University of Washington Affiliated Hospitals, United States of America.
Yale School of Medicine, United States of America.
Child Abuse Negl. 2022 Jun;128:105604. doi: 10.1016/j.chiabu.2022.105604. Epub 2022 Mar 24.
Although previous studies have examined differences in the characteristics of abusive versus non-abusive injuries, no study has focused on the differences in the circumstances surrounding these injuries, such as whether the event that caused the injury was witnessed or heard, or EMS was called.
To determine predictors related to the circumstances of the injury (COI) for distinguishing abusive versus non-abusive injuries.
PARTICIPANTS/SETTING: Children younger than 3-years-old who were hospitalized with either a head injury or a fracture and evaluated by the child abuse consultation service between June 1, 2008 and June 30, 2017.
In this case-control study, abusive (cases) and non-abusive (controls) injuries were determined by a consensus of two experts blinded to the COI. Multivariable logistic regression was used to identify COI predictors of abusive injuries.
We identified 302 children: 80 cases (26.5%) and 222 controls (73.5%). Abused children were less likely to have a clear event described (p < .001). Of the 251 with a clear event, we found that the significant variables for abuse were father's presence (adjusted odds ratio [aOR] 8.37; 95% CI 3.35-20.92), delay ≥24 h in seeking care (aOR 6.23; 95% CI 1.95-19.92) and calling EMS (aOR 3.21; 95% CI 1.10-9.36). In contrast, the event being heard (aOR 0.22; 95% CI 0.08-0.0.59) and the child being dropped (aOR 0.09; 95% CI 0.01-0.77) were less likely to be abusive.
We identified five COI predictors that may help clinicians in determining whether a child's injuries are due to abuse.
尽管先前的研究已经检查了虐待性和非虐待性损伤的特征差异，但尚无研究关注这些损伤周围环境的差异，例如导致损伤的事件是否被目击或听到，或是否呼叫了紧急医疗服务（EMS）。
确定与损伤环境（COI）相关的预测因素，以区分虐待性和非虐待性损伤。
参与者/设置：2008 年 6 月 1 日至 2017 年 6 月 30 日期间，因头部损伤或骨折住院且由儿童虐待咨询服务评估的 3 岁以下儿童。
在这项病例对照研究中，通过两位对 COI 不知情的专家的共识来确定虐待性（病例）和非虐待性（对照）损伤。多变量逻辑回归用于确定 COI 对虐待性损伤的预测因素。
我们共确定了 302 名儿童：80 例（26.5%）和 222 例对照（73.5%）。有明确事件描述的患儿更不可能发生虐待（p<0.001）。在 251 名有明确事件的患儿中，我们发现虐待的显著变量是父亲在场（调整后的优势比[aOR]8.37；95%置信区间[CI]3.35-20.92）、寻求治疗的延迟时间≥24 小时（aOR 6.23；95% CI 1.95-19.92）和呼叫 EMS（aOR 3.21；95% CI 1.10-9.36）。相比之下，事件被听到（aOR 0.22；95% CI 0.08-0.59）和患儿被跌落（aOR 0.09；95% CI 0.01-0.77）的可能性较小。
我们确定了五个 COI 预测因素，这可能有助于临床医生确定患儿的损伤是否是由虐待引起的。
