Truace
Subscribe

Non-medical drivers of health and pediatric no-show prediction: identifying modifiable risk factors

Background No-shows to medical appointments contribute to disruptions and inefficiencies in healthcare delivery. While artificial intelligence/machine learning (AI/ML) models are increasingly used to predict no-shows, few have incorporated non-medical drivers of health (NMDOH), particularly in pediatric populations. This study evaluated the inclusion of NMDOH in pediatric AI/ML no-show prediction and identified modifiable factors associated with no-shows. Methods A retrospective study was conducted of scheduled…

By the Truvace desk1 min read1 sourcePeer-reviewedTRV-2026-1614· cite & verify
Non-medical drivers of health and pediatric no-show prediction: identifying modifiable risk factors
Hospital Universitari Doctor Peset, València 09 by 19Tarrestnom65. CC BY-SA 4.0 · https://creativecommons.org/licenses/by-sa/4.0

In brief

Researchers reviewed 7,931 scheduled pediatric appointments across six general pediatrics clinics between April and August 2024, and 34.8% ended as no-shows. They tested a transformer-based AI/ML model with and without screening data on non-medical drivers of health, such as transportation and housing.

Adding those social variables left predictive accuracy essentially unchanged, so the main value of the data lay in explaining why families miss visits rather than in sharper forecasts. The authors suggest predictions are most useful when tied to outreach and barrier reduction, though the study does not test whether such interventions reduce no-shows.

Main points

  1. Of 7,931 appointments studied, 2,761 (34.8%) were no-shows.
  2. AUROC was 0.708 with and without NMDOH variables; AUPRC was 0.536 versus 0.535.
  3. Demographic and insurance variables gave a small gain, to AUROC 0.721 and AUPRC 0.542, with or without NMDOH.
  4. Unconfirmed appointments, inactive portal status, transportation needs and housing instability were linked to no-shows.

The gain

Analysis of the model's inputs identified actionable factors tied to missed pediatric visits, such as transportation needs, housing instability, unconfirmed appointments and inactive portal status, which clinics can target.

The rundown

The retrospective study drew on electronic health record data and NMDOH screening results from six general pediatrics clinics. Model performance was measured with AUROC and AUPRC, and multivariable logistic regression was used to examine which factors were associated with no-show status.

Appointment confirmation and patient portal activation are operational factors that clinics can change, alongside social barriers like transportation and housing. The authors frame these as modifiable targets for clinical and operational interventions.

The paper concludes that AI-based no-show prediction may be most effective when combined with targeted outreach and barrier-reduction strategies, rather than relying on better prediction alone.

Sources

  1. Peer-reviewedPediatric Research2026-10-08

The debate