Identifying child maltreatment during virtual medical appointments through the COVID-19 pandemic

Abstract

Background

Throughout the COVID-19 pandemic there has been a documented decline in reports to child protective services, despite an increased incidence of child maltreatment. This is concerning for increasing missed cases. This study aims to examine if and how Canadian paediatricians are identifying maltreatment in virtual medical appointments.

Methods

A survey was sent through the Canadian Paediatric Surveillance Program (CPSP) to 2770 practicing general and subspecialty paediatricians. Data was collected November 2021 to January 2022.

Results

With a 34% (928/2770) response rate, 704 surveys were eligible for analysis. At least one case of child maltreatment was reported by 11% (78/700) of respondents following a virtual appointment. The number of cases reported was associated with years in medical practice (P = 0.026) but not with the volume (P = 0.735) or prior experience (P = 0.127) with virtual care, or perceived difficulty in identifying cases virtually (Cramer’s V = 0.096). The most common factors triggering concern were the presence of social stressors, or a clear disclosure. The virtual physical exam was not contributory. Nearly one quarter (24%, 34/143) required a subsequent in-person appointment prior to reporting the case and 32% (207/648) reported concerns that a case had been identified late, or missed, following a virtual appointment. Some commented that clear harm resulted.

Conclusions

Many barriers to detecting child maltreatment were identified by paediatricians who used virtual care. This survey reveals that virtual care may be an important factor in missed cases of child maltreatment and may present challenges to timely identification.

Key Messages.

• Through the COVID-19 pandemic, a rise in child maltreatment cases was well established alongside a decline in reports to child protective agencies, suggesting missed cases.

• This study reveals that virtual care may be an important contributing factor to timely identification of child maltreatment with multiple missed cases reported.

• This work also highlights factors that may help physicians identify child maltreatment in future virtual care appointments.

• This study will help inform practice policy on virtual care visits, and advocacy for development of a tool in virtual care visits.

Since the onset of the COVID-19 pandemic, paediatricians have raised concerns about the increased risk of child maltreatment (1–5), with several studies corroborating these concerns (6–8). For many families, social isolation, financial insecurity, and psychosocial stressors have intensified through the pandemic (9–12), particularly when public health measures have required widespread lockdowns. Access to extended family, schools, social networks, community supports, and healthcare services have been disrupted, creating a particularly heavy burden on parents.

Despite this, many countries, including Canada, have reported a decline in calls to child protective services (13). In Ottawa, reports to the Children’s Aid Society decreased 30% to 40% in 2020, compared to 2019 (14,15). In New York City, precipitous drops in child maltreatment reports coincided with lockdowns, with up to 30% to 50% decreases in calls over such periods of time (16). Combined with the knowledge that child maltreatment typically increases at times of higher stress, this is concerning for a rise in missed cases of child maltreatment (17). A key factor in this under-reporting relates to children’s reduced contact with trusted adults outside the home, such as teachers (18,19), doctors, and social service workers.

Paediatricians and primary care providers also play an important role in identifying suspected child maltreatment (2,20). Children have had decreased access to healthcare providers through the pandemic lockdowns as evidence by a drop in vaccination rates and emergency department visits (21–25). In addition, access to care has adapted to the pandemic with a shift from in-person appointments to virtual medical appointments. This, too, has changed the assessment of child maltreatment, though it is unclear if it has contributed to missed cases or provided potential benefits.

Virtual appointments have included telephone calls, and video-based encounters. For some, it may increase access to care and offers physicians a different type of ‘window’ into the home through video-based appointments. However, virtual medical appointments preclude the ability to perform a full physical examination. In some instances, the paediatrician is not able to observe the child or the interactions with their caregivers. There is also often a lack of privacy and confidentiality at these appointments for families and patients to disclose sensitive issues.

To date, there is a paucity of literature around the impact of virtual health appointments on the detection of child maltreatment by physicians. This study aims to assess whether Canadian paediatricians are identifying cases of maltreatment during virtual medical appointments. We also examine the factors that allow for detection in virtual health appointments, and physician reported barriers.

METHODOLOGY

A one-time survey was sent to 2770 general paediatricians and paediatric subspecialists in Canada, using the Canadian Paediatric Surveillance Program (CPSP). The CPSP is a joint project of the Public Health Agency of Canada and the Canadian Paediatric Society with a mandate to conduct national surveillance on rare childhood disorders or rare complications of more common conditions.

This survey was developed with the support of the CPSP, by a team that included experts in survey design and paediatricians with expertise in child protection. The final survey contained 13 questions, divided into three parts, and took 10 to 15 min to complete. The first portion collected practice demographics (e.g., Academic affiliation, practice location), and virtual care experience. The second portion contained questions about reported cases of child maltreatment following a virtual care visit. Finally, it included questions assessing possible missed cases of child maltreatment initially seen by virtual care. Virtual care appointments included both telephone and video visits.

A survey was determined to be the most appropriate method to answer our question about cases of maltreatment identified during virtual care, because of the exploratory nature of the study, and the ability to reach a large sample of physicians covering a large geographic area.

Results were collected between November 2021 and January 2022.

Data was provided from the CPSP to the research team in an excel spreadsheet, and paper responses were collected and manually entered by one member of the study team.

The responses were analyzed for qualitative themes. Descriptive statistics was performed using SAS Version 9.4 (Cary, NC). P-values were derived using chi-squared or ANOVA F-test where appropriate. A P-values <0.05 was considered significant.

Patient and public involvement

Patients were not involved in this survey.

RESULTS

Respondent demographics and use of virtual care

We received 928 responses from 2770 distributed surveys. Our survey had a 34% (928/2770) response rate, and 76% of surveys (704/928) were eligible to be included in the analysis. Physicians who did not provide virtual care at all (n = 122) were excluded. A virtual care visit included a telehealth or video-based appointment. Those who had never reported a case of child maltreatment (n = 93), neither in-person, nor following a virtual healthcare appointment, were excluded from analysis. Surveys with multiple fields left blank (n = 9) were excluded. After exclusions, 704 surveys were analyzed.

Survey respondents had an average of 17.5 years in independent practice. There were more (69%, 486/700), general or consulting paediatricians than paediatric subspecialists. In terms of practice setting, 53% (343/644) worked in an academic setting at least half the time, and 74% (503/681) and 9% (64/681) spent at least half their time in urban areas and rural/remote areas, respectively.

Most reported child maltreatment following an in-person appointment at some point in their career, and approximately 70% (504/702) had seen at least three or more cases of child maltreatment.

Most, 69% (486/700), had no experience with virtual care prior to the pandemic.

Physician reported barriers to identifying cases

Identifying child maltreatment was rated as slightly or much more difficult during virtual appointments by 69% (481/700) of respondents. There was no significant association between the perceived difficulty of identifying cases of child maltreatment during a virtual appointment and the number of child maltreatment cases reported by the respondent (Cramer’s V = 0.096).

Respondents identified several challenges with identifying child maltreatment through virtual care, including a lack of interaction with the patient, inability to perform a physical examination, and concerns that the patient’s environment lacked privacy. The benefit of being able to observe the family in their own home environment was highlighted by 7% (39/596) of respondents.

Cases identified via virtual care

Cases of child maltreatment where the initial concern was recognized in a virtual appointment were reported by 11% (78/700) of respondents with most of these physicians reporting 1 or 2 cases during the pandemic (Figure 1).

Figure 1.

The number of reports to a child protection agency per physician from virtual care through the pandemic

The number of years in independent practice was associated with reporting more cases of child maltreatment following a virtual care appointment (P = 0.026). Neither experience with virtual care prior to the pandemic (P = 0.127), nor the volume of virtual encounters through the pandemic (P = 0.735), was significant.

The triggers to report a case to a child protection agency by frequency were analyzed. The most common trigger for a report to a child protective agency in a virtual visit was the presence of major social stressors, including concerns such as financial difficulty and the home environment. The next was parental capacity, such as cognitive capacity, mental health, and addictions (Figure 2). A clear disclosure from a child or youth prompted 30% of reports to a child protective service following a virtual care visit, with the most common concern being social stressors. When concerns for physical, sexual, emotional abuse or neglect were raised, these also most commonly resulted from a child’s disclosure.

Figure 2.

All factors triggering concern for child maltreatment for each case reported to a child protective service following a virtual care appointment. Multiple factors within one case were reported for some cases, and a total of 504 factors were reported for analysis. This graph demonstrates the most common factors, including whether the maltreatment was disclosed by a child/youth, parent, or noticed by a physician in the absence of an overt disclosure

The virtual physical exam, while being completed by many physicians, was not contributory to a single report of child maltreatment.

Missed and delayed reports of child maltreatment

Concerns that a case of child maltreatment had been identified late, or missed, from all causes following a virtual care visit were reported by 32% (207/648) of respondents. Harm resulting from a specific missed or delayed diagnosis was reported by 4% (28/704) of physicians. Examples of harm included cases of children with additional injuries, and delays in intervention while children remained in unsafe environments.

Of the cases where concerns were initially identified in a virtual care visit, 24% (34/143) were delayed in a report to a child protective service. The delay was due to the physician requiring a subsequent in-person visit following this virtual visit to confirm their concerns in person.

DISCUSSION

Child maltreatment is a widespread problem in Canada with 33% of adults reporting experiences of this in their childhood before the age of 15 (26). There are mandatory reporting duties in Canada, though variable by province, require physicians to report suspected child maltreatment to a provincial child protective service. Physicians are an important group for identifying and reporting child maltreatment.

A decline in child welfare reports, with a concurrent rise in maltreatment cases and public health measures isolating families and increasing financial hardships raised the concern of more missed cases of child maltreatment during the COVID-19 pandemic. For many physicians, virtual care was a novel and innovative method to maintain access to care during difficult periods of isolation through this pandemic. Detecting child maltreatment within a virtual care visit has not previously been studied. Some suspected that the window into a patient’s home would be beneficial, while others worried about the visits where the patient was not present at all.

Our survey revealed that Canadian paediatricians can identify cases of child maltreatment using virtual care. However, while almost all respondents had reported cases from in-person visits, 89% had not reported any from a virtual visit. Those who found that virtual care had greater barriers and challenges in the assessment of child maltreatment far outnumbered those who found opportunity to see the home environment beneficial. While debated prior to the publication of this survey, the benefit of being able to observe the family in their own home environment was described as a benefit by 7% of responding physicians compared to 68% of physicians who found their virtual assessments hindered their capacity to detect maltreatment.

The underlying barriers revealed in this survey are important. The greatest identified barrier was the limited ability to interact with the patient, including the performing of a physical exam. Not a single case reported to a child protective service was detected through a virtual physical exam. While it is unclear the number of cases annually detected through physical exam, sentinel injuries are well established to be an important flag for cases of child maltreatment in infants and young children (27). The second most commonly reported barrier was the lack of privacy and confidentiality. This is concerning considering patients, themselves, disclosed one third of the reported cases of child maltreatment - and were the most likely to report physical and sexual abuse, as well as emotional abuse and neglect.

Virtual medical encounters may be associated with delays in the identification of maltreatment. Of the cases where maltreatment was suspected from a virtual visit, nearly one quarter needed a subsequent follow-up visit in-person, prior to the physician reporting the case to a child protective authority. This points to additional delays, even in cases where child maltreatment was a concern through the virtual appointment. Alarmingly, 4% of physicians were able to identify a case where clear harm resulted from a child who had a missed or delayed report to a child protective service following a virtual care visit. Examples included reports of additional injuries during suspected delays in reporting. Recognizing that most cases of child maltreatment are never identified by or reported to professionals, this number likely represents an underestimate of the true number of cases missed.

STUDY LIMITATIONS AND NEXT STEPS

This survey was the first to evaluate physicians’ experiences in detecting child maltreatment through virtual medical visits. This exploratory study used survey methodology. As such, it is subject to possible recall and selection bias. The response rate was 34% and did not include other primary care providers for children and youth which limits the generalizability of the findings but is also consistent with other similar surveys using the CPSP survey methodology. As the data collected reflected the experiences of physicians and no specific patient identifying information was collected, analysis of case-based details, such as outcomes, could not be performed. Duplicate cases cannot be excluded. Given these limitations, this study cannot conclude that virtual care itself caused a delay in identification of child maltreatment. Finally, our findings do not compare virtual visits to in-person visits, as it is a survey of the identifying factors for child maltreatment in a virtual visit alone.

However, this survey establishes that cases of maltreatment can be detected during virtual care appointments. It also raises concerns, based on physician reports, that barriers relating to virtual care may delay the identification of child maltreatment. With the ongoing use of virtual care, this survey highlights the need for additional research in this area. Case details are needed to confirm child maltreatment and determine if a temporal or causal relationship exists with delays in the identification of child maltreatment in a virtual setting. This work should also be expanded to include other professionals for children and youth, including primary care providers.

CONCLUSIONS

During the COVID-19 pandemic lockdowns, a rise in unreported child maltreatment cases was established. Many barriers to detecting child maltreatment were identified by paediatricians who used virtual care. While virtual care does provide an additional access point to healthcare, it may be contributing to delays in identification and missed cases of child maltreatment, with resultant harm.

This survey is the first to report on the detection of child maltreatment by paediatricians in a virtual care setting. Given concerns about the difficulty identifying child maltreatment during virtual appointments, and the ongoing use of virtual care post-pandemic, it serves to highlight an important reason for in-person visits. Future work should seek to further understand the factors that help or prevent the detection of child maltreatment through virtual care, the impact of missed cases through virtual care and to develop guidance for health care professionals to improve detection of child maltreatment.

FUNDING

This research received no specific grant from any funding agency in the public, commercial or not-for-profit sectors.

POTENTIAL CONFLICT OF INTEREST

All authors: No reported conflicts of interest. All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.