2026 ECNP Congress

PS02-2154

Testing the validity of a digital diagnostic instrument: neurodevelopmental diagnoses in a South African sample of 178 children

Sara Landberg — G. Martin, L. Flusk, J. Afram, I. Michaelis, H. Minnis, M. Broadbent, M. Fordham, C. Gillberg · University of Gothenburg, Gillberg Neuropsychiatry Centre, Gothenburg, Sweden

2026 ECNP Congress

Background

Symptoms of neurodevelopmental conditions are highly prevalent, yet individuals are often not correctly identified and diagnosed for many years. This is particularly evident in low- and middle income settings, where access to specialised assessments is limited. Furthermore, despite growing knowledge that symptoms of these conditions overlap to a great extent, assessments are often narrow and only focused on one condition at the time, for example ADHD. The concept ESSENCE (Early Symptomatic Syndromes Eliciting Neurodevelopmental Clinical Examinations) was developed [1], highlighting the need for broad, transdiagnostic assessment approaches. On this foundation, the ESSENCE Diagnostic Aid (ESSENCE-D), a digital tool, was developed by members of this team [2]. The tool alerts clinicians to symptomatic areas, in advance of consultation. The tool has here been tested in a South African sample of 178 children, aged 2-17 years old.

Aims

The aim was to examine the predictive validity of ESSENCE-D by comparing its risk output with ESSENCE-D blinded clinician-assigned neurodevelopmental diagnoses in this population.

Methods

A prospective, cross-sectional validation study was conducted at a neurodevelopmental outpatient clinic in Eastern Cape, South Africa. Participants were 178 children aged 2–17 years, referred for neurodevelopmental assessment. The ESSENCE-D is an online tool, consisting of 144 items organised into functional domains and generates low, medium, or high risk flags across multiple neurodevelopmental conditions and commonly associated symptoms. Caregivers completed the ESSENCE-D online, prior to the child’s clinical evaluation. Clinicians, blinded to ESSENCE-D results, assigned diagnoses according to ICD-10 and/or DSM-5 criteria following standard clinical assessment. Logistic regression analyses were performed to examine associations between ESSENCE-D risk levels and clinical diagnoses for diagnostic categories with sufficient sample size, while controlling for age, sex, and socioeconomic status.

Results

ESSENCE-D significantly predicted subsequent clinical diagnoses of ADHD, autism, intellectual disability, and speech delay. For all examined conditions, higher ESSENCE-D risk levels were associated with increased odds of receiving the corresponding clinical diagnosis (see Table). Across diagnostic categories, ESSENCE-D flagged more children than those ultimately diagnosed, while relatively few clinically diagnosed cases were missed by the tool. Children who were flagged by ESSENCE-D but not diagnosed with the corresponding condition frequently received other neurodevelopmental diagnoses, reflecting substantial comorbidity and symptom overlap within the clinic population.

TableLogistic regression predicting diagnostic outcome based on ESSENCE-D

Condition OR “high risk” OR “medium risk” p-value
ADHD 5.73 1.49 0.001
Autism 20.7 7.82 0.004
Intellectual disability 3.08 1.45 0.047
Speech delay 29.8 2.71 <0.001

Note. Categories with n > 50.

Conclusions

These findings provide support for the predictive validity of ESSENCE-D in identifying neurodevelopmental conditions within a clinically complex and socioeconomically disadvantaged setting. The tendency toward over-flagging rather than under-identification is consistent with ESSENCE-D’s intended role as a clinical decision-support aid rather than an independent diagnostic instrument. ESSENCE-D may facilitate more comprehensive and systematic neurodevelopmental assessments, which can improve accessibility of specialised assessment and thereby access to appropriate care and support. Further randomised controlled studies are warranted to evaluate whether the use of ESSENCE-D influences diagnostic decisions, detection of comorbidities, and clinical outcomes in diverse healthcare settings.

References

  1. Gillberg C. The ESSENCE in child psychiatry: Early Symptomatic Syndromes Eliciting Neurodevelopmental Clinical Examinations. Res Dev Disabil. 2010;31(6):1543–51.
  2. Landberg S, Broadbent M, Fordham M, Chadwick G, Minnis H, Lang J, et al. The development of the ESSENCE Diagnostic aid (ESSENCE-D). Child & Youth Services. 2025:1–17.