Skip to main content
. 2019 Mar;224:127–137. doi: 10.1016/j.socscimed.2019.01.040

Table 4.

Decomposition of changes between 12 and 48 M waves in EI/CI and HI into mobility indices, by outcome.

PHI Medicine Use PHI Expenditures Medicine Expenditures Total Exp.
ΔCI/ΔEI −0.018 0.019 −0.019 0.071*** 0.042**
(0.022) (0.031) (0.022) (0.026) (0.018)
MH 0.044** −0.010 0.049*** −0.052** −0.019
(0.018) (0.029) (0.019) (0.024) (0.017)
p 0.201** 0.016 −6.768 0.168 0.117
(0.099) (0.047) (102.487) (0.104) (0.134)
q 0.218*** −0.622*** −0.007 −0.311*** −0.166***
(0.036) (0.053) (0.058) (0.069) (0.047)
MR 0.026 0.009 0.030** 0.019 0.023*
(0.016) (0.018) (0.015) (0.015) (0.013)
ΔHI −0.026 0.005 −0.019 0.062** 0.038**
(0.022) (0.030) (0.022) (0.025) (0.018)
MHIH 0.056*** 0.002 0.050*** −0.039* −0.014
(0.018) (0.029) (0.019) (0.023) (0.017)
MHIR 0.029* 0.007 0.031** 0.022 0.024*
(0.016) (0.017) (0.015) (0.016) (0.013)
N 2638 2638 1877 3145 2509

Notes: *, **, *** denotes p-values less than 10, 5, and 1%. ΔEI (for binary variables, PHI and Medicine Use), ΔCI, and ΔHI are respectively variations of EI, CI, and HI. MH and MR are income-related health care mobility and health-related income mobility for each of those cases. p represents the progressivity index and q is the factor scale. The standard errors of indices are in parentheses and were generated using bootstrapping with 300 replications. The subscript term HI represents the statistics for their respective cases.