Adults with Autism Spectrum Disorders: Challenges for Epidemiological and Outcome Research.
Guest blogger Ruth Carper, Ph.D., Asst. Research Scientist, Center for Human Development, University of California, San Diego
Adults with autism and their families may be interested to know that the scientific community is turning more attention toward understanding the full life course of autism spectrum disorders (ASDs). While a great deal of research is directed toward understanding the causes of the disorder and developing interventions for young children, we still know relatively little about how the disorder is expressed in adults and how autism changes as individuals on the spectrum get older. This year one of the Invited Educational Symposia at IMFAR was dedicated to this topic, and entitled: Adults with Autism Spectrum Disorders: Challenges for Epidemiological and Outcome Research.
The first two speakers Traolach “Terry” Brugha, M.D., and Fiona Scott, Ph.D. come from the UK where they have recently collaborated on studies of adults with Autism Spectrum Disorder. Dr. Brugha and his colleagues conducted a large epidemiological survey to assess the prevalence of ASD among adults living in the general community (published earlier this month in Archives of General Psychiatry, 2011; v. 68(5): pp. 459-466). The rates of ASD diagnosis have been rising rapidly in the last many years and it is generally accepted that at least part of this increase is due to increased awareness of the disorder among pediatricians, educators, and other child specialists, and to improvements in detection and diagnosis especially in individuals who do not have comorbid intellectual disability. However, it is important to remember that this also implies that there may be a large number of adults who could meet the criteria for an autism diagnosis but who have never been diagnosed with a disorder on the spectrum. To address this, Dr. Brugha and his colleagues contacted adults at more than 13,000 residential addresses in a door-to-door Adult Psychiatric Morbidity Survey in England. Using a stratified random sampling approach, 7,461 individuals participated in first phase interviews which included screening for a variety of psychiatric diagnoses. Individuals who met certain criteria on a 20 item screening questionnaire for autism spectrum disorders were selected to participate in more in-depth assessments including the ADOS (Autism Diagnostic Observation Schedule) and ADI (Autism Diagnostic Inventory). After all testing was completed, 19 individuals were determined to have previously undiagnosed autism spectrum disorders which the authors estimated represents of a rate of 9.8 per 1,000 in the general UK population. Importantly, they did not find evidence of any significant differences in rates across age. That is, it appeared that the oldest individuals were just as likely to have a previously undiagnosed ASD as the youngest.
The next speaker was Dr. Fiona Scott of the University of Cambridge, who is also part of the Adult Psychiatric Morbidity Survey in England. Dr. Scott focused on the particular challenges of accurately diagnosing ASD in higher functioning adults who did not have pre-existing diagnoses. ASD is a developmental disorder and accurate diagnosis requires knowledge of the individual’s social, communicative, and behavioral development during early childhood. The APMS study focused on a higher functioning population that may not have sought services during childhood and particularly wanted to include adults of all ages. Accurate diagnosis is particularly challenging in this population because it is often difficult or impossible to acquire accurate information about the early developmental period. Community-living adults in their 50s, 60s, or 70s may not have living parents who can describe their early development and if they do, these parents are being asked to recall subtle behavioral changes from many decades before. Accurate diagnosis may be further hindered by our limited knowledge of the life course of behaviors. The behaviors and abilities of individuals with ASD change as they get older, as part of normal maturation, as a result of the interventions and training that they partake in, and in response to the challenges that they face in daily life. For example, Dr. Scott pointed out the possibility of comorbid psychiatric disorders such as depression that may arise later in life and may make diagnosis less clear. This could be particularly problematic for previously undiagnosed populations such as those in the APMS study who would not have received outside support for dealing with social challenges.
Other diagnostic challenges include gender differences in ASD. The APMS study found a 9:1 ratio between males and females in the rates of ASD whereas most studies of children find ratios closer to 4:1. This may mean that the screening tools that were used need to be modified for use in women or it may be an artifact of the community-dwelling population that was examined. Women with ASD are often more severely affected cognitively than males and require more behavioral support, but individuals living in institutional housing were excluded from the APMS study.
The only member of the panel who was not from the UK was Marsha Seltzer, Ph.D. of the Waisman Center at the University of Wisconsin. Dr. Seltzer has been following more than 400 children and adults with ASD, and their families, for about 12 years. She is evaluating people who ranged in age between 10 and 52 years when they first entered the study and her research group continues to follow them. Dr. Seltzer reported several interesting findings from their series of studies. On average, individuals in her study showed improvement in social reciprocity and reductions in problematic repetitive behaviors and stereotyped interests across the first 6 years of the study. About 30% also showed significant improvements on the Scales of Independent Behavior-Revised which measure self-care and community living skills as well as cognitive and motor abilities. These improvements could be a result of the supports and interventions that the individuals receive in the community or could simply be the natural life-course of the disorder. However, Dr. Seltzer also reported a study that raises the question of whether even greater improvement might be possible if better support structures were provided. In a study that looked selectively at younger adult subjects who were exiting high school they found that, although repetitive behaviors decreased with time, the rate of improvement was much higher while students were still in school, but slowed substantially (or even regressed) after leaving school.
The final speaker was Patricia Howlin, Ph.D. of King’s College in London who provided a broader perspective of the natural life course of ASD across the life span. She reviewed outcome studies from 1967 to the present. These studies tend to classify individual outcomes as “good”, “fair”, “poor” or similar categories based on the level of independence achieved. For example, those who live in institutions are given poorer ratings than those who live by themselves and those who have part-time jobs rate better than those in sheltered employment or day programs. Over the years there appears to have been a slight decline in rates of “good” outcomes compared to “fair” or “poor” outcomes in such studies, but Dr. Howlin pointed out the subjectivity of these ratings and the difficulty of interpreting the effects of service changes such as the move from housing in large institutions to small group homes. Studies have also found that as much as 16% of adults with ASD develop additional psychiatric diagnoses such as depression and obsessive compulsive disorder which may be triggered by life stressors (citing Hutton et al., 2008). But those with good community support may have better outcomes (citing Farley et al., 2009).
The symposium on Adults with Autism Spectrum Disorders presented some important findings on the topic and also pointed out some of the unique challenges to this area of research. Identifying and recruiting this population is not a trivial task but one which must be addressed. Efforts to improve available tools in this area are moving us forward and it is clear not only from this symposium but from other presentations given at the IMFAR meeting — on diagnostic tools targeted at verbal and non-verbal adults, interventions for adults, and changing health status — that this area of research will continue to grow.