To describe the effects of the coronavirus disease 2019 (COVID-19) pandemic and associated practice shifts on consultation and referral patterns of an intimate partner violence program at a large, urban children’s hospital.
Secondary data analyses examined COVID-19–related variations in patterns of consultations and referrals in the 11 months before the COVID-19 pandemic (April 1, 2019–February 29, 2020) and those after its emergence (April 1, 2020–February 28, 2021). χ2 analyses were used to examine differences in categorical outcomes of interest by time and practice setting, as well as differences within practice settings. Poisson regressions were used to compare the number of reasons for consultation and the number of referrals during the 2 periods.
Analyses revealed significant decreases in face-to-face consults (28% to 2%; P < .001) during the period after COVID-19 emergence alongside significant increases in the total number of consults (240 to 295; P < .001), primarily for emotional abuse (195 to 264; P = .007). Psychoeducation referrals also increased significantly (199 to 273; P < .001), whereas referrals to community resources decreased significantly (111 to 95; P < .001). Setting-specific analyses revealed that primary care settings were the only practice settings to demonstrate significant differences in overall number of and specific reasons for consultation and associated referral types before and after COVID-19 emergence.
Even during a shift away from face-to-face care, there was an increase in intimate partner violence referrals after the start of the COVID-19 pandemic. These findings suggest the importance of pediatric primary care as a location for survivors to access support.
Although coronavirus disease 2019 has generated widespread concern about elevated intimate partner violence (IPV) risk, little is known about how the pandemic has affected specific types of IPV, the capacity of IPV prevention programs to deliver services, and the availability of community resources.
This study describes how the coronavirus disease 2019 pandemic affected consultation and referral patterns for an IPV program at a large, urban children’s hospital. Findings demonstrate the utility of telehealth as a modality for providing assessment, support, and referrals.
The World Health Organization defines intimate partner violence (IPV) as “any behavior within an intimate relationship that causes physical, psychological, or sexual harm to those in the relationship.”1 The coronavirus disease 2019 (COVID-19) pandemic has resulted in a significant disruption in the day-to-day lives of many across the nation. As different states declared stay-at-home orders and considered it a national state of emergency, the state of Massachusetts acted accordingly. Although for some, this may have been a welcome change, for many, the implications were far-reaching. Many IPV survivors were forced to stay at home with their abusive partners, isolated from colleagues, friends, family, and even neighbors.2–4 A recent meta-analysis of studies examining rates of IPV before and after implementation of lockdown measures suggests an average increase of 8.9% of domestic violence postimplementation in the United States, with some estimates as high as 38%.5 Although necessary to preserve the health of the public, these data suggest that COVID-19-related precautions presented an opportunity for IPV perpetrators to capitalize on social isolation, one of the most frequently employed tactics used to separate survivors from their support networks.6,7
For survivors who may be isolated within their home or relationship, the ability to seek support where their children receive medical care is invaluable.8 Moreover, the American Academy of Pediatrics has identified IPV as a pediatrics issue because identification of IPV may be one of the most effective ways to prevent child maltreatment.8 Although not every child will be impacted in the same way by IPV, integration of standard IPV screening protocols that are accompanied by appropriate clinical responses can mitigate the effects of IPV on child health.9–13 In particular, the use of standard IPV screening protocols may help to normalize to both patients and providers that IPV could happen to anyone in any situation.14–17
Within Boston Children’s Hospital, a large, urban pediatric hospital located in the northeastern United States, the Advocacy for Women and Kids in Emergencies (AWAKE) Program works to both provide educational training and support to providers to screen for IPV and offer direct, confidential support to survivors of IPV who are patients, family members, or employees of the hospital. Specifically, the AWAKE Program provides a suite of confidential and voluntary services to IPV survivors, including supportive counseling, referrals to community organizations, assistance with housing application processes, and legal referrals (Fig 1). The goal of this study was to examine the effects of the COVID-19 pandemic and associated practice shifts on consultation and referral patterns within the AWAKE program.
Drawing upon data collected as part of standard AWAKE program procedures, we used secondary data analyses to examine COVID-19–related variations in patterns of consultations and referrals among two cohorts of AWAKE patients: those occurring before the COVID-19 pandemic (April 1, 2019– February 29, 2020) and those occurring after emergence of the COVID-19 pandemic (April 1, 2020–February 28, 2021). The current study included all data collected between April 1, 2019, and February 28, 2021, by the AWAKE program social workers. Anonymized data were extracted from a program-specific database external to the hospital’s electronic health record system that is used to monitor population-level trends in AWAKE consultations and quality improvement purposes. As a component of standard protocol, all AWAKE clinicians are required to enter information about each consultation received by the program, including patient demographic information; consultation date, source, and modality; and specific types of consultation and referral classifications. This information is gathered during both consults with providers and clinical interventions with survivors and is entered into a secure Health Insurance Portability and Accountability Act-compliant database (see below) after the initial intervention or consult is complete. This information may be updated if there is ongoing involvement of the AWAKE social worker. Relevant variables are described below.
Abstracted demographic information included patient race, gender, age, and primary language. Consultation date was recorded as month, day, and year of consultation. Source of consultation spans 19 unique care settings at the host hospital, an option for self-referral, and an option to designate “other” for consultation source. Modality was composed of two options: face-to-face or telephone. Type of consultation was composed of 6 presenting-problem classifications: emotional abuse, physical abuse, sexual abuse, community violence, housing problem, legal problem, and financial problem. Type of referral was composed of 6 types of referrals provided to patients working with clinicians in the AWAKE program: psychoeducation, housing, emergency shelter, restraining order, legal, and community resources. Apart from demographic characteristics and consultation dates, each variable in the database was recorded as a binary response (1 = yes or 0 = no). Multiple reasons for consultation and referral could be indicated for each entry. Outcomes of interest for the current study included counts of overall consultations and referrals, specific types of consultation and referral categories, and modality of consultation.
Consultation-date data were recoded to compute the total number of consultations per month each year, and to indicate whether the consultation occurred in one of two pandemic-related time periods: before COVID-19 or after COVID-19 emergence. Before COVID-19 was defined as the period between April 1, 2019, and February 28, 2020, whereas after COVID-19 emergence was defined as April 1, 2020, to February 28, 2021. These time periods were selected on the basis of the date in which the stay-at-home order was issued in the host clinic’s state (March 11, 2020). Data collected during March in 2020 and 2021 were excluded from analyses because the pandemic may have started to affect behavior before the enactment of the formal statewide stay-at-home order, complicating both month-to-month and year-over-year comparisons. In addition, because of limited sample sizes in some consultation source categories, practice setting data were collapsed into 7 mutually exclusive categories: inpatient, outpatient, specialty, emergency, hospital-based primary care, community-based primary care, and other.
Descriptive analyses are reported using numbers and proportions for categorical variables and means and SDs for continuous variables. Monthly and cumulative counts of the total number of consultations and referrals, specific types of consultations and referrals, and modality of consultation were employed as outcomes to examine differences both by time and by practice setting. χ2 analyses were used to examine differences in categorical outcomes of interest by time and practice setting, as well as differences within practice settings. Poisson regressions were used to compare the number of reasons for consultation and the number of referrals during the two periods. All data were computed in either SPSS v.27 or Stata v.16. Graphs were created in GraphPad Prism 9.0. This protocol was reviewed by the Boston Children’s Hospital institutional review board and declared as exempt (IRB-P00040145).
Across both time periods, most individuals referred to the AWAKE program were adult women between the ages of 25 and 59. Most had one or more children aged <18 living in the home. The prepandemic population was 27% Black, 29% Hispanic/Latino, and 34% White. Data collected on race/ethnicity decreased significantly when care shifted to telehealth; in the period after the emergence of COVID-19, the percentage of consultations without recorded racial or ethnicity data quadrupled from 6% to 24% (P = <.001). Regardless of time period, most people spoke English as their primary language, but ∼10% of referrals spoke other languages, primarily Spanish (Table 1).
Overall Consultation and Referral Patterns Over Time
In the 11 months leading up to the COVID-19 pandemic, the AWAKE team received 240 consults, 197 of which were new. In the matched 11-month period after the emergence of COVID-19, there was a significant increase in consult volume to 295 consults (P < .001), 235 of which were new. Less than 2% of postemergence consults were face to face, compared with 28% before the pandemic (P < .001). The increase in consults was driven primarily by a significant increase in the number of consults for emotional abuse (195 [81%] to 264 [89%], P = .007). There were no significant changes in the number of consults for other reasons or in the number of reasons recorded for each consult.
In terms of referrals, psychoeducation referrals increased significantly during the pandemic period (199 [83%] to 273 [92%], P < .001), and there was a decrease in referrals to community resources (111 [46%] to 95 [32%], P < .001); however, there was no significant difference across time periods in the number of referrals provided to any individual (Table 2). Although no significant differences emerged in month-to-month comparisons for specific consultation sources or referral types, the psychoeducation referral types largely corresponded to the increase in consultations for emotional abuse (Fig 2).
Within-Setting Comparisons of Reasons for Consultation and Referral Type Over Time
The community-based and hospital-connected primary care settings were the only practice settings to demonstrate significant differences in overall number of reasons for consultation, specific reasons for consultation, and referral types across the two pandemic time periods. The number of reasons for consultation that were identified for patients sourced by community-based primary care increased significantly after the emergence of COVID-19, from 14 (2.36) to 36 (3.08) (P = .03). A significantly higher number of emotional abuse consultations sourced by hospital-based primary care was also observed in the period after the emergence of COVID-19, relative to those generated prepandemic (65 vs 42, P < .001). During the time after the pandemic emergence, significantly fewer housing consultations were sourced from hospital-based primary care compared with those sourced prepandemic (20 vs 28, P = .01), and as would be expected, this group of patients also received significantly fewer housing referrals (7 vs 16, P = .007).
Our analyses indicated a shift in IPV consult and referral patterns after the emergence of the COVID-19 pandemic. Overall, there was an increase in consults received by the AWAKE program after the emergence of COVID-19, primarily driven by an increase in consultations for emotional abuse from primary care settings. The pandemic changed both patient care and the social environment over time, which is evidenced in the location and quantity of both consultations and referrals. For example, within the first few months of the pandemic, elective procedures were canceled. Many families were fearful to come to the hospital because of the virus and clinics were expected to downsize the number of patients seen in-person. The decline in clinic capacity, as well as elective surgeries scheduled, reduced the number of patients and families interacting with the health system. The lack of change in volume of consults generated by specialty, inpatient, emergency, and other programs over time may represent missed cases rather than a true consistency across the study period.
The stay-at-home order had pervasive effects on individuals’ well-being and all aspects of life. Our data suggest that, during the pandemic, individuals continued to be seen in primary care and seek support. AWAKE consultations generated by both hospital- and community-based primary care programs increased during the period after emergence of COVID-19. We hypothesize that this increase may be attributable to the fact that, although many survivors were at home with partners who use abusive behaviors, obligatory pediatric primary care visits may have been a rare opportunity for them to leave their residence and seek support. In contrast, the decline in housing referrals observed in primary care settings may reflect the eviction moratorium that was put into place on April 21, 2020.18
The AWAKE program is managed by two clinical social workers (1.5 full-time equivalents). Any provider throughout the hospital system can make a referral to AWAKE. Although before the emergence of COVID-19, services were typically provided both via in-person meetings with survivors in outpatient, inpatient and emergency department settings, as well as telephonically, programming shifted after the emergence of COVID-19 to entirely remote care. All interactions shifted to telephone. AWAKE social workers do not e-mail, text, or use video calls with survivors to reduce the risk of compromising their privacy.
As expected, our analyses reveal a shift away from face-to-face care in the period after the emergence of COVID-19, which initially generated concern about a potential decrease in referrals to the AWAKE program. Within several weeks of the shift to primarily remote care delivered via telehealth, it was evident that this was not the case. Providers continued to refer to AWAKE, survivors consistently engaged in support, and AWAKE effectively assessed, safety planned, and managed risk remotely. Significant increases in psychoeducation and decrease in referrals to community resources provided by AWAKE clinicians during the period after the emergence of COVID-19 may be attributable to the limited capacity of community organizations during this time because of closures and reduced hours of operation.
Implications for Practice
Our data support the importance of a domestic violence program in pediatric hospitals and suggest that such support should be available as a standard part of care. They further suggest that support and assessment may be effective regardless of whether that care is performed face to face or using telehealth. The AWAKE program can potentially serve as a model for other pediatric institutions hoping to provide awareness and support for survivors of IPV who are patients, families, or staff. One of the most critical components of the program is the focus on not just providing clinical care to IPV survivors, but training and empowering providers. Specifically, AWAKE clinicians offer consultation and provide ongoing trainings to social workers and other clinical staff across the hospital on topics such as recognizing and responding to IPV, screening for IPV, utilizing AWAKE and connecting with community referrals, documenting in the electronic health record, and more. During the time frame of this study, AWAKE clinicians provided trainings to multidisciplinary teams in primary care, community-based primary care, and specialty services such as general surgery, urology, and adolescent medicine. When the health care system shifted to telehealth virtual visits, AWAKE was sought out to provide additional training to social workers for this specific type of encounter. Although virtual visits may continue to be standard practice, consideration for ability to screen and provide education about IPV safely to patients and families must continue to be considered. It is possible that trainings provided to clinicians in primary care settings may have affected the number of referrals during the latter half of the study period.
Pediatric care represents an important opportunity to reach those who have experienced IPV. When caregivers of pediatric patients access health care for their children, it represents an opportunity to engage with people who might not otherwise interact with the medical system. It is important to use this opportunity to provide awareness, education, and screening of IPV. We hope that other pediatric hospitals will recognize the impact of IPV on their patients and prioritize the opportunity to address and support the needs of patients and families when they present for health care.
The current study has several limitations. Data were specific to an urban hospital in Massachusetts. Because of the specificity of program services and small sample size, results may not be generalizable to all IPV programs in other hospital-based settings. Because of staffing limitations, it is often difficult for providers to update referral information in the database, which may result in inconsistencies between provided services and recorded data. Using aggregate monthly data meant that our analysis did not account for the timing of specific pandemic-related events. Although we attempted to account for the stay-at home order by removing the month of March from each period, a more detailed analysis of events such as stay-at-home orders and hospital policy changes around elective care may reveal more specific antecedents for some of the observed changes. Finally, an additional limitation that became evident over the course of this analysis was inconsistency in the collection of demographic information. Historically, it has been difficult to capture race/ethnicity data of survivors, but this was further complicated when services were provided remotely. To address this issue, the hospital is currently implementing a new protocol for collecting race and ethnicity data more consistently, but it had not yet been implemented at the time of the study.
After the emergence of COVID-19, we saw an increase in IPV referrals across our systems, particularly for emotional abuse. Our data suggest that the presence of an IPV program within the pediatric setting has the potential to identify numerous individuals in need of support and provide referrals to ongoing care. The fact that providers were able to identify and connect these patients/families even in the setting of a move to remote care should encourage clinicians that it is possible to provide these services across a range of settings and with a variety of resources.
Ms Rahman and Ms Huysman are the clinical social workers in the AWAKE Program and were responsible for drafting the initial manuscript; Drs Boskey and Ross developed the analysis plan and conducted the initial data analyses; and all authors reviewed and revised the manuscript; approved the final manuscript as submitted, and agree to be accountable for all aspects of the work.
FUNDING: No external funding.
CONFLICT OF INTEREST DISCLAIMER: The authors have indicated they have no conflicts of interest relevant to this article to disclose.
COMPANION PAPER: A companion to this article can be found online at www.pediatrics.org/cgi/doi/10.1542/peds.2022-056980.