2017年-世界发展银行全球_Effect_of_Stress_on_Later-Life_Health___Evidence_from_the_Vietnam_War_Draft_54页_1mb

报告摘要

Summary of "Effect of Stress on Later-Life Health: Evidence from the Vietnam War Draft"

Core Content

This paper investigates the long-term health effects of stress experienced during adolescence and early adulthood, using the risk of military induction during the Vietnam War as a proxy for stress. The authors argue that the uncertainty surrounding conscription caused significant stress, which may have lasting impacts on health outcomes in later life.

Main Findings

A 10 percentage point increase in induction risk (equivalent to 2 standard deviations) is associated with:
- A 0.2 unit increase in BMI (0.08%)
- A 1.5 percentage point increase in the probability of being obese
- A 1 percentage point increase in the probability of being in fair or poor health
- A 2 percentage point decrease in the probability of being in very good or excellent health
These associations are only observed for men who did not serve in the Vietnam War, suggesting that the effects are not due to cohort effects or other time trends.
The paper finds that induction risk does not appear to be a valid instrument for education or marriage, as it has a direct impact on health.

Key Mechanisms of Stress Impact

Chronic Stress Effects:
- Stress during the Vietnam War increased cortisol and epinephrine levels.
- These hormones affect bodily functions by diverting energy to muscles and the brain, reducing energy for eating, growth, and sexual activity.
- Chronic stress is linked to high blood pressure, cardiac hypertrophy, damaged arteries, suppressed immunity, and increased inflammation.
- Psychological effects include anxiety, depression, sleep problems, substance abuse, and increased smoking and alcohol consumption.
Impact on Eating Behavior:
- Chronic stress is associated with diminished self-control, leading to impulsive food choices and increased calorie intake.
- This mechanism is supported by Jackson et al. (2017), who found that hair cortisol levels correlate with obesity.
Draft Avoidance Behaviors:
- Some men used college enrollment or marriage to avoid induction.
- These behaviors may have improved health outcomes through better education and social support.
- However, the paper argues that the direct impact of induction risk on health remains significant even after controlling for these behaviors.

Data and Methodology

Data Sources:
- National Health Interview Survey (NHIS): Provides data on BMI, self-reported health, and military service.
- Selective Service Reports: Used to calculate induction risk based on monthly induction numbers and cohort size.
Induction Risk Calculation:
- The risk is calculated using the formula:
  $$
  R A _ {m, y} = \frac {\text {Inductions (12 Months)}}{\text {Size of Cohort at Risk}}
  $$
- It is computed for each month and year of birth from 18.5 to 26 years of age, as these were the legislated ages for induction.
Model Specifications:
- The authors estimate ordinary least squares (OLS) and probit models.
- They control for demographic and socioeconomic characteristics, race and ethnicity, time trends, and age and year fixed effects.
- Three model specifications are used:
  1. Basic model: Includes the most exogenous regressors.
  2. Second model: Adds family size and marital status.
  3. Third model: Includes education and log income.
Mortality Analysis:
- A Cox proportional hazards model is used to estimate mortality risk.
- The model is based on NHIS-Mortality Linked File data, which includes mortality information.

Falsification Tests

Veterans: The authors test the model on male veterans, who were no longer at risk of induction. The results show no significant association between induction risk and health, suggesting that the correlation is not due to cohort effects.
Women: Women had zero induction risk, but the paper assigns them the induction risk of men of the same birth month and year as a falsification test. The results show no significant association, further supporting the stress hypothesis.

Conclusion

The study provides evidence that stress during adolescence and early adulthood has lasting consequences on health.
The Vietnam War draft serves as a valid proxy for stress, and its impact is not due to other trends or cohort effects.
The results challenge the use of induction risk as an instrument for education or marriage, as it has direct health effects.
The paper highlights the complex pathways through which stress affects health, including biological, psychological, and behavioral mechanisms.

Additional Notes

The sample includes men born between 1937 and 1956, to avoid contamination from the Korean War.
The authors adjust for reporting error in BMI using NHANES III data.
The study is part of the World Bank's policy research initiative and is open-access.