Table 2.
Log-log specified multilevel regressions of deflated hospital expenditure (p-values in brackets)
| 2003 | 2004 | 2005 | 2006 | 2007 | 2008 | 2009 | 2010 | 2011 | 2012 | 2013 | 2014 | 2015 | Aggregate panel | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Intercept (fixed structural costs) | 9633.8 (0.000) | 7834.7 (0.000) | 9969.3 (0.000) | 13,271.0 (0.000) | 25,391.3 (0.000) | 30,900.0 (0.000) | 46,625.0 (0.000) | 37,512.9 (0.000) | 17,574.4 (0.000) | 10,836.5 (0.000) | 9387.4 (0.000) | 92,843.1 (0.000) | 15,636.9 (0.000) | 9798.2 (0.000) |
| Masculinity index | 0.57 (0.026) | 0.58 (0.066) | 0.53 (0.030) | 0.56 (0.013) | 0.61 (0.061) | 0.56 (0.038) | 0.52 (0.019) | 0.63 (0.131) |
0.63 (0.116) |
0.61 (0.090) | 0.77 (0.330) |
0.26 (0.063) | 0.54 (0.040) | 0.56 (0.000) |
| Ageing index | 0.95 (0.013) | 0.95 (0.031) | 0.96 (0.125) |
0.92 (0.001) | 0.94 (0.014) | 0.94 (0.020) | 0.97 (0.193) |
0.93 (0.016) | 0.91 (0.001) | 0.94 (0.024) | 0.99 (0.618) |
0.93 (0.019) | 0.96 (0.097) | 0.96 (0.000) |
| Over-ageing index 2 | 1.81 (0.108) |
2.11 (0.053) | 1.67 (0.182) |
2.41 (0.002) | 0.81 (0.368) |
1.15 (0.718) |
0.91 (0.637) |
1.24 (0.541) |
2.26 (0.003) | 2.37 (0.001) | 1.65 (0.070) | 2.20 (0.028) | 1.93 (0.013) | 1.96 (0.000) |
| Ln (morbidity) | 0.04 (0.381) |
0.02 (0.779) |
0.01 (0.881) |
0.12 (0.023) | 0.07 (0.213) |
0.07 (0.202) |
0.07 (0.203) |
0.13 (0.009) | 0.22 (0.002) | 0.11 (0.071) | −0.02 (0.783) |
0.24 (0.046 | 0.10 (0.068) | 0.07 (0.000) |
| Ln (outpatient case-day surgery) | 0.05 (0.006) | 0.06 (0.023) | 0.01 (0.638) |
−0.01 (0.563) |
0.02 (0.251) |
−0.01 (0.823) |
− 0.01 (0.652) |
0.01 (0.781) |
− 0.03 (0.244) |
0.00 (0.933) |
0.02 (0.631) |
0.01 (0.892) |
−0.06 (0.033) | 0.02 (0.018) |
| Ln (medical hospitalisations) | 0.33 (0.000) | 0.34 (0.000) | 0.41 (0.000) | 0.33 (0.000) | 0.31 (0.000) | 0.34 (0.000) | 0.27 (0.000) | 0.08 (0.260) |
0.29 (0.000) | 0.42 (0.000) | 0.42 (0.000) | 0.38 (0.000) | 0.36 (0.000) | 0.34 (0.000) |
| Ln (surgical hospitalisations) | 0.46 (0.010) | 0.42 (0.064) | 0.35 (0.104) | 0.25 (0.134) |
0.19 (0.289) |
0.14 (0.486) |
0.15 (0.395) |
0.77 (0.000) | 0.62 (0.002) | 0.47 (0.012) | 0.37 (0.080) | 0.30 (0.247) |
0.50 (0.005) | 0.30 (0.000) |
| Ln (surgical price of stay-complexity) | 0.10 (0.553) |
0.15 (0.487) |
0.20 (0.339) |
0.27 (0.095) | 0.33 (0.067) | 0.38 (0.061) | 0.46 (0.008) | −0-07 (0.736) |
− 0.16 (0.421) |
−0.05 (0.805) |
0.15 (0.468) |
0.00 (0.988) |
0.04 (0.836) |
0.23 (0.000) |
| Ln (average length of stay - medical hospitalisations) | 0.07 (0.448) | 0.19 (0.088) | 0.12 (0.327) |
0.13 (0.179) |
0.07 (0.495) |
0.00 (0.970) |
−0.07 (0.506) |
−0.19 (0.093) | 0.15 (0.200) |
0.33 (0.002) | 0.25 (0.025) | 0.30 (0.014) | 0.21 (0.015) | 0.12 (0.001) |
| Ln (average length of stay - surgical hospitalisations) | 0.29 (0.000) | 0.21 (0.021) | 0.23 (0.023) | 0.23 (0.008) | 0.31 (0.001) | 0.29 (0.007) | 0.21 (0.026) | 0.51 (0.000) | 0.34 (0.002) | 0.17 (0.101) | 0.13 (0.251) |
−0.09 (0.571) |
0.30 (0.003) | 0.20 (0.000) |
| Tertiary | 1.07 (0.001) | 1.08 (0.003) | 1.06 (0.039) | 1.08 (0.000) | 1.09 (0.000) | 1.07 (0.005) | 1.06 (0.023) | 1.06 (0.031) | 1.05 (0.049) | 1.07 (0.006) | 1.06 (0.015) | 1.04 (0.284) |
1.03 (0.200) |
1.06 (0.000) |
| Medical staff/functioning beds | 0.93 (0.541) |
0.87 (0.236) |
0.92 (0.498) |
1.15 (0.118) |
1.02 (0.852) |
0.99 (0.878) |
0.88 (0.186) |
1.14 (0.166) |
1.32 (0.009) | 1.33 (0.002) | 1.27 (0.032) | 1.04 (0.786) |
1.21 (0.040) | 1.11 (0.000) |
| Nursing staff/functioning beds | 1.19 (0.024) | 1.39 (0.000) | 1.38 (0.000) | 1.12 (0.039) | 1.21 (0.002) | 1.22 (0.003) | 1.30 (0.000) | 1.18 (0.014) | 1.21 (0.006) | 1.08 (0.248) |
1.05 (0.537) |
1.25 (0.032) | 1.12 (0.049) | 1.14 (0.000) |
| Medical professionals in training/medical professionals | 1.22 (0.014) | 1.11 (0.341) |
1.21 (0.094) | 1.24 (0.000) | 1.19 (0.084) | 1.25 (0.059) | 1.27 (0.046) | 1.18 (0.202) |
1.38 (0.007) | 1.38 (0.003) | 1.52 (0.000) | 1.38 (0.018) | 1.62 (0.000) | 1.29 (0.000) |
| VPC (Regions) Confidence interval 95% |
0.40 (0.20 0.63) |
0.28 (0.10 0.54) |
0.18 (0.04 0.46) |
0.48 (0.26 0.71) |
0.26 (0.08 0.55) |
0.33 (0.13 0.60) |
0.45 (0.23 0.69) |
0.28 (0.10 0.56) |
0.48 (0.26 0.70) |
0.53 (0.31 0.74) |
0.51 (0.29 0.72) |
0.41 (0.24 0.68) |
0.35 (0.15 0.61) |
0.42 (0.26 0.60) |
| Likelihood-ratio test of sigma(u) = 0 | 25.35 (0.000) | 12.84 (0.000) | 5.41 (0.000) | 25.43 (0.000) | 9.84 (0.001) | 17.58 (0.000) | 28.10 (0.000) | 11.59 (0.000) | 31.23 (0.000) | 29.58 (0.000) | 33.42 (0.000) | 26.11 (0.000) | 14.32 (0.000) | 531.24 (0.000) |
| Pseudo-adjusted R2 | 0.980 | 0.977 | 0.978 | 0.977 | 0.978 | 0.972 | 0.968 | 0.968 | 0.964 | 0.964 | 0.962 | 0.943 | 0.976 | 0.965 |
Partial beta coefficients in the Ln covariates can be directly interpreted as elasticities, measuring the relative change of health expenditure to variations in covariates as follows: a 1% variation in the Ln-covariate turns into a x% increase (decrease) in expenditure]; Beta coefficients in non-Ln covariates (Tertiary, Medical staff, Nursing staff and Medical professionals in training) have to be interpreted as follows: a 1 point increase (decrease) in the covariate turns into a x% increase (decrease) in expenditure. X equals the value of the coefficient