The researchers observed a clear pattern in which females begin to close the gap during adolescence. They say this trend points to an urgent need to better understand why females are often diagnosed later than males.

Rising Autism Diagnoses Over Time

The prevalence of autism spectrum disorder (ASD) has risen steadily over the past 30 years. Throughout this period, diagnosis rates have shown a pronounced imbalance, with males diagnosed about four times as often as females.

Experts believe the overall rise in autism diagnoses is linked to broader diagnostic definitions and social factors (eg, parental age). The large difference between male and female diagnoses has often been explained by the fact that girls tend to have stronger social and communication skills, which can make autism harder to identify. Until now, however, no large study had followed these patterns across different stages of life.

Following Millions Across the Lifespan

To fill this gap, researchers analyzed national health records covering 2.7 million individuals born in Sweden between 1985 and 2022. Participants were followed from birth for as long as 37 years.

Over more than 35 years of observation, autism was diagnosed in 78,522 individuals, representing 2.8% of the population studied. The average age at diagnosis was 14.3 years.

How Autism Diagnosis Rates Change With Age

Autism diagnosis rates rose with each five year age group throughout childhood. Among males, the highest rate occurred between ages 10-14 years, reaching 645.5 per 100,000 person years. For females, the peak came later, between ages 15-19 years, at 602.6 per 100,000 person years.

While males were more likely to be diagnosed during childhood, females showed a strong increase in diagnoses during adolescence. By about age 20 years, the ratio of males to females diagnosed with autism approached 1:1.

Study Limitations and Strengths

The authors noted that this research was observational. They did not account for other conditions often linked to autism, such as ADHD and intellectual disability. The study also could not fully adjust for shared genetic or environmental influences, including parental mental health.

At the same time, the researchers emphasized that the scale and duration of the study made it possible to analyze data from an entire population. This allowed them to separate the influence of age, calendar period, and birth cohort.

Autism Rates May Equalize by Adulthood

Based on their analysis, the authors wrote: “These findings indicate that the male to female ratio for autism has decreased over time and with increasing age at diagnosis. This male to female ratio may therefore be substantially lower than previously thought, to the extent that, in Sweden, it may no longer be distinguishable by adulthood.”

They added that “These observations highlight the need to investigate why female individuals receive diagnoses later than male individuals.”

Missed Diagnoses and Real-World Consequences

The findings are consistent with recent research suggesting that autism in women is frequently missed or identified much later in life. In a linked editorial, patient and patient advocate Anne Cary said the results support concerns about gaps in current diagnostic practices.

She emphasized that studies like this help challenge the long-standing belief that autism is more common in males than in females. However, she also warned that while autistic female individuals wait for accurate diagnosis, “they are likely to be (mis)diagnosed with psychiatric conditions, especially mood and personality disorders, and they are forced to self-advocate to be seen and treated appropriately: as autistic patients, just as autistic as their male counterparts.”

One of the biggest drivers was a sharp drop in hydroxyl radicals, which are the main chemicals responsible for breaking methane down in the air. During 2020-2021, this atmospheric clean-up process slowed dramatically. According to the research team, which includes Boston College Professor of Earth and Environmental Science Hanqin Tian, this decline explains about 80 percent of the year-to-year changes in how quickly methane accumulated.

Wet Conditions Fueled Methane Production

At the same time, a prolonged La Niña phase from 2020 to 2023 brought wetter-than-average weather to large parts of the tropics. These conditions expanded flooded landscapes, which are ideal environments for microbes that produce methane. As a result, emissions increased from wetlands, rivers, lakes, and farmed land, adding to the buildup of methane, the second-most important greenhouse gas after carbon monoxide.

Measurements show that atmospheric methane rose by 55 parts per billion between 2019 and 2023, reaching a record level of 1921 ppb in 2023. The fastest growth occurred in 2021, when methane levels increased by nearly 18 ppb. That jump was 84 percent higher than the increase seen in 2019.

“As the planet becomes warmer and wetter, methane emissions from wetlands, inland waters, and paddy rice systems will increasingly shape near-term climate change,” said Tian. “Our findings highlight that the Global Methane Pledge must account for climate-driven methane sources alongside anthropogenic controls if its mitigation targets are to be achieved.”

Natural and Managed Systems Both Matter

The surge was not limited to natural wetlands. Managed environments such as paddy rice fields and inland waters also contributed significantly. According to Tian, who serves as Director of the Center for Earth System Science and Global Sustainability in the Schiller Institute for Integrated Science and Society, these sources are often underrepresented in global methane models.

The largest increases in emissions were observed in tropical Africa and Southeast Asia. Arctic wetlands and lakes also showed notable growth as warmer temperatures boosted microbial activity. In contrast, methane emissions from South American wetlands dropped in 2023 during an extreme El Niño-related drought. This contrast highlights how sensitive methane release is to climate extremes, the report notes.

How Researchers Tracked the Methane Spike

Tian and his colleagues played a key role in identifying and measuring how wetlands, rivers, lakes, reservoirs, and global paddy rice farming contributed to the rapid rise in atmospheric methane. By linking land, freshwater, and atmospheric processes in advanced Earth system models, the Boston College team showed how climate variability amplified emissions across connected ecosystems.

The study also found that fossil fuel use and wildfires played only a small role in the recent methane increase. Chemical fingerprinting indicates that microbial sources, including wetlands, inland waters, reservoirs, and agriculture, were responsible for most of the observed changes.

“By providing the most up-to-date global methane budget through 2023, this research clarifies why atmospheric methane rose so rapidly,” said study lead author Philippe Ciais of the University of Versailles Saint-Quentin-en-Yvelines. “It also shows that future methane trends will depend not only on emission controls, but on climate-driven changes in natural and managed methane sources.”

Key Findings From the Study

• This early-2020s methane surge was mainly caused by a weakened atmospheric chemistry sink, not runaway emissions.
• A temporary drop in hydroxyl (OH) radicals — the atmosphere’s primary methane “cleanser” — during 2020-2021 explains about 80-85 percent of the year-to-year variability in methane concentration growth.
• COVID-19-related air pollution changes played a central role.
• Reductions in nitrogen oxides (NOₓ) during pandemic lockdowns reduced OH levels, allowing methane to accumulate faster in the atmosphere.
• Climate-driven wetland emissions amplified the surge.
• Exceptionally wet conditions during a prolonged La Niña (2020-2023) boosted methane emissions from wetlands and inland waters, especially in tropical Africa and Southeast Asia, with additional increases in Arctic regions.
• Fossil fuel and fire emissions were not the main drivers.
• Changes in fossil fuel and biomass-burning methane emissions were comparatively small and cannot explain the observed global methane spike.
• Current bottom-up emission models for natural flooded ecosystems miss critical dynamics.
• Many widely used models underestimated wetland and inland-water emissions and their dynamics during the surge, highlighting urgent gaps in monitoring flooded ecosystems and microbial methane emission processes.

Research on this ancient ancestor shows that many features seen in modern life were already in place at that time. Cells already had membranes, and genetic information was stored in DNA. Because these essential traits were already established, scientists seeking to understand how life first took shape must look even further back in time, to evolutionary events that occurred before this shared ancestor existed.

Studying Life Before the First Common Ancestor

In a study published in the journal Cell Genomics, researchers Aaron Goldman (Oberlin College), Greg Fournier (MIT), and Betül Kaçar (University of Wisconsin-Madison) describe a way to explore that earlier period of evolution. “While the last universal common ancestor is the most ancient organism we can study with evolutionary methods,” said Goldman, “some of the genes in its genome were much older.” The team focuses on a special group of genes called “universal paralogs,” which preserve evidence of biological changes that took place before the last universal common ancestor.

A paralog is a group of related genes that appear multiple times within a single genome. Humans provide a clear example. Our DNA contains eight different hemoglobin genes, all of which produce proteins that carry oxygen through the blood. These genes all originated from a single ancestral globin gene that existed around 800 million years ago. Over long periods of time, repeated copying errors produced extra versions of the gene, and each copy gradually developed its own specialized role.

What Makes Universal Paralogs Unique

Universal paralogs are much rarer. These gene families appear in at least two copies in the genomes of nearly all living organisms. Their widespread presence suggests that the original gene duplication occurred before the last universal common ancestor emerged. Those duplicated genes were then passed down through countless generations and remain present in life today.

Because of this deep evolutionary reach, the authors argue that universal paralogs are a critical yet often overlooked resource for studying the earliest history of life on Earth. This approach is becoming more practical as new AI-based techniques and AI-optimized hardware make it easier to analyze ancient genetic patterns in detail.

“While there are precious few universal paralogs that we know,” says Goldman, “they can give us a lot of information about what life was like before the time of the last universal common ancestor.” Fournier adds, “The history of these universal paralogs is the only information we will ever have about these earliest cellular lineages, and so we need to carefully extract as much knowledge as we can from them.”

Clues to the First Cellular Functions

In their analysis, Goldman, Fournier, and Kaçar reviewed all known universal paralogs. Every one of these genes plays a role in either building proteins or moving molecules across cell membranes. This finding suggests that protein production and membrane transport were among the first biological functions to evolve.

The researchers also emphasize the importance of reconstructing the ancient forms of these genes. In one study from Goldman’s lab at Oberlin, scientists examined a universal paralog family involved in inserting enzymes and other proteins into cell membranes. Using standard methods from evolutionary biology and computational biology, they reconstructed the protein produced by the original ancestral gene.

Their results showed that this simpler, ancient protein could still attach to cell membranes and interact with the machinery that makes proteins. It likely helped early proteins embed themselves into primitive membranes, offering insight into how the earliest cells may have operated.

A New Window Into Life’s Earliest History

The authors hope that continued advances in computational tools will allow scientists to identify additional universal paralog families and study their ancient ancestors in greater detail. “By following universal paralogs,” says Kaçar, “we can connect the earliest steps of life on Earth to the tools of modern science. They provide us a chance to transform the deepest unknowns of evolution and biology into discoveries we can actually test.” Their goal is to build a clearer picture of evolution before the last universal common ancestor, shedding light on how life as we know it first emerged.