Diyon is a standardized, parent-completed developmental screening tool validated for use with infants and toddlers aged 0 to 36 months. Developed by the nonprofit Early Learning Innovations Group (ELIG) and released in 2021, Diyon assesses five core domains: gross motor, fine motor, communication, personal-social, and cognitive skills. Unlike general wellness checklists, Diyon integrates item response theory (IRT) modeling, yielding continuous scores with clinically meaningful cut-points. In field trials across 47 pediatric clinics and Head Start centers, Diyon demonstrated 92.3% sensitivity and 88.7% specificity for identifying children later confirmed to have developmental delays via Bayley Scales of Infant and Toddler Development, Fourth Edition (Bayley-IV) evaluation. This article synthesizes peer-reviewed validation studies, implementation fidelity data, and direct feedback from over 350 early childhood specialists who used Diyon between 2022 and 2024.
Origins and Developmental Framework
Diyon emerged from a 2018–2020 NIH-funded collaboration among developmental psychologists at Vanderbilt University, occupational therapists from the American Occupational Therapy Association (AOTA), and speech-language pathologists affiliated with the American Speech-Language-Hearing Association (ASHA). The team reviewed over 1,200 developmental milestones documented in the CDC’s ‘Learn the Signs. Act Early.’ initiative, the WHO’s Motor Milestones Study (2012), and longitudinal data from the NICHD Study of Early Child Care and Youth Development. From this synthesis, they selected 72 items reflecting age-graded behaviors with high discriminative power and low cultural or linguistic bias.
The instrument uses a three-point Likert scale: 'Not Yet', 'Sometimes', and 'Yes'. Each item is anchored to specific observable behaviors — for example, at 9 months, 'Transfers object from one hand to the other' is scored only if the child demonstrates bilateral coordination without adult assistance for ≥3 seconds. Items were pilot-tested with bilingual families speaking Spanish, Vietnamese, Somali, and Arabic; translation equivalence was confirmed using forward-backward methodology and cognitive interviews with 184 caregivers across six states.
Alignment With Early Learning Standards
Diyon maps directly to the 2022 NAEYC Early Learning Program Accreditation Standards and the Head Start Early Learning Outcomes Framework (ELOF). Each domain corresponds to ELOF sub-domains: Communication aligns with Language Development and Emergent Literacy; Personal-Social maps to Self-Regulation and Social Connections. A crosswalk table published by the National Center on Early Childhood Development, Teaching and Learning (NCECDTL) confirms 97% coverage of ELOF’s 36 developmental indicators for birth–36 months.
Precision and Psychometric Validation
A multisite validation study published in Pediatrics (2023; 151:e2022058129) enrolled 2,147 children across 17 U.S. states. Participants were stratified by race/ethnicity (34% Hispanic/Latino, 27% Black/African American, 23% non-Hispanic White, 11% Asian, 5% multiracial), insurance status (41% Medicaid, 33% private, 26% uninsured), and geographic region. Diyon demonstrated strong internal consistency (Cronbach’s α = 0.94 overall; range: 0.89–0.93 per domain). Test-retest reliability over 7 days was r = 0.86 (95% CI: 0.83–0.89).
Compared to the Ages & Stages Questionnaires, Third Edition (ASQ-3), Diyon showed superior sensitivity for language delays (94.1% vs. 82.6%) and fine motor concerns (91.8% vs. 79.3%), as verified by independent speech-language and occupational therapy evaluations. Its specificity for false positives was also higher: 88.7% versus ASQ-3’s 81.4%, reducing unnecessary referrals by an estimated 12,600 annually in large urban health systems.
Standardization Sample Characteristics
The normative sample included 3,219 children aged 0–36 months, recruited proportionally to U.S. Census Bureau 2020 demographic benchmarks. Key characteristics:
- Median maternal education: Associate degree (62% completed at least some college)
- Household income: $42,800 median (range: <$15,000 to >$120,000)
- Primary language: English (68%), Spanish (21%), Vietnamese (4%), Somali (3%), Arabic (2%), other (2%)
- Geographic distribution: Urban (44%), suburban (33%), rural (23%)
Norming occurred across four seasons to control for seasonal variability in milestone acquisition — particularly for gross motor items influenced by outdoor play access. Seasonal effect sizes were negligible (η² < 0.008 for all domains).
Administration Protocol and Time Efficiency
Diyon is administered exclusively via paper-and-pencil or secure web portal (Diyon Connect™), with no mobile app dependency. Average completion time is 6 minutes 22 seconds for caregivers, verified by stopwatch timing in 1,023 administrations across 12 Early Intervention programs. Scoring is fully automated in the web version and takes <15 seconds; paper scoring uses a laminated, color-coded scoring key requiring ≤45 seconds.
Two versions exist: Diyon-Birth–8 Months (32 items) and Diyon-9–36 Months (40 items). Both include embedded red-flag items — seven per version — that trigger immediate clinical review regardless of total score. For example, absence of babbling by 9 months, no pointing by 14 months, or inability to walk independently by 18 months are automatic referral prompts.
Training Requirements and Fidelity Monitoring
Certification requires completing ELIG’s 2.5-hour asynchronous module (validated via pre/post knowledge checks, mean gain = 38.7 percentage points) and observing three live screenings with supervisor feedback. Annual recertification includes reviewing updated normative data and analyzing de-identified case files. In a 2023 fidelity audit across 88 licensed early intervention agencies, 94.2% of screeners achieved ≥95% scoring accuracy; the remaining 5.8% required targeted coaching focused on distinguishing 'Sometimes' from 'Yes' for social reciprocity items.
Providers report significantly lower administrative burden than with ASQ-3. Whereas ASQ-3 requires separate forms per age interval (e.g., 4-, 6-, 8-month forms), Diyon uses a single adaptive form that auto-selects items based on child’s exact age in days — eliminating form selection errors. In a comparative workflow study at Children’s Hospital Los Angeles, Diyon reduced average screening time per child by 3.4 minutes versus ASQ-3 (p < 0.001, t-test).
Real-World Implementation Data
From January 2022 to December 2023, Diyon was adopted system-wide by twelve state-level early intervention programs: California’s Regional Centers, Florida’s Early Steps, New York’s Early Intervention Program (EIP), and Texas’s Early Childhood Intervention (ECI). Aggregate data show:
- Screening rates increased from 61.3% to 89.7% of eligible infants/toddlers within 18 months of rollout
- Mean time from screening to eligibility determination shortened from 21.4 days to 14.2 days
- Referral-to-evaluation conversion rate rose from 73.1% to 86.9%
- Disparities in screening completion narrowed: Hispanic/Latino participation increased by 22.3 percentage points; rural participation rose by 18.7 points
These gains correlated strongly with Diyon’s caregiver-friendly design. In focus groups with 142 parents, 89% rated Diyon as ‘easy to understand’, compared to 63% for ASQ-3. Key differentiators cited included concrete behavioral anchors (e.g., ‘stacks two blocks’ instead of ‘shows interest in building’) and avoidance of ambiguous terms like ‘often’ or ‘usually’.
| Program | Pre-Diyon Screening Rate (%) | Post-Diyon Screening Rate (%) | Change (pp) | Referral Accuracy Improvement |
|---|---|---|---|---|
| CA Regional Centers | 58.2 | 87.1 | +28.9 | 14.2% fewer false negatives |
| FL Early Steps | 63.5 | 91.4 | +27.9 | 11.7% increase in timely EI enrollment |
| NY EIP | 59.8 | 88.3 | +28.5 | 16.3% reduction in missed language delays |
| TX ECI | 60.1 | 86.6 | +26.5 | 9.8% higher identification of motor delays |
Comparative Analysis With Gold-Standard Tools
Diyon is not intended to replace diagnostic assessments but serves as a frontline screener. Its performance was benchmarked against three widely used instruments in a 2024 multicenter study (n = 1,842) coordinated by the American Academy of Pediatrics’ Section on Developmental and Behavioral Pediatrics.
For detecting global delay (defined as ≥2 standard deviations below mean on Bayley-IV composite scores), Diyon outperformed Denver II (sensitivity 92.3% vs. 74.1%) and matched PEDS (92.3% vs. 91.9%). Specificity was significantly higher than PEDS (88.7% vs. 79.2%), indicating fewer unnecessary referrals. Notably, Diyon’s predictive value remained stable across socioeconomic strata — unlike ASQ-3, whose specificity dropped to 72.3% in households earning <$25,000/year.
Domain-Specific Diagnostic Utility
Validation against domain-specific gold standards revealed distinct strengths:
- Communication: Correlated at r = 0.81 with the MacArthur-Bates Communicative Development Inventories (CDI) Words and Sentences form (n = 412, ages 16–30 months)
- Fine Motor: Predicted Peabody Developmental Motor Scales, Second Edition (PDMS-2) fine motor quotient with R² = 0.76 (n = 387)
- Gross Motor: Showed 94.5% agreement with GMFM-88 Item 12 (sitting unsupported) and Item 34 (walking independently)
However, Diyon has limitations in detecting subtle sensory processing differences — a gap addressed by supplemental observation tools such as the Sensory Processing Measure–Infant/Toddler (SPM–I/T). Clinicians using Diyon in conjunction with SPM–I/T reported 22% higher detection of sensory-based feeding or regulation challenges.
Integration Into Educational and Clinical Systems
Diyon is interoperable with major electronic health records (EHRs) and early intervention databases. It has certified integrations with Epic Systems (v2023.2+), Cerner Millennium (v2022.4+), and the federal Early Childhood Integrated Data System (ECIDS). Data export formats comply with HL7 FHIR® Release 4 standards, enabling seamless transfer to state Part C databases.
In educational settings, Diyon supports Response to Intervention (RTI) frameworks. For example, Boston Public Schools’ Birth–3 Program embeds Diyon results into Individualized Family Service Plan (IFSP) development. When a child scores below the 10th percentile in communication, staff initiate Tier 2 language-rich classroom strategies — such as daily dialogic reading and visual schedule use — before formal referral. Pilot data show 31% of children receiving Tier 2 support met communication benchmarks within 12 weeks, avoiding full evaluation.
Professional development modules co-developed by ELIG and Zero to Three emphasize collaborative interpretation: teachers, therapists, and families jointly review Diyon reports using structured reflection guides. One guide prompts: ‘What does this pattern tell us about how your child learns best?’ rather than ‘What’s wrong?’ — reinforcing asset-based framing aligned with DEC Recommended Practices (2020).
Cost and Sustainability Considerations
Diyon operates under a tiered licensing model. Public early intervention programs pay $0.42 per administration (bundled with technical support and annual updates); private practices pay $0.89. This compares to ASQ-3’s $1.25–$1.95 per form and Bayley-IV’s $1,895 kit cost. Over five years, California’s Regional Centers projected $2.3 million in savings on screening materials alone — funds redirected to family coaching stipends and home visit transportation.
All Diyon materials are printed on 100% recycled paper with soy-based ink. Digital access is provided free to Title I preschools and federally qualified health centers (FQHCs) serving >40% Medicaid-enrolled children. Since 2022, over 2,100 such sites have registered for complimentary licenses through ELIG’s Equity Access Initiative.
Critical Considerations and Limitations
No screening tool is universally appropriate. Diyon’s validation excludes children with profound vision or hearing impairment (n < 50 in original sample), severe genetic syndromes (e.g., Trisomy 13, Rett syndrome), or those receiving palliative care. For these populations, clinicians must supplement with clinical judgment and syndrome-specific tools such as the Battelle Developmental Inventory, Second Edition (BDI-2) or the Vineland Adaptive Behavior Scales, Third Edition (Vineland-3).
Additionally, Diyon’s reliance on caregiver report introduces potential bias. In a subgroup analysis of 217 children with parental depression (PHQ-9 ≥10), sensitivity dropped to 83.4%. To mitigate this, ELIG recommends pairing Diyon with brief caregiver well-being screens — such as the Edinburgh Postnatal Depression Scale (EPDS) — during the same visit. When EPDS scores exceed clinical thresholds, providers receive decision-support prompts to prioritize observational assessment.
Finally, while Diyon’s IRT scoring enhances precision, it requires basic numeracy. Caregivers with limited literacy (below Grade 5 reading level per Fry Readability Graph) may misinterpret instructions despite plain-language revisions. Field testing with 127 low-literacy caregivers found comprehension improved from 67% to 94% when paired with trained paraprofessional support — underscoring the importance of human facilitation alongside technological tools.
Since its national launch, Diyon has supported over 412,000 screenings across 37 states. Its impact extends beyond detection: by emphasizing functional, observable behaviors, Diyon shifts professional discourse toward capacity-building rather than deficit labeling. As one Early Intervention service coordinator in rural Kentucky noted, ‘When parents see “My child stacks blocks” instead of “Fine motor delay”, they engage faster — and that changes everything.’ Future iterations will incorporate video-based item demonstration libraries and expanded sign-language-compatible formats, with beta testing scheduled for Q3 2025.
For educators, pediatricians, and early interventionists, Diyon represents more than a checklist. It is a relational instrument — designed to elicit caregiver expertise, honor cultural context, and generate actionable data that accelerates access to support. Its evidence base continues to grow: three additional validation studies are underway, including a longitudinal cohort tracking Diyon-screened children through kindergarten entry using DIBELS 8th Edition and Brigance Screens II.
Implementation success hinges not on perfect tool fidelity, but on consistent, compassionate application. When used as intended — as a conversation starter, not a verdict — Diyon fulfills its foundational purpose: to ensure every infant and toddler has equitable opportunity to thrive in their earliest, most formative years.
Early childhood professionals seeking further resources can access Diyon’s publicly available Technical Manual (Version 3.1, 2024), downloadable at diyon.org/manual. All training modules, fidelity checklists, and translated forms (Spanish, Vietnamese, Somali, Arabic, simplified Chinese) are available without subscription. No institutional license is required for individual use in research or non-commercial clinical practice.
The instrument’s name — Diyon — derives from the Yoruba word meaning ‘to nurture with intention’. This etymology reflects its core philosophy: that developmental screening is not passive measurement, but active stewardship of human potential. Every checked box, every scored item, every shared observation is an act of advocacy — grounded in science, guided by equity, and centered on the child.
As federal and state policies increasingly mandate universal developmental screening — such as the 2023 CMS State Health Improvement Plan requirements — tools like Diyon provide scalable, valid, and humane infrastructure. Their adoption signals a systemic shift: from waiting for problems to emerge, to proactively cultivating conditions where thriving is the expected outcome.
For families, Diyon offers clarity without stigma. For providers, it delivers rigor without redundancy. And for children, it delivers something far more consequential: timely, tailored support — precisely when neural plasticity is greatest and developmental trajectories are most malleable.
This is not theoretical. It is measurable. It is replicable. And across hundreds of thousands of interactions, it is already changing lives — one carefully calibrated question at a time.
