Association of Attention-Deficit/Hyperactivity Disorder With Teenage Birth Among Women and Girls in Sweden

IMPORTANCE Attention-deficit/hyperactivity disorder (ADHD) is associated with a plethora of adverse health outcomes throughout life. While Swedish specialized youth clinics have carefully and successfully targeted risk of unplanned pregnancies in adolescents, important risk groups, such as women and girls with ADHD, might not be identified or appropriately assisted by these interventions. OBJECTIVES To determine whether women and girls with ADHD are associated with increased risk of teenage birth compared with their unaffected peers and to examine the association of ADHD with risk factors for adverse obstetric and perinatal outcomes, such as smoking, underweight or overweight, and substance use disorder. DESIGN, SETTING, AND PARTICIPANTS This nationwide cohort study included data from 6 national longitudinal population-based registries in Sweden. All nulliparous women and girls who gave birth in Sweden between January 1, 2007, and December 31, 2014, were included. Data analyses were conducted from October 7, 2018, to February 8, 2019. EXPOSURES Women and girls treated with stimulant or nonstimulant medication for ADHD (Anatomic Therapeutic Chemical classification code N06BA) in the Swedish Prescribed Drug Register between July 1, 2005, and December 31, 2014. MAIN OUTCOMES AND MEASURES Maternal age at birth. Secondary outcome measures were body mass index, smoking habits, and psychiatric comorbidities. RESULTS Among 384 103 nulliparous women and girls aged 12 to 50 years who gave birth between 2007 and 2014 included in the study, 6410 (1.7%) (mean [SD] age, 25.0 [5.5] years) were identified as having ADHD. The remaining 377 693 women and girls without ADHD (mean [SD] age, 28.5 [5.1] years) served as the control group. Teenage deliveries were more common among women and girls with ADHD than among women and girls without ADHD (15.3% vs 2.8%; odds ratio [OR], 6.23 [95% CI, 5.80-6.68]). Compared with women and girls without ADHD, those with ADHD were more likely to present with risk factors for adverse obstetric and perinatal outcomes, including smoking during the third trimester (OR, 6.88 [95% CI, 6.45-7.34]), body mass index less than 18.50 (OR, 1.29 [95% CI, 1.12-1.49]), body mass index more than 40.00 (OR, 2.01 [95% CI, 1.60-2.52]), and alcohol and substance use disorder (OR, 20.25 [95% CI, 18.74-21.88]). CONCLUSIONS AND RELEVANCE This study found that women and girls with ADHD were associated with an increased risk of giving birth as teenagers compared with their unaffected peers. The results suggest that standard of care for women and girls with ADHD should include active efforts to prevent teenage pregnancies. JAMA Network Open. 2019;2(10):e1912463. doi:10.1001/jamanetworkopen.2019.12463 Key Points Question Is attention-deficit/ hyperactivity disorder (ADHD) associated with increased risk of teenage birth? Findings This nationwide cohort study of 384 103 women and girls in Sweden who gave birth for the first time between 2007 and 2014, including 6410 women and girls with ADHD, found that teenage deliveries occurred at a significantly higher rate among women and girls with ADHD than among those without ADHD (15.2% vs 2.8%). Meaning This study suggests that women and girls with ADHD may have an increased risk of giving birth as teenagers compared with their unaffected peers. Author affiliations and article information are listed at the end of this article. Open Access. This is an open access article distributed under the terms of the CC-BY License. JAMA Network Open. 2019;2(10):e1912463. doi:10.1001/jamanetworkopen.2019.12463 (Reprinted) October 2, 2019 1/11 Downloaded From: https://jamanetwork.com/ by a Uppsala University User on 12/19/2019


Introduction
Attention-deficit/hyperactivity disorder (ADHD) is a prevalent neurodevelopmental disorder characterized by symptoms of hyperactivity, impulsivity, and inattention, and it is associated with a plethora of adverse health outcomes throughout life. 1,2 The worldwide prevalence of childhood and adolescent ADHD is estimated to be approximately 5%. 3,4 Importantly, children with ADHD have 2-fold the likelihood of mortality, 2-fold again among individuals whose symptoms persist into adulthood, which is associated with a significantly reduced estimated life expectancy. 5,6 Adolescents with ADHD are at increased risk of externalizing and risk-taking behaviors compared with their unaffected peers and more often engage in risky sexual behavior, such as earlier initiation of sexual activity and more sexual partners. 7,8 Consequently, they are at risk of sexually transmitted diseases and unplanned pregnancies. 9,10 Most previous studies exploring the risk of teenage pregnancies associated with ADHD have been conducted in clinical settings using selfreported measures. The clinical context may render outcomes vulnerable to low power and with limited generalizability to community populations. Furthermore, studies using self-reported measures may be sensitive to recall bias. However, a 2017 Danish nationwide cohort study 5 that addressed these limitations found increased likelihood of teenage parenthood among individuals with ADHD compared with those without.
Teenage pregnancies are associated with several long-and short-term adverse outcomes for both parents and children. Young parents are at risk of low educational attainment, single habitation, and use of public assistance. 11,12 Risks for the children include perinatal morbidity and mortality, low socioeconomic status, and low quality of life. [13][14][15][16] In Sweden, teenage birth rates have decreased from 15.3% of all births in 1973 to 2.4% in 2014, 17 one of the lowest rates internationally. 18,19 To our knowledge, the prevalence of teenage pregnancies in women and girls with ADHD has not been fully explored from a clinical perspective. Given the easy access to counseling and contraception, Sweden represents an ideal setting for investigating teenage births in women and girls with ADHD. Independent of socioeconomic status, all women and girls with ADHD have access to adequate contraception and counseling. Thus, this large-scale epidemiological study was designed to explore the prevalence of birth in young women and teenage girls with ADHD and to address modifiable risk factors associated with adverse obstetric and perinatal outcomes, such as smoking, body mass index (BMI), and substance use disorder in these women and girls. 6

Methods
This nationwide cohort study was based on data from 6 Swedish national population-based registries. The personal identity number assigned to every Swedish citizen at birth or immigration facilitated information linkage across registries. 20 The Swedish National Board of Health and Welfare provided data from the Swedish Medical Birth Register, the Patient Register, the Swedish Prescribed Drug Register, and the Cause of Death Register. Statistics Sweden provided data from the Education Register and the Total Population Register.
The Medical Birth Register includes 98% of all births in Sweden, including prospectively collected clinical variables, demographic data, and information on reproductive history, as well as complications during pregnancy, birth, and the neonatal period. 21 Information was compiled from the standardized antenatal care records, at the first antenatal visit (conducted in approximately gestational week 9), and in gestational week 32 for smoking habit. 22 Information on year and month of birth and gestational length at birth was collected from standardized birth records.

The Patient Register includes information on dates of hospital admissions and International
Statistical Classification of Diseases and Related Health Problems, Tenth Revision (ICD-10) 23 diagnoses, with full national coverage since 1987. 24 The Patient Register has also covered specialized

Outcomes
The Medical Birth Register was used for information on maternal age at birth, height, weight, and smoking habits. Age at birth was categorized as teenagers (women and girls aged <20 years) and nonteenagers (women aged Ն20 years We collected information on psychiatric comorbidities from the Patient Register between

Maternal Characteristics
Data on maternal education in 2014 were collected from the Education Register. Maternal education was categorized as less than 10 years completed, 10 to 12 years completed (ie, high school), and more than 12 years completed (ie, college). Data on maternal country of birth were collected from the Total Population Register. Maternal country of birth was classified as Nordic (ie, Denmark, Finland, Iceland, Norway, or Sweden), European, or other.

Statistical Analysis
Characteristics of the population were described according to exposure with absolute and relative frequencies. Logistic regression models were used to estimate the magnitude of the associations between age at first pregnancy, risk factors for adverse obstetric and perinatal outcomes, psychiatric comorbidities, and ADHD diagnosis, presented as odds ratios (ORs) with 95% CIs. Missing data were excluded. As the emphasis in this study was on the total burden of disease, no adjustments were made. In addition, we refrained from any adjustment on any of the outcomes by ADHD treatment during the year preceding pregnancy, as we would not be able to ascertain whether participants had stopped taking their medication when planning a pregnancy or not.
Furthermore, differences in risks factors for adverse obstetric and perinatal outcomes (ie, smoking, underweight, overweight, obesity, and substance use disorder) were explored in distinct age-at-birth subgroups among women and girls with ADHD (<20 years vs Ն20 years). All analyses were performed using SPSS statistical software version 25.0 (IBM). P values were 2-tailed, and statistical significance was set at less than .05.

Results
The total cohort included 384 103 women and girls, including 6410 women and girls with ADHD compared with 14.5% of women and girls without ADHD, and 93.2% of women and girls with ADHD were born in Nordic countries compared with 79.2% of women and girls without ADHD.

Discussion
This population-based cohort study examined the association of ADHD with age at first childbirth and associated medical and psychiatric risks. Our results showed an increased likelihood for teenage childbirth in women and girls with ADHD. Pregnant women and girls with ADHD presented with a number of medical and psychiatric comorbidities, among which substance use disorder was the most common.
Although teenage pregnancies are a rare occurrence in women and girls in Sweden with or without ADHD, as suggested by an overall rate of teenage deliveries of 3.0% in this study, women and girls with ADHD were associated with a 6-fold increased risk for teenage birth compared with women and girls without ADHD and contributed 8.4% to all teenage births. This is evident despite the widespread availability of contraception in Sweden. Becoming a mother at such early age is associated with long-term adverse outcomes for both women and their children. [11][12][13][14][15][16] Consequently, our findings argue for an improvement in the standard of care for women and girls with ADHD, including active efforts to prevent teenage pregnancies and address comorbid medical and psychiatric conditions. In addition, antenatal care should focus on adequate measures to reduce effect of obstetrics risk factors in these women and girls. Unfortunately, owing to a lack of understanding and specific research addressing sex differences, ADHD in women and girls is still underrecognized, misdiagnosed, and, once appropriately diagnosed, suboptimally treated. 28 Some important but unexplored hypotheses that may explain our results are that women and girls with ADHD receive inadequate contraceptive counseling, inadequately respond to counseling, fail to access or act on counseling, or experience more adverse effects from hormonal contraceptives.
As to contraceptive counseling, Swedish youth clinics have made counseling and contraception easy to access at low cost for this population. However, collaboration between psychiatric care clinics for youths and specialized youth clinics is needed to specifically address and provide adequate care, including contraception, for women and girls with ADHD. Thus, it is possible that a lack of transdisciplinary knowledge and a subsequent team effort failure may be associated with the high proportion of teenage births among women and girls with ADHD. However, even if provided with such assistance, women and girls with ADHD may be less likely to act on or implement such counseling owing to their disorder and its associated deficits in self-regulation.
Adverse mental health effects from hormonal contraceptives are increasingly reported by young users, 29 and an increasing proportion of younger women and girls abstain from hormonal contraceptives owing to a fear of future adverse effects. 30 Previous placebo-controlled randomized clinical trials have suggested minor mood disturbances with combined hormonal contraceptive use in healthy women, 31-33 especially among women with a history of mental health conditions. 34 As women with chronic psychiatric conditions are excluded from most clinical trials on hormonal contraceptives, these studies provide limited guidance in selecting contraceptives most suitable for this group. Further studies on contraceptives in women and girls with ADHD are needed.
Our study found that young women and girls with ADHD were more likely to present with a number of risk factors for adverse obstetric and perinatal outcomes, such as underweight and smoking. Our findings suggest that delaying first childbirth until after age 20 years may be advantageous for women with ADHD in terms of the risks associated with smoking and underweight for the offspring. This is in line with the current conceptualization of ADHD as a neurodevelopmental disorder with delayed brain maturation associated with age-inappropriate symptoms of disinhibiting, risk taking, and impulsivity. 35 However, while our findings may imply that the obstetric risk factor profile is somewhat normalized with increased age at first birth, this may also be due to reverse causation, ie, women and girls with ADHD with a lower symptom load and high functioning may simply choose to delay their first birth.
Pharmacological ADHD treatment in the year preceding the first pregnancy was not frequent in any of the age groups. Stimulant drug treatment has been associated with a decrease of core symptoms and adverse outcomes associated with ADHD. 36 Indeed, studies controlling for confounding by indication by using individuals as their own comparisons during periods with and without treatment suggest that stimulant medication may reduce or even ameliorate relevant adverse outcomes, such as educational failure, 37,38 and decrease risk for unwanted outcomes   associated with ADHD, such as substance use disorder, 39 motor vehicle crashes, 40 suicidal behavior, 41 and criminal behavior. 42 While it may be hypothesized that not using stimulant medication is associated with increased risk of unplanned pregnancy, this study cannot confirm any such association. Low rates of ADHD drug treatment may simply be owing to some women discontinuing psychotropic medication when planning for pregnancy.

Limitations
Our study has limitations. The ascertainment of ADHD diagnosis was based on prescribed medication unique for the treatment of ADHD rather than ICD-10 or Diagnostic and Statistical Manual of Mental Disorders 43 diagnoses. However, Swedish national guidelines state that medication should be reserved for ADHD treatment when other supportive interventions have failed, 27 indicating that our proxies most likely underestimated the incidence of ADHD and identified the most severe cases.
Thus, while our definition of exposure probably could not avoid false-negatives, we considered bias due to false-positives more unlikely. Also, it should be noted that our findings did not rule out the possibility of an association of age at first pregnancy with more refined ADHD-related neurocognitive deficits. To further explore this, studies using more detailed measures of ADHD symptom severity are warranted. Additionally, as in all observational studies, we could not fully rule out selection bias due to a lack of intact information on exposure and outcome variables.

Conclusions
In conclusion, our data replicated the findings from a 2017 Danish cohort study by Ostergaard et al 5 suggesting increased likelihood of teenage parenthood among women and girls with ADHD by showing that women and girls with ADHD were associated with a 6-fold increased risk of giving birth as teenagers compared with their unaffected peers. Furthermore, our results expanded the findings from the prior Danish study 5 by suggesting that women and girls with ADHD are a significant and underrecognized group with several obstetric risk factors and comorbid medical and mental health concerns, among which substance use disorder was the most common. Standard of care in women and girls with ADHD should include active efforts to prevent teenage pregnancies to reduce longterm adverse consequences for both mothers and children. Transdisciplinary collaboration between psychiatric clinics for youths and specialized youth clinics, as well as further studies on tolerability of hormonal contraception in women and girls with ADHD, are warranted to provide adequate care and suitable contraception for youth with neurodevelopmental disorders.