Perspectives on Adverse Childhood Experiences

More than half a century ago, I was a graduate student in a research methods course. To illustrate “face validity” in predictive assessment, the instructor used this example (which may well have been apocryphal): In WW II, researchers were tasked with identifying soldiers who could be assigned effectively to work in military bases in the Arctic. After considering several alternatives, they settled on one question: “Do you like to work outside in the cold?” Problem solved—at least for the moment. When word got out that a “yes” answer could get you posted to Alaska in the winter, reluctant individuals began to “fake” their response (saying “no” even when they should have answered “yes”).

For the past 30 years I have served as a psychological expert witness in murder cases. I think of the “working in the cold” question, especially when I am in a jail or prison administering the Adverse Childhood Experiences Scale to an inmate seeking resentencing or release (or maybe even avoiding the death penalty in favor of life in prison). I do so for two reasons. First, they have transparent face validity and predictive power. They predict a variety of important outcomes beyond conventional health concerns such as high blood pressure (accounting for ∼50% of the variation in substance abuse, depression, and suicidal thoughts and behaviors, and 30% of violent behaviors against others).

Unlike the Minnesota Multiphasic Personality Inventory (MMPI), which has 567 questions, many of which are opaque, with 10 questions, the Adverse Childhood Experiences Scale (ACE) is simple. Most importantly, and in a sense like some of the MMPI's predictive questions, are “behavioral” rather than “evaluative.” Although the goal in the first half of the ACE questions is to elicit reports of child maltreatment, the words “abuse” and “neglect “do not appear. Despite the “behavioral” rather than “judgmental” nature of the questions, they seek to find out these important adverse experiences: Were you degraded and verbally threatened (Question #1)? Were you beaten (Question #2)? Were you sexually molested (Question #3)? Did you feel emotionally abandoned (Question #4)? Did you feel you suffered basic material and social deprivation (Question #5)?

The questions then turn to “facts” about elements of your family: Question #6 asks whether your were parents together when you were growing up? It makes no mention of parental abandonment. Question #7 does not use the word “domestic violence” as it asks whether you witnessed your mom being assaulted? The final three questions simply ask about the status of household members: Were the people you lived with drunk or high (Question #8)? Were the people you lived with were crazy or suicidal (Question #9)? Did someone you lived with go to prison (Question #10)? The inclusion of each question is supported by a body of research that validates its status as a “risk factor.” It is an elegant tool in that sense.

In addition to predicting his or her likely developmental pathway, the ACE “score” serves to “map” the respondent's place in America. Roughly speaking, two-thirds of American kids grow up in a social environment with an ACE score of 0 or 1. Only 1 in 100 grows up with a score of 7. These are seriously discrepant “social addresses” (to use a term that my mentor Urie Bronfenbrenner promoted in the latter stage of his career). When I administer the ACE to my “clients” in the criminal justice system ordinarily working for the defense, the average score is 7.

I started using the ACE with these guys (they are mostly males, of course) to counter the state's refrain, “Lots of kids have difficult childhoods, but they don't kill anyone.” Having the ACE score allows me to say, “We are not talking about some generic ‘difficult childhood,’ we are talking about more adversity than 99 out of 100 kids growing up.” The average mean score being 7, many guys have scores of 8, 9, or 10 (putting them in the worst 0.01% of the population). And, very few have scores less than 5; almost none has the score typical of American kids (66% of whom have a score of 0 or 1).

I usually ask these guys this question: “If I gave these ten questions to 100 guys you knew growing up, how many do you think would have a score of seven or more?” The answer is usually something like “all of them” or “80 out of 100.” I then ask the same question about 100 kids drawn at random from the U.S. population—often going to great lengths to explain what “randomly selected” means. The answers vary quite a bit, but most indicate they think it would be more than half, and are often stunned to hear that it is 1 in 100. In a sense, no one lives in “the United States”; everyone lives in a specific social environment, and some of these environments are socially toxic.

These guys think an extraordinary level of adversity is the norm rather than the exception it is. And, they tend to think that a high level of adversity is normal. Consider this response by a 30-year-old man sentenced to a life for a murder he committed at age 17 years to the question about witnessing domestic violence (Question #7): “No because it didn't happen often. I only know of one time my grandma hit my mom in the eye and one time my mom's boyfriend tried to cut her throat with a knife.”

A study done several decades ago created scenarios that a mainstream adult would recognize as child maltreatment. These were posed to abused kids, who were asked, “Is this abuse?” “Whose responsibility is this?” and “Will this have any long-standing effects on the child?” Before undergoing a treatment program—a kind of consciousness raising project—the abused kids typically responded that it was not child abuse, that it was the child's fault, and that it would not have any long-term consequences. After the treatment experience, they were tested again, and now saw the parent's behavior as abusive, that it was the parent's responsibility, and that this kind of treatment of a child would have long-term consequences.

In recent years, I have worked on ∼150 cases in which guys who committed murders as adolescents and who were sentenced to life in prison are eligible for resentencing. These resentencing cases often provide a kind of “experiment of nature” to observe this consciousness raising at work. Youth who told a probation officer or other professional during a Presentence Investigation that their family was ok “change their story,” when as mature men they are asked about these same issues as adults. They now are able to see what is in front of their newly opened eyes.

The ACE questions are brilliantly elegant and transparent. However, like the “like to work outside in the cold” Arctic question, they can be too transparent for the “in the know” respondent who has a stake in portraying his (or perhaps her) life in a false light to gain some advantage in the courts. To combat this risk, I routinely warn respondents against lying—reminding them that the state will likely detect their lies. I also ask them whether there are people who can corroborate their reports of childhood adversity (and sometimes have the protective service records and social history documents to do so). I also ask them to write an explanation, with examples, for each “yes” answer they provide when responding to the ACE questions.

This brings us to this special issue. Each of the three articles contributes something important to the process of refining the developmental dynamics of adversity, refining the “gross” insight that led Dr. Felitti to put together the Adverse Childhood Experiences Scale in the first place. The study by Meghan Clifford, Amanda Nguyen, and Catherine Bradshaw (“Patterns of Adverse Childhood Experiences and Associations with Externalizing Problems: A Latent Class Analysis”) reports that “…we identified three distinct ACE classes: 1) Low ACEs (77%), 2) Moderate Multiple ACEs (11%), and 3) Hostile Maltreatment (12%). Compared to the Low ACEs Class, the other two classes were associated higher levels of externalizing problems overall, with the Moderate Multiple ACEs Class exhibiting the greatest risk of endorsing externalizing outcomes during adolescence and adulthood.” This kind of fine tuning is important.

The study by Angela Moe, Catherine Kothari, and Homer Stryker illustrates the need for behavioral transparency that is generally met by the ACE questions. They report that of the screened 301 participants in a community study, 64 “screened positive” for “intimate partner violence (IPV).” Of the 40 of those mothers who agreed to be interviewed, 25 “described perinatal IPV.” Assuming that the 40 who were contacted to interview are representative of the entire group of 64, this suggests that some 37% of the women were under-reporting their experience of domestic violence.

The authors conclude that “While they reported various forms of IPV and recognized that their children had been exposed to it, the women did not often exhibit a clear understanding of how their own victimization impacted their children, prenatally or upon birth. More recognition came in retrospect for those with older children.” As noted earlier, the more we learn about the differences between observing and evaluating adversity the better.

The study by Sylvie Levesque, Dominic Julien, Katrina Joubert, Mariet-Eve clement, Genevieve Lessard and Jasline Flores (“Exposure to intimate partner violence in children aged 6 months to 8 years: Factors associated with mothers' awareness of children's exposure to this violence”) helps to illuminate this issue of observation versus evaluation. They report that “Findings revealed that 11.1% of young children were exposed to at least one form of IPV. However, that proportion decreases to 5.9% when the mother is asked whether the child is aware of or a witness of the violence.” The observation/evaluation gap is real and significant.