Colorful illustration in red and bright aqua blue, of a pizza / pie chart with researcher's hand taking a slice.
Illustration by Laurène Boglio

By the Numbers: Coronavirus infection odds, Medicaid waivers, correlating conditions

In this edition of By the Numbers, we discuss the outsize burden of COVID-19 among some groups of autistic people, expansions in Medicaid waivers for autism and how an autistic person’s age at diagnosis may affect their risk for other conditions.

Welcome to this month’s edition of the By the Numbers Newsletter! At Spectrum, we do our best to summarize the latest autism research findings — and sometimes the best summary comes in the form of a chart or map. In this newsletter, we boil down interesting new research that is conveyed most succinctly by way of data visualizations.

As a housekeeping note, if you want to continue receiving this newsletter after today, please click here to update your preferences.

Let us know what you think of the newsletter, or tell us about your own data-rich work, at [email protected].

Vulnerable ages for SARS-CoV-2 infection among autistic people

People with autism age 16 and younger and those aged 40 to 60 have higher odds of being infected with SARS-CoV-2, the virus that causes COVID-19, than their non-autistic peers do, according to a new study.

The researchers analyzed data from 16,406 autistic people enrolled in Israel’s largest health care organization, which provides supplemental insurance to more than half of all Israelis. The team matched each participant with a non-autistic person of the same age and gender.

Autistic people aged 16 and younger had 1.3 times higher odds of being infected with SARS-CoV-2, and those in the 40 to 60 age range had double the odds, the researchers found. The results appeared in the Journal of Autism and Developmental Disorders in July.

The findings suggest that people with autism or other neurodevelopmental disorders should be prioritized for the COVID-19 vaccine, according to study investigator Dana Tzur Bitan, clinical psychologist at Ariel University in the West Bank.

Israeli officials issued guidance a few weeks ago to start vaccinating children with neurodevelopmental disorders from the age of 5, Tzur Bitan says. “There is awareness towards the increased risk of morbidity among people with neurodevelopmental disorders, but I can’t really connect the link and say, ‘OK, they’ve taken our findings and turned it into a policy.’”

It’s unclear why certain people with autism have elevated odds of contracting COVID-19, and the dataset doesn’t include information that might help answer that question, such as each participant’s housing situation — rural or urban, sheltered or unsheltered.

“The differences we see in ages 40 to 60 could be associated with familial interactions, but we don’t know yet,” Tzur Bitan says. “This is something we want to look into more carefully.”

Expansion of Medicaid waivers for autism care

The number of autism-specific Medicaid waivers grew more than fivefold from 2004 to 2015, according to a new study. States use these waivers to provide services to people who might not otherwise be covered under Medicaid, a government program that provides health coverage to millions of people in the United States.

Researchers compiled a list of 269 Medicaid programs that provide services to people with autism or intellectual disability. During the 11-year study period, 26 states changed their waivers to increase care options for autistic people; 9 of those states added a type of waiver called 1915(c) specifically for people with autism, far outpacing new 1915(c) waivers for intellectual disability; and 21 states made no changes.

“There’s a big difference in how states use Medicaid to meet the needs of autistic individuals throughout their life,” says Lindsay Shea, associate professor and director of the Policy and Analytics Center at the A.J. Drexel Autism Institute at Drexel University in Philadelphia.

Three states — Arizona, Rhode Island and Vermont — do not offer 1915(c) waivers, the main type assessed in the study, and were not included in the data. Those states use a different waiver, called 1115, to provide long-term patient care. Shea says, “1115 programs operate differently from traditional waivers. Ideally, we would be able to put everything together and compare apples and oranges so that we can better assess the effectiveness of policy choices.”

The researchers also used data from autism prevalence studies to estimate the percentage of autistic people in each state who could be served using 1915(c) Medicaid waivers. Wisconsin could serve more than 20 percent of people with autism or intellectual disability, they estimated, and Minnesota could serve about 14 percent. For most states, though, the figure was far lower: The average state could serve just 4 percent of its autistic or intellectually disabled residents with Medicaid waivers.

The findings appeared in Health Services Research in July.

Spectrum story spotlight: Sex, age of diagnosis correlate with autism comorbidities

The likelihood that an autistic person has another condition correlates strongly with the age at which they received their autism diagnosis, according to a new study. Also, autistic girls are more likely than non-autistic girls to have other conditions, to a degree not seen in autistic boys.

The study assessed whether an autistic person’s age, age at diagnosis or birth sex changed their chance of having any of 11 commonly co-occurring conditions, including epilepsy, anxiety and attention deficit hyperactivity disorder (ADHD). It drew on data from roughly 16,000 people with autism and more than 650,000 people without autism up to 16 years old in the Danish National Patient Registry, a large dataset that records the birth date, sex and diagnoses of people in the Danish hospital system.

Among people diagnosed late, at 11 to 15 years old, 26 percent of girls and 13 percent of boys were also diagnosed with an affective disorder, the study found. The trend for intellectual disability was the opposite: 40 percent of people with an early autism diagnosis had intellectual disability, compared with just 10 percent of people with a late autism diagnosis.

For the 11 co-occurring conditions considered in the paper, the age of autism diagnosis was the single biggest predictor of whether a participant had that condition. But sex was another major factor.

Among autistic individuals, girls were 2.2 times more likely to have anxiety than boys, the study found. By contrast, anxiety is about 1.4 times higher in non-autistic girls than in non-autistic boys. And whereas non-autistic boys are 2.6 times more likely to have ADHD than non-autistic girls, that ratio dropped within the autism population. Autistic boys are just 1.6 times more likely than autistic girls to have ADHD. The findings appeared in Acta Psychiatrica Scandinavica in July.

Spectrum index:

86 percent: The proportion of people with autism who show “a fair to very good level of objective psychosocial functioning,” according to a study that tracked the jobs, happiness and close friendships of 917 adults — 425 men and 492 women — over a six-year period. The results appeared in Autism in June.

98.8 percent: The proportion of autism diagnoses independently verified using electronic medical records among a subset of participants in SPARK, an ongoing genetic study of autism. (SPARK is funded by the Simons Foundation, Spectrum’s parent organization.) People who enroll in SPARK must have a professional autism diagnosis, but that diagnosis typically has not been independently verified. The results appeared in the Journal of Autism and Developmental Disorders in July.

Tool spotlight: Autism_genepheno, a text-mining tool built using the programming language Python, can sift through thousands of autism papers and determine which genes (such as FMR1) are linked with which phenotypes (such as epilepsy or anxiety). The tool was described in Scientific Reports in July and is available for free on GitHub.