TABLE 2. Coefficients and Variance Components from Mixed Effects Linear Growth Models for Health Spending Measures (natural log) in Latin America and the Caribbean, 26 countries, 1995–2009, N = 390
Graphic RepresentationTime Trend CoefficientConstantVariance in Time TrendVariance in InterceptCovariance (Time Trend, Intercept)Chi-squared for Covariance1
Health expenditure per capita (constant 2005 international dollars)Embedded Image0.052***5.575***0.0007***0.522*0.55
Health expenditure, total (% of GDP)Embedded Image0.008*1.764***0.0003***0.0856***0.0073***19.40***
Health expenditure, public (% of total health expenditure)Embedded Image0.0043.903***0.0002***0.0658**0.65
Health expenditure, public (% of government expenditure)Embedded Image-0.0042.508***0.0004***0.1318***−0.0027+3.03+
  • 1Likelihood ratio chi-square versus model that constrains variance components to be independent (df = 1). The preferred model according to the X2 test is presented in the table.

  • *** p < 0.001, ** p < 0.01, * p < 0.05, + p < 0.1