Les séminaires de recherche de l'Irdes constituent un outil de présentation, de discussion scientifique et de diffusion de travaux de recherche en cours de réalisation dans le domaine de l'économie de la santé.
Durée d'une séance : 1h30
- 40 minutes de présentation par les auteurs
- 25 minutes de discussion par un chercheur senior
- 25 minutes de discussion avec la salle
Ces séminaires se tiennent dans les locaux de l'Irdes et sont ouverts à l'ensemble de la communauté scientifique.
Auteurs : Mousquès J. (Irdes), Colla C. (The Dartmouth Institute for Health Care Policy and Clinical Practice)
Discutant : Forest P.-G. (School of Public Policy, University of Calgary-Canada)
Jeudi 10 novembre 2016
Rationale: Achieving health care delivery efficiency gains to support the maintenance or expansion of health care coverage is a key aim of the post-Affordable Care Act (ACA) era in the United States. Toward this aim, numerous policies have been implemented to strengthen primary health care delivery, improve coordination and integration with other levels of care, and introduce payment mechanisms to improve quality of care and foster coordination. These efforts include development of the Accountable Care Organization (ACO) model, a payment and delivery system reform experiment using pay-for-performance and risk-sharing mechanisms intended to improve integration and coordination between networks of primary and secondary providers held responsible for the total quality and cost of care for a defined population. Currently, there is a lack of research evaluating the impact of ACOs on performance with consideration of regional and local contexts, organizational characteristics, level of financial risk bearing, and patient characteristics.
Objective: This research aims to analyze ACO integration in combination with ACO pay-for-performance and risk-sharing mechanisms on the efficiency of health care delivery through the measurement of health outcomes, utilization of care by beneficiaries and quality of care delivered. The research hypothesis is that integration associated with mixed payment systems is effective but also that results depend on the context, degree, breadth and nature of integration, and on patient characteristics.
Methods: The analytical and empirical framework is based on a mixed-method design that combines three main sequential and embedded steps. First, qualitative interviews of executives in 16 high-performing ACOs is used to identify a framework of key dimensions of integration. Second, a quantitative exploratory framework, with factor and cluster analysis of mixed data based on waves one to three of the National Survey of ACOs (NSACO) merged with Medicare claims is used to determine ACO characteristics, commonalities and differences. Third, a quantitative explanatory design with econometric analysis is used to estimate if different clusters of ACOs identified in the previous step have a differential impact on efficiency outcomes.
Data sources: Waves one to three of the NSACO were collected by The Dartmouth Institute for Health Policy and Clinical Practice and The School of Public Health at the University of Berkeley. 398 ACOs completed the full survey and 248 were merged with Medicare claims data to obtain information on beneficiaries.
Results: ACO context, integration, and characteristics of patient populations influence the impact of ACOs on efficiency outcomes. ACOs with complete integration are more efficient in terms of quality of care and health care utilization, but less efficient in terms of activity or productivity.
Auteurs : Pierre A. (Irdes, CESP Equipe Economie de la santé), Jusot F. (Université Paris-Dauphine, PSL Research University, Leda-Legos & Irdes), Raynaud D. (Irdes), Franc C. (Inserm, CESP Equipe Economie de la santé, Irdes)
Discutant : Geoffard P.-Y. (Ecole d'économie de Paris, PSE)
Jeudi 10 novembre 2016
In France, despite the existence of Public Health Insurance, access to health care depends greatly on having Complementary Health Insurance (CHI). Thus policy makers want to generalize access to good CHI for the whole population. The first measure, on January 1st, 2016, extended employer-sponsored CHI compulsory to all private sector employees and extended its coverage for 12 months after the loss of employment by an employee. Taking into account the Expected utility theory, we simulate the likely effects of this law on the welfare of the population. This research is based on the 2012 Health, Health Care and Insurance survey, which provides information on socio-economic characteristics and risk preferences.
Under the assumption that individual contract premiums will remain identical, the law increases collective welfare. This result is mainly explained by the increased welfare of employees covered by an individual contract before the law was introduced (and therefore by an employer sponsored contract after the law). The results are less clear for employees not insured before the law as Half of them lose welfare. If we now assume that individual contract premiums will increase by 20%, collective welfare finally decreases. The decrease of welfare mainly concerns the poorest, the oldest and those with poor health status. However, this assumption is quite probable as the law deteriorates the risk structure of people covered by an individual contract.
Lundi 30 mai 2016
Hospital use is often a critical and decisive episode of the individual health and healthcare trajectory, particularly among the older people. Heart failure is a common chronic condition and a leading cause of hospitalization for elderly. Avoiding repeated hospital use for this condition is a major concern in France as in many other countries. It is often suggested in the healthcare literature that better monitoring and follow-up of patients in primary care settings can both increase the quality of care for patients and decrease hospital use. However, there is little quantitative evidence of this.
In this paper we test the hypothesis that the risk of readmission after a hospital episode for heart failure decreases with appropriate ambulatory care follow-up. More precisely, we propose to estimate the impact of GP follow-up on the readmission risk during the first month after discharge using a discrete time duration model and an instrumental variable strategy to account for omitted variables affecting both readmissions and primary care use.
The results from IV regression show that the readmission risk decreases significantly with GP follow-up in the weeks after the hospital discharge (OR=0,4). This confirms GPs role in determining patients' care pathway.
Auteurs : Or Z. (Irdes), Rococo E. (Institut Gustave Roussy) et Bonastre J. (Institut Gustave Roussy)
Discutant : Choné P. (Centre de recherche en économie et statistique)
Lundi 30 mai 2016
In France, minimum activity thresholds for cancer related procedures were introduced in 2008 in order to improve care quality. This policy, together with the activity based payment used since 2004/2005 has impacted significantly the level of competition in the market for cancer services. In this study, we question the impact of competition on quality by examining the variations in surgical treatments considered as “good practice” or “better quality” as a function of hospital and market characteristics.
Pooling data from 2005 and 2012, we use multilevel regression models for estimating the impact of market concentration on the propensity to receive better quality treatments. We extend the existing literature on hospital competition and quality by measuring hospital competition for a specific service which is not much studied in the literature: breast cancer surgery, and by calculating measures of competition and quality specific to breast cancer surgery market.
Auteurs : Buchmueller T. (Michigan University), Dourgnon P. (Irdes), Dumontet M. (Irdes), Jusot F. (Leda-Legos, Paris-Dauphine University), Wittwer J. (Isped, Bordeaux University)
Présenté par : Dumontet M. (Irdes)
Discutant : Grignon M. (McMaster University)
Jeudi 14 avril 2016
Background: In a context of high health care spending, the 2004 French health care reform modified the principles of care coordination by introducing a system of non-compulsory coordinated care pathways for patients. Since the 1st January 2005, the health insurance fund invites every 16 year old and over person to select a doctor (in most of case it is a General practitioner (GP)) who becomes his “Preferred doctor”. This doctor handles the first point of contact with the health system, provides care or makes referral to a specialist. The reform also defines three direct access specialties: gynecology, ophthalmology, psychiatrists for patients between 16 and 25 years old. Patients stay free to consult doctor without referral. However, if patients see a doctor who is not is “preferred doctor” or a doctor without a referral of their “preferred doctor” (except for direct access specialties), they are outside the coordinated care pathways and have financial penalties. These financial penalties (increase in co-payment rate, possibility for specialists with fixed fees to practice overbillings) have been introduced by the health care reform to encourage patients to respect the coordinated care pathway. The first goal of this reform was to improve the care coordination and also reduce unnecessary specialists' consultations.
Aims: In this paper we want to evaluate the effect of the French health care reform on access to care. How the reform impacts the number of doctor visits? Is the reform impact differently access to GPs and specialists?
Methods: In order to study the reform impact, we use national sickness fund panel databases for the years 2000, 2002, 2004, 2006, 2008 including nearly 36 333 individuals. We conduct before and after analysis with count data models to estimate the effect of the reform on different outcomes variables: number of GPs, specialists, gynecologists, number of different GPs seeing during the year …
Main Results: We find that the reform has reduced by 2.6% the number of GPs' visits and by 7.6% the number of specialists' visits (not including direct access specialties). This reduction was even more important on specialties (11%) which were often consulted without doctor's referrals before the reform (dermatology, otolaryngology, and rheumatology). The effect of the reform was also a 5.6% reduction in the number of different GPs seen during the year.
Conclusion: Given these results, the reform has contained the demand for specialist care and seems to improve coordination of care because patients see less different general practitioners.