India needs a national programme on autism Premium

There needs to be consultation with different stakeholders, with a primary focus on the end-users within the Indian autism community

One of the most frequent questions I faced at academic conferences as soon as people realised that I was an Indian researcher working on autism was this: ‘How many children in India have autism?’ At this point, India did not have any systematic estimates of autism prevalence. My collaborators and I set out to answer this question, and found ourselves telephoning a number of government hospitals in India. We were looking for available records of the children who came to them, and were eventually given a diagnosis of autism. In the absence of a central medical registry, this method seemed reasonable to estimate the prevalence of autism — and had been successfully used in some other countries.

Unfortunately, it did not work. None of the hospitals kept any records of how many children got a diagnosis of autism. It made us change our strategy from focusing on hospitals to schools. Instead of relying on existing medical records from hospitals, we translated and validated widely used autism assessment tools — and used these to estimate the prevalence of autism in nearly 12,000 schoolchildren. Interestingly, none of the children who met the diagnostic criteria for autism spectrum in our study had a prior clinical diagnosis. All of these children were in mainstream schools, and reported facing more challenges than their counterparts. There have been a few other prevalence estimates since our initial study, and a conservative estimate based on pooling the results from different studies suggests that well over one crore Indians are on the autism spectrum.

While biochemical pathways involved in core autism symptoms are unlikely to be different between cultures, there are notable cultural differences in who gets a clinical diagnosis of autism. The majority of children with an autism spectrum diagnosis within the United States and the United Kingdom are likely to be verbal, with average or higher than average IQ, and attending mainstream schools. In contrast, a significant majority of children in India who get a clinical diagnosis of autism often also have intellectual disability, and limited verbal ability.

This difference is likely to be driven by a range of sociological factors, such as access to appropriate clinical expertise, the allowance of provisions for inclusion in mainstream schools, as well as availability of medical insurance coverage for autism interventions. It raises the question of why one chooses to get an autism diagnosis in the first place. If a clinical diagnosis is unlikely to open doors to appropriate services and support provisions, and is instead going to add to societal stigma, then rolling out a nationwide screening and diagnostic programme for autism across all schools is not likely to be useful. Even if it were theoretically possible to do so, how we assess autism presents its own set of challenges.

Autism is assessed behaviourally, and behavioural assessment tools (i.e. questionnaires or interviews with professionals) are the starting point for all research and clinical work on autism. Yet, most of the widely used autism assessment tools have limited availability in Indian languages. While some of the more widely used tools have been translated and validated locally, recent years have seen a rise in the development of indigenous autism assessment tools.

The plethora of these tools can create challenges in comparing across them. Beyond questionnaires and interviews, observational measures (e.g. where individuals are asked to perform a task) can contribute meaningfully to assessing autism-related features. Observational measures rely less on verbal information, and more on task performance, and are thus more easily translatable across cultures and settings. A combination of questionnaire/ interview measures alongside observational measures is ideal for assessing autism. Crucially, assessment must not remain limited to measures of behaviour alone; autism is increasingly viewed as a systemic condition. Greater focus on areas that have historically been under-researched in autism, such as sleep, diet, sensory symptoms, and immune function need to be included within routine assessments.

While considering the components of assessments we also need to consider the assessors. Most available autism assessment tools need to be administered by a specialist mental health professional. According to the latest estimates, India has less than 10,000 psychiatrists, a majority of whom are concentrated in big cities. While the number of mental health professionals continues to grow, the current gap between demand and supply cannot be met directly by the specialists alone. This gap is not relevant for behavioural assessments alone but also for providing psychological interventions. Parallel efforts to widen the reach of diagnostic and intervention services through involving non-specialists, similar to a stepped-care model for psychological therapies, is required in order to bridge this chasm. To this end, emerging lines of evidence suggest the feasibility of such an approach for both autism identification and intervention in an Indian context. Two pillars for such an approach to succeed are the availability of a suitable non-specialist workforce (e.g. Accredited Social Health Activist (ASHA)/Anganwadi workers, parents/caregivers) and the appropriate digital technology (e.g. apps, smartphones) that can capture both self/caregiver report as well as observational data.