Table 2.

Multivariable Cox proportional hazards regression on the association between childhood and adolescence factors and multiple sclerosis

Variable	Multivariable Cox proportional hazards modela
	HRb	95% CIb
Sex
Female	Ref
Male	0.48	0.41 – 0.56
Number of siblings
Only child	Ref
1-2 sibling(s)	0.93	0.73 – 1.17
≥ 3 siblings	0.81	0.61 – 1.09
Premature birth (> 4 weeks before due date)
No	Ref
Yes	0.94	0.61 – 1.47
Cesarean section
No	Ref
Yes	1.10	0.80 – 1.52
Birth weight
Low	0.99	0.74 – 1.33
Average	Ref
High	1.14	0.90 – 1.45
Ever breastfed
No	Ref
Yes, ≤ 4 months	1.02	0.79 – 1.31
Yes, > 4 months	0.88	0.69 – 1.11
Contact with pets and/or livestock during childhood
No	Ref
Yes	1.04	0.69 – 1.58
Attended daycare
No	Ref
Yes, 1st attendance at age 3-6 years	0.93	0.77 – 1.12
Yes, 1st attendance at age 1- < 3 year(s)	0.77	0.58 – 1.03
Yes, 1st attendance at age < 1 year	0.97	0.63 – 1.50
Weight at the age of 10 years compared to peers
Lower	0.86	0.70 – 1.05
Average	Ref
Higher	0.93	0.73 – 1.19
BMIb at the age of 18 years (kg/m2)
Underweight (< 18.5)	1.12	0.89 – 1.42
Normal weight (18.5 - < 25)	Ref
Overweight (25 - < 30)	2.03	1.41 – 2.94
Obesity (≥ 30)	1.89	1.02 – 3.48
Childhood Traumac (per 5 units)	1.08	0.95 – 1.24

^aAdjusted for education and migration status, stratified by birth year (categorized as: < 1955, 1955–1964, 1965–1974, 1975–1984, ≥ 1985)

^bHR Hazard Ratio, CI Confidence Interval, BMI Body Mass Index

^cAssessed with the Childhood Trauma Screener (5 – 25 points)