Catastrophic and Impoverishing Health Spending and Financial Protection in Latin America: The Case of Colombia
Posted: 21 Jun 2007
Date Written: June 2007
Starting in 1993, Colombia has implemented a national social health insurance policy financed by a mixture of fiscal resources and payroll taxes reaching today more than half of the poorest and planning to achieve universal coverage by year 2010. Those able to pay are covered under the so called contributory regime whereas the poor are affiliated to the so called subsidized regime. Those affiliated to the contributory regime are covered by a comprehensive benefits package (premium=US$ 208) covering all levels of care. The benefits package is more limited in the subsidized regime (about US$ 117) but legislation calls for it to become similar to the contributory regime depending on the mobilization of additional resources. By year 2007, most low level complexity care and most high cost interventions related to catastrophic illnesses such as cancer and AIDS were covered under the subsidized regime even though most remaining hospital care is not included in the benefits package. Financial protection has been one central motivation of the implementation of the current health insurance scheme. This study presents descriptive statistics, uses an instrumental variable (IV) approach and Propensity Score Matching (PSM) to evaluate whether the Colombian social health insurance scheme has helped to mitigate the financial impact of illness.
Descriptive data indicates a lower incidence of catastrophic expenditure among those insured when ill: 58% of all uninsured households have out of pocket spending amounting to at least 10% of their ability to pay (non subsistence income). This incidence drops to 35% among those insured under the subsidized regime and to 14% for those covered by the contributory regime. However, based on the former data, no causal relationship between health insurance and the incidence of catastrophic expenditure can be established due to in potential differences in observed and unobserved characteristics between affiliates and non affiliates that may be biasing results. Given the differences in affiliation processes in the SR and CR and the data available, two different strategies were used to correct for potential selection bias in both regimes: PSM for the SR and a IV approach for the CR. Results indicate that, compared to the unaffiliated, those affiliated to the SR have a 21% lower probability of suffering from a catastrophic health expenditure (10% threshold) when ill. Formal workers affiliated to the CR have a 40% lower probability to do so, and informal workers have a 72% lower probability. Finally, in both regimes, the financial protection provided by health insurance decreases as the size of the catastrophe increases. In the SR, the difference in the incidence of catastrophic expenditure between the uninsured and the insured drops from 21% to 4% when rising the threshold from 10% to 40% of a households' ability to pay. Among independent workers belonging to the CR, the same number drops from 72% to 8% indicating, in both cases, that the insurance's capacity to mitigate the impact of a health shock decreases as the catastrophe increases. Overall then, there is evidence that the Colombian Social Health Insurance scheme is providing substantial financial protection by mitigating significantly the financial impact of health shocks on households.
Keywords: catastrophic expenditure, financial protection, health insurance, developing countries, Colombia
Suggested Citation: Suggested Citation