Table 3

Lightest participants (body-weight < 60 kg) initiating smoking at ≤ 20 years: relative risk of hospital-treated pneumonia by vitamin E supplementation

Intervention

Effect of vitamin E

Vitamin E

No vitamin E

Subgroup

No. of participants

No. of cases

No. of cases

RR (95% CI)*

P-value for interaction

All

935

41

25

1.84 (1.11–3.0)

BMI †

< median

467

25

17

1.87 (0.99–3.5)

0.8

≥ median

468

16

8

2.12 (0.90–5.0)

Height †

< median

461

16

10

1.91 (0.85–4.3)

0.9

≥ median

474

25

15

1.86 (0.97–3.6)

Dietary vitamin E †

< median

467

15

15

1.30 (0.63–2.7)

0.2

≥ median

468

26

10

2.70 (1.30–5.6)

Dietary vitamin C †

< median

467

15

16

0.98 (0.48–2.0)

0.026

≥ median

468

26

9

3.48 (1.61–7.5)

Residual of fruit, vegetables, berries †

< median

467

19

15

1.53 (0.76–3.1)

0.6

≥ median

468

22

10

2.27 (1.06–4.9)

β-Carotene supplementation

No

476

23

12

2.20 (1.06–4.5)

0.7

Yes

459

18

13

1.62 (0.78–3.4)

* Proportional hazards regression model comparing participants who received vitamin E with those who did not. The regression models were adjusted for age, baseline smoking, intake of coffee and alcohol, BMI and employment. Participants with missing data on confounders (n = 119) are excluded from this table. The sizes of all compared intervention groups are the same within 25% accuracy. RR, risk ratio; CI, confidence interval.

† The medians for the light-weight group are: weight 57.0 kg; BMI 20.0 kg/m²; height 168 cm; dietary vitamin E intake 9.1 mg/day; dietary vitamin C intake 75.3 mg/day; residual of fruit, vegetable, and berry intake -2.9 g/day.

Hemilä and Kaprio Nutrition Journal 2008 7:33   doi:10.1186/1475-2891-7-33