Knowledge and Intentions of French Adults Regarding Abusive Head Trauma in Infants: An Online Survey

Overview

Abusive head trauma (AHT) is a leading cause of infant mortality and long-term neurodevelopmental disability, yet existing prevention programmes have shown limited effectiveness in reducing its incidence. This may reflect insufficient understanding of upstream determinants of behaviour, including population-level knowledge of AHT. This cross-sectional online study aims to assess AHT-related knowledge, behavioural intentions, and their determinants in a sample representative of the French adult population aged 18-80 years. Participants will be recruited through a quota-based online panel and social media. Participants will complete a standardised procedure including exposure to short videos depicting either a shaking gesture (dangerous) or non-dangerous infant-handling behaviors (playing, rocking), followed by questionnaires assessing interpretation of the gesture, knowledge of its consequences, and AHT-related knowledge. After standardised information on AHT is provided, behavioral intentions and theoretical knowledge will be reassessed. This study will provide the first nationally representative data on AHT-related knowledge in France and inform the development of targeted, evidence-based prevention strategies.

Abusive Head Trauma (AHT) in infants is the most severe form of physical child abuse. It is defined as "an injury to the skull or intracranial contents of an infant or young child (under 5 years of age) resulting from inflicted blunt impact and/or violent shaking" (Parks, CDC). AHT is a leading cause of infant mortality, with mortality rates ranging from 7% to 26% (Niederkrotenthaler, Child Abuse Negl 2013; Hobbs, Arch Dis Child 2005; Ferguson, Crit Care Med 2017), and a major cause of long-term neurodevelopmental disability among survivors (Jayawant, Arch Dis Child 2007; Ferguson, Crit Care Med 2017). AHT predominantly affects infants younger than 2 years, with peak incidence between 2 and 4 months of age (Lee, J Dev Behav Pediatr 2007; Emrick, Child Abuse Negl 2019). Annual incidence has been estimated at 14 to 34 cases per 100,000 infants (Paget, Child Abuse Rev 2020; Rebbe, Child Abuse Rev 2020). Adult males are the most frequently identified perpetrators (Edwards, Child Abuse Rev 2020). Over the past three decades, prevention programs have primarily focused on educating parents and caregivers about infant crying, the dangers of shaking, and strategies for stress regulation (Chang, Trauma Violence Abuse 2024; Scott, Child Abuse Negl 2022; Prokofevia, Child Abuse Negl 2026). However, several meta-analyses have reported no significant reduction in AHT incidence following such interventions (Viswanathan, JAMA 2024). These findings underscore the need to better understand upstream determinants of behavior, including knowledge, which is a key construct in many behavior change theories (Ajzen, 2005). International studies (eg, Germany (Berthold, Pediatr Res 2019), Saudi Arabia (AlOmran, Pan Afr Med J 2022), China (Gao, Child Abuse Negl 2021), Ireland (Mann, Eur J Pediatr 2015)) report that 41% to 69% of respondents had never heard of AHT. However, these studies have several limitations. Most are single-center, involve small sample sizes, and exhibit heterogeneous methodological quality. Recruitment has typically been restricted to parents, excluding other adults who regularly care for or interact with infants. In addition, most studies assess knowledge of AHT and its consequences without providing a clear definition of AHT or explicitly evaluating recognition of the shaking behavior that can lead to AHT. Only one study has used video-based stimuli to assess recognition of shaking; however, it relied on a neutral prompt without confirmation of the nature of the depicted action, thereby limiting the validity of the knowledge assessment. Observed levels of knowledge vary widely across studies, likely reflecting both true population heterogeneity and methodological inconsistency. Data on determinants of AHT-related knowledge remain scarce. In France, in particular, no nationally representative data are available; existing studies are limited to small, single-center samples recruited in maternity settings. This cross-sectional online survey aims to assess AHT-related knowledge, behavioral intentions, and determinants of knowledge in a sample representative of the French adult population. Recruitment will be conducted by IPSOS BVA (organization specialized in survey and panel research) using two channels: (1) a quota-based online panel, representative of the French adult population, and (2) social media. This approach will yield three independent samples of adults aged 18-80 years residing in France. Sample 1 (n = 1,000; general population, representative) will be recruited via the online panel and supplemented with a targeted oversample of 200 men aged 18-40 years, the demographic group most frequently identified as perpetrators of AHT, recruited through social media. Sample 2 (n = 1,000; general population, representative) and Sample 3 (n = 1,000; general population, representative) will both be recruited via the online panel. All participants will complete the same sequential procedure. First, participants will complete an 8-item sociodemographic questionnaire. Each sample will then be exposed to one short video stimulus (6-8 seconds; motion design): Sample 1 will view Video 1 (depicting a shaking gesture, representing a situation that may cause AHT); Sample 2 will view Video 2 (depicting a knee-bouncing gesture, representing a non-dangerous play activity); and Sample 3 will view Video 3 (depicting a rocking gesture, representing a non-dangerous soothing activity). After viewing their assigned video, participants will complete a 5-item questionnaire assessing their perception of the depicted gesture, their knowledge about the potential consequences of the gesture, and their knowledge about AHT. Subsequently, all participants will then view Video 1 (shaking gesture), accompanied by explicit information indicating that the depicted behaviour may cause AHT, along with a standardised operational definition of AHT. Participants will then complete a 4-item questionnaire assessing behavioral intentions and theoretical knowledge related to AHT. AHT-related knowledge will be assessed by combining questionnaire responses into predefined categories: "perfect responder", graded levels of "intermediate responders", "incorrect responder", and "incoherent responder". Classification rules will be established a priori by a panel of experts in AHT. Descriptive analyses will be conducted to characterise the study population and to describe levels of AHT-related knowledge and behavioral intentions. Associations between participant characteristics (potential determinants) and AHT-related knowledge will be examined using appropriate statistical methods. This study will provide the first nationally representative data on AHT-related knowledge among adults in France and will inform the development of targeted, evidence-based AHT prevention strategies.