Table 3

Comparisons between quantitative MWS and corresponding NSHD variables

Difference

Scaled 95% limits

Intraclass

Pearson

MWS

NSHD

(MWS - NSHD)

of agreement

correlation

correlation


Variable (units)

N

mean (SD)

mean (SD)

mean (SD)

P*

lower to upper†

ICC (SE) ‡

r (SE)


Self-reported at recruitment in the MWS

Height (cm)

449

162.8 (6.3)

162.0 (5.7)

0.9 (2.7)

<0.001

-0.8 to 1.1

0.89 (0.01)

0.91 (0.01)

Weight (kg)

443

68.8 (13.1)

71.0 (14.1)

-2.2 (4.2)

<0.001

-0.7 to 0.4

0.94 (0.006)

0.95 (0.004)

Body mass index (kg/m2)

437

25.9 (4.7)

27.0 (5.2)

-1.1 (2.0)

<0.001

-1.0 to 0.5

0.90 (0.01)

0.92 (0.01)

Age at menarche (yr)

394

13.0 (1.4)

12.9 (1.0)

0.1 (1.2)

0.093

-2.1 to 2.3

0.54 (0.04)

0.57 (0.03)

Age at menopause (yr)

134

48.0 (4.5)

48.2 (4.2)

-0.3 (1.8)

0.091

-0.9 to 0.8

0.92 (0.01)

0.92 (0.01)

Self-reported at resurvey in the MWS

Birth weight (kg)

234

3.2 (0.7)

3.3 (0.5)

-0.1 (0.4)

<0.001

-1.8 to 1.4

0.75 (0.03)

0.78 (0.03)

Mother's height (cm)

294

160.4 (7.1)

161.4 (6.5)

-0.9 (5.2)

0.002

-1.7 to 1.4

0.70 (0.03)

0.71 (0.03)

Father's height (cm)

240

174.1 (7.6)

173.7 (7.9)

0.3 (6.3)

0.451

-1.5 to 1.6

0.67 (0.04)

0.67 (0.04)

Waist circumference (cm)

206

76.0 (9.3)

82.0 (10.6)

-6.0 (7.2)

<0.001

-1.9 to 0.8

0.59 (0.05)

0.74 (0.03)

Hip circumference (cm)

204

100.0 (8.3)

103.1 (8.5)

-3.2 (5.3)

<0.001

-1.6 to 0.9

0.74 (0.03)

0.80 (0.03)

Waist-to-hip ratio

197

0.76 (0.06)

0.79 (0.06)

-0.03 (0.06)

<0.001

-2.4 to 1.3

0.41 (0.06)

0.52 (0.05)

Waist-to-height ratio

204

0.47 (0.06)

0.51 (0.07)

-0.04 (0.04)

<0.001

-1.9 to 0.7

0.60 (0.05)

0.75 (0.03)


MWS: Million Women Study. NSHD: National Survey of Health and Development. N: numbers of women with non-missing values for both MWS and NSHD variables. r: Pearson correlation. SE: standard error.

* P-value for t-test of mean difference = 0.

† Lower to upper Bland-Altman limits of agreement divided by the SD of the NSHD data.

‡ ICC(1,1) in the notation of Shrout and Fleiss [26].

Cairns et al. BMC Medical Research Methodology 2011 11:7   doi:10.1186/1471-2288-11-7

Open Data