The Dutch Lifelines cohort: the risk of Developmental Coordination Disorder runs in families

Developmental Coordination Disorder (DCD) means the presence of motor problems interfering with daily activities in the absence of evident neurological pathology.1 It affects 5-6% of children and it is often accompanied by other neurodevelopmental disorders, such as attention deficit hyperactivity disorder or autism spectrum disorder.1 It often has been suggested that DCD runs in families, but only recently some evidence of a possible genetic background of DCD has been presented.2 The Dutch Lifelines Cohort study3 offered the opportunity to address the question whether the risk of DCD runs in families.
Lifelines is a multi-disciplinary prospective population-based cohort study examining in a unique three-generation design the health and health-related behaviours of 167,729 persons living in the North of the Netherlands. It employs a broad range of investigative procedures in assessing the biomedical, socio-demographic, behavioural, physical and psychological factors which contribute to the health and disease of the general population, with a special focus on multi-morbidity and complex genetics.3 The Lifelines Cohort study has been approved by the ethical committee of the University Medical Center Groningen, the Netherlands (METC 2007/152).
In 2017 all parents participating in Lifelines, with children aged 5-12 years (n=5479) received the Dutch translation of the DCD Questionnaire 2007.4 This is a widely used instrument to screen for DCD with 15 questions addressing control during movement, fine motor activities, including handwriting, and general coordination. We used available age-based percentiles to classify children as at risk of DCD (rDCD).4 Children with diagnoses not compatible with DCD, e.g., cerebral palsy, were excluded from the analysis, resulting in a total of adequately completed questionnaires on 1722 children (31.4%), coming from 1255 different families (each having 1-4 children in the sample; see Table 1 and Appendix S1). From the 1722 children 223 were classified as rDCD. The proportion of rDCD in children with no siblings in the dataset was 13.38% (see Appendix S1). 
We calculated the probability of observing the number of children classified as rDCD and  non-rDCD in each of the 418 families with two or more children present in our data under the assumption of rDCD not running in the family. That is, we tested whether the probability of a child having rDCD was independent of having a sibling with rDCD. For mathematical details on the statistical approach and the use of one-sample Chi-square tests to arrive at an overall p-value for this probability, we refer to Appendix S1. Our analysis resulted in an overall probability of rDCD occurring independently in our data of p = 0.008. This is a very small p-value, implying that our data do not support the assumed hypothesis of rDCD occurring independently within families. As our data shows a tendency towards observing more rDCD children within families than expected, we conclude that our data suggests being at risk of DCD runs in families. 
This means that our data support evidence of the existence of familial predisposition of DCD, which might include genetic variants playing a pathogenetic role.2 In other words, genetic susceptibility for motor impairment may be one of the steps in the multifactorial pathway of DCD. Well-known risk factors in the etiological pathway are male gender and preterm birth.5 Conceivably, the multifactorial pathway varies among children with DCD, a diagnosis that by itself is an umbrella term covering children with various types of motor problems, e.g. fine motor impairments or balance problems.1 
The strength of our study is its two generations design and the use of a large population-based cohort. But the study also has limitations. First, our response rate was 31.4%, which reduces generalizability of the findings. We cannot rule out the potential effect of selection bias either, although the direction of its effect on our conclusion is not clear. In general, parents may be more inclined to respond when already suspecting their offspring to be at risk for DCD. This would increase the observed proportion of children with rDCD overall in the data. However, this effect would simultaneously imply that the benchmark of 13.38% against which we tested our hypothesis of independence would be positively biased as well, which in turn would make it harder to reach a significant result in our test. Lastly, we did not diagnose DCD, but only were able to evaluate whether children were at risk of DCD. 
In conclusion: our findings support emerging evidence that the risk of DCD may run in families.