Autism Spectrum Disorders (henceforth 'Autism') are conceptualised as a set of lifelong neurodevelopmental conditions characterised by atypical social communication skills and rigid and repetitive behaviours and interests. Autism is increasingly seen as lying at the end of a dimension of socio-communicative difficulties, with overlapping genetic influences operating on diagnosed autism and on subclinical autistic traits in the general population. Taking this dimensional trait-wise approach increases statistical power by examining individuals with high levels of autistic traits and characteristics who nonetheless fall below the diagnostic threshold for autism. This technique may be particularly useful for exploring autism-related issues in under-diagnosed/under-researched groups, such as in older age. One reason for autism being under-researched beyond middle adulthood is that the changes to diagnostic criteria over the past 50 years have made it challenging to recruit older autistic adults; thus, we know relatively little about the health, cognition, and life experiences of this ageing population. This thesis uses cross-sectional data from the PROTECT cohort to examine the health and cognitive profiles of those ageing with high autistic traits. PROTECT is an online study of healthy ageing, with approximately 30,000 adults age 50 years and older taking part. PROTECT collects a broad range of health information and self-report questionnaire data, as well as cognitive task performance using their online test platform. Autistic traits were measured using a bespoke 5-item screener, which was validated by the present author in a separate sample and found to have high sensitivity and specificity for identifying those with an autism diagnosis. Approximately 1% of PROTECT participants met criteria for the high autistic traits group, with those reporting no autistic traits in childhood or at present forming an age/gender matched comparison older adult group. In Chapter 1, an overview of Autism is given, including an outline of the historic changes to the conceptualisation of Autism's diagnostic criteria and how a dimensional approach to autistic traits can be used to examine understudied populations, such as older adults. In Chapter 2, the scarcity of research examining Autism in midlife and older age is explored, and the disparity between the number of publications that examine childhood vs. later life is highlighted. In Chapter 3, the mental and physical health profile of middle-aged and older adults high vs. low in autistic traits from the PROTECT cohort is examined. In Chapter 4, self-reported sleep problems are examined in the same PROTECT groups. In Chapter 5, experiences of trauma and current symptoms of post-traumatic stress disorder are examined. In Chapter 6, experiences of self-harm and suicidality are examined. In Chapter 7, cognitive abilities are examined from online tests of memory and executive functions. Overall, study findings indicate that middle-aged and older adults with high autistic traits experience more mental (but not physical) health problems, more sleep problems, higher rates of trauma and post-traumatic stress symptoms, higher rates of self-harm and suicidality, and lower task performance in several cognitive domains than an age/gender-matched low trait comparison group. A general discussion of the findings across the empirical chapters is presented in Chapter 8, including the implications of the key results, the strengths and limitations of the research methods, and suggestions for future directions for this under-researched topic.