Is prenatal care really ineffective? Or, is the 'devil' in the distribution?

Area of Science:

• Perinatal Health
• Biostatistics
• Maternal Health

Background:

• Prenatal care is crucial for infant health, yet studies often report minimal effects.
• This discrepancy may arise from combining 'normal' and 'complicated' pregnancies in analyses.
• Existing research often overlooks the heterogeneity in pregnancy types.

Purpose of the Study:

• To investigate the differential effects of prenatal care on infant health outcomes.
• To determine if accounting for pregnancy complexity improves the estimation of prenatal care effectiveness.
• To re-evaluate the impact of prenatal care using advanced statistical modeling.

Main Methods:

• Utilized data from the National Maternal and Infant Health Survey (NMIHS).
• Employed a finite mixture model to differentiate between pregnancy types.
• Compared results with the standard 2SLS (Two-Stage Least Squares) approach.
• Conducted a Monte Carlo simulation to validate findings.

Main Results:

• The standard 2SLS method produced bimodal residuals and insignificant prenatal care coefficients.
• Finite mixture models demonstrated a substantial positive effect of prenatal care on 'normal' pregnancies.
• Ignoring 'complicated' pregnancies led to an underestimation of prenatal care's importance.

Conclusions:

• Prenatal care effectiveness is masked when 'normal' and 'complicated' pregnancies are analyzed together.
• Finite mixture models offer a more accurate assessment of prenatal care's impact on infant health.
• Future research should account for pregnancy heterogeneity to better understand prenatal care benefits.