Cognitive development in children up to age 11 years born after ART - a longitudinal cohort study

Since the first IVF baby was born in 1978, there has been a rapid increase in ART use, with more than 5 million children conceived to date (Präg and Mills, 2015). This has raised questions regarding the long-term impacts of ART on offspring. Results to date remain highly contradictory. Many studies report an increased risk of an impaired behavioural, socio-emotional and cognitive development (Sutcliffe and Ludwig, 2007; Hart and Norman, 2013a; Spangmose et al., 2017) and mental disorders (Svahn et al., 2015) in ART children. Some studies have found a delay in cognitive development of ART offspring, particularly with ICSI children (Knoester et al., 2008; Goldbeck et al., 2009). ART offspring have higher risks of adverse birth outcomes, such as low birthweight (LBW), preterm delivery and birth defects (van Balen, 1998; Romundstad et al., 2008; Wilson et al., 2011; Hansen et al., 2013; Hart and Norman, 2013a; Kondapalli and Perales-Puchalt, 2013; Präg and Mills, 2015). Higher risks of poor health at birth are partially related to the higher incidence of multiple births, a historical consequence of ART (Pison et al., 2015). A series of systematic reviews have concluded, however, that there are no developmental differences between ART offspring and those from a natural conception (NC) once perinatal problems such as preterm birth, LBW and birth defects are taken into account (Wilson et al., 2011; Pandey et al., 2012; Hansen et al., 2013).

Others have likewise found no differences in the cognitive, motor and language development of 2-year-olds (Balayla et al., 2017) and in the intellectual development of 5-year-olds (Ponjaert-Kristoffersen et al., 2004), cognitive outcomes in 3 and 5-year-olds (Carson et al., 2009, 2011), mental health and developmental outcomes at age 7–8 years (Carson et al., 2013; Punamaki et al., 2016), overall, verbal and IQ cognitive ability in 8–10 year-olds (Leunens et al., 2006, 2008) or in academic performance of adolescents (Spangmose et al., 2017). Other studies have found not only comparable but also even higher mental health and socio-emotional development of ART children (Wagenaar et al., 2008, 2009; Mains et al., 2010; Hart and Norman, 2013b).

Research to date has generally examined development in different age groups in isolation, rarely scrutinising how differences in cognitive development in ART versus NC children develop over time. Cognitive ability during childhood has been shown to be a pivotal determinant of adolescent and adult outcomes such as educational attainment, earnings, and also crime, participation in risky behaviour, depression and teenage parenthood (Feinstein, 2003; Heckman, 2007). These studies have underlined the importance of looking at progression across different stages of childhood rather than one point in time (Heckman, 2006; Ermisch, 2008; Cunha et al., 2010). Some studies have suggested that infertility and conception via ART may have stronger detrimental impacts in early childhood, related to elevated anxiety, delayed mother–infant attachment, diminished maternal confidence and overprotecting parenting (Bernstein, 1990; Golombok and Brewaeys, 1996; Gibson et al., 2000; Hammarberg et al., 2008). Others have argued that perinatal problems may have long-term negative effects on children's development at older ages, from health conditions to educational attainment and income (Black et al., 2007; Myrskylä et al., 2013).

Recent studies have argued that findings may be confounded by the characteristics of ART parents (Golombok, 2015; Spangmose et al., 2017), yet few have attempted to empirically investigate this selectivity. There is consensus that the early formation of children's cognitive ability is positively related to older, highly educated and high socio-economic status (SES) parents (Feinstein, 2003; Liu et al., 2011; Goisis and Sigle-Rushton, 2014; Myrskylä et al., 2013). Since ART treatments are often costly, access is frequently related to individuals who are able to bear these treatment costs (Chambers et al., 2014a,b; Präg and Mills, 2017).

The aim of this study is to examine whether the cognitive development of children conceived after ART (IVF and ICSI) – measured as cognitive skills at age 3, 5, 7 and 11 years – differs over time compared to children born after NC. A secondary aim is to examine how parental characteristics serve as confounders in these effects.