📄 medicine

Prenatal and Perinatal Factors Associated with Autism Spectrum Disorder Diagnosis: An Exploratory Population-Based Registry-Linkage Study from Slovenia, 2005–2017

This exploratory population-based registry-linkage study of 302,439 singleton births in Slovenia (2005–2017) identified significant crude associations between ASD diagnosis and male sex, preterm birth, postpartum hemorrhage, cesarean delivery, and neonatal respiratory morbidity, though the authors caution that the lack of multivariable adjustment and potential confounding limit causal inference.

Original authors: Joško Osredkar, Uroš Godnov

Published 2026-07-09

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Joško Osredkar, Uroš Godnov

Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

The Big Picture: A "Digital Detective" Story

Imagine the researchers are like digital detectives in Slovenia. They didn't interview families or run new medical tests. Instead, they went into the country's massive digital filing cabinet (the National Perinatal Information System) and pulled out the records for every single baby born between 2005 and 2017. That's over 302,000 babies.

Their goal was to see if there were any "clues" in the birth records that showed up more often for children who were later diagnosed with Autism Spectrum Disorder (ASD) compared to children who were not.

The Cast of Characters

The "ASD Group": The researchers found 117 children in the system who had a specific medical diagnosis of autism (either "childhood autism" or "Asperger syndrome").
- Important Note: The paper is very careful to say this is not a count of all autistic children in Slovenia. Think of this group as a specific "club" of children whose records were successfully linked. It's a sample, not the whole population.
The "Comparison Group": The remaining 302,322 children who were born during that time but did not have that specific diagnosis in the linked records.

The Investigation: What Did They Find?

The researchers compared the birth records of the two groups to see what happened differently. They found several "red flags" that appeared more often in the ASD group's birth stories.

Here are the main findings, explained with analogies:

1. The Gender Gap (The "Boys' Club" Effect)

The Finding: About 84% of the children in the ASD group were boys, compared to about 51% of boys in the general group.
The Analogy: If you walked into a room of 100 random people, about half would be men. But if you walked into this specific group of children with autism, nearly 8 or 9 out of 10 would be boys. The study confirms that boys are much more likely to be in this group, though the researchers note this might be partly because boys are easier to spot or diagnose than girls.

2. The "Early Arrival" (Prematurity)

The Finding: Babies born early (before 37 weeks) and very early (before 31 weeks) were more common in the ASD group.
The Analogy: Imagine a race where most runners finish at the standard time. However, in the ASD group, there were significantly more runners who crossed the finish line early. The study found that being born very early was about 3 times more likely in the ASD group than in the general group.

3. The "Breathing Struggles" (Respiratory Issues)

The Finding: Babies in the ASD group had more trouble breathing right after birth. This included conditions like Respiratory Distress Syndrome (RDS), needing extra oxygen, or needing a machine to help them breathe.
The Analogy: Think of a newborn's lungs like a brand-new engine. In the ASD group, these engines were more likely to sputter or need a "jump start" (like surfactant medicine or a ventilator) to get running smoothly.
The Catch: The researchers warn that this might not be the cause of autism. Instead, it's like finding that people who arrive at the hospital early (premature) are also more likely to have engine trouble. The breathing issues might just be a sign that the baby was born too early, rather than the breathing trouble causing the autism.

4. The "Birth Day Drama" (C-Sections and Bleeding)

The Finding: There was more bleeding after birth (postpartum hemorrhage) and more Cesarean sections (C-sections) in the ASD group.
The Analogy: The birth process for these families was often more "eventful" or complicated. It was like a car trip that required a detour (C-section) or had a flat tire (bleeding) more often than the average trip.

5. The "Medication Mystery" (Dexamethasone)

The Finding: Mothers in the ASD group were more likely to have taken a steroid called dexamethasone during pregnancy.
The Analogy: This is a classic "chicken or egg" situation. The researchers explain that this medicine is usually given when a baby is in danger of being born too early. Since the ASD group had more early births, they naturally had more mothers taking this medicine. It's likely the medicine didn't cause the autism; rather, the reason the mother took the medicine (a threatened early birth) is what links them together.

What the Study Does NOT Say (The "Fine Print")

The authors are very careful not to overstate their results. Here is what they explicitly say they cannot conclude:

No Causation: They did not prove that being born early, having a C-section, or having breathing trouble causes autism. They only found that these things happened together more often.
No "Fixing" the Data: Because they only had summary numbers (like "16 babies were born early") rather than individual files for every baby, they couldn't run a complex math check to see if one factor caused another. They call their results "crude" (unadjusted).
Not a National Census: The 117 children are just the ones the computer successfully linked. There are likely many more autistic children in Slovenia who weren't in this specific digital link, so you can't use this number to say "1 in X children in Slovenia has autism."

The Bottom Line

Think of this study as a first draft or a hypothesis generator.

The researchers looked at a huge pile of birth records and found a pattern: Children in this specific ASD group were more likely to be boys, born early, have breathing trouble at birth, and have complicated deliveries.

However, the paper concludes that we need more sophisticated detective work (using individual-level data and complex math) to figure out if these birth events are the cause, just a symptom of other underlying issues, or simply things that happen to go hand-in-hand. For now, these findings are a map of where to look next, not the final destination.

Technical Summary: Prenatal and Perinatal Factors Associated with Autism Spectrum Disorder Diagnosis in Slovenia (2005–2017)

Problem Statement
Autism spectrum disorder (ASD) is a heterogeneous neurodevelopmental condition with a multifactorial etiology involving the interaction of genetic susceptibility and prenatal/perinatal exposures. While population-based registry studies offer a mechanism to describe associations between routinely recorded perinatal characteristics and later ASD diagnoses, they are often constrained by administrative case ascertainment and a lack of individual-level data for multivariable adjustment. This study addresses the need to describe these associations within the Slovenian population while explicitly acknowledging the limitations of unadjusted, aggregate data and the specific constraints of the linked clinical records used for case identification.

Methodology
The authors conducted an exploratory, retrospective, population-based registry-linkage study covering all singleton live births in Slovenia between January 1, 2005, and December 31, 2017 ( $N = 302,439$ ).

Data Sources: The study utilized the Slovenian National Perinatal Information System (NPIS) for maternal, obstetric, and neonatal variables, deterministically linked to national clinical diagnostic records using a single national identifier.
Case Definition: ASD cases ( $n=117$ ) were identified via linkage to clinical records classified as childhood autism ( $n=90$ ) or Asperger syndrome ( $n=27$ ), consistent with DSM-5 unification. Pervasive developmental disorder not otherwise specified (PDD-NOS) was excluded due to coding specificity concerns.
Comparison Group: The reference group comprised all other singleton live births in the registry ( $n=302,322$ ).
Statistical Analysis: Due to data availability being limited to aggregate cross-tabulations (group-level counts) rather than individual-level records, the study employed simple logistic regression to calculate unadjusted odds ratios (OR) with 95% confidence intervals (CI). No multivariable adjustment, model selection, or multicollinearity assessment was performed. Chi-square tests and ANOVA were used for group comparisons.
Ethical Oversight: The study was approved by the National Institute of Public Health and the National Medical Ethics Committee, utilizing fully anonymized data with waived individual consent.

Key Contributions

Population-Level Association Mapping: The study provides a comprehensive, albeit unadjusted, profile of perinatal factors associated with a clinically linked ASD group in Slovenia, a dataset distinct from existing Nordic or U.S. cohorts.
Explicit Limitation Framework: A significant contribution is the rigorous framing of results as "exploratory" and "descriptive." The authors explicitly delineate that the linked group does not represent the total national prevalence of ASD and that the absence of individual-level data precludes causal inference.
Differentiation of Respiratory Markers: The paper contributes to the literature by distinguishing neonatal respiratory morbidity (RDS, TTN, surfactant use) as likely markers of prematurity rather than independent evidence of hypoxic-ischemic injury, given the strong confounding by gestational age.

Results
The analysis identified several crude associations between the linked ASD group and specific perinatal characteristics:

Demographics: The ASD group was predominantly male (83.8% vs. 51.4% in controls; OR 4.88, 95% CI 2.99–7.98). Maternal age did not differ significantly between groups.
Gestational Age: Preterm birth (<37 weeks) was nearly twice as frequent in the ASD group (13.7% vs. 7.6%; OR 1.92), and very preterm birth (<31 weeks) was approximately three times as frequent (4.3% vs. 1.4%; OR 3.12).
Obstetric Complications: Postpartum hemorrhage (>500 mL) was more common in the ASD group (6.0% vs. 2.3%; OR 2.74), as was cesarean delivery (26.5% vs. 18.0%; OR 1.65).
Neonatal Respiratory Morbidity: The ASD group showed elevated rates of respiratory distress syndrome (OR 3.39), transient tachypnea of the newborn (OR 3.86), surfactant administration (OR 5.93), and mechanical ventilation (OR 3.29).
Medication Exposure: Antenatal dexamethasone exposure was ~3.4-fold higher in the ASD group (9.6% vs. 2.8%), which the authors attribute to confounding by indication (threatened preterm labor) rather than a direct etiologic effect.
Rare Events: Severe intraventricular hemorrhage showed a high crude OR (18.62) but was based on only 2 events in the ASD group, rendering the estimate unstable.

Significance and Claims
The paper positions its findings as hypothesis-generating rather than definitive evidence of causality. The authors explicitly state that the results describe "exploratory associations" within a specific, clinically linked subgroup, not the entire population of children with ASD in Slovenia.

The significance of the study lies in its utility as a baseline for future research. The authors emphasize that the observed associations—particularly regarding respiratory morbidity and obstetric complications—are likely confounded by gestational age and shared familial factors. Consequently, the paper concludes that adjusted analyses, multicollinearity assessments, and broader outcome capture using individual-level datasets are necessary before these associations can inform clinical or public health recommendations. The study serves as a descriptive foundation, highlighting the need for more granular data to disentangle the effects of prematurity from specific perinatal insults in the etiology of ASD.