether the arrival of new and well-tolerated HCV treatments (direct-acting antivirals: DAA) has improved HCV treatment uptake in French PWID compared with former treatments (pegylated interferon-based treatments: Peg-IFN).

DESIGN

Using discrete-time Cox proportional hazards models based on exhaustive care delivery data, we tested for associations between AUD (defined by AUD-related long-term illness status, diagnosis coding during hospitalization and/or AUD pharmacological treatment) and first HCV treatment delivery, after adjusting for gender, age, complementary universal health cover, liver disease severity and type of opioid agonist therapy (OAT) received. Separate analyses were performed for 2012-13 (Peg-IFN era) and 2014-16 (DAA era).

SETTING

France.

PARTICIPANTS

All French people chronically HCV-infected who received OAT at least once during 2012-16 and were covered by the national health insurance (n = 24 831).

MEASUREMENTS

Incidence rate of HCV treatment uptake, hazard ratios associated with AUD and other covariates.

FINDINGS

Incidence rate (IR) of HCV treatment uptake per 100 person-years was 6.56, confidence interval (CI) = 6.30-6.84; and IR = 5.70, 95% CI = 5.51-5.89 for Peg-IFN-based treatment (2012-13) and DAA (2014-16), respectively. After multiple adjustment, people with AUD not receiving related medication had 30 and 14% lower Peg-IFN-based treatment and DAA uptake, respectively, than those without AUD [hazard ratio (HR) = 0.70, 95% CI = 0.62-0.80 and HR = 0.86, 95% CI = 0.78-0.94]. No difference was observed between those treated for AUD and those without AUD.

CONCLUSIONS

Despite the benefits of direct-acting antiviral treatment, untreated alcohol use disorder appears to remain a major barrier to hepatitis C virus treatment access for people who inject drugs in France.