Visits to Alberta Emergency Departments for Child Mental Health Concerns During the COVID-19 Pandemic

An Examination of Visit Trends in Relation to School Closures and Reopenings

Abstract

Objective 

We examined emergency department (ED) mental health visit trends by children in relation to periods of school closure and reopening during the COVID-19 pandemic in Alberta, Canada.

Methods 

Mental health visits by school-aged children (5 to <18 years) were extracted from the Emergency Department Information System, a province-wide database, from March 11, 2020, to November 30, 2021 (pandemic period; n = 18,997) and March 1, 2019, to March 10, 2020 (1-year, prepandemic comparator period; n = 11,540). We calculated age-specific visit rates and compared rate differences between periods of school closure (March 15–June 30, 2020; November 30, 2020–January 10, 2021; April 22–June 30, 2021) and reopening (September 4–November 29, 2020; January 11–April 21, 2021; September 3–November 30, 2021) to matched prepandemic periods. We used a ratio of relative risk to examine the risk of a visit during closures versus reopenings.

Results 

The cohort included 11,540 prepandemic visits and 18,997 pandemic visits. Compared with prepandemic periods, ED visit rates increased across all ages during the first (+85.53%; 95% confidence interval [CI], 73.68% to 100.41%) and third (+19.92%; 95% CI, 13.28% to 26.95%) school closures, and decreased during the second closure (−15.37%; 95% CI, −22.22% to −7.92%). During school reopenings, visit rates decreased across all ages during the first reopening (−9.30%; 95% CI, −13.94% to −4.41%) and increased during the third reopening (+13.59%; 95% CI, 8.13% to 19.34%); rates did not change significantly during the second reopening (2.54%; 95% CI, −3.45% to 8.90%). The risk of a visit during school closure versus reopening was only higher for the first closure with 2.06 times the risk (95% CI, 1.88 to 2.25).

Conclusions 

Emergency department mental health visit rates were highest during the first school closure of the COVID-19 pandemic, and the risk of a visit during this closure period was twice compared with when schools first reopened.

When the World Health Organization classified transmission of the SARS-CoV-2 virus as a global pandemic,¹ major disruption to education and social norms for children—school and extracurricular program closures, and social distancing—followed to protect the health of the population. While the incidence of mortality and severe illness due to the virus are now known to be low among children,² the disruptions to education and social interactions that occurred during the first 2 years of the pandemic are proposed to have had profound effects on child mental health.^3,4

Unintended harms and consequences incurred by children from school closures during the COVID-19 pandemic have been the focus of several published studies. Proposed negative effects include an increase in mental health symptoms,^5,6 maladaptive and dysregulated behaviors,^6,7 as well as stress and distress.^6–8 Other American- and Canadian-based studies have focused on changes in the number of visits to emergency department (EDs) for mental health care during the pandemic—because of the reality that the EDs are a point of first contact in times of acute distress and are a “safety net” when mental health care access and availability are limited.⁹ Reported across these studies are initial decreases in the number of mental health visits at pandemic onset,^10–15 followed by an increase in the proportion of ED visits for mental health concerns in studies that examined trends further along in the pandemic.^10–14 The authors of 2 ED studies noted that increases in mental health visits may reflect pandemic-related stress and unintended consequences of pandemic measures, including the loss of school-based mental health supports^10–12; however, an examination of visit trends in relation to school closures has yet to be conducted.

Our aim was to examine trends in ED mental health visits by children in relation to periods of school closure and reopening during the first 4 waves of the COVID-19 pandemic in Alberta, Canada.

METHODS

Study Design and Setting

In this retrospective cohort study, we used administrative data from the Emergency Department Information System (EDIS), an administrative database for the 112 EDs in Alberta, Canada. The data are collected by Alberta Health Services, the sole provider of ED care in the province, and submitted to the National Ambulatory Care Reporting System, a product of the Canadian Institute of Health Information. Submission of EDIS data for National Ambulatory Care Reporting System is mandated in Alberta. Standards for data elements are followed,¹⁶ and reported data are subjected to Canadian Institute of Health Information content and quality standards.^17,18 The study was approved by the health research ethics board of the University of Alberta. Findings are reported in accordance with the STrengthening the Reporting of OBservational studies in Epidemiology statement.¹⁹

Population and Periods

We included school-aged children (5 to <18 years) who presented to an Alberta ED for mental health care during the first 4 COVID-19 pandemic waves (March 11, 2020, to November 30, 2021), and those that made an ED visit during a 1-year, prepandemic comparator period (March 1, 2019, to March 10, 2020) to account for seasonality trends.²⁰ The start date of the comparator period was chosen to align with the start date of the first school closure.

Included children had a main ambulatory care diagnosis for a mental disorder (International Classification of Diseases, Tenth Revision, Clinical Modification, diagnostic codes F20–F25, F28–F29, F30–F34, F38–F39, F40–F43, F50, F55, F59, F60–F69, F90–F94, F99) or mental and behavioral disorders secondary to substance use (F10–F19), or a visit for intentional self-harm (X60–X69, X70–X84) that was recorded in any discharge diagnosis field as the injury (eg, laceration, ingestion) is typically recorded as the main ambulatory care diagnosis.

We used publicly available data on pandemic-related initiatives in Alberta from the Centre for Health Informatics COVID Dashboard²¹ to identify periods of school closure (March 15–June 30, 2020; November 30, 2020–January 10, 2021; April 22–June 30, 2021) and reopening (September 4–November 29, 2020; January 11–April 21, 2021; September 3–November 30, 2021). Periods of school closure were defined as periods of remote instruction (ie, online learning) and no instruction (ie, school shutdown) during a school year. Periods of school reopening were defined as periods of in-person learning (ie, in the classroom) during a school year.

Demographic and ED Visit Variables

Variables included the child's age at the time of the visit, discharge diagnoses (up to nine), acuity level for the presenting concern (Canadian Triage and Acuity Scale levels: 1: resuscitation, 2: emergent, 3: urgent, 4: less urgent, 5: nonurgent),²² and disposition (discharged from hospital, transferred, admitted, left ED, died). All data were checked to ensure cohort accuracy—age and diagnostic data met cohort parameters, all values assigned for acuity and disposition data were valid, and visit dates occurred within the study period. We created a new variable to group diagnostic data by assigning a code according to International Classification of Diseases, Tenth Revision, diagnostic clusters²³; each new entry was cross-checked with the original to ensure accuracy.

Statistical Analyses

Demographic and visit characteristics were summarized using descriptive statistics. Within school closure and reopening periods, we quantified the number of ED visits per month and calculated age-specific ED visits per 100,000 Alberta population (5 to <18 years) per day using census data.²⁴ We grouped results according to age: elementary school (5 to <12 years), junior high (12 to <16 years), and high school (16 to <18 years) to understand the impact of school closures and reopenings on different age groups. We used 2 proportion tests to compare visit rates per day between matched prepandemic and pandemic periods. Rate differences and the percent change between periods are provided with 95% confidence intervals (CIs). We used a ratio of relative risk to examine the risk of a visit during school closures versus reopenings and report findings with 95% CIs. Two-tailed P less than 0.05 were considered statistically significant.

RESULTS

Study Population

The cohort included 30,537 ED mental health visits by children; of these, 11,540 visits were made the year before the pandemic, and 18,997 visits were made during the first 4 COVID-19 pandemic waves. The median age of all children and adolescents was 15.0 years (interquartile range, 13.0–16.0); 95.3% (29,110/30,537) of visits were made by children 10 years or older.

The most common diagnoses assigned as the main reason for the ED visit were anxiety, adjustment and stress-related disorders (prepandemic: 41.7%, 4811/11,540; during the pandemic: 40.5%, 7690/18,997; difference: 1.2%; 95% CI, 0.07% to 2.3%), and mood disorders (prepandemic: 20.7%, 2392/11,540; during the pandemic: 21.6%, 4094/18,997; difference: −0.8%; 95% CI, −1.8% to 0.1%). Intentional self-harm was identified as a reason for 13.4% of visits (1549/11,540) before the pandemic, and for 17.1% of visits (3246/18,997) during the pandemic (difference: −3.7%; 95% CI, −4.5% to −2.8%).

Most ED visits were for emergent (prepandemic: 40.2%, 4638/11,540; during the pandemic: 44.3%, 8425/18,997) or urgent (prepandemic: 41.7%, 4807/11,540; during the pandemic: 40.2%, 7639/18,997) level concerns. Of note, the proportion of visits for emergent level concerns increased by 10.3% (95% CI, 7.4% to 13.4%) during the pandemic. The proportion of visits for intentional self-harm visits considered emergent and urgent in acuity also changed during the pandemic with increases of 3.3% (95% CI, −0.81 to 7.68) and 3.4% (95% CI, −7.89 to 16.06), respectively.

Visit Trends During School Closure and Reopening

Compared with prepandemic periods, ED visit rates during school closures increased across all ages during the first (+85.53%; 95% CI, 73.68% to 100.41%) and third (+19.92%; 95% CI, 13.28% to 26.95%) closures, and decreased during the second closure (−15.37%; 95% CI, −22.22% to −7.92%) (Table 1). Across all closure periods, the greatest changes in visits for mental health concerns were during the first closure among high school–aged children (108.6% increase; 95% CI, 85.74% to 134.62%).

TABLE 1 - Age Group-Specific ED Visit Rates for Mental Health Care Per 100,000 Children Per Day in Alberta, Stratified by School Closure Periods, and Rate Differences Between Closure Periods and Matched, Prepandemic Periods

	ED Visits, n		ED Visits Per 100,000 Children Per Day, n		Rate Difference (95% CI)	P	% Change (95% CI)
	Prepandemic Mar 15, 2019–Jun 30, 2019	Pandemic Closure 1 Mar 15, 2020–Jun 30, 2020	Prepandemic Mar 15, 2019–Jun 30, 2019	Pandemic Closure 1 Mar 15, 2020–Jun 30, 2020
Visits by age group, y
5 to <18 (all)	1173	2188	1.70	3.11	1.44 (1.28 to 1.61)	<0.001	86.53 (73.68 to 100.41)
5 to <12	159	253	0.40	0.64	0.24 (0.14 to 0.34)	<0.001	59.12 (29.98 to 95.27)
12 to <16	585	1040	2.87	5.09	2.23 (1.84 to 2.62)	<0.001	77.78 (60.50 to 97.07)
16 to <18	429	895	4.16	8.68	4.52 (3.83 to 5.21)	<0.001	108.62 (85.74 to 134.62)
	Prepandemic Nov 30, 2019–Jan 10, 2020	Pandemic closure 2 Nov 30, 2020–Jan 10, 2021	Prepandemic Nov 30, 2019–Jan 10, 2020	Pandemic closure 2 Nov 30, 2020–Jan 10, 2021
Visits by age group, y
5 to <18 (all)	1204	1019	4.47	3.79	−0.69 (−1.03 to −0.34)	0.001	−15.37 (−22.22 to −7.92)
5 to <12	152	92	1.00	0.61	−0.40 (−0.60 to −0.19)	0.001	−39.47 (−53.80 to −21.07)
12 to <16	578	534	7.39	6.83	−0.56 (−1.4 to 0.27)	0.190	−7.61 (−18.02 to 4.1)
16 to <18	474	393	12.00	9.95	−2.05 (−3.51 to −0.59)	0.006	−17.09 (−27.65 to −5.03)
	Prepandemic Apr 22, 2019–Jun 30, 2019	Pandemic closure 3 Apr 22, 2021–Jun 30, 2021	Prepandemic Apr 22, 2019–Jun 30, 2019	Pandemic closure 3 Apr 22, 2021–Jun 30, 2021
Visits by age group, y
5 to <18 (all)	2204	2643	4.86	5.83	0.97 (0.67 to 1.27)	<0.001	19.92 (13.28 to 26.95)
5 to <12	324	294	1.27	1.15	−0.12 (−0.31 to 0.074)	0.23	−9.26 (−22.78 to 6.59)
12 to <16	1091	1399	8.29	10.62	2.34 (1.60 to 3.08)	<0.001	28.23 (18.39 to 38.92)
16 to <18	789	950	11.86	14.29	2.42 (1.19 to 3.7)	<0.001	20.41 (9.44 to 32.49)

During school reopenings, visit rates decreased across all ages during the first reopening (−9.30%; 95% CI, −13.94% to −4.41%) with the greatest decreases for high school–aged children (−15.51%; 95% CI, −22.34% to −8.08%) and increased during the third reopening (+13.59%; 95% CI, 8.13% to 19.34%). Rate differences were statistically insignificant for the second reopening (2.54%; 95% CI, −3.45% to 8.90%) (Table 2).

TABLE 2 - Age Group-Specific ED Visit Rates for Mental Health Care Per 100,000 Children Per Day in Alberta, Stratified by School Reopening Periods, and Rate Differences Between Reopening Periods and Matched, Prepandemic Periods

	ED Visits, n		ED Visits Per 100,000 Children Per Day, n		Rate Difference (95% CI)	P	% Change (95% CI)
	Prepandemic Sep 4, 2019–Nov 29, 2019	Pandemic Reopening 1 Sep 4, 2020–Nov 29, 2020	Prepandemic Sep 4, 2019–Nov 29, 2019	Pandemic Reopening 1 Sep 4, 2020–Nov 29, 2020
Visits by age group, y
5 to <18 (all)	2969	2693	5.26	4.77	−0.49 (−0.75 to −0.23)	<0.001	−9.30 (−13.94 to −4.41)
5 to <12	359	305	1.13	0.96	−0.17 (−0.33 to −0.01)	0.036	−15.04 (−27.31 to −0.76)
12 to <16	1404	1369	8.55	8.34	−0.21 (−0.84 to 0.42)	0.506	−2.49 (−9.55 to 5.12)
16 to <18	1206	1019	14.55	12.29	−2.26 (−3.37 to −1.14)	<0.001	−15.51 (−22.34 to −8.08)
	Prepandemic Jan 11, 2020–Mar 11, 2020*	Pandemic reopening 2 Jan 11, 2021–Mar 11, 2021*	Prepandemic Jan 11, 2020–Mar 11, 2020*	Pandemic reopening 2 Jan 11, 2021–Mar 11, 2021*
Visits by age group, y
5 to <18 (all)	2128	2182	5.40	5.54	0.14 (−0.19 to 0.46)	0.411	2.54 (−3.45 to 8.90)
5 to <12	252	226	1.14	1.02	−0.12 (−0.31 to 0.08)	0.234	−10.32 (−25.39 to 7.75)
12 to <16	1159	1235	10.12	10.79	0.66 (−0.17 to 1.50)	0.120	6.56 (−1.73 to 15.55)
16 to <18	717	721	12.40	12.47	0.07 (−1.22 to 1.35)	0.916	0.56 (−9.44 to 11.67)
	Prepandemic Sept 3, 2019–Nov 30, 2019	Pandemic reopening 3 Sept 3, 2021–Nov 30, 2021	Prepandemic Sept 3, 2019–Nov 30, 2019	Pandemic reopening 3 Sept 3, 2021–Nov 30, 2021
Visits by age group, y
5 to <18 (all)	3009	3418	5.21	5.91	0.71 (0.44 to 0.98)	<0.001	13.59 (8.13 to 19.34)
5 to <12	365	341	1.12	1.05	−0.07 (−0.23 to 0.09)	0.366	−6.58 (−19.63 to 8.58)
12 to <16	1424	1787	8.48	10.64	2.16 (1.50 to 2.82)	<0.001	25.49 (16.99 to 34.63)
16 to <18	1220	1290	14.38	15.21	0.83 (−0.33 to 1.98)	0.162	5.74 (−2.30 to 14.44)

*This monthly period for school reopening 2, January 11 to April 21, was truncated to January 11 to March 11 to ensure that the prepandemic period did not overlap with the pandemic period (pandemic declared March 11, 2020).

The risk of a visit during school closure versus reopening was only higher for the first closure with 2.06 times the risk (95% CI, 1.88 to 2.25) (Table 3). The largest risk during this period was for high school–aged children (2.47; 95% CI, 2.14 to 2.85). During the second period of school closure and reopening, there was a reduced risk for all ages except junior high–aged children (Table 3).

TABLE 3 - ED Mental Health Visit Rate (Per 100,000 Children Per Day) and Relative Risk of a Visit (Pandemic vs Prepandemic School Closure and Reopening Periods), and Ratio of Relative Risk (Visit During School Closure vs Reopening)

	School Closure			School Opening
	ED Visit Rate (95% CI)	ED Visit Rate (95% CI)	Relative Risk (95% CI)	ED Visit Rate (95% CI)	ED Visit Rate (95% CI)	Relative Risk (95% CI)	RRR (95% CI)
First closure and reopening	Pandemic Mar 15, 2019–Jun 30, 2019	Prepandemic Mar 15, 2020–Jun 30, 2020	Pandemic period vs prepandemic period	Pandemic Sep 4, 2020–Nov 29, 2020	Prepandemic Sep 4, 2019–Nov 29, 2019	Pandemic period vs prepandemic period
Visits by age group, y
5 to <18 (all)	3.11 (2.98 to 3.25)	1.67 (1.58 to 1.77)	1.87 (1.74 to 2.00)	4.77 (4.59 to 4.95)	5.26 (5.07 to 5.45)	0.91 (0.86 to 0.96)	2.06 (1.88 to 2.25)
5 to <12	0.64 (0.56 to 0.72)	0.40 (0.34 to 0.47)	1.59 (1.30 to 1.95)	0.96 (0.86 to 1.07)	1.13 (1.02 to 1.25)	0.85 (0.73 to 0.99)	1.87 (1.45 to 2.42)
12 to <16	5.09 (4.79 to 5.41)	2.86 (2.64 to 3.11)	1.78 (1.60 to 1.97)	8.34 (7.91 to 8.8)	8.55 (8.11 to 9.01)	0.98 (0.90 to 1.05)	1.82 (1.61 to 2.07)
16 to <18	8.68 (8.12 to 9.27)	4.16 (3.78 to 4.57)	2.09 (1.86 to 2.35)	12.29 (11.55 to 13.07)	14.55 (13.74 to 15.39)	0.84 (0.78 to 0.92)	2.47 (2.14 to 2.85)
Second closure and reopening	Pandemic Nov 30, 2020–Jan 10, 2021	Prepandemic Nov 30, 2019–Jan 10, 2020	Pandemic period vs prepandemic period	Pandemic Jan 11, 2021–Mar 11, 2021	Prepandemic Jan 11, 2020–Mar 11, 2020	Pandemic period vs prepandemic period
Visits by age group, y
5 to <18 (all)	3.78 (3.56 to 4.02)	4.47 (4.22 to 4.73)	0.85 (0.78 to 0.92)	5.54 (5.31 to 5.78)	5.40 (5.17 to 5.64)	1.03 (0.97 to 1.09)	0.83 (0.74 to 0.92)
5 to <12	0.61 (0.49 to 0.75)	1.00 (0.85 to 1.18)	0.61 (0.46 to 0.79)	1.02 (0.89 to 1.16)	1.14 (1.00 to 1.29)	0.90 (0.75 to 1.08)	0.67 (0.49 to 0.93)
12 to <16	6.82 (6.26 to 7.43)	7.39 (6.80 to 8.01)	0.92 (0.82 to 1.04)	10.79 (10.19 to 11.4)	10.12 (9.55 to 10.72)	1.07 (0.98 to 1.16)	0.87 (0.75 to 1.00)
16 to <18	9.95 (8.99 to 10.98)	12.00 (10.94 to 13.13)	0.83 (0.72 to 0.95)	12.47 (11.57 to 13.41)	12.40 (11.51 to 13.34)	1.01 (0.91 to 1.12)	0.82 (0.69 to 0.98)
Third closure and reopening	Pandemic Apr 22, 2021–Jun 30, 2021	Prepandemic Apr 22, 2019–Jun 30, 2019	Pandemic period vs prepandemic period	Pandemic Sep 3, 2021–Nov 30, 2021	Prepandemic Sep 3, 2019– Nov 30, 2019	Pandemic period vs prepandemic period
Visits by age group, y
5 to <18 (all)	5.83 (5.61 to 6.06)	4.86 (4.66 to 5.07)	1.20 (1.13 to 1.27)	5.91 (5.72 to 6.12)	5.21 (5.02 to 5.4)	1.14 (1.08 to 1.19)	1.06 (0.98 to 1.14)
5 to <12	1.15 (1.03 to 1.29)	1.27 (1.14 to 1.42)	0.91 (0.77 to 1.07)	1.05 (0.94 to 1.17)	1.12 (1.01 to 1.24)	0.93 (0.80 to 1.09)	0.97 (0.78 to 1.21)
12 to <16	10.62 (10.07 to 11.2)	8.29 (7.80 to 8.79)	1.28 (1.18 to 1.39)	10.64 (10.15 to 11.15)	8.48 (8.04 to 8.93)	1.25 (1.17 to 1.35)	1.02 (0.92 to 1.14)
16 to <18	14.29 (13.39 to 15.22)	11.86 (11.05 to 12.72)	1.20 (1.09 to 1.32)	15.21 (14.39 to 16.06)	14.38 (13.59 to 15.21)	1.06 (0.98 to 1.14)	1.14 (1.01 to 1.29)

RRR indicates ratio of relative risk.

DISCUSSION

In our study, the greatest change across all ages, in ED visits for child mental health care, occurred when schools closed during the first months of the pandemic. This result raises important questions about the impact this particular closure had on child mental health and well-being. This closure involved children responding to, for the first time, abrupt changes in social interactions with peers, the demands of online learning (eg, remote teacher support, access to technology and Internet), academic stressors (eg, lack of in-person support for learning needs), and changes in access to school-based mental health supports. The effects of first school closure may have been experienced more intensely by high school–aged children, who may have navigated changes more independently and were more likely to experience unique academic stressors (eg, final examinations, considering transitions to higher levels of education). As visit trends during other school closure periods varied, as did trends for school reopenings, our use of the relative risk ratio provided perspective on these variations—the ratio for the first closure/reopening periods indicated the first closure as a period of greatest risk for an ED visit for a mental health crisis.

Concerns about the impact of school closures during the COVID-19 pandemic have been discussed by others, with particular emphasis placed on loss of access to school-based resources and programs for general health and well-being.²⁵ While the increase in ED visits and visit acuity levels may reflect this loss, we were unable to examine these factors in relation to ED visits given the data that were available to us. However, inaccessibility to school-based mental health services is a plausible mediating factor for the changes in ED mental health visits that we observed. Results from recent meta-analyses suggest that school-based services are effective in reducing depression and anxiety symptoms²⁶ and that programs for at-risk children can reduce psychological stress symptoms.²⁷ Children seeking emergency mental health care may be the children these programs are most effective and tailored for.

Another finding from this study that deserves discussion relates to the urgency of the concerns seen in the ED during the pandemic. Others have commented that the increase in ED visits for mental health care during the pandemic may reflect the reliance on EDs for routine care and crisis support due to changes to community-based and outpatient mental healthcare access.^4,12 That 84.5% of ED visits in this study were classified as emergent or urgent in acuity, and that the number of visits for higher acuity levels increased, particularly for intentional self-harm, during the pandemic underscores the severity of the mental health concerns during this time. If access to mental health services across multiple settings will be disrupted during subsequent public health crises, school and healthcare leaders must consider how such disruptions can be mitigated or resolved so that children and their caregivers have access to care. In this regard, virtual assessment and care delivery models warrant further investigation.^28–31 For some children, this access may prevent acute crises requiring ED care.

STRENGTHS AND LIMITATIONS

While retrospective in design, this study is the largest study to focus on ED visit trends in relation to school closures and reopenings. We feel that the generalizability of our results is plausible. Our study captured data that were very broad in scope. Data represented reporting by more than 100 EDs, with visits made to both general (n = 110) and pediatric (n = 2) EDs throughout the province.

School closures and reopenings were phenomena that occurred globally, although there was variability in their use as a public health measure. Data from the UNESCO Institute for Statistics suggest that government-mandated school closures that included remote instruction or no instruction were fewer in Canada compared with the United States (52 vs 77 weeks); however, in Canada, there were more weeks of no instruction compared with the United States (13 vs 0 weeks) and fewer weeks of remote instruction (39 weeks in Canada vs 77 weeks in the United States).³²

Although the data indicate changes in ED visit rates in relation to school closures and reopenings, we were unable to account for the potential impacts of other factors and conditions that could explain our findings including the role of school-based mental health services in supporting child mental health and well-being. An investigation of the role of these services in mitigating acute crises (including ED visits) could help clarify the role of school-based mental health care in the continuum of healthcare services.

There are also known limitations associated with using EDIS data including the possibility of inaccurate documentation and underreporting. It is possible that some mental disorders recorded as a main ambulatory care diagnosis reflect what Mitchell and Westerdium note as a “close but inaccurate” diagnosis.³³ For example, this could occur if a child presented with both depressive symptoms and suicidal ideation, and the physician recorded suicidal ideation as the main concern due to its acuity rather than the child's depression for which they were being treated long-term in the community. The possibility of underreporting mental health ED visits can occur if a medical concern was also presented during the visit.³⁴ For example, if a child presents with lacerations due to intentional self-harm, the laceration would likely be coded in the main ambulatory field and self-harm in a subsequent diagnostic field. For this reason, we collected self-harm diagnoses in any of the diagnostic fields. As both potential limitations would be present in prepandemic and pandemic era data, the impact on our ED visit rate comparisons may be minimal; however, underreporting implies that rates could be higher than what we found.

CONCLUSIONS

Study findings demonstrate that the greatest changes, across all ages, in visits for child mental health care to Alberta's EDs occurred when schools closed during the first months of the COVID-19 pandemic. This was a period of rapid change to the mode of delivery of education, interactions with peers, and access to mental health supports. Future policies surrounding school closures by public health decision makers should account for access to mental health services.