WP n° 42

Duration of sick leaves, income and health insurance : A microeconometric application from database Hygie
Ben Halima M.A., Debrand T. (Irdes)
Irdes working paper n°42. 2011/09

The aim of this article is to establish the relationship between the duration of sick leaves, income and health insurance. The analysis is based on a modified version of the model developed by Allen (1981), taking into account the specificity of the French model.
The model's equilibrium properties indicate an indeterminate effect of wages on the duration of sick leaves, and that National Health Insurance rules equally modify their duration.
This property was subject to an econometric estimation using the Hygie database, constructed from the merger of different administrative files concerning private sector employees in France in 2005. It allows the employer/employee relationship to be taken into consideration, together with the impact of company characteristics on the health of their employees and the interactions between work and health. Estimations are carried out using a discrete time proportional hazard model allowing for unobserved heterogeneity in the male and female samples. In order to remove the indeterminate wage effect on the duration of duration of sick leaves, four levels are taken into account: the current wage level, wage progression since the beginning of the career (return on education), wage increases over the last two years (monetary recompense) and the efficiency wage. Estimation results show that current wage has a negative effect on the duration of sick leaves. On the contrary, high wage increases over the long-term tends to reduce the duration of sick leaves among men and to increase them among women. Moreover, the different Health Insurance modalities appear to modify employee behaviours concerning work absences.

WP n° 41

The Impact of Difficult Working Conditions on Health Expenditures
Debrand T. (Irdes)
Irdes working paper n° 41. 2011/03

The aim of this study is to estimate the impact of certain difficult working conditions on health expenditure indicators and to measure the aggregated effect on public health expenditures. The empirical analysis is based on a sample of employees aged from 18 to 65 using data from the 2002-2003 French Health survey. Three health expenditure indicators are retained: the number of GP or specialist consultations over the last twelve months, prescribed sick leave over two consecutive months and hospitalisations over the last twelve months. Impacts are estimated using two different methods: the intuitive estimation method and the matching method. Results clearly confirm that difficult working conditions have a direct impact on increased health expenditures.
The three attributes describing difficult working conditions (current physical risk, past physical risk and psychosocial risk) cause modifications in ambulatory care consumption, sick leave and hospitalisations. Results also reveal an additional impact on health expenditure induced by the cumulative exposure to occupational risks. According to the estimation methods retained, individuals subject to exposure to the three occupational risks studied record 22.4% to 25.1% more GP or specialist consultations than employees without exposure to risk, 46.3% to 56.1% more sick leave and 27.2% to 35.9% more hospitalisations.

WP n° 40

Social Capital and Health of Older Europeans
Sirven N., Debrand T. (Irdes)
Irdes working paper n° 40. 2011/02.

This research uses a time-based approach of the causal relationship (Granger-like) between health and social capital for older people in Europe. We use panel data from waves 1 and 2 of SHARE (the Survey of Health, Ageing, and Retirement in Europe) for the analysis. Additional wave 3 data on retrospective life histories (SHARELIFE) are used to model the initial conditions in the model. For each of the first 2 waves, a dummy variable for involvement in social activities (voluntary associations, church, social clubs, etc.) is used as a proxy for social capital as involvement in Putnamesque associations; and seven health dichotomous variables are retained, covering a wide range of physical and mental health measures. A bivariate recursive Probit model is used to simultaneously investigate (i) the influence of baseline social capital on current health - controlling for baseline health and other current covariates, and (ii) the impact of baseline health on current participation in social activities - controlling for baseline social capital and other current covariates. As expected, we account for a reversed causal effect: individual social capital has a causal beneficial impact on health and vice versa. However, the effect of health on social capital appears to be significantly higher than the social capital effect on health. These results indicate that the sub-population reaching 50 years old in good health has a higher propensity to take part in social activities and to benefit from it (social support, etc.). Conversely, the other part of the population in poor health at 50, may see its health worsening faster because of the missing beneficial effect of social capital. Social capital may therefore be a potential vector of health inequalities.

Published in: Social Science & Medicine, vol 75, n° 7, 2012/10, 1288-1295. Social capital and health of older Europeans: causal pathways and health inequalities.

WP n° 39

Sick Leaves: Understanding Disparities between Departments
Ben Halima M.A., Debrand T., Regaert C. (Irdes)
Irdes working paper n° 39. 2011/02

The purpose of this publication is to better understand disparities of proportions of sick leaves granted in French Departments. The Hygie database was used for this, built by merging a number of administrative files of employees in the private sector in France in 2005. This database enables the determination of "employers/ employees" relations, the impact of the characteristics of firms on the health of their employees and interactions between health and work.
After briefly reviewing the various determinants, between the effect of composition and effect of context, sick leaves and their importance for understanding geographic differences, we present a three-phase empirical analysis: a descriptive analysis to detect differences between Departments, a multivariate analysis to highlight explanatory factors of probability of being on sick leave and finally an analysis of determinants of differences between Departments.
Our different models explain a large part of the disparities between Departments. The effects of composition and effects of context account for about half of the absolute difference and two-thirds of the mean square error. These are the variables describing the medical supply (density of general practitioners), verifications by National Health Insurance and patient age when the professional career started, which best explain disparities between Departments concerning sick leave. In contrast to other composition or context included in our model, the percentage of sick leaves verified and the density of general practitioners are important levers of health policies. Our research shows that they could be used as public policy instruments aimed at reducing geographic disparities.

Which gave rise to: Issues in Health Economics (Questions d'économie de la santé) n° 177, 2012/06.
Sick Leave: What Explanation for Disparities between French Departments? First exploitation of the Hygie database.
Ben Halima M. A., Debrand T., Regaert C. (Irdes)

Published in: Revue française d'économie, vol 26, n°4, 2012/04.

WP n° 38

Disability and Social Security Reforms: The French Case
Behaghel L. (Paris School of Economics, Inra), Blanchet D. (Insee-D3E),
Debrand T. (Irdes), Roger M. (Paris School of Economics, Inra, Insee-D3E)
Irdes working paper n° 38. 2011/02.

The French pattern of early transitions out of employment is basically explained by the low age at “normal” retirement and by the importance of transitions through unemployment insurance and early-retirement schemes before access to normal retirement. These routes have exempted French workers from massively relying on disability motives for early exits, contrarily to the situation that prevails in some other countries where normal ages are high, unemployment benefits low and early-retirement schemes almost non-existent. Yet the role of disability remains interesting to examine in the French case, at least for prospective reasons in a context of decreasing generosity of other programs.
The study of the past reforms of the pension system underlines that disability routes have often acted as a substitute to other retirement routes. Changes in the claiming of invalidity benefits seem to match changes in pension schemes or controls more than changes in such health indicators as the mortality rates. However, our results suggest that increases in average health levels over the past two decades have come along with increased disparities. In that context, less generous pensions may induce an increase in the claiming of invalidity benefits partly because of substitution effects, but also because the share of people with poor health increases.

Published in: NBER Working Paper n° 17055. Cambridge : NBER, 2011/05, 33 p.

Published in : Working paper (PSE) n° 2011 – 02. Paris : PSE, 2011/01, 32 p. Disability and Social Security Reforms: The French Case.
Behaghel L., Blanchet D., Debrand T., Roger M.
On Archives-ouvertes site