Table 4.

Health outcomes measured at six months according to allocation to extra postnatal support or control group. Figures are mean (SD) scores^* unless stated otherwise

	Intervention			Control		Mean difference (95% CI)	P value†
	No of women	Mean (SD)		No of women	Mean (SD)
SF-36 physical functioning	258	89.8 (16.8)		230	91.2 (15.1)	−1.5 (−1.2 to 4.2)	0.23
Social functioning	257	83.6 (22.0)		233	84.0 (23.6)	−0.4 (−4.7 to 4.0)	0.36
Role limitation—physical	259	80.2 (32.5)		229	82.1 (32.6)	−1.9 (−7.2 to 3.5)	0.34
Role limitation—emotional	257	82.4 (31.7)		228	79.5 (35.5)	2.8 (−3.4 to 8.3)	0.57
Mental health	254	72.8 (17.3)		227	74.0 (17.5)	−1.2 (−4.3 to 1.8)	0.30
Vitality	252	56.1 (21.1)		228	54.7 (21.3)	1.4 (−2.5 to 5.1)	0.49
Pain	256	81.0 (22.7)		232	82.8 (23.2)	−1.9 (−5.8 to 2.2)	0.22
General health perception	255	76.0 (19.4)		230	76.9 (20.4)	−0.9 (−4.5 to 2.7)	0.38
Health change	259	67.4 (23.0)		232	64.8 (24.2)	2.6 (−1.6 to 6.7)	0.26
Duke functional social support	240	17.1 (6.8)		225	16.7 (7.3)	0.4 (−0.9 to 1.8)	0.29
Edinburgh postnatal depression scale	252	6.6 (5.1)		229	6.7 (5.6)	−0.1 (−1.0 to 1.9)	0.73
No who breast fed only	260	33		233	28		0.86§
No who fed formula milk only	260	208		233	185
No who mixed feeding	260	19		233	20

^*For all measures higher scores indicate better health except for Duke functional social support and Edinburgh postnatal depression scale where higher scores indicate poorer health. 

^†All P values are from Mann-Whitney test, which compares distributions of two groups, except ^§χ²=0.30 on 2 df, P=0.86. Bootstrap 95% confidence intervals calculated for characteristic of distributions (for example, mean difference) by centile method. Groups may have differences in distributions but similar characteristics (for example, means).