Research by Topic

Emergency care for children is provided predominantly in community emergency departments, where abusive injuries frequently go unrecognized. There are substantial differences between pediatric and community EDs in the access to subspecialty expertise, quality of pediatric care, and outcomes of pediatric patients. Following qualitative interviews with providers, quality improvement studies, and implementation of a clinical guideline, our targeted interventions have increased compliance with national guidelines for abuse evaluations and mitigated differences between pediatric and community EDs in the evaluation of high-risk injuries.

Our team has conducted various interventions to improve the care of infants presenting with abuse-associated injuries using a clinical guideline. Our child abuse guideline links ED providers with child abuse experts and has improved provider decision-making, decreased bias, and standardized care between pediatric and community EDs. We have developed natural language processing technology to identify high-risk injuries in the electronic medical record and are in the process of implementing electronic clinical decision support to sustain and augment the positive impact of the guideline across multiple EDs in the Yale-New Haven Health System.

Our Community Advisory Board consists of members from local IPV organizations, the Department of Children and Families, the Yale child abuse team, Yale emergency medicine department, New Haven police department, Yale Child Study Center, and community representatives. Board members ensure a trauma-focused environment for abused caregivers and children exposed to IPV. One project that arose from this collaboration was the idea to link a child’s medical evaluation with an opportunity to connect the child’s caregiver to a community-based IPV advocate. The board’s next project is to implement the evidence-based CUES (Confidentiality, Universal Education & Empowerment, Support) intervention in the Yale-New Haven Health adult and pediatric EDs.

As part of the New Haven Family Violence Community Advisory Board, the FRESH Collaborative recently collaborated on the development of an integrated care program at the South-Central Child Advocacy Center for families living with IPV. Since July 2020, the program has provided medical evaluations for young children exposed to IPV and real-time linkage with an IPV advocate for caregivers. Addressing the child’s safety while optimizing the safety and advocacy of a caregiver may ultimately decrease the child’s exposure to physical harm and reduce the occurrence of IPV. We are examining the acceptability and feasibility of this model and planning a clinical trial to compare the effectiveness of the family-centered care model with usual care.

Children are witnesses to 50% of IPV episodes in the US and are often evaluated after exposure to IPV. Abusers may access the electronic health records (EHRs) of victims and their children and learn of the IPV disclosure, leading to potential threats to safety. Implementation of the 21st Century Cures Act, a federal law mandating that healthcare providers give patients immediate access to electronic clinical notes, has amplified concerns of worsening safety for victims of IPV and their children. We recently completed a study demonstrating that discussions about EHR safety are infrequent when adults and children present after IPV and notes are infrequently unshared even when safety concerns are present. We are evaluating and developing interventions to improve documentation practices for IPV-related care in emergency departments.

We aim to understand the impact of racism on Asian Americans in healthcare. Currently, we seek to understand the impact of anti-Asian racism on Asian American medical students.

We aim to decrease restraint use, both physical and chemical, in the behavioral health population in the pediatric emergency department. More specifically, we have found that people of color are disproportionately more restrained compared to their white counterparts, so our team is aiming to reduce the disparities of restraint use as well as bring awareness to this issue.

Pediatric emergency department (PED) visits are opportunities to identify children living with or exposed to caregiver intimate partner violence and initiate timely intervention, yet one-third of related PED visits lack associated ICD-10 codes. Through a pilot grant award from the Yale Department of Pediatrics, Salahudeen Mirza is testing the accuracy of large language models in the identification of youth exposed to intimate partner violence in their families. If accurate, LLMs have the potential to identify youth IPV exposure at the population level. Future directions include external validation and testing the ability of LLMs to generate clinical recommendations for families living with IPV.

Children in foster care have disproportionately high rates of chronic health conditions, developmental delay, and mental health needs. When children enter DCF care in Connecticut, they receive a comprehensive health evaluation to identify these health needs. Arielle Shibi Rosen's project examines whether these needs are subsequently addressed during routine primary care visits, and whether pediatric providers acknowledge a child's foster care status. Findings will inform future interventions to improve care delivery and reduce health disparities for children in foster care.

Injuries due to child abuse are frequently missed in emergency departments (EDs), leading to escalating abuse and even death. Digital child abuse clinical decision support systems (CA-CDSS) have the potential to address this by providing evidence-based guidance to ED providers, particularly when using accurate identification methods such as natural language processing (NLP). Dhatri Abeyaratne’s project evaluates the implementation of a CA-CDSS that uses a previously validated NLP algorithm across 8 general and 1 pediatric ED. The study will evaluate uptake and usability of the CA-CDSS, as well as its impact on guideline compliance and bias in abuse evaluation and reporting.

Intimate partner violence (IPV) affects an estimated 1 in 3 US women, disproportionately impacting those experiencing poverty. Studies from low- and middle-income countries suggest that cash transfers (CTs) may reduce rates of IPV. However, evidence from the US is very limited. Dhatri Abeyaratne's project explored the experiences of IPV survivors who received unconditional CTs ($500-1000) through the Yale Hospital Violence Intervention Program (HVIP). Findings suggest that CTs may help survivors address urgent needs, enhance safety for themselves and their children, and foster empowerment and engagement with advocacy services. These insights can guide the selection of patient-informed outcomes for future clinical trials and help inform the design of survivor-centered CT models within healthcare settings.

Exposure to intimate partner violence (IPV) between caregivers is an important risk factor for the physical abuse of children, which is reported to occur in 30–60% of homes with IPV. However, most IPV-exposed children are never evaluated for abuse, marking a critical gap in interventions that could potentially reduce future family violence. The Trauma- and Violence-Informed Care (TVIC) model offers an effective and integrated treatment approach for the child-caregiver dyad, providing advocacy support to the caregiver during a child's medical evaluation, as supported by the lab’s prior research. Currently, two interns, Jayde Ma and Kate Guittari, are developing a manual that details the TVIC model, including its rationale and its operations, to inform dissemination and generalizability.