Your SlideShare is downloading. ×
Boudreaux essen
Boudreaux essen
Boudreaux essen
Boudreaux essen
Boudreaux essen
Boudreaux essen
Boudreaux essen
Boudreaux essen
Boudreaux essen
Boudreaux essen
Boudreaux essen
Boudreaux essen
Boudreaux essen
Boudreaux essen
Boudreaux essen
Boudreaux essen
Boudreaux essen
Boudreaux essen
Boudreaux essen
Boudreaux essen
Boudreaux essen
Boudreaux essen
Boudreaux essen
Boudreaux essen
Boudreaux essen
Boudreaux essen
Boudreaux essen
Upcoming SlideShare
Loading in...5
×

Thanks for flagging this SlideShare!

Oops! An error has occurred.

×
Saving this for later? Get the SlideShare app to save on your phone or tablet. Read anywhere, anytime – even offline.
Text the download link to your phone
Standard text messaging rates apply

Boudreaux essen

64

Published on

Published in: Healthcare
0 Comments
0 Likes
Statistics
Notes
  • Be the first to comment

  • Be the first to like this

No Downloads
Views
Total Views
64
On Slideshare
0
From Embeds
0
Number of Embeds
2
Actions
Shares
0
Downloads
0
Comments
0
Likes
0
Embeds 0
No embeds

Report content
Flagged as inappropriate Flag as inappropriate
Flag as inappropriate

Select your reason for flagging this presentation as inappropriate.

Cancel
No notes for slide
  • I like this one.
  • Misspelling of contemporaneous
  • Transcript

    • 1. The Long-Term Impacts of Medicaid Exposure in Early Childhood Michel Boudreaux University of Minnesota June 1, 2014 1
    • 2. Medicaid • Medicaid provides health insurance coverage to low income mothers and children in the US • Joint effort of the states and federal government – Eligibility and benefits vary by state and year – Finances 48% of all childbirths – Covers 35% of all children under 19 • We have little information on the long term impacts of Medicaid 2
    • 3. Acknowledgments • Funding – UofM Interdisciplinary Doctoral Fellowship – AHRQ Dissertation Grant • All errors and opinions are my own • Dissertation committee – Donna McAlpine & Ezra Golberstein (co-advisers) – Bryan Dowd – Julian Wolfson 3
    • 4. Aim and approach • Research Question – Does exposure to Medicaid in early childhood improve health and economic outcomes in adulthood? • Empirical Challenge – Unobserved selection into the program • Use state-by-time variation in Medicaid using difference-in-differences to measure the effect of exposure to policy 4
    • 5. Medium-term impacts of Medicaid • Mechanisms – Improvements in child health that persist over- time – Improvement in family economic resources • Medicaid exposure earlier in childhood improves outcomes later in childhood – Self Reported Health, mortality, academic achievement (Currie et al., 2008; Meyer & Wherry, 2012; Levine et al., 2009 ) 5
    • 6. Study setting • Staggered timing of Medicaid’s introduction across the states – Created variation in the amount of early life exposure to Medicaid • Enacted in 1965 – Roll out mainly occurred between 1966-1970 – By 1972, all but 1 state had a program 6
    • 7. Medicaid adoption by quarter and year 7
    • 8. Components of Medicaid • Prior to Medicaid health services for the poor were limited • Mandated that all people receiving cash welfare (AFDC) were to be automatically enrolled – Income thresholds for eligibility were low • Covered services with no copay – Physician services, hospital stays, lab and x-ray • 8.5 million participants by 1970 8
    • 9. Previous evidence on Medicaid’s introduction • Two papers find substantial effects to short- run health – In groups targeted by Medicaid • 60% reduction in the incidence of low birth weight • 24% reduction in child mortality • Other work from this project – 51% increase in hospital utilization among young low-income children (Decker and Gruber 1993; Goodman-Bacon 2013; Boudreaux, 2014) 9
    • 10. Data • 1968-2009 Panel Study of Income Dynamics – Follows respondents and their descendants – A large oversample of low income families • Key measures – Health (chronic conditions) – Economic status (education, income, wealth) – Demographic information (place and time of birth) 10
    • 11. Sample selection • 1955-1980 birth cohorts – 6 cohorts with no Medicaid exposure prior to age 6 – 10 cohorts w/exposure starting in early childhood – 10 cohorts exposed starting in utero • I drop AZ (< 1% of sample) • A sample of adults (Age 18-54 ) – Average age is 37 at follow-up – Attach childhood characteristics • Define childhood exposure and isolate subgroups targeted by the program 11
    • 12. Medicaid exposure • Share of months from conception to 6th birthday (“early childhood”) • Calculated based on state of birth • Range is [0,1] • A function of time and place of birth • Measures exposure to policy, not participation 12
    • 13. Average months exposed 13
    • 14. Outcome Indexes • Indexing improves precision and reduces problems associated with multiple comparisons – Equally weighted mean of z-scores • Chronic Condition Index ( 𝑥=-0.03; sd=0.6) – Hypertension, diabetes, heart disease, obesity • Economic index ( 𝑥=0.3; sd=0.9) – Years of education, poverty level, family wealth (Anderson, 2008; Hoynes et al. 2013) 14
    • 15. Non-random adoption timing • Many things were changing around Medicaid’s introduction – Other social programs and changing health care markets • 3 strategies – Control for changes in social policy and health markets – State specific time trends – Compare target and placebo groups 15
    • 16. Models 𝑦𝑖𝑛𝑠𝑡 = 𝜆𝑀𝐶𝐴𝐼𝐷𝑆𝐻𝐴𝑅𝐸𝑠𝑡 + 𝛽𝑋𝑖𝑛𝑠𝑡 + 𝜙𝑍𝑠𝑡 + 𝜌 𝑛 + 𝛿𝑡 + 𝛾𝑠 + (𝛾𝑠∗ 𝑡) + 𝑒𝑖𝑛𝑠𝑡 • OLS • Sample weights adjust for initial selection and attrition • Standard errors clustered on state of birth (Bertrand et al. 2004) 16
    • 17. Subgroups targeted by Medicaid • Low income (< 150% Poverty Line) – Average level in the early childhood period – Any AFDC in childhood: 40% • Placebo group: Moderate Income (175-300) • Estimates are robust to using the predicted probability of AFDC participation 17
    • 18. The effect of Medicaid in childhood on adult health and economic status Low Income Moderate Income Effect of Medicaid Exposure SE Effect of Medicaid Exposure SE Condition Index -0.36** 0.13 0.05 0.12 Sample Size 5,926 5,695 Mean of Y -0.01 -0.14 R2 0.21 .15 Economic Index -0.11 .21 -0.03 0.14 Sample Size 5,973 11,210 Mean of Y -0.24 -0.2 R2 0.33 .12 *p<0.1; **p<0.05; ***p<0.01 Models control for demographics, contextual controls, fixed effects 18
    • 19. Robustness • School and Hospital desegregation – Results robust to removing southern born non- whites • Selective migration – Robust to removing cases that moved states in the early childhood period • Contextual controls – Robust to removing contextual controls 19
    • 20. Summary • Exposure to Medicaid in early childhood appears to decrease the prevalence of adult chronic conditions – Point estimate implies a reduction in the probability of having one chronic condition of 0.4 • No evidence of improved economic status – Wide confidence intervals cannot exclude a large range of potential effects 20
    • 21. Limitations • I can not identify exact mechanisms – Economic resources – Health • But what services, specifically? • Can not determine if there is a critical period embedded in the early childhood years 21
    • 22. Implications • This study suggests that providing health insurance at early ages produces long term benefits for low income children – Improves individual health – Could produce down-stream savings • 5.5 million U.S. children uninsured in 2011 – The benefit of covering these children through recent expansions could go beyond contemporaneous measures 22
    • 23. Thank You Michel Boudreaux boudr019@umn.edu 23
    • 24. Backup 1 24
    • 25. Backup 2 25
    • 26. Backup 3 26
    • 27. Backup 4 Condition Index Economic Index Coef. SE Coef. SE Medicaid Exposure -0.03 0.08 -0.16 0.14 Exposure*Predicted Participation -0.88* 0.45 -0.07 1.10 Sample Size 18,094 38,442 Mean of Y -0.04 0.19 R2 0.12 0.25 *p<0.1; **p<0.05; ***p<0.01 Models control for demographics, contextual controls, fixed effects 27

    ×