Problems with the recommendation to implement ACEs screening

Thanks to the Journal for extending the opportunity for discussion about the topic of adverse childhood experiences (ACEs) screening and to Dr. Watson for her response to our letter (1,2). It is important to highlight that we are NOT contesting the need for health care providers to be aware of, and knowledgeable about, ACEs. Rather, our concern is specific to the recommendation to implement ACEs screening in clinical practice.

The evidence used to inform any screening recommendation needs to be systematically evaluated using well-defined criteria. When current screening criteria, such as those outlined in a recent synthesis published in the Canadian Medical Association Journal (3), are applied to ACEs screening, the majority of criteria are simply not met. Recommendations to systematically screen for any particular exposure, symptom, sign, or illness, require clear operationalization and justification of each step in the proposed process. Identifying an important issue and a related tool is not sufficient. Furthermore, there is no evidence that systematic screening for individual items on ACEs questionnaires, or using an aggregate ACEs score, leads to better health outcomes. Additionally, the potential for more harm than benefit, from any well-intentioned screening, should not be ignored but considered across multiple domains (4).

Although it is possible that having a child or parent complete a checklist and having a health care provider review it could “… prompt and inform a subsequent conversation” (2), what evidence is there that such conversations will routinely lead to accurate identification of need and then systematic linkage to an evidence-based intervention? These are steps that need to be evaluated before recommending the implementation of a screening program. Dr. Watson includes citations providing conceptual support for ACEs screening, but none showing benefits to children from ACEs screening. It is not adequate to justify a proposed screening approach solely based on the idea that it will identify previously unknown information.

It is informative to consider evaluations of other types of screening in primary care that overlap with ACEs (5). History of maternal exposure to intimate partner violence (IPV) is one of the original ACEs and there are services to which women can be referred (and their children) if such exposure is identified. Yet, IPV screening when evaluated with three randomized controlled trials, has not been shown to reduce IPV or improve health benefits (6). It cannot be assumed that identification of an adverse experience through universal screening and referral for services results in health benefits; this is precisely why we need rigorous evaluation in making decisions about any type of screening.

Finally, ‘no screening’ does not mean that questions should not be asked as part of a diagnostic assessment, based on signs and symptoms (i.e., case finding). Specific and relevant ACE questions can be integrated, as needed, into history-taking in a way that is sensitive, safe, and considers the needs of the patient, along with a discussion about the limits of confidentiality before such inquiry occurs. These are already the expectations of good practice.

Potential Conflicts of Interest: HLM reports being involved in guideline development with the World Health Organization on the topics of screening for child maltreatment and intimate partner violence, which have some overlap with this topic. She has also conducted a trial on intimate partner violence screening and written about this topic. JM reports that she was a technical advisor for the World Health Organization, which included doing systematic reviews related to screening for child maltreatment. She is also funded by a postdoctoral fellowship for the VEGA (Violence Evidence Guidance Action) project, which involves conducting systematic reviews related to screening for child maltreatment. There are no other disclosures. All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.