With an estimated 1 out of every 68 children in the US (CDC, 2014) and 1 out of every 100 children in the UK (NAS, 2014) diagnosed with Autism Spectrum Disorders (ASD), it is clear that ASD now poses a substantial health burden. Although these statistics are striking, they are taken from the top two countries producing ASD research publications. The US and the UK are two countries of many, and both are prominent developed nations—what is the global prevalence of ASD beyond the US and the UK? As of 2012, only a few studies have investigated prevalence in middle-income countries and no prevalence estimates have been reported for low-income countries, resulting in an incomplete global prevalence figure. While much scientific progress has been made in the past decades in terms of the genetic underpinnings, environmental risk factors, and early detection of ASD, the lack of understanding of the worldwide ASD prevalence presents a hurdle both for assessing the generalizability of these scientific findings and for further research progress.
An important step in determining the global prevalence of ASD is to investigate prevalence in developing nations, as these countries have been studied less frequently. Some evidence suggests a lower prevalence of ASD in developing nations compared to developed countries. For example, in Oman the reported figure is only 0.014% and in South Korea, the government reports that 0.046% of the population has been diagnosed with ASD (although the most recent experimental findings suggest a prevalence closer to 1.89%, see SFARI.org for an interactive global prevalence map). Why is there such a large discrepancy between these countries and the US and the UK? One problem is that these figures usually come from medical records obtained from the state, which are often incomplete and depend on how well-structured the healthcare system is. Even when statistics come from more controlled epidemiological prevalence studies, a potential roadblock is that culture and language differences may affect diagnosis rates. For example, the Korean language differs from the English language in that it involves complex relational suffixes that distinguish the speaker from the subject, an aspect of speech that is often impaired in Korean individuals with ASD. The Autism Diagnosis Observation Schedule (ADOS), one of the most widely used diagnostic tool for autism in the US, does not account for such differences, nevertheless the ADOS is often the diagnostic tool used in developing countries. In order to conduct thorough and comparable epidemiological research, ASD must be studied across cultures to find common markers that consistently differentiate children with ASD and typically developing children around the world. At the same time, diagnostic tools must be developed that are sensitive to cultural differences that may not be easily translated.
In addition to the lack of a culturally sensitive diagnosis, the presence of fewer ASD advocacy organizations and less community support/awareness in developing countries skews prevalence rates, as well as perpetuates cultural stigmas that are detrimental to individuals with ASD and their families. For example, in South Korea having a child with ASD reflects negatively on the family such that it affects siblings’ prospects of marriage and even parents’ jobs. Because of this intense stigma, clinicians will intentionally misdiagnose children with a different disorder—aechak changae, or reactive attachment disorder—to avoid family shame. Not only does this hide the true prevalence of ASD in South Korea, but more importantly, misdiagnosis eliminates the chance for children with ASD to get the care they need. Even within the US, diagnosis rates correlate with states that have the best autism support services, as well as with socioeconomic factors such as income and education. This suggests that good community support and awareness is important for the accurate assessment of global prevalence.
One thought discussed at the 2013 International Meeting for Autism Research (IMFAR) is that because of the royalties associated with common diagnostic tools and the fact that diagnosis can only be made with access to trained clinicians, having an autism diagnosis is analogous to having access to a new technology — access that is often limited in developing countries. One way around this hurdle may therefore be to follow recent trends in the tech industry: develop inexpensive, open source diagnostic tools that are accessible to many. Recent experimental work has shown promising results on this front already. In a 2014 peer-reviewed journal article, minimally trained undergraduate students watched and scored youtube.com videos to diagnose autism; the students achieved over 94% diagnostic accuracy. This method is low cost, in that the training is minimal enough for undergraduates without clinical degrees to do the scoring, and is easy access, in that individuals can remotely upload videos from all over the world without making a trip to a clinician. The validity of this work is of course not without debate and will require additional follow-up study.
Nevertheless, we are in the first stages of acquiring an actual understanding of the global prevalence of autism. However, as is always important when implementing technologies for the developing world, communities need to dictate their own agendas according to their needs – only in this way can the best solutions be achieved. This must be taken into account not only when assessing prevalence but also when focusing on the next step of developing ASD health programs across the globe.