Childhood exposure to parental smoking and life-course overweight and central obesity

Figures & data

Table 1. Characteristics of the study population among the participants exposed and non-exposed to parental smoking.

	Exposure to parental smoking during 3–21 years		Exposure to parental smoking during 3–20 years
Participants	YFS		STRIP
	Exposed N = 1512–1595	Non-exposed N = 642–673	Exposed N = 282	Non-exposed N = 350
Sex (women), %	55.3	55.9	50.4	49.1
Own smoking until the age of 20/ 21 years, %^a	28.6 (1429)***	18.3 (618)***	26.2***	12.3***
Own smokingduring the follow-up in YFS, %^a	58.3***	41.4***	–	–
In the latest study visit
the prevalence of overweight/obesity, % (n)^b	58.7 (759)*	52.1 (272)*	20.4 (35)	15.1 (39)
the prevalence of central obesity, % (n)^c	36.6 (474)*	30.3 (159)*	17.0 (32)	9.5 (25)
Own educational level, %^d	5.4 / 72.3 / 22.4***	3.4 / 66.0 / 30.5***	–	–
Birth weight, g	3506 (546)	3551 (497)	3547 (536)	3607 (491)
Maternal age, years^e	26.3 (5.48)***	27.6 (5.41)***	29.8 (4.59)	30.2 (4.68)
Paternal age, years^e	28.9 (6.50)**	29.8 (6.12)**	31.6 (5.81)	32.2 (5.37)
Parental educational level, %^d	66.9 / 22.5 / 10.6*	62.6 / 22.7 / 14.6*	3.2 / 72.3 / 24.5***	2.6 / 53.1 / 44.3***

a Ever daily smoking during follow-up. Participants own smoking during 12–21 years in the YFS and 10–20 years in the STRIP.

b Body mass index ≥ 25kg/m² for adults and according to the Cole in children; exposed N = 1292, non-exposed N = 522 in YFS; exposed N = 188, non-exposed N = 262 in STRIP.

c In women >90cm/in men >100cm/in children <18 years a waist circumference to height ratio >0.50; exposed N = 1292, non-exposed N = 522 in YFS; exposed N = 188 non-exposed N = 262 in STRIP.

d Comprehensive school/secondary or non-academic education/academic education. The participant’s own highest educational level queried in 2001, 2007 or 2011 (during ages 24–49 years) in the YFS. The highest educational level of either mother or father queried at the child’s age of 6–30 years (in 1983–1992, YFS) and at the age of 1, 5 or 9 years (in 1990–2001, STRIP).

e At participant’s birth values are means (standard deviations) for continuous variables and percentages for categorical variables. P-values are from T-tests or Qhi-Square tests (for educational level) Boldface indicates statistical significance (*p < .05, **p < .01, ***p < .001).

Table 2. Exposure to parental smoking and the risk for overweight/obesity (Panel A), and central obesity (Panel B).

	Exposure to parental smoking during 3–21 years		Exposure to parental smoking during 3–20 years		Exposure to parental smoking prior to age 3 years
	YFS		STRIP		STRIP
Panel A
Risk of overweight/obesity^d	RR aged 9–49 years (y)	95% CI^a	RR aged 3–20 y	95% CI^b	RR aged 2–20 y	95% CI^c
Model 1	1.16	1.05–1.27**	1.67	1.17–2.39**	1.51	1.06–2.16*
Model 2	1.15	1.04–1.26**	1.61	1.11–2.34*	1.44	0.99–2.08
Model 3	1.13	1.02–1.24**	1.57	1.10–2.26*	1.38	0.96–1.97
Panel B
Risk of central obesity^g	RR aged 24–49 y	95% CI^a	RR aged 7–20 y	95% CI^e	RR aged 7–20 y	95% CI^f
Model 1	1.26	1.08–1.47**	1.68	1.12–2.51*	1.86	1.26–2.75**
Model 2	1.22	1.05–1.43*	1.52	1.02–2.25*	1.67	1.12–2.48*
Model 3	1.18	1.01–1.38*	1.45	0.98–2.15	1.58	1.06–2.33*

a Exposed N = 1522 and non-exposed N = 647.

b Exposed N = 233 and non-exposed N = 304.

c Exposed N = 186 and non-exposed N = 365.

d Body mass index ≥25 kg/m².

e Exposed N = 232 and non-exposed N = 302.

f Exposed N = 185 and non-exposed N = 363.

g in YFS: waist circumference >100cm for men and >90cm for women; in STRIP: waist-to-height ratio > 0.50.

RR = Risk Ratio; 95% CI = Confidence Interval.

Model 1 is adjusted for age and sex.

Model 2 is adjusted additionally for family socioeconomic status (SES) and own SES since the age of 24 years in YFS.

Model 3 is adjusted additionally for birth weight, own smoking status, parental ages, diet and physical activity.

Boldface indicates statistical significance (*p < .05, **p < .01); analyses conducted by Generalised Estimating Equations.

Table 3. Exposure to parental smoking during aged 3–21 years and skinfold thickness^a values in childhood and young adulthood.

YFS	Exposed N = 1621	Non-exposed N = 682
Subscapular (mm)^b
Model 1	11.0 (0.13)	10.5 (0.19)
Model 2	10.9 (0.15)	10.4 (0.20)
Model 3	11.0 (0.18)	10.5 (0.21)
Triceps (mm)^b
Model 1	11.4 (0.12)*	10.9 (0.17)*
Model 2	11.3 (0.13)*	10.8 (0.18)*
Model 3	11.2 (0.15)*	10.8 (0.19)*
Biceps (mm)^b
Model 1	6.4 (0.07)**	6.0 (0.10)**
Model 2	6.4 (0.08)**	6.0 (0.11)**
Model 3	6.3 (0.11)**	6.0 (0.13)**

a Measured by Harpenden callipers.

b Age 9–24 years.

Values are means (standard error).

Model 1 is adjusted for age and sex.

Model 2 is adjusted additionally for family socioeconomic status (SES and own SES since the age of 24 years in YFS.

Model 3 is adjusted additionally for own smoking status, birth weight, parental ages, diet and physical activity.

Boldface indicates statistical significance (*p < .05, **p < .01), analyses conducted for log-transformed values and p-values are from Repeated Mixed Model.

Table 4. Exposure to parental smoking and adolescence^a abdominal fat thickness (at xiphoid process/at navel) using ultrasound.

STRIP	Exposure to parental smoking during 3–20 years		Exposure to parental smoking prior to age of 3 years
	Exposed N = 226	Non-exposed N = 297	Exposed N = 182	Non-exposed N = 355
at xiphoid process (mm)
Model 1	23.7 (0.50)	22.3 (0.44)*	24.1 (0.56)	22.3 (0.40)*
Model 2	24.1 (0.85)	22.9 (0.78)	24.5 (0.93)	23.0 (0.76)*
Model 3	24.6 (0.88)	23.7 (0.84)	25.1 (0.95)	23.8 (0.81)
at navel(mm)
Model 1	21.7 (0.73)	18.9 (0.64)*	21.8 (0.82)	19.1 (0.59)*
Model 2	21.4 (1.15)	20.3 (1.09)	22.7 (1.23)	20.5 (1.06)
Model 3	23.1 (1.18)	21.3 (1.16)	23.4 (1.25)	21.6 (1.12)

a Age 13, 15, 17, 19.

Values are means (standard error).

Model 1 is adjusted for age and sex.

Model 2 is adjusted additionally for family socioeconomic status (SES) and own SES since the age of 24 years in YFS.

Model 3 is adjusted additionally for own smoking status, birth weight, parental ages, diet and physical activity.

Boldface indicates statistical significance (*p < .05); analyses conducted for log-transformed values by Repeated Mixed Model.