Out-of-pocket health spending and equity implications in Tunisia^†

† An earlier version of this paper was presented at Economic Research Form annual conference (March 2014).

Abstract

Using data from 2000, 2005 and 2010 national household budget and consumption surveys, this paper examines the scope of catastrophic out-of-pocket (OOP) health expenditures in Tunisia and their distribution. It analyses their evolution since 2000 in relation to policy reforms in Tunisia especially since a new national health insurance fund (Caisse Nationale d’Assurance Maladie (CNAM)) became effective in 2007. The paper assesses the catastrophic dimension of OOP payments by calculating their incidence and their intensity at several thresholds and by analyzing their distribution throughout concentration indices. Estimating a probit model, the paper explores the determinants of catastrophic payments and identifies the characteristics of households coping with higher risk of catastrophic health expenditures. Finally, questioning whether the catastrophic payments have impoverishing effects on households, the paper estimates the percentage of households who move below the poverty line when accounting for OOP health payments. The paper concludes that the lasting importance of health OOP expenditures in Tunisia, their catastrophic dimension and their impoverishing effect make it necessary to look for a better allocation of the existing resources through more efficient financing mechanisms and better institutional arrangements.

Keywords:

• out of pocket payment
• catastrophic health expenditures
• impoverishing health care
• health risk protection
• health insurance
• Tunisia

JEL Classification:

• H510
• I100
• I140
• I320

Acknowledgements

We thank the referees for their comments and Kmar Inel Makni for proof reading.

Disclosure statement

No potential conflict of interest was reported by the authors.

ORCID

Salma Zouari http://orcid.org/0000-0001-5685-4900

Notes

† An earlier version of this paper was presented at Economic Research Form annual conference (March 2014).

1. Data were weighted according to the probability of each household unit being sampled to reflect the entire Tunisian population. The weighting factor is provided by NIS.

2. Caisse Nationale de Retraite et de Prévoyance Sociale (CNRPS)/ Caisse Nationale de Sécurité Sociale (CNSS).

3. Health Care System Law No. 91–63 of 29 July 1991.

4. The public health system includes in 2014 three levels across its 24 regions – (a) 2091 primary health centers and 108 local hospitals, (b) 32 regional hospitals and (c) 23 university hospitals.

5. Private sector employs 55.6% of specialists and 42% of general practitioners.

6. Eastern Mediterranean Regional Office (World Health Organization).

7. Implemented in late 2007.

8. The working people and retirees are covered by national health insurance. They are subject respectively to 6.75% health insurance premiums on their salary (2.75% payable by the employee and 4% payable employer) and a 4% tax on their pension.

9. The insured’s spouse, his minor children and in some conditions his parents benefit from his health insurance.

10. Enrollees under mandatory health insurance have increased from about 900 thousand persons in 1987 to 1700 thousand persons in 1998. They are estimated to 2817 thousand in 2008 and 3237 thousand in 2013 (CNAM report, 8-5-2014).

Beneficiaries’ of free or subsidized access to public facilities grow from about 600 thousand households in 2001 to 720 thousand in 2009 and 852 thousand in 2014 (Ministry of social affairs).

11. Coverage rate is 9% by AMG1 and 22% by AMG2.

12. Because most resources are absorbed by items essential to sustenance, such as food.

13. The concentration index is defined as twice the area between the concentration curve and the line of equality. The convention is that the index takes a negative value when the curve lies above the line of equality, indicating disproportionate concentration of the health variable among the poor, and a positive value when it lies below the line of equality.

14. At the 10% threshold, the incidence of catastrophic OOP we calculate using 2005 NIS data (19.5%) is much higher than the one calculated by Elgazzar et al. (2010) using a Health Survey (12%).

15. O’Donnell et al. (2005) suggests that the positive association between total expenditures and the incidence of catastrophic health spending is not due to the ability of the easiest to devote a larger share of their resources to health care. Instead, the explanation could be that households facing unusually high medical payments are forced to cover these expenses by borrowing or selling assets. This results in both an abnormal increase in the costs of care and a temporary increase in total household expenditures.

16. Such as the three options for access to ambulatory care or the APCI coverage.

17. Obviously, since total expenditures exceed nonfood expenditures, the curve representing OOP share defined with respect to total expenditures lies upon the curve representing OOP share defined with respect to nonfood expenditures. At any threshold, catastrophic payment gap (i.e. intensity) and percentage exceeding threshold (i.e. incidence) are more important when calculated with reference to NFE.

Additional information

Funding

This work was supported by ERF.