External validation, recalibration and updating of the OxSATS risk model for suicide after self-harm in England - Scorecard - MDSpire
Coming Soon: Introducing MDSpire News. Learn more
Conexiant’s news site is now MDSpire News. Learn more

Validation and Adjustment of the OxSATS Suicide Risk Model Following Self-Harm in an English Context

  • By

  • Tyra Lagerberg

  • Denis Yukhnenko

  • Maria D L A Vazquez-Montes

  • Thomas R Fanshawe

  • Seena Fazel

  • September 12, 2026

Share

Clinical Scorecard: Validation and Adjustment of the OxSATS Suicide Risk Model Following Self-Harm in an English Context

At a Glance

CategoryDetail
ConditionSuicide Risk Following Self-Harm
Key MechanismsExternal validation and recalibration of the OxSATS model to predict suicide risk.
Target PopulationIndividuals aged ≥10 years presenting to emergency departments for self-harm.
Care SettingEmergency departments and secondary mental health services.

Key Highlights

  • OxSATS demonstrated good discrimination with a c-index of 0.73.
  • Recalibration improved model calibration with an O:E of 1.00.
  • The study is the first external validation of OxSATS in an English clinical setting.
  • Initial calibration showed overprediction of suicide risk due to lower baseline event rates.
  • Structured risk assessment tools can aid in recognizing individuals needing further risk management.

Guideline-Based Recommendations

Diagnosis

  • Utilize the OxSATS model for estimating suicide risk following self-harm.

Management

  • Consider structured risk assessment tools for comprehensive clinical assessments.

Monitoring & Follow-up

  • Implement closer monitoring and follow-up for individuals identified at high risk.

Risks

  • Be aware of the potential for overprediction in different clinical settings.

Patient & Prescribing Data

Individuals presenting to hospitals after self-harm.

Structured assessments may facilitate better resource allocation and follow-up care.

Clinical Best Practices

  • Conduct external validation of risk models before implementation in new settings.
  • Perform sensitivity analyses to address missing predictors.
  • Recalibrate models as necessary to ensure accurate risk predictions.

Related Resources & Content

    Original Source(s)

    Related Content