Table 4. Adjusted expenditures by category for publicly insured children with a chronic condition (N = 6,060).
Any expenditure | Expenditure amount if any | Combined incremental effect | ||||
---|---|---|---|---|---|---|
|
|
|
||||
OR | 95 % CI | β | 95 % CI | IE ($) | 95 % CI | |
Total expenditures | 0.71* | (0.51–0.98) | 0.12 | (−0.08–0.31) | 281.67* | (270.40–292.94) |
ED expenditures | 1.26* | (1.03–1.54) | −0.18 | (−0.39–0.02) | 0.86* | (0.76–0.96) |
Office visit expenditures | 0.85 | (0.67–1.09) | 0.14 | (−0.07–0.34) | 54.30* | (53.28–55.32) |
Inpatient hospital | ||||||
Expenditures | 1.07 | (0.85–1.34) | –0.07 | (−0.50–0.36) | −40.10* | (−41.35 to –38.80) |
Outpatient expenditures | 0.95 | (0.76–1.20) | –0.09 | (−0.59–0.41) | –30.16* | (–30.62 to –29.71) |
Prescription | ||||||
Expenditures | 0.96 | (0.78–1.17) | 0.03 | (−0.15–0.21) | 12.42* | (11.97–12.86) |
Dental expenditures | 0.81* | (0.66–0.98) | 0.16 | (−0.08–0.40) | 4.54* | (4.42–4.66) |
The combined incremental effect was calculated by combining both parts of the model with the formula IE = [Pr(y > 0lx = 1) -Pr(y >0|x = 0)] 9 E[y|y>0] + Pr(y>0) × (E[y|y>0, x = 1] – E[y|y>0, x = 0), where y is the expenditure, x is whether or not the mother reported depressive symptoms, Pr is the probability from the logit model and E[y] is the expected value of expenditures from the GLM model. The 95 % CI was estimated by bootstrapping for 1,000 repetitions
p <.05