What Is the ASQ? A Child Development Screening Tool

The Ages and Stages Questionnaires, commonly called the ASQ, are a set of parent-completed screening tools designed to flag potential developmental delays in children from about one month through five and a half years of age. Rather than a single test, the ASQ is actually two related instruments: the ASQ-3, which screens broad developmental milestones, and the ASQ:SE-2, which focuses specifically on social and emotional behavior. Pediatricians, early childhood programs, and public health agencies worldwide use these questionnaires as a first-pass check during well-child visits, and the American Academy of Pediatrics endorses this kind of structured screening as standard practice.

Two Tools Under One Name

People often refer to “the ASQ” as though it were a single questionnaire, but there are actually two separate instruments that serve different purposes. The ASQ-3 (the “3” marks the third edition) looks at five developmental domains: communication, gross motor, fine motor, problem solving, and personal-social skills.¹ Each questionnaire interval has six questions per domain, totaling 30 items. A parent reads each question and marks whether their child performs the described activity “yes,” “sometimes,” or “not yet.”

The ASQ:SE-2 (Social-Emotional, Second Edition) is a companion tool that zeroes in on emotional regulation, social interactions, autonomy, and related behaviors. Research using item-response modeling has found that the ASQ:SE-2 actually captures two related but distinct constructs, social competence and emotional competence, rather than a single blended trait.² In practice, a clinic might use the ASQ-3 at every well-child visit and add the ASQ:SE-2 when there are behavioral concerns or at specific age checkpoints.

How Parents Fill It Out

One of the defining features of the ASQ is that it is designed to be completed by a parent or primary caregiver, not by a clinician. Parents receive an age-appropriate version of the questionnaire, usually in the waiting room, through the mail, or via an online portal. Each form is tied to a specific age interval, so a 12-month questionnaire asks about skills typical for that stage, while a 36-month version asks about more advanced behaviors. The whole thing takes most parents about 10 to 15 minutes.

This parent-driven approach was baked into the ASQ from its earliest versions. When the questionnaire was first revised in the late 1990s, analyses on over 7,000 completed forms showed strong test-retest reliability, good interobserver reliability, and an overall agreement of about 85% when compared against standardized professional assessments.³ The rationale is straightforward: parents observe their child every day across a wider range of settings than any clinician sees during a brief office visit. That said, parent reporting is not without complications, which come up later in accuracy discussions.

A systematic review comparing parent-completed screening tools against the Bayley Scales of Infant Development, considered a gold-standard professional assessment, found that the ASQ was one of the two most commonly evaluated parent-report instruments, alongside the PARCA-R.⁴ Its wide adoption reflects both the practical advantages and the decades of validation data behind it.

Scoring and What the Numbers Mean

Each of the ASQ-3’s five domain scores is compared against population-based cutoff values. A score that falls more than two standard deviations below the mean for that age interval is flagged as abnormal, and a child is considered to have a concerning result if even one domain crosses that threshold.⁵ There is also a “monitoring zone” between one and two standard deviations below the mean, where the recommendation is typically to watch the child more closely and rescreen at the next visit rather than immediately refer for evaluation.

It is worth understanding what a screening result actually tells you. A score below the cutoff does not diagnose anything. It flags a child for further professional evaluation, ideally by a developmental pediatrician, psychologist, or early intervention team. Conversely, a passing score does not guarantee typical development; it means the child is unlikely to have a significant delay based on the skills sampled by the questionnaire at that moment.

Accuracy in Practice

The question most parents and providers care about is: how well does the ASQ actually catch kids who need help, and how often does it raise false alarms? The answer depends on what kind of delay you are looking for and how severe it is.

A large systematic review and meta-analysis published in JAMA Pediatrics pooled data from multiple studies and found that using the standard two-standard-deviation cutoff, the ASQ-3 had a pooled sensitivity of about 77% and specificity of about 81% for detecting any developmental delay. For severe delays, sensitivity rose to roughly 84%, though specificity dipped slightly.⁶ In plain terms, the ASQ catches most children with meaningful delays but misses roughly one in five, and it also flags some children who turn out to be developing normally.

Domain-specific accuracy varies considerably. That same meta-analysis found that the ASQ’s motor and cognitive domain scores, when used individually, had much lower sensitivity, around 41-44%, even though specificity stayed above 90%.⁶ This means if your child has a motor-specific delay but is developing well otherwise, the individual motor domain score may miss it nearly 60% of the time. The tool works better as a broad sweep than as a domain-level diagnostic instrument.

A study of high-risk infants found a similar pattern with gross motor screening: at four months, sensitivity using the standard cutoff was only about 33%, improving to around 80-86% at 8 and 12 months when providers used the lower “monitoring” cutoff instead of the strict failure threshold.⁷ The takeaway for parents of premature or medically complex babies: a clean ASQ at a very young age is less reassuring than a clean ASQ at 12 months or later.

Where the ASQ consistently performs well is in ruling out delays. Studies regularly find high specificity and strong negative predictive values, meaning that when the ASQ says your child is fine, there is a high probability that the child genuinely does not have a significant delay.⁸ A Singapore cohort study comparing the ASQ-3 against the Bayley Scales confirmed specificity values ranging from 72% to 99% and negative predictive values from 69% to 98% across different domains, though sensitivity and positive predictive values were lower.⁹

The Social-Emotional Side

The ASQ:SE-2 has its own accuracy profile. In a Dutch validation study comparing the ASQ:SE against the Bayley Scales social-emotional subscale, specificity ranged from about 71% to 89% and negative predictive values reached above 87%, but sensitivity was sufficient only for the 18-months-and-older age versions and positive predictive values hovered around 33-35%.¹⁰ That low positive predictive value means that of the children flagged as at risk by the ASQ:SE, only about a third actually had confirmed social-emotional difficulties on the gold-standard assessment. That is a lot of false alarms, which matters for families who may experience unnecessary anxiety or be put through evaluations that turn out to be unneeded.

A Singapore study using a cut-off score of 51 at 24 months found more encouraging numbers: an area under the curve of 0.82, with 70% sensitivity and 79% specificity. The ASQ:SE scores also correlated moderately with behavioral problem scores measured two years later, suggesting the tool captures something meaningful about a child’s trajectory even when it does not perfectly predict a clinical diagnosis.¹¹

Researchers have also explored whether the ASQ:SE-2 could help with early autism screening. One study found that the standard ASQ:SE-2 identified nearly all children in a high-risk clinical sample, but its specificity for autism spectrum disorder specifically was very low, around 4%. An experimental ASD-focused subscale derived from the same items raised that specificity to 52% and correctly predicted an ASD diagnosis in 70% of cases.¹² The ASQ:SE-2 is not designed or validated as an autism screener, and dedicated tools like the M-CHAT remain the standard for that purpose.

How the ASQ Stacks Up Against Other Screening Tools

The ASQ is not the only developmental screening questionnaire available to pediatricians. The Parents’ Evaluation of Developmental Status (PEDS) and the Survey of Wellbeing of Young Children (SWYC) are common alternatives. A head-to-head comparison published in JAMA Pediatrics found that for children under about 42 months, the ASQ-3 and SWYC Milestones both had higher specificity than the PEDS (roughly 89% versus 80%), while sensitivity differences were not statistically significant across tools. For older children, the ASQ-3’s specificity stayed high above 92%, but its sensitivity to mild delays dropped sharply to around 24%, while the PEDS and SWYC did better at catching milder problems in that age range. Across all tools, sensitivity exceeded 70% only for severe delays.¹³

An interesting earlier finding showed that the ASQ and PEDS do not even flag the same children. In a sample of 60 children, about 52% passed both screens and 15% failed both, but a third failed one tool and passed the other. Agreement between the two tests was statistically no better than chance.¹⁴ This underscores that no single screening questionnaire is a perfect net. Each tool has different questions, different sensitivities to different types of delays, and different patterns of false positives and false negatives. Some clinics use more than one tool at different ages to compensate.

Using the ASQ in Different Countries and Cultures

The ASQ was developed and normed on English-speaking populations in the United States, which raises obvious questions about whether it works the same way in other cultural contexts. The short answer is: it can, but it usually needs adjustments.

When researchers adapted the ASQ for use in Iran, they found that several items, particularly in the communication domain, required modification to reflect local language and cultural practices. Similar modifications were needed for Korean and Turkish adaptations. By contrast, Norwegian and Dutch versions needed little content change.¹⁵ The researchers concluded that the ASQ cannot be considered entirely culture-free, especially in countries outside the Western sphere. A Korean validation study found that internal consistency and concurrent validity remained strong after translation, and Rasch analyses showed that most items functioned similarly across cultures, though some differences persisted.¹⁶

A Taiwanese adaptation of the 36-month ASQ showed initial cultural appropriateness and acceptable reliability and validity.¹⁷ In lower-resource settings, the ASQ has also been administered by trained providers rather than by parents, which changes the dynamic. A study in Guatemala found that provider-administered ASQ results had reasonable validity against the Bayley Scales, demonstrating the tool’s potential even in contexts where parent literacy or familiarity with formal questionnaires is limited.¹⁸

What Happens After a Positive Screen

A positive ASQ result is supposed to trigger a referral for comprehensive developmental evaluation, and often an early intervention referral for children under three. In practice, the referral pipeline is leakier than most parents realize. A study of community primary care clinics found that only about 42% of children with positive ASQ-3 screens actually received an early intervention referral.¹⁹ The likelihood of referral was higher when children had low scores in communication or gross motor domains specifically, suggesting that providers may take some types of delays more seriously than others, or that certain flag patterns feel more clinically urgent.

This referral gap is not mainly a problem with the ASQ itself; it reflects broader challenges in how developmental screening fits into busy primary care workflows. A mixed-methods study of urban practices identified several obstacles: physicians who preferred their own clinical judgment over standardized tools, limited time for counseling families about results, and insufficient training on how to connect families with services.²⁰ Parents in that study said they wanted more discussion about their child’s development and more time with the physician, while physicians often did not fully trust parental input. The irony is that the ASQ was specifically designed to capture parental observations, and some of the physicians using it were discounting the very data it collects.

Cost and Practical Advantages

One reason the ASQ has been so widely adopted is cost. Because parents do most of the work, administration costs are essentially zero when no staff assistance is required, compared to up to $67 for lengthy provider-administered assessments like the Neonatal Behavioral Assessment Scale.²¹ The cost-effectiveness appeal was recognized from the tool’s earliest development.³

This low cost also makes the ASQ attractive for large-scale public health programs, not just individual pediatric practices. A cluster-randomized trial in China that integrated a parenting intervention with primary health care used the ASQ to measure outcomes and estimated the cost of improving a child’s overall ASQ score by one standard deviation at roughly $203 per child.²² In those contexts, the ASQ serves double duty as both a screening instrument and a population-level outcome measure.

The Preterm Adjustment

For children born prematurely, most providers adjust the child’s age before selecting which ASQ form to use. If a baby was born eight weeks early, at a chronological age of 12 months the provider would typically use the 10-month questionnaire instead. This “corrected age” approach is standard practice, but it introduces its own wrinkles. A study examining parent-reported ASQ accuracy in preterm infants found that the tool had strong negative predictive value for gross and fine motor screening at 6 and 12 months of corrected age, but was less reliable for screening language skills at 18 months.²³ Parents of preemies should know that the ASQ performs differently at different ages and across different skill areas for their children, and a single clean screen at one visit does not substitute for ongoing monitoring.

Moving to Digital Formats

Increasingly, parents complete the ASQ on a tablet in the waiting room or through an online link before the appointment. A realist review of digitally delivered developmental screening tools found that web-based and paper versions of the ASQ were generally equivalent, and that digital administration could be supported effectively by features like examiner availability, time accommodations, and family comfort with technology.²⁴ For practices trying to integrate screening into electronic health records, digital ASQ completion is appealing because scores can auto-calculate and flag results for the provider before the visit even begins. Families who might lose a paper form in a diaper bag or skip the questionnaire when it arrives in the mail are sometimes more likely to complete it if a text message sends them a direct link.

That said, digital delivery is not a cure-all. Families without reliable internet access, those with lower digital literacy, and those who speak a language the digital platform does not support may be disadvantaged by the shift away from paper. Public health programs rolling out digital ASQ systems need to keep paper alternatives available to avoid inadvertently widening the screening gap for already-underserved populations.

Maternal Education and Score Patterns

One underappreciated factor in ASQ results is the education level of the parent completing the form. Research has found that maternal education influences ASQ performance, which could mean that the same child’s skills get reported differently depending on how the parent interprets the questions.⁵ A parent with more formal education might recognize a skill described in the questionnaire more readily, or might have more experience with the types of activities the questions reference, like stacking blocks or scribbling with crayons. This does not mean the tool is unreliable for families with less education, but it does suggest that clinicians should be aware of the potential for scoring differences and offer help completing the forms when needed rather than assuming all parents will interpret every item the same way.