Blame diagnostic inflation for the NDIS’ $10b Autism bill

As recently reported in The Australian Financial Review, autism costs the National Disability Insurance Scheme more than $10 billion annually. A record 62,500 people diagnosed with autism were added to the scheme last year.

Autism now accounts for 43 per cent of all NDIS participants. Those are staggering figures for a condition once regarded as rare. They also raise an uncomfortable question: how did a scheme designed to support people with severe and permanent disabilities come to be dominated by a single diagnostic category?

The standard explanation is greater awareness. Doctors are better at recognising autism. Families are more willing to seek assessment. Society is more open about psychiatric conditions. This may explain some of the increase. But it does not explain the pattern.

If awareness were the principal driver, growth would be distributed across all diagnostic categories. Instead, my recent analysis of Australia’s mental health system found growth in serious conditions (psychoses, profound intellectual disability) is stable or declining. Expansion is concentrated precisely where diagnostic thresholds are most ambiguous: high-functioning autism, developmental delay, and attention disorders.

Environmental ‘toxins’, vaccines, and prenatal medications, have also been blamed for the growth in autism, but research has failed to support these claims. The most persuasive explanation is diagnostic inflation.

Autism was once narrowly defined. In the 1960s it was estimated to affect roughly one in 2500 children and required early onset and severe impairment. In 1994, Asperger’s syndrome, a milder version, was added to the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, as a distinct diagnosis.

In 2013, the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition, merged Asperger’s and autism into autism spectrum disorder, broadening the category again.

Allen Frances, the psychiatrist who led development of autism spectrum disorder, has since acknowledged that the criteria became “exceedingly broad”, capturing children whose behaviours would once have been described as shy, awkward or eccentric.

The policy consequences are significant. Non-verbal children requiring lifelong supervision now share a diagnostic label with those almost certain to live independently. As the Lancet Commission on the future of care and clinical research in autism observed, when all autism is treated as equivalent, resources are spread thinly and “those with the most severe impairments” go without adequate support.

Explosive growth is not unique to autism. ADHD prescriptions have risen from roughly 500,000 annually to 4.5 million in 14 years. Around one in five Australians now meet the criteria for a mental disorder. Each condition follows the same trajectory: diagnostic boundaries expand, numbers rise, resources flow to those able to navigate assessment systems rather than those with greatest functional impairment.

Why does this continue?

Because diagnosis has become the gateway to funding. Clinicians are paid to assess and certify. Families unlock NDIS packages averaging about $70,000 a year. Schools access additional funding. Governments appear responsive and compassionate.

The system rewards entry and continuation, not recovery or independence. A provider whose client becomes self-sufficient loses revenue. A family whose child improves leaves the NDIS scheme. Success reduces funding; persistence sustains it.

For those with mild or developmental conditions, this is not only expensive; it can be corrosive. Labelled individuals are encouraged to see themselves as permanently impaired rather than capable of growth. Agency gives way to identity.

The Productivity Commission found that about half of publicly funded care flows to people with mild or moderate symptoms while those with severe, life-threatening conditions receive fragmented support. Current reform efforts do not address this structural issue.

The Grattan Institute notes that fewer than 20 per cent of NDIS participants with autism are under nine and eligible for the government’s proposed Thriving Kids program. Even if successful, more than 80 per cent of participants will remain within the same diagnostic framework, competing for the same pool of resources.

Tightening eligibility criteria alone will not solve the problem. Without altering incentives, it simply produces more elaborate diagnostic justifications.

The NDIS was created to provide support for people with severe disabilities. Restoring that purpose requires redesign, not administrative reform. Support should be allocated according to functional impairment – whether a person can manage daily living, communicate effectively or sustain employment – not diagnostic criteria alone. This is the method NDIS uses for physical disabilities, and it should apply to psychological ones as well.

Providers should be rewarded for achieving measurable improvement and independence, not ongoing enrolment. Assistance for milder presentations should be time-limited and regularly reviewed. As functioning improves, support should step down, freeing resources for people with profound and enduring disabilities.

The $10 billion autism bill is not the disease. It is a symptom of a system that rewards diagnosis over disability and expansion over recovery. Until funding is severed from psychiatric diagnoses, the NDIS will continue to expand while those it was designed to protect compete for resources in an increasingly crowded scheme. That is not a design flaw. It is the system working exactly as built.

Steven Schwartz, a senior CIS fellow and a former vice-chancellor and professor of clinical psychology, is the author of CIS paper Drowning in a Sea of Diagnoses.

Photo by Artem Podrez.