Summary
Background
Cardiovascular diseases (CVDs) constitute the second leading cause of death in France. The Système national d’information interrégimes de l’assurance maladi e (SNIIRAM; national health insurance information system) can be used to estimate the national medical and economic burden of CVDs.
Objectives
To describe the rates, characteristics and expenditure of people reimbursed for CVDs in 2013.
Methods
Among 57 million general health scheme beneficiaries (86% of the French population), people managed for CVDs were identified using algorithms based on hospital diagnoses either during the current year (acute phase) or over the previous 5 years (chronic phase) and long-term diseases. The reimbursed costs attributable to CVDs were estimated.
Results
A total of 3.5 million people (mean age, 71 years; 42% women) were reimbursed by the general health scheme for CVDs (standardized rate, 6.5%; coronary heart disease, 2.7%; arrhythmias/conduction disorders, 2.1%; stroke, 1.1%; heart failure, 1.1%). These frequencies increased with age and social deprivation, and were higher in Northern and Eastern France and Réunion Island. The total sum reimbursed by all schemes for CVDs was € 15.1 billion (50% for hospital care and 43% for outpatient care [including 15% for drugs and 12% for nurses/physiotherapists]); coronary heart disease accounted for € 4 billion, stroke for € 3.5 billion and heart failure for € 2.5 billion (i.e. 10% of the total expenditure reimbursed by all national health insurance schemes for all conditions).
Conclusion
CVDs constitute the leading group in terms of numbers of patients reimbursed and total reimbursed expenditure, despite a probable underestimation of both numbers and expenditure.
Résumé
Contexte
Les maladies cardioneurovasculaires (MCV) constituent la seconde cause de décès en France. Le Système national d’information inter-régime de l’assurance maladie (Sniiram) permet d’en estimer le poids médical et économique au niveau national.
Objectifs
Décrire les taux, caractéristiques et dépenses des personnes prises en charge pour MCV en 2013.
Méthodes
Parmi 57 millions d’assurés du régime général (86 % de la population française), les personnes prises en charge pour les MCV ont été repérées à l’aide d’algorithmes basés sur les diagnostics hospitaliers soit sur l’année en cours (phase aiguë), soit sur les cinq dernières années (phase chronique) et les affections de longue durée. Les coûts remboursés attribuables aux MCV ont été estimés.
Résultats
On dénombrait 3,5 millions de personnes (âge moyen, 71 ans ; femmes 42 %) prises en charge par le régime général pour une ou des MCV (taux standardisé, 6,5 % ; maladie coronaire, 2,7 % ; troubles du rythme/conduction, 2,1 % ; AVC, 1,1 % ; insuffisance cardiaque, 1,1 %). Ces taux augmentaient avec l’âge et le désavantage social et étaient plus élevées dans le Nord et l’Est et aussi à la Réunion. Le montant remboursé tous régimes pour les MCV était de 15,1 milliards d’euros (hôpital 50 %, soins de ville 43 % dont médicaments 15 % et infirmiers/kinésithérapeutes 12 %) dont la maladie coronaire : 4 milliards d’euros, AVC : 3,5 milliards, insuffisance cardiaque : 2,5 milliards, soit au total 10 % des remboursements versés à toute la population.
Conclusion
Les MCV constituent le premier groupe en termes d’effectifs de personnes prises en charge et de dépenses affectées, malgré une sous-estimation probable à la fois des effectifs et des dépenses.
Background
Cardiovascular diseases (CVDs) constitute the second leading cause of death in France, accounting for 27% of the 535,000 deaths in 2011 (i.e. a standardized rate of 216/100,000) . More precisely, ischaemic heart disease accounted for 7% of deaths (standardized death rate of 55/100,000) and cerebrovascular disease accounted for 6% of deaths (standardized death rate of 48/100,000); the male/female ratios for these last two causes of death were 2.4 and 1.3, respectively. However, age-standardized death rates decreased by one-half between 1990 and 2009: by 55% for ischaemic heart disease and by 48% for cerebrovascular disease . In 2012, the crude hospitalization rates were 339/100,000 for ischaemic heart disease, 188/100,000 for acute coronary syndromes, 94/100,000 for myocardial infarction, 224/100,000 for cerebrovascular disease and 239/100,000 for heart failure (HF) . Standardized hospitalization rates for each of these diseases also decreased between 2002 and 2012 for both sexes, but with different amplitudes. However, standardized hospitalization rates for stroke in subjects aged < 65 years increased over the same period (+13.3% for men and +16.3% for women). The observed variations can be attributed to lifestyle changes and improvements in primary/secondary prevention and management of patients at the acute phase, and to an increased prevalence of certain risk factors among younger subjects.
Despite the importance of these diseases, and apart from disease registries, few large-scale prevalence studies and cost estimates have been performed in France. Self-reporting Handicap-Santé (health and disability) surveys conducted on samples of households and institutionalized subjects (2008–2009) have reported a prevalence of 3.7% for ischaemic heart disease, 2.3% for HF and 2% for stroke, including 0.8% for stroke with sequelae, in the adult population .
The objective of this study was to estimate, on the basis of algorithms using data derived from the Système national d’information interrégimes de l’assurance maladie (SNIIRAM; national health insurance information system), the prevalence rates of the most common CVDs among people covered by the main French national health insurance scheme (86% of the population), as well as their variations according to region and social deprivation index, associated comorbidities and the expenditure reimbursed by national health insurance specifically related to these diseases.
Methods
Population and information system
The national health insurance general health scheme covered about 76% of the 66 million inhabitants of France in 2014, as well as various population groups covered by local mutualist sections for students, civil servants, etc. (i.e. 86% of the French population). The Mutualité Sociale Agricole (agricultural workers’ health insurance fund) and the Régime Social des Indépendants (self-employed health insurance fund) each cover 5% of the population, and the remaining 4% is covered by other schemes. The population of the present study was therefore composed of general health scheme beneficiaries, including local mutualist sections, after exclusion of those individuals for whom no hospital or outpatient expenditure was reimbursed during the year, representing about 4–5% of the population, varying according to age and sex.
The SNIIRAM database comprehensively and individually records all outpatient prescriptions and health care procedures reimbursed to beneficiaries of the various health schemes , but does not comprise any clinical information concerning results related to consultations, prescriptions or examinations. However, attribution of long-term disease (LTD) status, validated by a national health insurance physician at the request of the attending physician, allowing exemption of copayment, can provide information about the nature of the diseases treated. An anonymous and unique identification number for each beneficiary allows this information to be linked to the data collected by the Programme de médicalisation des systèmes d’information (PMSI; a the French national hospital discharge database) during hospital stays in the various types of health care institutions. Hospital diagnoses are coded according to the International Classification of Diseases 10th Edition (ICD-10), in the same way as the diagnoses allowing attribution of LTD status.
Methodology
The Caisse nationale d’assurance maladie (CNAM) general health scheme fund has developed a tool designed to identify beneficiaries reimbursed for chronic diseases and common, serious or expensive diseases and treatments, in order to study these diseases in terms of numbers, prevalence rates, expenditure and annual growth . This tool, based on SNIIRAM data, uses algorithms to distribute beneficiaries into 56 non-exclusive groups of patients, pooled into 13 main categories. The details of these algorithms are available and updated annually, and have been subjected to expert review . The algorithms used are based on principal diagnosis (PD), related diagnosis (RD) or significant associated diagnoses (AD) for short-stay and psychiatric hospital stays; diagnoses eligible for LTD cover; dispensing of specific drugs, identified by their Code identifiant de présentation (pack identification codes; the products and services reimbursed to the patient as medical device according to the liste des produits et prestations [LPP]); and specific procedures identified by their code according to the Classification commune des actes médicaux (French medical classification for clinical procedures).
Schematically, for some CVDs (coronary heart disease, stroke, HF), the term “acute” refers to a hospital stay during the year corresponding to the appropriate codes (with or without LTD), while the chronic phase is defined by the presence of LTD cover during the year and/or a hospital stay with a specific code during the previous 5 years. The acute episode always takes precedence over a chronic phase, and these two groups are mutually exclusive for a given disease, and are sometimes combined in statistical analyses. PDs were investigated in the various ward summaries when the stay comprised transfers to different wards.
The following CVD algorithms were used: for coronary heart disease, acute coronary syndrome (codes = I21 to I24 for a PD for a hospital stay in year N) and chronic ischaemic heart disease (codes = I20 to I25 for an LTD in year N; or codes I20 to I25 for PD/RD for hospital stays in years N to N-4 or for AD for hospital stays in year N); for stroke, acute stroke (codes I60 to I64 for a PD for a hospital stay in year N) and sequelae of stroke (codes = I60 to I64 or I67 to I69 for an LTD in year N; or codes = I60 to I64 or I67 to I69 for PD/RD for hospital stays in years N to N-4 or for AD for hospital stays in year N); for HF, acute phase (hospital stays with PD = I50 or AD/RD = I50 with PD = I11.0, I13.0, I13.2, I13.9, K76.1, J81 in year N) and chronic phase (LTD = I50, I11, I13 in year N; or hospital stays with PD/RD = I50 or AD = I50 and PD/RD = I11.0, I13.0, I13.2, I13.9, K76.1, J81 in years N to N-4; or hospital stay with AD = I50 with no restrictions concerning the PD/RD in year N); peripheral artery disease (PAD) (LTD = I70, I73, I74 in year N; or hospital stay with PD/RD = I70.2, I73.9 in years N to N–4; or hospital stay with AD = I70.2 or I73.9 in year N); arrhythmias/conduction disorders (LTD = I44 to I49 in year N; or hospital stay with PD/RD = I44 to I49 in years N to N-4; or hospital stay with AD = I44 to I49 in year N); valvular heart disease (LTD = I05 to I08 or I34 to I39 in year N; or hospital stay with PD/RD = I05 to I08 or I34 to I39 in years N to N–4; or stay with AD = I05 to I08 or I34 to I39 in year N, with the exception of congenital valvular heart disease); pulmonary embolism (hospital stay with PD = I26 in year N). The other CVDs corresponded to patients with LTD cover for other CVDs, i.e. a disease with an ICD-10 code corresponding to the CVD section, but not described above (I codes, especially myocarditis, endocarditis, pericarditis, occlusion and stenosis of cerebral arteries, not resulting in cerebral infarction, and cerebral aneurysm) or congenital cardiovascular malformations (Q codes) or cardiovascular disorders originating in the perinatal period (P29), vascular disorders of the intestine (K55), abnormalities of heart beat (R00), complications of cardiac and vascular prosthetic devices, implants and grafts (T82) or the presence of cardiac and vascular implants and grafts (Z95). Patient groups treated with antihypertensives were defined by the presence of at least three reimbursements for specific drugs at different dates in year N. Finally, the identification of patients reimbursed for other diseases (diabetes; cancers; neurological or degenerative diseases; chronic lung diseases; inflammatory or rare diseases or human immunodeficiency virus/acquired immunodeficiency syndrome; end-stage renal disease; or liver and pancreatic diseases) or patients treated with psychotropic drugs is described elsewhere . Only expenditure directly attributed to individual patients was considered for calculation of the costs of disease , corresponding to health care (mandatory health scheme) provided in 2013 and reimbursed up until 30 June 2014 for all types of insurance (disease; maternity; work accident/occupational disease; invalidity). Expenditure was classified into three main categories: outpatient care; hospital care; and cash payments. Outpatient care included visits to doctors, dentists, physiotherapists, nurses, laboratory examinations, drugs/other health products and transport costs. Hospital care included hospital stays in public and private hospitals, in short-stay, psychiatry, post-acute and rehabilitation hospital. Cash payments included sick leave, work accident and occupational disease allowances, maternity leave and invalidity pensions.
Except for hospital short-stay, allocation of reimbursed expenditure to a specific disease is based on a “top down” methodology. This methodology permits us–for people with several diseases in particular–to split their expenditure between each of their diseases on a pro rata basis, thus avoiding double counting. Furthermore, to avoid overestimating the expenditure attributed to a disease, a sum of expenses corresponding to “usual care” consumption was deducted for each individual, according to age and sex, and according to the various types of ambulatory care considered. So, for beneficiaries with only one of the 56 chronic diseases identified through our algorithms, all of their expenditures (excluding “usual care”) were attributed to the disease. For beneficiaries with several of the 56 identified diseases, their expenditures (excluding “basic care”) were distributed between each of their diseases using, as pro rata coefficient, the mean expenditure by disease as observed for beneficiaries with only one disease. Hospital (short-stay) expenditures were attributed directly to diseases according to short-stay PD or RD, using the same ICD codes as those used by the CVD algorithms. Thus, the total expenditure attributed to each CVD disease corresponds to the sum of: the expenditure of beneficiaries presenting only one CVD, the proportion of expenditure attributable to each CVD disease for patients with several diseases and hospital expenditures for each CVD short-stay. All expenditures, initially extracted for the study population (general health insurance scheme), were extrapolated to the entire French population (i.e. the population of all health insurance schemes) and matched to the main aggregates (or subtargets) of the maximum national ceiling for statutory health insurance expenditure (ONDAM), which is voted on each year by parliament .
Statistical analysis
The social deprivation index used, expressed in deciles, was constructed according to the communes (the smallest administrative units in France, 30,500 units) of residence by using four factors derived from Institut national de la statistique et des études écononomiques (national institute of statistics and economic studies) data: mean household income, percentage of high-school graduates among inhabitants aged ≥ 15 years, percentage of labourers in the employed population and unemployment rate . This index does not include overseas departments. Another marker of individual deprivation was used: the presence or absence of Couverture maladie universelle-complémentaire (CMU-C; complementary universal medical cover), a means-tested benefit granted for 1 year. In 2014, the means test limit was an annual income of about € 8,645 for a single person, and increased according to the number of people in the household. This means limit is lower than the poverty limit, defined as 50% of the median income. CMU-C covers the share of expenditure not reimbursed by national health insurance, and allows beneficiaries to access health care by waiving prepayment and top-up fees. However, over the age of 60 years, other aids are also available, such as the Allocation de solidarité aux personnes âgées (solidarity allowance for the elderly), which is higher than the means test limit for CMU-C. Analysis of CMU-C beneficiaries was therefore limited to people aged < 60 years.
The numbers and rate of general health scheme beneficiaries reimbursed for CVD in 2013 are reported in the “Results” section, together with the prevalence rates of the main comorbidities and other health states available in the health care tool . The reference French population at 01/01/2014 published by National Institute of Statistics and Economic Studies was used to establish sex- and age-standardized rates, and to perform regional comparisons. All analyses were performed with SAS Enterprise Guide software (version 4.3; SAS Institute Inc., Cary, NC, USA).
Results
In 2013, the SNIIRAM database comprised more than 3.5 million general health scheme beneficiaries (mean age, 71 years; 42% women) who were reimbursed for care associated with one or more CVD (i.e. almost 4.3 million people extrapolated to all of France; Table 1 ). The standardized rate for the overall population of all ages was 6.5% (8.1% in men; 5.2% in women) and the standardized rate for subjects aged ≥ 20 years was 8.5% (10.8% in men; 6.7% in women). The two most common groups of CVD identified, for all subjects aged ≥ 20 years, were coronary heart disease (3.6%) and arrhythmias/conduction disorders (2.7%), followed by stroke (1.5%), HF (1.4%) and PAD (1.3%). More specifically, in terms of acute episodes, identified exclusively by a hospital stay during the year, standardized rates in subjects aged ≥ 20 years were 0.15% for acute coronary syndrome, 0.23% for stroke and 0.33% for HF. Hospitalization rates for coronary heart disease and PAD were two-fold higher among men compared with women. A slightly smaller difference was observed for arrhythmias/conduction disorders and valvular heart disease, as well as HF and stroke. In contrast, the hospitalization rate for pulmonary embolism was similar for men and women.
| Diseases | N a | Mean age (years) | Women (%) | Rates (%) | N a for France | ||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Crude | Standardized, all ages | Standardized, age ≥ 20 years | |||||||||
| All | Men | Women | All | Men | Women | ||||||
| Coronary heart disease | 1467.3 | 71.2 | 34.3 | 2.59 | 2.71 | 3.93 | 1.67 | 3.59 | 5.31 | 2.17 | 1783.3 |
| Acute | 59.9 | 67.7 | 32.1 | 0.11 | 0.11 | 0.16 | 0.07 | 0.15 | 0.22 | 0.09 | 72.6 |
| Chronic | 1407.4 | 71.3 | 32.4 | 2.48 | 2.60 | 3.77 | 1.60 | 3.45 | 5.09 | 2.08 | 1710.8 |
| Stroke | 624.6 | 70.2 | 49.4 | 1.10 | 1.15 | 1.25 | 1.07 | 1.51 | 1.67 | 1.38 | 757.8 |
| Acute | 93.2 | 72.4 | 50.7 | 0.16 | 0.17 | 0.18 | 0.17 | 0.23 | 0.24 | 0.21 | 113.4 |
| Sequelae | 531.4 | 69.8 | 49.1 | 0.94 | 0.98 | 1.07 | 0.91 | 1.28 | 1.43 | 1.17 | 644.4 |
| Heart failure | 578.5 | 77.6 | 51.6 | 1.02 | 1.07 | 1.12 | 1.05 | 1.42 | 1.50 | 1.36 | 706.4 |
| Acute | 136.0 | 79.6 | 52.4 | 0.24 | 0.25 | 0.26 | 0.25 | 0.33 | 0.35 | 0.33 | 166.6 |
| Chronic | 442.5 | 76.9 | 51.3 | 0.78 | 0.82 | 0.86 | 0.80 | 1.08 | 1.15 | 1.03 | 539.9 |
| Peripheral artery disease | 541.3 | 71.9 | 32.5 | 0.96 | 1.00 | 1.45 | 0.61 | 1.33 | 1.96 | 0.80 | 658.1 |
| Arrhythmias/conduction disorders | 1124.7 | 74.2 | 47.7 | 1.98 | 2.09 | 2.34 | 1.88 | 2.74 | 3.13 | 2.43 | 1372.5 |
| Valvular heart disease | 0.30 | 72.9 | 49.8 | 0.54 | 0.57 | 0.61 | 0.53 | 0.74 | 0.82 | 0.69 | 372.3 |
| Pulmonary embolism | 0.32 | 68.0 | 56.0 | 0.06 | 0.06 | 0.06 | 0.06 | 0.08 | 0.08 | 0.08 | 39.1 |
| Other diseases | 0.22 | 61.3 | 40.2 | 0.39 | 0.40 | 0.52 | 0.31 | 0.48 | 0.64 | 0.35 | 265.7 |
| All CVDs | 3526.2 | 70.6 | 42.2 | 6.22 | 6.51 | 8.08 | 5.19 | 8.52 | 10.80 | 6.68 | 4280.5 |
The prevalence of the various CVDs increased markedly with age ( Table 2 ), especially in men: 43% of men aged ≥ 75 years had at least one identified reimbursement for CVD versus 29% of women in the same age group. The CVDs most commonly observed after the age of 75 years were arrhythmias/conduction disorders (17% in men and 12% in women), followed by coronary heart disease (20 and 9%, respectively). The prevalence of HF was 9% for men and 8% for women.
| Age 0–14 years | Age 15–34 years | Age 35–54 years | Age 55–64 years | Age 65–74 years | Age ≥ 75 years | |
|---|---|---|---|---|---|---|
| Males | ||||||
| Coronary heart disease | 0.01 | 0.05 | 1.70 | 7.41 | 12.74 | 20.12 |
| Acute | 0.00 | 0.01 | 0.14 | 0.32 | 0.42 | 0.63 |
| Chronic | 0.01 | 0.05 | 1.56 | 7.08 | 12.33 | 19.49 |
| Stroke | 0.06 | 0.13 | 0.60 | 1.92 | 3.43 | 7.17 |
| Acute | 0.00 | 0.02 | 0.09 | 0.26 | 0.47 | 1.13 |
| Sequelae | 0.06 | 0.11 | 0.52 | 1.66 | 2.96 | 6.04 |
| Heart failure | 0.02 | 0.03 | 0.27 | 1.26 | 2.71 | 8.99 |
| Acute | 0.00 | 0.00 | 0.05 | 0.23 | 0.56 | 2.36 |
| Chronic | 0.02 | 0.03 | 0.23 | 1.03 | 2.15 | 6.63 |
| Peripheral artery disease | 0.00 | 0.01 | 0.46 | 2.86 | 4.83 | 7.69 |
| Arrhythmias/conduction disorders | 0.06 | 0.18 | 0.66 | 2.68 | 6.47 | 17.02 |
| Valvular heart disease | 0.02 | 0.04 | 0.22 | 0.78 | 1.72 | 4.09 |
| Pulmonary embolism | 0.00 | 0.01 | 0.04 | 0.09 | 0.15 | 0.26 |
| Other diseases | 0.18 | 0.12 | 0.24 | 0.78 | 1.42 | 2.38 |
| All cardiovascular diseases | 0.32 | 0.52 | 3.52 | 13.91 | 24.47 | 43.30 |
| Women | ||||||
| Coronary heart disease | 0.00 | 0.02 | 0.42 | 1.64 | 3.46 | 9.39 |
| Acute | 0.00 | 0.00 | 0.03 | 0.07 | 0.12 | 0.36 |
| Chronic | 0.00 | 0.02 | 0.39 | 1.57 | 3.34 | 9.03 |
| Stroke | 0.05 | 0.11 | 0.48 | 1.01 | 1.83 | 5.63 |
| Acute | 0.00 | 0.01 | 0.06 | 0.11 | 0.25 | 1.01 |
| Sequelae | 0.04 | 0.10 | 0.42 | 0.90 | 1.59 | 4.62 |
| Heart failure | 0.02 | 0.02 | 0.13 | 0.49 | 1.31 | 7.61 |
| Acute | 0.00 | 0.00 | 0.02 | 0.08 | 0.26 | 1.94 |
| Chronic | 0.02 | 0.02 | 0.11 | 0.40 | 1.05 | 5.67 |
| Peripheral artery disease | 0.00 | 0.01 | 0.16 | 0.62 | 1.13 | 3.55 |
| Arrhythmias/conduction disorders | 0.05 | 0.15 | 0.37 | 1.14 | 3.18 | 11.95 |
| Valvular heart disease | 0.01 | 0.03 | 0.14 | 0.44 | 1.04 | 3.07 |
| Pulmonary embolism | 0.00 | 0.01 | 0.03 | 0.05 | 0.12 | 0.31 |
| Other diseases | 0.16 | 0.09 | 0.15 | 0.31 | 0.53 | 1.17 |
| All cardiovascular diseases | 0.28 | 0.41 | 1.62 | 4.70 | 9.77 | 28.62 |
| Total | ||||||
| Coronary heart disease | 0.01 | 0.04 | 1.01 | 4.27 | 7.68 | 13.30 |
| Acute | 0.00 | 0.00 | 0.08 | 0.19 | 0.26 | 0.46 |
| Chronic | 0.01 | 0.03 | 0.93 | 4.09 | 7.43 | 12.84 |
| Stroke | 0.05 | 0.12 | 0.54 | 1.43 | 2.56 | 6.19 |
| Acute | 0.00 | 0.01 | 0.07 | 0.18 | 0.35 | 1.05 |
| Sequelae | 0.05 | 0.10 | 0.47 | 1.24 | 2.21 | 5.14 |
| Heart failure | 0.02 | 0.03 | 0.19 | 0.84 | 1.95 | 8.11 |
| Acute | 0.00 | 0.00 | 0.03 | 0.15 | 0.40 | 2.09 |
| Chronic | 0.02 | 0.02 | 0.16 | 0.69 | 1.55 | 6.02 |
| Peripheral artery disease | 0.00 | 0.01 | 0.30 | 1.64 | 2.82 | 5.06 |
| Arrhythmias/conduction disorders | 0.06 | 0.17 | 0.50 | 1.84 | 4.68 | 13.80 |
| Valvular heart disease | 0.01 | 0.04 | 0.17 | 0.60 | 1.35 | 3.45 |
| Pulmonary embolism | 0.00 | 0.01 | 0.03 | 0.07 | 0.13 | 0.29 |
| Other diseases | 0.17 | 0.10 | 0.19 | 0.52 | 0.93 | 1.61 |
| All cardiovascular diseases | 0.30 | 0.46 | 2.50 | 8.90 | 16.46 | 33.98 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
