Skip to main content

Population vs Individual Prediction of Poor Health From Results of Adverse Childhood Experiences Screening [jamanetwork.com]

 

By Jessie R. Baldwin, Avshalom Caspi, Alan J. Meehan, et al., JAMA Pediatrics, January 25, 2021

Key Points

Question Can screening for adverse childhood experiences (ACEs) accurately predict individual risk for later health problems?

Findings In 2 population-based birth cohorts (with a total of 2927 individuals) growing up 20 years and 20β€―000 km apart, ACE scores were associated with mean group differences in health problems independent of other information available to clinicians. However, ACE scores had low accuracy in predicting health problems at the individual level.

Meaning ACE scores can forecast mean group differences in later health problems; however, ACE scores have poor accuracy in identifying individuals at high risk for future health problems.

[Please click here to read more.]

Add Comment

Comments (3)

Newest · Oldest · Popular

Excellent, thought provoking article taking a different tact.

I have a question. They apparently looked at individuals when grown and their health issues and said it does not directly correlate or predict that a particular ACE score will predict health outcomes. What confuses me is that it seems all ACE studies say that a higher ACE score is, on average, associated with a higher risk of physical and mental health issues, not a one to one correlation.

Secondly, does this study suggest we change what we do because a specific ACE score is no guarantee of an adverse outcome or, as I do, say that a specific ACE score implies a higher likelihood or risk for adverse outcomes, not a one to one correlation.  Seems like smoking. All smokers have higher risk for lung cancer, COPD, etc., but not all get it.

Perhaps this study informs us to be sure of our language with patients and others. If you have a high ACE score, your risk of adverse outcomes is higher, but it does not prove you will have adverse outcomes.

Would love to hear others' thoughts. Thanks

  Yes, the study underlines the point Dr. Anda made that ACE data provide useful population level predictions but very poor individual level predictions. For example, the ADVERSE CHILDHOOD EXPERIENCES DATA REPORT: Behavioral Risk Factor Surveillance System (BRFSS), 2011-2017: An Overview of Adverse Childhood Experiences in California (https://osg.ca.gov/sg-report/ )  reports that 1.3% of those with 4 or more ACEs experience heart disease compared to only .9% of those with no ACEs. This means an individual with four or more ACEs is β€œ1.5 times as likely to have heart disease” as is someone with no ACEs. That is useful at a population level – reducing childhood adversity would reduce the risk and the incidence of heart disease.  But predicting an individual adult would have heart disease by their high ACE score would be in error almost 99% of the time.

  Dr. Robert Anda observed that: β€œThe ACE score is a powerful tool for describing the population impact of the cumulative effect of childhood stress and provides a framework for understanding how prevention of ACEs can reduce the burden of many public health problems and concerns. However, the ACE score is neither a diagnostic tool nor is it predictive at the individual level.” (https://www.ajpmonline.org/art...(20)30058-1/fulltext )

Hi Ellen:
I also read this study and think it does / should shift how we look at using ACE scores. While those of us with high ACE scores, as a group, are more likely to have poor health outcomes, that doesn't mean we actually will. Just as important, it means that those without an ACE score will escape disease (though as a group, the risk is lower). My view is that it's too crude to use at the individual level or in a medical or clinical setting but that doesn't mean it's not an amazing, powerful, empowering tool that helps gives wider context - as do social determinants of health. Knowing risks is important but it's not deterministic.

My worry, given the biases we know exist in medicine, historically, and currently, and the risk for misuse of info. by insurance companies, I think we have to remember the value is in the  population level and not for use in individual medical settings. This doesn't mean we can't talk about our ACEs and their relevance to our own health issues if we feel it's relevant, helpful, and we have providers we trust.

But we can't predict and project health outcomes - positive or negative - on individuals, though we might be informed by the info.

I like your analogy to lung cancer as we also know people who don't smoke can get it and not all who smoke will.  Therefore, universal measures are needed for all patients.

I think there's a good amount of discussion in the book coming out by Dr. Bruce Perry & Oprah Winfrey, entitled, What Happened to You?

There's also a great discussion between @Jane Mulcahy & Dr. Perry below.

Warmly,

Cissy

Post
Copyright Β© 2021, PACEsConnection. All rights reserved.
×
×
×
×
Link copied to your clipboard.
×