How ADHD affects pregnancy outcomes in women and their babies

23 March 2018
Women on medication for ADHD should not stop treatment without consulting a doctor, says Dr Alison Poulton.

Women on medication for attention deficit hyperactivity disorder (ADHD) should not stop treatment without consulting a doctor, say the authors of a new study examining pregnancy outcomes among women treated with ADHD medications.

Published today in CNS Drugs the study of 5,056 Australian (NSW) women and their newborns is the largest of its kind to assess the impact of ADHD and commonly prescribed stimulant medications (dexamphetamine or methylphenidate) on mothers and their babies born between 1994 and 2012.

The study led by the University of Sydney, assessed the impacts of ADHD and its treatment on outcomes including: spontaneous labour (normal vaginal delivery without medical intervention), caesarean delivery, active neonatal resuscitation, premature birth (before 37 weeks), and a 1-minute Apgar score less than seven (7 to 10 is normal).

The researchers compared these outcomes in a control group, that is, the babies and pregnancy outcomes of women who did not have ADHD and who were not treated with medications for ADHD.

This meant the researchers could determine whether ADHD, its treatment with stimulant medication, or both might adversely affect pregnancy outcomes. 


Women diagnosed with ADHD at any stage were 20 to 30 per cent more likely to have a caesarean delivery, and their babies had a similarly increased rate of needing support to start breathing or admission to a neonatal unit.

These increases even affected women (and their babies) who were not diagnosed or treated for ADHD until after giving birth, suggesting that ADHD itself is a significant predictor of adverse pregnancy and perinatal outcomes.

Women who were diagnosed with ADHD and treated with stimulant medications during childhood or pregnancy also had an elevated risk for pre-eclampsia (high blood pressure, protein in the urine and swelling), having a premature birth, or their baby having a lower than normal Apgar score (less than 7).

“These adverse outcomes were seen even in women not yet treated for ADHD and in women who stopped taking stimulant medication several years before becoming pregnant, suggesting that the increased risk isn’t caused by medication,” said the study’s lead author Alison Poulton, a paediatrician and senior lecturer at the University of Sydney.

“Based on the evidence of this study, the potential benefit of ceasing treatment for ADHD during pregnancy may be limited.”

About ADHD

ADHD is more prevalent in childhood but may persist into adult life or be first diagnosed in adults. It is estimated that ADHD affects 4.5 per cent of the adult population globally.

In New South Wales (NSW) adult ADHD became increasingly recognised and treated with stimulants during the last decade of the 20th century, with statewide prescription records showing a 27-fold increase from 1993 to 2003. The proportion of women among those treated with stimulants increased from 17 to 36 per cent over the same period. In NSW, Dexamphetamine was the prescribed drug in 60 per cent of cases.

Study details

The study linked and analysed data from two datasets: the NSW Pharmaceutical Drugs of Addiction System and the NSW Perinatal Data Collection (PDC). Linkage of these two datasets identified a cohort of 5,056 women treated with stimulants for ADHD from 1982-2012 who gave birth between 1994 and 2012. An untreated cohort of women who did not have NSW prescription records for stimulant medication were recruited by selecting at random from the PDC five women matching each treated woman by maternal age and infant year of birth.

Of 5056 women treated for ADHD with stimulant medication, 3351 (66.3 per cent) had stimulant treatment documented before the index pregnancy but not within 1 year before the expected date of delivery, 175 (3.5 per cent) had stimulant treatment before and possibly during pregnancy, and 1530 (30.2 per cent) had no stimulant treatment until after the index pregnancy.

Women in each treatment category were separately compared with all untreated women and odds ratios calculated for the perinatal outcomes using unconditional multiple logistic regression while adjusting for matching variables (baby’s year of birth and mother’s age) and the potential confounding effects of maternal age, parity, multiple pregnancy, cigarette smoking, pre-existing diabetes and hypertension.

Related articles