2005年-世界发展银行全球_Health_Insurance_Impacts_on_Health_and_Nonmedical_Consumption_in_a_Developing_Country_27页_420kb

Summary of "Health Insurance Impacts on Health and Nonmedical Consumption in a Developing Country"

Core Content

This paper evaluates the impact of Vietnam's Health Insurance (VHI) program, introduced in 1993, on health outcomes, health care utilization, and nonmedical household consumption. The study uses panel data from the Vietnam Living Standards Survey (VLSS) collected before and after the program's introduction, and combines propensity score matching (PSM) with a double-difference estimator to reduce biases from model specification and unobserved heterogeneity.

Main Viewpoints

• Health Outcomes: VHI had a positive impact on health outcomes, particularly for young school children and adults. It improved height-for-age and weight-for-age among children, and increased body mass index (BMI) among adults.

• Health Care Utilization:

  • Among young children, VHI increased the use of primary care facilities and shifted usage from pharmacists (for non-prescribed medicines) to health professionals (for prescribed medicines).
  • For older children and adults, VHI led to a marked increase in the use of hospital inpatient and outpatient services.

• Out-of-Pocket Expenditures: VHI significantly reduced annual out-of-pocket health expenditures, which is consistent with the idea that households with insurance are less likely to face large unexpected health costs.

• Nonmedical Consumption: The program also led to an increase in nonmedical household consumption, especially non-food consumption. This suggests that households use the reduced financial burden of health care to spend more on other goods and services, which may have indirect health benefits.

• Household Behavior: The study highlights that health insurance reduces uncertainty and creates a wealth effect, allowing households to increase consumption in other areas. This is particularly relevant for poor households, where health insurance may help alleviate financial strain and improve overall well-being.

Key Information

Context and Background

• VHI Introduction: Launched in 1993 in response to high out-of-pocket health spending, which accounted for over 70% of total health expenditure and 6% of household total consumption.
• Coverage and Enrollment:
  • VHI was initially compulsory for civil servants, state enterprise workers, pensioners, and people of merit.
  • Voluntary enrollment was primarily among school children, encouraged by local governments and schools.
  • Coverage varied by consumption quintile and age group, with the lowest coverage in the poorest quintile and among the elderly.

Methodology

• Data: Panel data from the VLSS, collected in 1992/93 and 1997/98.
• Estimation Approach: A combination of propensity score matching (PSM) and double-difference estimation is used to isolate the effect of VHI.
• Propensity Score Model: Includes individual-level variables such as schooling, employment status, and income, as well as province-level fixed effects to account for unobserved community influences.

Results Overview

• Health Outcomes:
  • Positive effects on height-for-age and weight-for-age of young children.
  • BMI increase among adults.
• Health Care Utilization:
  • Primary care usage increased among children.
  • Hospital use rose significantly among older children and adults.
• Out-of-Pocket Spending:
  • A substantial reduction in out-of-pocket health expenditures.
• Nonmedical Consumption:
  • Increase in nonmedical consumption, especially non-food items.
  • The reduction in out-of-pocket spending is smaller than the increase in nonmedical consumption, suggesting that households hold back consumption when facing health risks.

Implications

• The findings support the idea that health insurance can lead to improved health outcomes and better resource allocation in households.
• The wealth effect from reduced health risk is significant, especially in a context where a single hospital visit could cost up to 41% of a household's non-food consumption.
• The methodological approach is robust to selection biases and unobserved time trends, though it relies on the assumption that unobserved trends are the same for insured and uninsured groups.

Conclusion

The study concludes that Vietnam's health insurance program had favorable effects on both health outcomes and nonmedical consumption, with stronger impacts on children and adults. It underscores the importance of health insurance in reducing financial barriers to health care and enhancing household welfare through consumption smoothing. The use of propensity score matching and double-difference estimation provides a reliable framework for impact evaluation in the absence of randomization.