. 2024 Jun 4;7(6):e2414709. doi: 10.1001/jamanetworkopen.2024.14709

Table 4. Risk of Autism Spectrum Disorders Among Children Born of Fathers Who Used Valproate During Spermatogenesis.

Comparison	Valproate-exposed children^a	Reference children^a	Autism spectrum disorder		Follow-up time, median (IQR), y		HR (95% CI)
Comparison	Valproate-exposed children^a	Reference children^a	Valproate-exposed children	Reference children	Valproate-exposed children	Reference children	Unadjusted	Adjusted^b
Main analysis^c	1336	1 234 017	34	24 479	10.5 (5.5-15.2)	10.5 (5.3-15.7)	1.31 (0.93-1.83)	0.92 (0.65-1.30)
Valproate dose response^d
High dose of valproate	715	1 234 017	13	24 479	10.6 (5.9-15.6)	10.5 (5.3-15.7)	0.90 (0.52-1.55)	0.66 (0.38-1.15)
Low dose of valproate	621	1 234 017	21	24 479	10.1 (4.9-14.6)	10.5 (5.3-15.7)	1.81 (1.18-2.77)	1.20 (0.78-1.86)
Sibling analyses^e	303	381	7	17	10.5 (5.1-14.5)	10.4 (6.5-15.4)	0.39 (0.10-1.47)	0.54 (0.14-2.12)
Restriction analysis 1^f	1052	11 308	27	298	10.7 (6.0-15.1)	8.5 (4.1-13.5)	0.80 (0.54-1.19)	0.85 (0.57-1.26)
Restriction analysis 2^g	828	7860	27^h	219	11.4 (6.9-15.6)	9.3 (4.5-14.2)	0.83 (0.54-1.28)	0.87 (0.56-1.33)
Active comparator analysis 1ⁱ^,^j	1336	1663	34	43	10.5 (5.5-15.2)	6.1 (2.9-10.1)	0.56 (0.36-0.89)	0.68 (0.41-1.13)
Active comparator analysis 2^k	1043	1043	21	31	8.4 (4.3-12.4)	8.5 (4.2-12.4)	0.68 (0.39-1.18)	0.76 (0.42-1.38)
Analysis with negative exposure control^l	1336	690	34	21	10.5 (5.5-15.2)	9.1 (4.4-14.8)	0.77 (0.45-1.33)	0.83 (0.45-1.54)

Abbreviation: HR, hazard ratio.

^{^a}

Shows the number for the whole cohort; 6 exposed and 8291 unexposed died, emigrated, got the outcome, or reached the end of follow-up before age 1 year (start of follow-up) and are not included in these analyses.

^{^b}

Adjusted for sex of the child, year of birth, paternal and maternal age at the time of the child’s birth, and paternal and maternal psychiatric diagnosis, psychotropic medication use, epilepsy diagnosis, and highest completed educational level at the time of the child’s birth (sibling analysis is adjusted for sex of the child, year of birth, and paternal and maternal age at the time of the child’s birth).

^{^c}

Valproate-exposed children compared with unexposed children.

^{^d}

Children exposed to a high dose and low dose of valproate during spermatogenesis compared with unexposed children.

^{^e}

Valproate-exposed children compared with unexposed paternal siblings (264 exposure-discordant sibling sets).

^{^f}

Valproate-exposed children of fathers with epilepsy compared with unexposed children of fathers with epilepsy.

^{^g}

Valproate-exposed children of fathers with epilepsy of unknown cause compared with unexposed children of fathers with epilepsy of unknown cause.

^{^h}

This number is valproate-exposed children of fathers with epilepsy, as the difference to valproate-exposed children of fathers with epilepsy of unknown cause is small (<5) and therefore cannot be shown.

^ⁱ

Valproate-exposed children compared with lamotrigine-exposed children.

^{^j}

There were 160 children whose fathers had filled prescriptions for both valproate and lamotrigine during spermatogenesis. In the analyses, these children were included in the group of children whose fathers had filled prescriptions for valproate during spermatogenesis.

^{^k}

Valproate-exposed children compared with lamotrigine-exposed children, equal number of exposed and unexposed per birth year.

^{^l}

Valproate-exposed children compared with children of fathers who filled prescriptions for valproate 2 years prior to the exposure period, but not during the exposure period.