Two intersecting planes (1952) M.C Escher

What can we learn from research that looks at many traumatic experiences at once? 

Childhood adversity rarely comes in neat, separate categories. A child who is physically abused may also experience emotional abuse, neglect, parental mental illness, poverty, family violence, or instability at home. For clinicians, this overlap is familiar: patients’ histories are usually complex, and one experience cannot always be separated cleanly from the wider family and social context.

Researchers face the same problem. A common solution is to create a cumulative adversity score.

In the well-known Adverse Childhood Experiences, or ACE, approach, researchers count how many categories of adversity a person experienced. Someone reporting physical abuse, parental divorce, and household substance misuse receives a score of three. Higher scores are consistently associated with poorer mental and physical health.

This approach is simple and clinically intuitive. It communicates that repeated or multiple adversities matter. It can also create a false sense, however, that combining experiences automatically solves the methodological problems caused by their co-occurrence. The crucial question is why they occur together. I recently wrote a manuscript about it.

Consider physical abuse and sexual abuse. Both may be associated with later depression, and they may occur in the same families. A study focusing on sexual abuse alone could therefore overestimate its impact if some of the observed association is actually due to co-occurring physical abuse.

Combining both into one adversity score may appear to solve that problem. Yet the two experiences may co-occur because they share a common cause. Family instability, harsh parenting, socioeconomic disadvantage, parental mental illness, or unsafe living conditions may increase the risk of several forms of adversity at once.

Those same factors may also influence depression directly. Socioeconomic disadvantage, for example, can affect mental health through chronic stress, reduced access to care, educational opportunities, housing conditions, and many other pathways. Adding several adversities together does not remove these pathways. The resulting score may still be confounded by the family and social conditions that produced the adversities.

For clinicians, this matters because research findings are often translated into statements about what is most harmful, what should be screened for, or what should be targeted in treatment. A strong association between a cumulative adversity score and depression does not necessarily tell us what would happen if one particular experience were prevented or treated.

The problem becomes even more difficult when adversities unfold over time and influence one another. Early abuse may contribute to substance use in adolescence. Substance use may then increase the risk of hospitalisation, school exclusion, homelessness, or further victimisation. If all these events are added to the same cumulative score, the score combines early causes, later consequences, possible mediators, and new exposures.

Suppose childhood abuse contributes to later depression partly through substance use. Adjusting statistically for substance use may remove part of the pathway through which abuse has its effect. But including substance use in the adversity score blends the mediator with the original exposure. Neither approach is automatically correct. The appropriate analysis depends on the question being asked.

A clinician might ask, “What experiences should I assess because they help me understand this patient’s current difficulties?” A policymaker might ask, “Which preventable exposure should receive priority?” A researcher might ask, “Through what mechanism does early abuse affect depression?” These are different questions and may require different ways of representing adversity.

Grouping experiences can also be valuable when there is a clear theoretical rationale. For example, several experiences may share features of threat, deprivation, humiliation, or loss and may affect development through related mechanisms. In that case, grouping them reflects a specific hypothesis about how adversity affects the child.

The broader lesson is that there is no universally correct way to combine childhood adversities. Researchers should begin with the clinical or causal question, map out how the experiences may be related, and then decide whether individual adversities, theoretically defined dimensions, developmental pathways, or a cumulative score provide the clearest answer.

For clinicians reading adversity research, a useful question is: what does this score actually contain, and what conclusion can it legitimately support?

A cumulative score may capture the weight of a difficult childhood. It may also conceal which experiences occurred, why they clustered, how they unfolded over time, and where intervention could make the greatest difference.