Table 6.
Efficiency of multi-attribute utility measures to detect differences in postnatal depression
| Measure | Categorisation of postnatal depression risk score | Utility score | t-testa | Relative efficiencyb | ROC curve | |||
| Mean | (SD) | t-statistic | p-value | Areac | 95% CI | |||
| All women (n = 493) | ||||||||
| EQ-5D | EPDS score < 13 | 0.885 | (0.147) | 4.332 | <0.001 | 1.000 | 0.696* | (0.615, 0.777) |
| EPDS score ≥ 13 | 0.738 | (0.244) | ||||||
| SF-6D | EPDS score < 13 | 0.830 | (0.129) | 7.008 | <0.001 | 2.617 | 0.767* | (0.697, 0.837) |
| EPDS score ≥ 13 | 0.696 | (0.132) | ||||||
| EQ-5D | EPDS score < 10 | 0.896 | (0.142) | 5.404 | <0.001 | 1.000 | 0.679* | (0.619, 0.739) |
| EPDS score ≥ 10 | 0.780 | (0.208) | ||||||
| SF-6D | EPDS score < 10 | 0.843 | (0.125) | 8.192 | <0.001 | 2.298 | 0.749* | (0.696, 0.802) |
| EPDS score ≥ 10 | 0.724 | (0.133) | ||||||
| Women for whom both utility scores were between 0.296 and 1.0 (n = 481) | ||||||||
| EQ-5D | EPDS score < 13 | 0.891 | (0.130) | 3.699 | <0.001 | 1.000 | 0.664* | (0.579, 0.750) |
| EPDS score ≥ 13 | 0.806 | (0.152) | ||||||
| SF-6D | EPDS score < 13 | 0.833 | (0.126) | 6.509 | <0.001 | 3.096 | 0.760* | (0.685, 0.835) |
| EPDS score ≥ 13 | 0.708 | (0.123) | ||||||
| EQ-5D | EPDS score < 10 | 0.901 | (0.128) | 5.110 | <0.001 | 1.000 | 0.662* | (0.600, |
| EPDS score ≥ 10 | 0.821 | (0.141) | 0.724) | |||||
| SF-6D | EPDS score < 10 | 0.846 | (0.122) | 7.800 | <0.001 | 2.330 | 0.743* | (0.688, 0.798) |
| EPDS score ≥ 10 | 0.735 | (0.126) | ||||||
EPDS denotes Edinburgh Postnatal Depression Scale. SD denotes standard deviation. ROC denotes receiver operating characteristic. CI denotes confidence interval.
a Not assuming equality of variance as Levene test showed statistically significant differences in variances between self-reported health status groups.
b Relative efficiency statistic is referenced to 1.0 for the EQ-5D measure. A value higher than 1.0 indicates that the SF-6D is more efficient than the EQ-5D in detecting differences between women in terms of their self-reported health status.
c Area under receiver operating characteristic (ROC) curves; * p < 0.05 indicates that area under the ROC curve was statistically significantly greater than 0.5 and that measure has discriminatory power.