In the scientific literature, natural disasters, periods of economic decline, and emergencies have been associated with an increase of the risk of CAN. In 2021, a meaningful review conducted by Seddighi et al. pointed out that “violence increases after many emergencies compared to the conditions prior to emergencies and disasters. Polyvictimization or exposure to multiple types of abuse like physical violence, neglect, and mistreatment seems more common in disasters” [10].
Until now, there are no clear indications about the possible effects determined by the SARS-CoV-2 pandemic on CAN. It is not known if stay at home policies and/or the closure of the most part of extra-familiar activities could have influenced the CAN phenomenon [6–9, 11].
In this study, four variables differed between pandemic and pre-pandemic samples: country of origin; origin of the referral; imaging; outcome (Table 1). The first variable was characterized by a significant increase of African and South American children in the pandemic period (GA). The second variable showed the increase of referral from hospitals. The third variable pointed out that a major number of imaging tests were executed during the pandemic. Finally, the number of hospitalizations after the ambulatory evaluation increased (fourth variable). The present study did not allow us to identify the reasons for different distributions of these variables due to its retrospective and observational nature. However, for the second variable the increase of referrals from hospitals could have been determined by the stay-at-home policies and/or the closure of the most part of extra-familiar activities. Due to pandemic restrictions, hospitals were the few accessible extra-familiar places in which CAN cases could have been suspected and, thus, submitted to the specialized ambulatory. Similar considerations can be suggested for the increase in imaging tests’ execution and hospitalization during the SARS-CoV-2 pandemic. The latter could have been influenced by the fact that the most part of children had been referred to the specialized ambulatory by hospitals. Thus, they were already characterized by clinical conditions with mild/high levels of complexity. This could have determined the need of a major number of imaging tests’ execution and of hospitalizations after the specialized ambulatory evaluation.
It is important to note that these hypotheses can be neither confirmed nor excluded throughout the comparison with the scientific literature due to the few available articles about this topic and the differences in study sampling.
Indeed, talking about the execution of imaging tests, in 2022 Henry et al. proposed a meaningful study in which they analyzed “child abuse imaging and findings in the time of COVID-19” [12]. They highlighted “a > 20% decrease in skeletal survey performance early in the pandemic” [12]. This result could appear in contrast with our study. Nevertheless, it is important to note that the above-mentioned authors evaluated a sample characterized by children younger than “2 years undergoing skeletal surveys because of concern for physical abuse at a tertiary children's hospital” [12]. Thus, the results of the two manuscripts are hardly comparable.
One of the most important results of this study is that the distributions of the most part of frequencies between pandemic (GA) and pre-pandemic (GB) samples did not differ. Some manuscripts reported that the occurrence of some forms of CAN - especially physical maltreatment - were more common during the SARS-CoV-2 pandemic. In 2021, Sharma et. al stated that their findings “add weight to existing concerns regarding increased rates of child maltreatment under mandatory stay-at-home orders” [13]. The present manuscript pointed out that distributions of CAN diagnoses were not different between the pandemic (GA) and the pre-pandemic (GB) periods. We identified neither a major number of physical maltreatment cases in GA nor an increase or a decrease of other forms of CAN. However, it is important to highlight that Sharma’s study was focused on sentinel injuries evaluated on < 6 months of age children. Thus, also in this case the two studies are hardly comparable [13].
The present study did not identify an increase of intra-familiar perpetrators despite the stay-at-home policies could have influenced this outcome. Moreover, in the pandemic period (GA) it was identified neither a major number of cases characterized by the presence of at least one sentinel injury, nor an increase of reporting to the prosecutor. These findings seem to be in contrast with the ones of Massiot et al. who reported: “fewer violence cases were perpetrated outside the family as compared with 2018 and 2019”; “the increase in incidence of severe abuse cases during the lockdown and the next 3 months” [6]. However, despite the similarity of two manuscripts’ samples, in our study we included patients with suspected physical maltreatment above 15 years of age while Massiot et al. excluded them [6]. Moreover, we evaluated a longer period of the pandemic. This is a significant limitation because it does not allow an accurate comparison between the two manuscripts.
The most important result of the present analysis is that in the pandemic period the number of observed cases was significantly lesser than the ones evaluated in the pre-pandemic one: 116 and 229 cases respectively (Fig. 1). This finding is confirmed by many manuscripts which talk about the correlation of CAN and SARS-CoV-2 pandemic, demonstrating that the volume of observed CAN cases decreased during stay-at-home policies and lockdowns [6–9, 11–13]. The present study allows us to confirm this finding because the major decrease in referred cases was detected during March - June 2020 (Fig. 1). These months correspond to the period in which the Italian Government ordered a first lockdown in which school, recreational, and not indispensable activities were forbidden [14]. Substantially, the Government allowed only indispensable working, economic, and healthcare activities. Citizens without a role in them were forced to stay at home [14]. Thus, children could only be at home or, if necessary, they could undergo healthcare evaluations at ambulatories and/or hospitals. Then, due to the reduction of SARS-CoV-2 cases and deaths, the Government imposed a progressive reduction of the above-mentioned policies from the first days of June 2020. Similar considerations can be proposed for the period October, November, December 2020 - January 2021 in which the Italian Government ordered new restrictions [14].
Many authors interpreted the reduction of case referral not as the result of a decrease in CAN phenomenon but as child protection system’ less capability to intercept and/or manage CAN cases during the pandemic. For example, Nguyen pointed out that “the COVID-19 pandemic has led to a precipitous drop in CAN investigations where almost 200,000 children are estimated to have been missed for prevention services and CAN in a 10-month period” [15]. Garstang et al. reported that “there are approximately 1500 (95% CI 538 to 2192) potentially abused or neglected children in England who remain hidden from services” [16]. Katz et al. stated that “the initial data presented and discussed among the international teams pointed to the way COVID-19 has hampered CPS responses and the protection of children more generally in most jurisdictions” [17]. Our analysis cannot ascertain the specific reasons for CAN cases’ decrease. However, we cannot exclude that also in our geographical area some cases of CAN remained unseen due to the negative impact of pandemic on our child protection system. This seems to be suggested by the fact that the major decrease was precisely registered during the two periods in which stay at home policies were stricter.
We think that, confirming the data available in the scientific literature about the reduction of CAN referrals during pandemic, the present manuscript highlights the need for collecting useful and updated indications on a phenomenon which is still far from clear explanation to identify corrective measures and eventually adjust child protection systems’ activities. This could be useful not only to mitigate the possible negative effects caused by SARS-CoV-2 pandemic on children but also to program corrective interventions for future pandemics and/or disasters. For example, due to the difficulty to evaluate if child protection systems can be negatively influenced by a pandemic, researchers should identify new tools capable of objectifying a potential impairment of the afore-mentioned systems during pandemics and/or disaster with the aim to timely implement corrective measures.
The limitations of the present manuscript are related to its monocentric, retrospective, and observational nature. Thus, our analysis cannot be used to clearly identify and explain the reasons for similar or different distributions of the frequencies between pre-pandemic and pandemic samples. In addition, our results cannot be generalized referring to other populations and other geographical areas. Finally, we cannot exclude that in the pandemic period the distributions of frequencies observed between pre-pandemic and pandemic samples had been strongly affected by the fact that some CAN cases could have not reached and properly activated our child protection system.