Table 1

Example of calculation: Estimated effect of time trend in reported intake of pasta 1986–99 on mean hip-circumference in men:

(ii) Δlevel1–8


Reported intake frequency

level

Proportion 19991

Proportion 19861

(i) partial regression coefficientsβlevel1–8 2

(iii) Estimated effect on mean HC: Σ(Δi * βi) i = level 1–8


Never

1

(3/631 -

7/777)*

-5.7

= 0.024 mm

1/year

2

(19/631 -

59/777)*

-5.4

= 0.247 mm

1–3/month

3

127/631

408/777

reference category3

1/week

4

(243/631 -

262/777)*

-0.2

= -0.010 mm

2–3/week

5

(199/631 -

40/777)*

4.3

= 1.135 mm

4–5/week

6

(31/631 -

1/777)*

7.8

= 0.373 mm

1/day

7

(8/631 -

0/777)*

10.6

= 0.134 mm

2–3/day

8

(1/631 -

0/777)*

13.4

= 0.021 mm


total

1,925 mm


1number of valid answers reporting specific number of intakes divided by total number of valid answers in respective survey-year.

2intake-level specific partial β-coefficients for association with hip-circumference estimated from all valid answers '86+'90+'94+'99.

3Level 3 is the reference category and therefore not included in the calculation.

Krachler et al. Nutrition Journal 2006 5:34   doi:10.1186/1475-2891-5-34

Open Data