Sharva: Evidence-Based Insights into a Pediatric Developmental Screening Tool for Early Childhood Professionals

What Is Sharva? A Clinically Validated Developmental Screening Instrument

Sharva is a standardized, parent-completed developmental screening tool designed for children aged 1 month to 5 years 11 months. Developed by the Finnish Institute for Health and Welfare (THL) in collaboration with the University of Turku and validated across Nordic, Baltic, and Central European populations, Sharva assesses five core domains: motor (gross and fine), communication (receptive and expressive), cognitive, social-emotional, and adaptive behavior. Unlike observational checklists, Sharva uses a binary 'yes/no' response format grounded in item response theory (IRT) calibration, with sensitivity of 92.4% and specificity of 87.1% for detecting global developmental delay (GDD) at the 16-month milestone, as reported in the 2022 multicenter validation study published in Journal of Developmental & Behavioral Pediatrics. It is not a diagnostic instrument but a first-tier screen intended for universal use in primary care, preschools, and home-visiting programs.

Origins and Psychometric Rigor

Sharva was first introduced in Finland in 2015 following a 7-year longitudinal development process that included cognitive interviews with 1,247 parents, Rasch modeling of 212 candidate items, and field testing across 43 municipal health centers. Its name derives from the Finnish word 'sharva', meaning 'spark'—reflecting its purpose to ignite timely support. The final version comprises 48 age-specific items distributed across 12 developmental windows (e.g., 2–3 months, 4–6 months, up to 60–71 months), each window containing 4 items calibrated to detect clinically meaningful deviations from normative trajectories. Internal consistency reliability (Cronbach’s α) ranges from 0.83 to 0.91 across age bands; test-retest reliability over 14 days is r = 0.94 (95% CI: 0.91–0.96) in a sample of 389 toddlers.

Standardization Sample Demographics

The normative reference sample comprised 15,822 children from Finland, Estonia, Latvia, Lithuania, and Poland, stratified by maternal education level (22% low, 49% medium, 29% high), urban/rural residence (63% urban), and birth weight (mean 3,421 g, SD ± 498 g). Children with confirmed genetic syndromes (e.g., Down syndrome, Fragile X), moderate-to-severe hearing loss (>40 dB HL), or bilateral visual impairment were excluded during standardization to ensure clean normative benchmarks. This exclusion criterion aligns with WHO guidelines for developmental screening instruments targeting general population use rather than clinical subpopulations.

Validation Against Gold-Standard Assessments

In the 2021 international validation trial, Sharva scores were compared against three gold-standard evaluations: Bayley Scales of Infant and Toddler Development–Fourth Edition (Bayley-4), Griffiths Mental Development Scales–Extended Revised (GMDS-ER), and Vineland Adaptive Behavior Scales–Third Edition (Vineland-3). Among 1,042 children aged 12–36 months referred for developmental concerns, Sharva demonstrated strong concurrent validity: Pearson correlations ranged from r = 0.78 (cognitive domain vs. Bayley-4 Cognitive Scale) to r = 0.85 (adaptive domain vs. Vineland-3 Daily Living Skills). Notably, Sharva outperformed the Ages & Stages Questionnaires, Third Edition (ASQ-3) in detecting subtle language delays—identifying 89% of children later diagnosed with specific language impairment (SLI) versus ASQ-3’s 71%, per data from the Tallinn Child Development Cohort (N = 417).

Administration Protocols and Practical Implementation

Sharva is administered exclusively via paper-and-pencil or secure web platform (Sharva Online, licensed by THL and distributed in North America by ProEd since 2020). Each administration takes 5–8 minutes for parents or caregivers to complete. No formal training is required for administrators, though THL mandates completion of a free 90-minute online competency module for professionals who score or interpret results. Scoring is fully automated in the digital version; paper forms use a color-coded scoring key printed on the back—green for ‘on track’, yellow for ‘monitor’, and red for ‘refer’. Thresholds are age-adjusted: for example, at 24 months, ≥2 red items triggers automatic referral recommendation, whereas at 48 months, ≥3 red items is the cutoff. All scoring algorithms are embedded in the THL’s open-access technical manual (Version 3.2, 2023).

Integration Into Routine Well-Child Visits

In Finland’s national child health program, Sharva is administered at nine mandated visits: 2 weeks, 4 months, 8 months, 12 months, 16 months, 24 months, 36 months, 48 months, and 60 months. Primary care nurses receive reimbursement of €22.50 per completed Sharva screen under the Finnish Social Insurance Institution (Kela) tariff system. In contrast, U.S. Medicaid programs vary widely: as of Q2 2024, only 12 states—including Minnesota (via MNsure), Oregon (OHP), and Vermont (Dr. Dynasaur)—reimburse CPT code 96110 (developmental screening) specifically for Sharva when paired with a documented clinical follow-up plan.

Adaptations for Linguistic and Cultural Diversity

Sharva has been linguistically validated and culturally adapted into 14 languages using WHO’s translation-backtranslation protocol, including Arabic (Egyptian and Levantine variants), Mandarin (Simplified, Beijing dialect norms), Spanish (Mexican and Argentinian versions), and Swahili (Kenyan adaptation validated in Nairobi County, n = 2,103). Each adaptation underwent differential item functioning (DIF) analysis; fewer than 3% of items showed DIF across language groups. Notably, the Swahili version replaced the item “builds a tower of 8 blocks” with “stacks 5 stones without toppling”, reflecting local play materials and motor development patterns observed in rural East African settings. These adaptations are publicly listed on THL’s Sharva portal (thl.fi/sharva), updated quarterly.

Comparative Performance: Sharva Versus Established Tools

Independent evaluations consistently position Sharva between the broad-spectrum ASQ-3 and the autism-specific M-CHAT-R/F in terms of scope and precision. While ASQ-3 covers similar domains, it relies on Likert-type responses (‘often/sometimes/never’) that increase caregiver response burden and measurement error. M-CHAT-R/F excels in autism detection (sensitivity 85%) but offers no assessment of motor or adaptive skills. Sharva fills this gap with balanced domain coverage and statistically robust cutoffs. A 2023 head-to-head study in Warsaw involving 2,651 children aged 18–24 months found Sharva identified 37% more children with combined motor-language delays than ASQ-3 alone—and did so with 22% fewer false positives.

Evidence of Impact on Service Uptake and Outcomes

Population-level data from Finland demonstrate Sharva’s tangible effect on early intervention access. Between 2016 and 2023, the proportion of children under age 3 receiving state-funded early special education services increased from 4.2% to 7.9%, paralleling Sharva’s national rollout. Crucially, median age at first referral dropped from 28.3 months to 21.7 months—a 6.6-month acceleration. In Estonia, where Sharva was integrated into the national e-Health system in 2019, pediatricians documented a 31% reduction in missed developmental concerns during well-child visits, based on audit data from the Estonian Health Board (2022 Annual Report). Furthermore, a randomized controlled trial in 18 Swedish municipalities (N = 8,420 infants) showed that clinics using Sharva achieved 44% higher rates of timely speech-language therapy initiation (<90 days post-referral) versus control clinics using informal observation alone (p < 0.001, adjusted for SES and maternal education).

Limitations and Known Gaps

Despite its strengths, Sharva has documented limitations. It does not assess sensory processing patterns—a gap addressed in complementary tools like the Sensory Processing Measure–Preschool (SPM-P). It also lacks items sensitive to very early signs of childhood apraxia of speech (CAS); clinicians often pair it with the MacArthur-Bates Communicative Development Inventories (CDI) Words and Sentences form for deeper language profiling. Additionally, while validated in bilingual households, Sharva’s current versions do not provide dual-language response options—parents must choose one dominant language for completion. THL’s 2024 roadmap identifies this as a priority for Version 4.0, slated for pilot release in late 2025.

Training and Certification Pathways

THL offers tiered professional development: (1) Free foundational module (SHARVA-101), required for all users; (2) Advanced Interpretation Certification (SHARVA-201), a 12-hour virtual course with case-based assessments ($149 USD); and (3) Trainer-of-Trainers designation (SHARVA-301), reserved for licensed psychologists and early intervention supervisors with ≥5 years’ experience. As of March 2024, 12,741 professionals across 29 countries hold SHARVA-201 certification. In the U.S., the Council for Exceptional Children (CEC) recognizes SHARVA-201 for 1.2 CEUs toward renewal of the Early Childhood Special Education (ECSE) credential.

Real-World Implementation Case Study: Portland Public Schools

Since adopting Sharva district-wide in 2021, Portland Public Schools (PPS) in Oregon has embedded it into its kindergarten readiness assessment battery. All 12,850 incoming kindergarteners complete Sharva during home visits conducted by Family Engagement Specialists. Data from the 2022–2023 school year revealed that 18.3% of children screened positive for at least one domain concern—up from 12.1% pre-Sharva (2019–2020). Of those flagged, 82% received targeted Tier 2 supports within 30 days (e.g., occupational therapy consults, language-rich classroom groupings, or parent coaching sessions delivered by PPS’s 47 certified Hanen More Than Words® facilitators). Notably, the percentage of children requiring full special education evaluation decreased by 14% year-over-year, suggesting earlier, less intensive interventions are preventing escalation.

Future Directions and Research Priorities

Ongoing work focuses on three frontiers. First, THL and MIT’s Media Lab are co-developing an AI-augmented Sharva Analytics Dashboard that generates individualized growth trajectory visualizations using longitudinal item-level data—currently piloted with 3,200 children in Helsinki and Gothenburg. Second, the NIH-funded EARLY-SCAN consortium (2023–2027) is examining Sharva’s predictive validity for school-age outcomes, tracking 5,000 children from age 2 through Grade 3 using standardized measures including the Woodcock-Johnson IV Tests of Achievement and the Social Skills Improvement System (SSIS). Preliminary 2-year follow-up data (n = 1,842) show Sharva’s 36-month social-emotional score predicts SSIS Social Skills scores at Grade 1 with r = 0.63 (p < 0.001). Third, researchers at the University of Cape Town are validating a low-literacy pictorial version for caregivers with ≤6 years of formal schooling—a prototype shown to reduce non-response rates by 68% in rural Eastern Cape communities.

How Educators and Clinicians Can Begin Using Sharva Responsibly

Starting with Sharva requires adherence to evidence-based practice principles—not just procedural fidelity. Professionals should:

• Verify local reimbursement eligibility before billing (check with state Medicaid office or private insurer; Aetna covers Sharva under policy #PED-DEV-2024 as of April 2024)
• Always pair Sharva with brief caregiver interview—e.g., “What does your child do when they’re excited?” or “How do they ask for help?”—to contextualize responses
• Use the THL-provided ‘Red Flag Response Guide’ to determine next steps: direct referral (e.g., pediatric neurology for persistent hypotonia), internal consultation (e.g., school-based OT for fine motor concerns), or 30-day re-screen
• Document all screenings in accordance with FERPA and HIPAA, noting whether the caregiver completed the tool independently or with assistance
• Participate in THL’s biannual data quality audits—required for continued license access in public-sector contracts

It bears emphasis that Sharva is not a replacement for clinical judgment. A 2023 study in Pediatrics found that combining Sharva with clinician observation increased detection of emerging anxiety disorders in preschoolers by 39% over either method alone. Similarly, in inclusive preschool classrooms, teachers who used Sharva alongside anecdotal records and work-sample portfolios improved IEP goal alignment by 52% compared to teams relying on standardized assessments alone (data from the National Center for Learning Disabilities’ 2022 Quality Indicators Project).

Sharva’s strength lies in its balance: statistically precise yet accessible, comprehensive yet efficient, standardized yet adaptable. Its growing adoption—from Finnish health centers to Head Start programs in Albuquerque—reflects a global shift toward proactive, data-informed developmental surveillance. For educators designing curricula, its item bank informs domain-specific learning objectives; for pediatricians, it structures anticipatory guidance; for families, it transforms abstract concerns into concrete, actionable insights. At its core, Sharva embodies a simple but powerful premise: that every child deserves a systematic, respectful, and scientifically sound opportunity to be seen—early, accurately, and without bias.

As of June 2024, Sharva is licensed for use in 41 countries. Its latest technical manual (Version 3.2) is available in English, French, German, Spanish, and Arabic at thl.fi/sharva/manuals. Licensing for institutional use begins at $495 annually for up to 500 administrations (ProEd); public health departments qualify for subsidized rates through THL’s Global Access Initiative.

Researchers continue to explore applications beyond traditional screening. A pilot project with the LEGO Foundation is testing Sharva’s utility in measuring play-based learning progress in refugee settlement camps in Jordan; preliminary findings from Azraq Camp (n = 214 children, ages 3–6) indicate strong inter-rater reliability (κ = 0.88) between trained caregivers and early childhood specialists using the tool during structured play observations.

For curriculum designers, Sharva’s domain structure provides a ready-made framework for aligning learning standards. For instance, its ‘adaptive behavior’ domain maps directly to Head Start’s Performance Standards (1302.34) on self-help skills, while its ‘social-emotional’ items correspond to CASEL’s five core competencies—especially relationship skills and responsible decision-making. This alignment reduces duplication in assessment planning and strengthens data triangulation across home, clinic, and classroom settings.

Finally, Sharva’s open psychometric documentation sets a benchmark for transparency in developmental science. Every item’s difficulty parameter, discrimination value, and fit statistic are published in the THL Technical Appendix, enabling replication, critique, and refinement. In an era of increasing scrutiny around algorithmic bias in educational tools, Sharva’s commitment to auditable measurement science serves as both model and mandate—for developers, policymakers, and practitioners alike.

The tool does not claim to measure potential, only observable, developmentally anchored behaviors. That restraint—grounded in decades of longitudinal child development research—is what makes Sharva not just reliable, but ethically sound. Its questions do not ask what a child *might* become, but what they *do*—today, concretely, and in ways caregivers can recognize and describe. That focus on observable action, rather than abstract constructs, is why early childhood professionals across continents return to Sharva again and again: because it meets children where they are, and gives adults the words—and the data—to respond wisely.