Population Versus Individual Prediction of Poor Health from Adverse Childhood Experiences (ACEs) Screening

• In Event: Investigating the Value of Screening for Adverse Childhood Experiences (ACEs) in Predicting Poor Health

Abstract

Introduction: Adverse childhood experiences (ACEs) show a dose-response relationship with health problems. Based on this evidence, ACE screening has been implemented to identify at-risk individuals who may benefit from remedial interventions. However, concerns have been raised about the utility of ACE screening in preventing poor health, due to unanswered questions about forecasting, incremental prediction, discrimination, and measurement. Regarding forecasting, it is unclear whether ACE scores are associated with future health problems, as most previous research has been cross-sectional. Regarding incremental prediction, it is unclear whether ACE scores are associated with future health problems beyond other information typically available to clinicians, such as history of health problems. Regarding discrimination, it is unclear whether ACE scores differentiate between individuals who do and do not develop later health problems, because individual differences exist in children’s responses to stress. Regarding measurement, it is unclear whether the ability of ACE scores to predict health outcomes differs depending on whether ACEs are assessed prospectively in childhood or retrospectively in adulthood. This study directly addressed these questions to inform policy-makers and practitioners about the value of screening for ACEs in improving health.

Research questions: To examine forecasting, we tested whether individuals with higher ACE scores had greater average risk of later mental and physical health problems. To examine incremental prediction, we tested whether individuals with higher ACE scores had greater risk of later health problems above other clinically available information. To examine discrimination, we tested the predictive accuracy of ACE scores in identifying individuals with or without later health problems. To examine measurement, we tested the above questions using ACE scores assessed both prospectively in childhood and retrospectively in adulthood.

Study population: Two birth cohorts: The Environmental Risk (E-Risk) Longitudinal Twin Study followed 2,232 participants born in 1994-1995 until age 18 (2012-2014), and the Dunedin Multidisciplinary Health and Development Study followed 1,037 participants born in 1972-1973 until age 45 (2017-2019).

Methods: ACEs were measured prospectively in childhood through repeated interviews and observations in both cohorts. ACEs were also measured retrospectively in the Dunedin cohort through interviews at age 38. Health outcomes were assessed at age 18 in E-Risk and age 45 in Dunedin. Mental health problems were assessed through clinical interviews using the Diagnostic Interview Schedule. Physical health problems were assessed through interviews, anthropometric measurements, and blood collection.

Results: Regarding forecasting, children from the E-Risk and Dunedin cohorts with higher ACE scores had greater risk of later mental and physical health problems than those with lower ACE scores. Regarding incremental prediction, ACE scores were associated with health problems independently of other information typically available to clinicians (i.e., sex, socioeconomic disadvantage, and history of health problems). Regarding discrimination, ACE scores had poor accuracy in predicting an individual’s risk of later health problems (Area Under the ROC Curve<0.63 for mental health problems and <0.60 for physical health problems). Regarding measurement, findings were consistent for ACE scores obtained both prospectively in childhood and retrospectively in adulthood.

Authors

• Jessie R Baldwin, University College London (UCL)
  Presenting Author

• Avshalom Caspi, Duke University
  Non-Presenting Author

• Alan Meehan, King's College London
  Non-Presenting Author

• Antony Ambler, King's College London
  Non-Presenting Author

• Louise Arseneault, King's College London
  Non-Presenting Author

• Helen L Fisher, King's College London
  Non-Presenting Author

• HonaLee Harrington, Duke University
  Non-Presenting Author

• Timothy Matthews, King's College London
  Non-Presenting Author

• Candice L. Odgers, University of California, Irvine
  Non-Presenting Author

• Richie Poulton, University of Otago
  Non-Presenting Author

• Sandhya Ramrakha, University of Otago
  Non-Presenting Author

• Terrie E Moffitt, King's College London
  Non-Presenting Author

• Andrea Danese, King's College London
  Non-Presenting Author