There are a number of problems and challenges in relating the science of epidemiology to mental retardation (MR). These relate to how MR is defined and classified and how these definitions may change over time. These as well as other differences in ascertainment sources and methods need to be considered when comparing MR prevalence over time and place. On the other hand, advances in technology also provide new and efficient methods of data collection both by data linkage and by use of web-based methods to study rare diseases. While prevalence studies have not been individually reviewed, we have examined the range of data including recent studies relating to how prevalence differs according to age, gender, social class and ethnicity. Some problems with available etiological classification systems have been identified. Recent etiological studies, most of which use different classification systems, have been reviewed and explanations have been postulated to account for differences in results. Individual risk factors for MR are considered whilst the option of considering a population as opposed to a high risk strategy to MR prevention is raised. This might well involve improving the social milieu surrounding the occurrence of individual risk factors. The impact of biotechnological advances such as antenatal and neonatal screening and assisted reproduction on MR are discussed. The issue of how inequalities in access to technology may impact on case identification and even have the potential to further widen inequalities is raised. The importance of extending the use of epidemiological tools to study the social, health and economic burden of MR is also emphasized. However, in order to apply to MR the "prevention-intervention-research" cycle, which surely underpins all epidemiology, it is vital to ensure that the methodological challenges we raise are adequately addressed.

The clinical understanding of the CDKL5 disorder remains limited, with most information being derived from small patient groups seen at individual centres. This study uses a large international data collection to describe the clinical profile of the CDKL5 disorder and compare with Rett syndrome (RTT). Information on individuals with cyclin-dependent kinase-like 5 (CDKL5) mutations (n ¼ 86) and females with MECP2 mutations (n ¼ 920) was sourced from the InterRett database. Available photographs of CDKL5 patients were examined for dysmorphic features. The proportion of CDKL5 patients meeting the recent Neul criteria for atypical RTT was determined. Logistic regression and time-to-event analyses were used to compare the occurrence of Rett-like features in those with MECP2 and CDKL5 mutations. Most individuals with CDKL5 mutations had severe developmental delay from birth, seizure onset before the age of 3 months and similar non-dysmorphic features. Less than one-quarter met the criteria for early-onset seizure variant RTT. Seizures and sleep disturbances were more common than in those with MECP2 mutations whereas features of regression and spinal curvature were less common. The CDKL5 disorder presents with a distinct clinical profile and a subtle facial, limb and hand phenotype that may assist in differentiation from other early-onset encephalopathies. Although mutations in the CDKL5 gene have been described in association with the early-onset variant of RTT, in our study the majority did not meet these criteria. Therefore, the CDKL5 disorder should be considered separate to RTT, rather than another variant.

IMPORTANCE The origins and development of autism spectrum disorder (ASD) remain unresolved. No individual-level study has provided estimates of additive genetic, maternal, and environmental effects in ASD across several countries. OBJECTIVE To estimate the additive genetic, maternal, and environmental effects in ASD.