Sage Journals: Discover world-class research

Abstract

Objectives:

COVID-19–related stay-at-home orders (SAHOs) created an immediate physical barrier between children and professionals such as pediatricians and teachers, who are often first to identify and report signs of child maltreatment. Our objective was to determine how the SAHO in a southern state was associated with reports of child maltreatment and whether this association was modified by sociodemographic characteristics.

Methods:

We linked data on reports of child maltreatment from a southern state in the United States from October 1, 2018, through September 30, 2020, to data from the US Census Bureau to obtain data on county-level socioeconomic characteristics. We fit a segmented regression model to evaluate changes in reports before and after the SAHO, March 20, 2020. We evaluated potential disparities by child age, case and allegation severity, and socioeconomic characteristics.

Results:

Of 374 885 hotline calls, 276 878 (73.9%) were made before the SAHO and 98 007 (26.1%) after it. Although an immediate decrease in reports of child maltreatment occurred on the day of the SAHO, the rates of reporting within socioeconomic groups started increasing thereafter. While we found no significant change in the overall rate of change in hotline calls after versus before the SAHO (0.23; 95% CI, −0.11 to 0.58), stratified analyses indicate that the rates at which reporting increased varied by education level, health insurance coverage, median annual household income, and unemployment.

Conclusions:

Evaluating these trends is important for policy makers and practitioners to understand how policies enforced during the pandemic influence child maltreatment reporting and how these policies may affect reporting differently across socioeconomic groups.

Keywords

COVID-19 hotline calls child maltreatment child neglect child abuse

Child maltreatment is a serious public health issue and has major consequences for individuals, families, and society as a whole.¹ Every year, 1 in 10 children are neglected or psychologically abused, and 4% to 16% are physically abused in high-income countries such as the United States.² In the United States alone, >3 million referrals are received annually that involve >6 million children.¹ Child maltreatment is determined by complex factors related to individuals, families, communities, and cultures.³ It has lifelong cascading effects on children, negatively altering current and future physical and mental health, education, work, social relationships, and violent behavior.² Child maltreatment also imposes an economic burden on individuals and society. Among individual victims of child maltreatment, per-victim lifetime costs can vary from $210 012 to $16.6 million.⁴ About $80.3 billion each year is spent in the United States in direct and indirect costs related to cases of child abuse or neglect.¹

Child-level outcomes can be affected by broader societal structures.^5
-7 Economic factors in the household may directly and indirectly affect child and family health outcomes.^7-12 Specifically, greater rates of poverty, lower incomes and education levels, and higher unemployment have been positively associated with rates of child maltreatment at the community level.^8,9,13-20 In addition, lower rates of health insurance in a community have been found to increase the risk of a child being placed into out-of-home care.²¹ Other socioeconomic characteristics of a community related to economic disadvantages, such as the percentage of female-headed households, have been positively associated with rates of child abuse and maltreatment.^{8,10,19,20,22-25}

Reports of child maltreatment come in many forms. The 1974 Federal Child Abuse Prevention and Treatment Act requires some individuals to be mandatory reporters of child maltreatment.²⁶ The list in each state may vary, although the most common groups required to report child maltreatment include social workers, school personnel, health care workers, mental health professionals, child care providers, medical examiners or coroners, and law enforcement officers. In some states, all people are required to report suspected abuse or neglect, regardless of profession.²⁶ Reporting behavior among professionals versus nonprofessionals can vary, given that each group of potential reporters is involved in different aspects of a child’s life.²⁷ While institutions such as schools have clear internal policies and procedures for reporting and handling maltreatment,²⁶ reporting among the general population is less structured and may be influenced by the relationship to the child or family, community context, perceived level of collective community efficacy, personal experiences with child maltreatment, ethnicity, and culture, among others.^27
-30 In 2019, 21.0% of all reported child maltreatment in the United States was reported by education personnel; 19.1% by legal and law enforcement personnel; 16.3% by social services or mental health personnel; 15.3% by parents, friends, neighbors, or other relatives; and 11.0% by medical personnel.³¹

By the end of 2020, the COVID-19 pandemic was the leading cause of death in the United States³² and was associated with increased economic uncertainty among families and societies,³³ higher unemployment rates,³⁴ and greater perceived parental stress,³⁵ all of which are known risk factors for child maltreatment.³⁶ In addition, COVID-19–related stay-at-home orders led to social isolation for children and families,³⁷ creating an immediate physical barrier between children and the most common mandatory reporters of child maltreatment during a time of heightened risk of maltreatment.

The negative impact of COVID-19 on child maltreatment reporting has been documented in Australia,³⁸ Brazil,³⁹ Japan,⁴⁰ South Africa,⁴¹ the United Kingdom,⁴² France,⁴³ Mexico,⁴⁴ and Korea,⁴⁵ among others.⁴⁶ Major cities such as New York City⁴⁷ and Los Angeles, California,⁴⁸ saw a decrease in child maltreatment reporting after stay-at-home orders, as did states such as Florida,⁴⁹ California,⁵⁰ Colorado,⁵¹ North Carolina,⁵² and Georgia.⁵³ The impact of COVID-19 stay-at-home orders on child maltreatment reporting has been investigated by reporter type,^51,54 race,⁵⁴ maltreatment type,⁵⁴ and association to allegations of domestic violence.⁵⁰

Social determinants of health should not be ignored while conducting COVID-19 policy–related research.⁵⁵ However, to our knowledge, no study has evaluated the impact of COVID-19–related stay-at-home orders on child maltreatment or disproportionality with socioeconomic characteristics. We evaluated whether the stay-at-home order during the COVID-19 pandemic resulted in a significant change in the number and rate of change in reports of child maltreatment, focusing on differences related to socioeconomic characteristics such as income, poverty, unemployment, education, health insurance, and head of household.

Methods

Data Sources and Study Population

We used administratively collected data on child maltreatment allegations (also referred to as hotline call data) from a southern state in the United States during the 2-year period from October 1, 2018, through September 30, 2020 (N = 374 885 calls). Reports were made to trained case managers by calling a statewide hotline when child abuse and neglect were suspected. The data included variables such as child age (<5, 5-9, 10-14, 15-17, ≥18 y), allegation and case severity (severe, not severe), and county where the call occurred. Allegation severity was determined by the case manager who received the call. Each case was then assigned to a specific assessment or investigation group where the case severity was then confirmed. We merged data from the US Census Bureau’s American Community Survey from 2018 to 2020 to hotline call data at the county level to obtain data on socioeconomic characteristics.⁵⁶ The Vanderbilt University Institutional Review Board (#212114) and the state Department of Children’s Services Review Board reviewed and approved this study. The Vanderbilt Institutional Review Board determined that the study posed minimal risk to participants and met criteria for exempt review per 45 CFR 46.104(d)(4)(ii). The state requested that it not be identified; thus, we omitted this information from the article.

The primary outcome for our analysis was 7-day moving average of the number of hotline calls. County-level socioeconomic characteristics obtained from US Census data included the median annual household income and the percentage of households below the federal poverty level, total population unemployed, college-educated residents, female-headed households, and residents with private or public health insurance. All categories were adopted from published standards of the Harvard Public Health Disparities Geocoding Project if available.⁵⁷ No recommended cutoffs for percentage of total population unemployed were found in the literature, so we grouped it by quintiles across all counties. We chose these socioeconomic characteristics a priori according to subject matter knowledge and existing literature on child maltreatment risk factors.^{8,13,15,16,21} While individual child demographic information such as race may also be associated with child maltreatment,^54,58 the data source used in this study did not include that information; therefore, we did not include race as a study variable.

Data Analysis

We conducted a segmented regression analysis for an interrupted time series to model trends in hotline calls before and after March 20, 2020, the day of the statewide stay-at-home order due to COVID-19. The analysis model included report date, indicator for statewide stay-at-home order (before vs after), and an interaction between the date and the indicator. We used model results to estimate the average number of hotline calls and corresponding 95% CIs on the day before and the day of the stay-at-home order. We determined significant differences in these 2 values by fitting a corresponding model with a centered report date and by examining the significance of the coefficient on the indicator for statewide stay-at-home order at the α = .05 level (eTable in Supplemental Material). We also examined the rate of change in hotline calls before versus after the stay-at-home order, represented by the model coefficient on the interaction term. If the coefficient was significant (ie, the 95% CI for the coefficient did not include 0), we determined the rate of change to be significant at the α = .05 level. Additionally, to examine potential disparities, we stratified models by the following variables: case severity, allegation severity, and child age, as well as, at the county level, median annual household income and percentage of households below the federal poverty level, total population unemployed, college-educated residents, female-headed households, and residents with private or public health insurance. We created displays of piecewise segments to visualize trends of the average number and change of rate of calls before and after the stay-at-home order overall and by case severity, allegation severity, child age, and socioeconomic characteristics. We used R version 4.0.3 (R Core Team) for all analyses.

Results

From October 1, 2018, through September 30, 2020, a total of 374 885 hotline calls were made: 276 878 (73.9%) before the statewide stay-at-home order on March 20, 2020, and 98 007 (26.1%) after it (Table 1). On average, 3702 and 3224 calls per week were being made before and after the stay-at-home order, respectively. The median (IQR) child age decreased from 8.0 (3.0-12.0) to 7.0 (3.0-12.0) before and after the stay-at-home order, while the mean (SD) child age decreased from 7.9 (5.1) to 7.6 (5.2). The percentage of severe allegations increased from 9.3% to 10.3%, while the percentage of severe cases increased from 14.3% to 16.7% before and after the stay-at-home order.

Table 1.

Characteristics of hotline calls to report allegations of child maltreatment before and after stay-at-home orders during the COVID-19 pandemic, on March 20, 2020, in a southern state in the United States, October 1, 2018, through September 30, 2020^a

	Hotline calls, no. (column %)
Variable	Before stay-at-home order (n = 276 878)	After stay-at-home order (n = 98 007)	Total (N = 374 885)
Case severity
Not severe	237 187 (85.7)	81 633 (83.3)	318 820 (85.0)
Severe	39 691 (14.3)	16 374 (16.7)	56 065 (15.0)
Allegation severity
Not severe	251 125 (90.7)	87 889 (89.7)	339 014 (90.4)
Severe	25 753 (9.3)	10 118 (10.3)	35 871 (9.6)
Age of child, y
<5	83 447 (30.1)	33 064 (33.7)	116 511 (31.1)
5-9	77 580 (28.0)	26 378 (26.9)	103 958 (27.7)
10-14	71 547 (25.8)	23 728 (24.2)	95 275 (25.4)
15-17	33 628 (12.1)	11 733 (12.0)	45 361 (12.1)
≥18	459 (0.2)	187 (0.2)	646 (0.2)
Missing	10 217 (3.7)	2917 (3.0)	13 134 (3.5)
Socioeconomic characteristics of county in which hotline call originated
Households living below federal poverty level, %
0.0-4.9	3211 (1.2)	932 (1.0)	4143 (1.1)
5.0-9.9	70 864 (25.6)	30 521 (31.1)	101 385 (27.0)
10.0-19.9	201 848 (72.9)	65 883 (67.2)	267 731 (71.4)
≥20.0	955 (0.3)	671 (0.7)	1626 (0.4)
Total population unemployed, %
0.0-4.3	53 819 (19.4)	21 158 (21.6)	74 977 (20.0)
4.4-4.8	51 465 (18.6)	23 512 (24.0)	74 977 (20.0)
4.9-6.0	59 417 (21.5)	15 560 (15.9)	74 977 (20.0)
6.1-7.1	59 536 (21.5)	15 441 (15.8)	74 977 (20.0)
≥7.2	52 641 (19.0)	22 336 (22.8)	74 977 (20.0)
Median annual household income, $
26 472-33 162	958 (0.3)	348 (0.4)	1306 (0.3)
33 163-39 286	18 395 (6.6)	4446 (4.5)	22 841 (6.1)
39 287-47 124	71 152 (25.7)	21 949 (22.4)	93 101 (24.8)
≥47 125	186 373 (67.3)	71 264 (72.7)	257 637 (68.7)
Female-headed households, %
0.0-19.9	141 934 (51.3)	47 122 (48.1)	189 056 (50.4)
≥20.0	134 944 (48.7)	50 885 (51.9)	185 829 (49.6)
Residents with private or public health insurance, %
0.0-79.9	5874 (2.1)	5022 (5.1)	10 896 (2.9)
≥80.0	271 004 (97.9)	92 985 (94.9)	363 989 (97.1)
College-educated residents, %
0.0-39.9	252 275 (91.1)	89 451 (91.3)	341 726 (91.2)
≥40.0	24 603 (8.9)	8556 (8.7)	33 159 (8.8)

Data sources include administratively collected data on child maltreatment allegations for information on case severity, allegation severity, and age of child, as well as the US Census Bureau American Community Survey from 2018 to 2020 for information on county-level socioeconomic characteristics.⁵⁶

Day-to-day variability in average hotline calls decreased after the stay-at-home order, with the SD decreasing from 147 to 100. Piecewise segments suggest that the average number of hotline calls tended to decrease on the day of the stay-at-home order as compared with the day before, while the change in rate of calls increased after the stay-at-home order overall (Figure 1B) and by case severity, allegation severity, child age, and socioeconomic characteristics (Figure 2).

Figure 1.

Trends and variability in 7-day moving average of hotline calls for child maltreatment allegations (A) and segmented regression graph with piecewise segments (B) before (SD = 147) and after (SD = 100) stay-at-home orders during the COVID-19 pandemic, from October 1, 2018, through September 30, 2020, in a southern state in the United States. The dashed vertical line indicates the first day of the statewide COVID-19 stay-at-home order on March 20, 2020. Data source: administratively collected data on reports of child maltreatment from a southern state in the United States.

Figure 2.

Segmented regression graph with piecewise segments by case severity (A), allegation severity (B), age of child (C), county-level percentage of households below the federal poverty level (D), county-level percentage of total population unemployed (E), county-level median annual household income (F), county-level percentage of female-headed households (G), county-level percentage of residents with private or public health insurance (H), and county-level percentage of college-educated residents (I) from October 1, 2018, through September 30, 2020, in a southern state in the United States. The dashed vertical line indicates the first day of the statewide COVID-19 stay-at-home order on March 20, 2020.

We found a significant difference between the overall average number of hotline calls on the day of the stay-at-home order (March 20, 2020) (466; 95% CI, 428-505) and the day before (March 19, 2020) (509; 95% CI, 486-532). We also found a significant drop in the average number of hotline calls on the day of the stay-at-home order across all variable categories, except among those who were aged ≥18 years, those who resided in a county with a median annual household income of $26 472-$33 162, and those who lived in a county in which ≥20% of the households lived below the federal poverty level (Table 2). Despite these decreases, after 6 months the number of hotline calls eventually reached the levels seen before the stay-at-home order.

Table 2.

Estimated average number of hotline calls to report allegations of child maltreatment before and on the day of a statewide stay-at-home order during the COVID-19 pandemic in a southern state in the United States, March 2020^a

	Estimated no. of hotline calls, average (95% CI)
Variable	Day before COVID-19 stay-at-home order (March 19, 2020)	Day of COVID-19 stay-at-home order (March 20, 2020)
Overall	509 (486-532)	466 (428-505)^b
Case severity
Not severe	453.6 (444.0-463.3)	280.8 (265.0-296.7)^b
Severe	82.3 (80.2-84.5)	56.7 (53.1-60.2)^b
Allegation severity
Not severe	482.9 (472.7-493.0)	301.6 (284.8-318.4)^b
Severe	53.1 (51.8-54.4)	35.9 (33.7-38.0)^b
Age of child, y
<5	157.7 (154.7-160.7)	133.6 (128.7-138.5)^b
5-9	150.8 (146.9-154.7)	84.0 (77.6-90.4)^b
10-14	141.5 (137.7-145.3)	70.2 (63.9-76.5)^b
15-17	67.0 (65.1-68.8)	37.9 (34.9-40.9)^b
≥18	1.1 (0.9-1.3)	2.1 (1.7-2.4)^c
Socioeconomic characteristics of county in which hotline call originated
Households living below federal poverty level, %
0.0-4.9	6.2 (5.7-6.6)	2.3 (1.6-3.1)^b
5.0-9.9	172.0 (167.5-176.4)	107.1 (99.7-114.4)^b
10.0-19.9	355.8 (347.8-363.9)	225.5 (212.2-238.7)^b
≥20.0	2.9 (2.5-3.3)	3.0 (2.3-3.7)
Total population unemployed, %
0.0-4.3	136.8 (132.3-141.7)	71.7 (64.3-79.0)^b
4.4-4.8	128.9 (125.9-132.0)	85.3 (80.3-90.3)^b
4.9-6.0	84.5 (81.4-87.6)	49.4 (44.3-54.5)^b
6.1-7.1	96.9 (93.4-100.3)	54.0 (48.2-59.7)^b
≥7.2	88.9 (84.3-93.5)	77.1 (69.6-84.7)^b
Median annual household income, $
26 472-33 162	1.5 (1.2-1.8)	2.2 (1.7-2.8)^c
33 163-39 286	22.9 (21.6-24.2)	16.8 (14.7-19.0)^b
39 287-47 124	126.4 (123.0-129.8)	68.1 (62.6-73.7)^b
≥47 125	385.6 (377.3-393.9)	250.9 (237.3-264.5)^b
Female-headed households, %
0.0-19.9	261.7 (255.6-267.7)	153.2 (143.2-163.1)^b
20.0-39.9	274.3 (268.4-280.2)	184.3 (174.6-194.1)^b
Residents with private or public health insurance, %
0.0-79.9	19.3 (18.3-20.3)	15.5 (13.9-17.1)^b
≥80.0	517.4 (506.4-528.4)	322.0 (303.9-340.1)^b
College-educated residents, %
0.0-39.9	485.3 (474.8-495.8)	304.4 (287.1-321.6)^b
≥40.0	50.7 (49.3-52.1)	33.1 (30.8-35.4)^b

Estimated averages and 95% CIs were obtained by fitting stratified segmented regression models, including report date, indicator for statewide stay-at-home order, and an interaction between the date and the indicator. A significant difference was determined if the 95% CI for the coefficient on the indicator did not contain 0 for the corresponding model with centered date. See Supplemental Table for details.

A significant drop in means within that category at the α = .05 level.

A significant increase in means within that category at the α = .05 level.

We found no significant change in the overall rate of change in hotline calls after versus before the stay-at-home order (0.23; 95% CI, −0.11 to 0.58) (Table 3). Stratified analyses indicated a significant increase in the rate of change in hotline calls after the stay-at-home order among all variables, except among those who resided in a county in which ≥20% of the household lived below the federal poverty level or a county with a median annual household income of $26 472-$33 162. This finding indicates a faster increase in the day-to-day change in hotline calls after the stay-at-home order (vs before) in all other groups. We observed the rate of change in hotline calls to differ most notably among groups in the following variables: case severity, allegation severity, percentage of county with college-educated residents, percentage of residents with private or public health insurance, and median annual household income. The rate of change in average hotline calls increased with median annual household income and percentage of residents with private or public health insurance and decreased with case severity, allegation severity, and percentage of county with college-educated residents. The rate of change in average hotline calls increased with the percentage of the population unemployed, albeit not as strongly as other socioeconomic characteristics.

Table 3.

Estimated change in rate of hotline calls to report allegations of child maltreatment in a southern state after a statewide COVID-19 stay-at-home order on March 20, 2020, versus before the stay-at-home order, stratified by case severity, allegation severity, age, and socioeconomic characteristics^a

Category	Estimated change in rate (95% CI)
Overall	0.23 (–0.11 to 0.58)
Case severity
Not severe	1.07 (0.92 to 1.21)^b
Severe	0.19 (0.16 to 0.22)^b
Allegation severity
Not severe	1.15 (1.00 to 1.31)^b
Severe	0.11 (0.09 to 0.13)^b
Age of child, y
<5	0.25 (0.21 to 0.30)^b
5-9	0.40 (0.34 to 0.46)^b
10-14	0.40 (0.34 to 0.46)^b
15-17	0.17 (0.14 to 0.20)^b
≥18	–0.01 (−0.01 to −0.01)^b
Socioeconomic characteristics of county in which hotline call originated
Households below federal poverty level, %
0.0-4.9	0.02 (0.02 to 0.03)^b
5.0-9.9	0.24 (0.17 to 0.31)^b
10.0-19.9	0.99 (0.87 to 1.11)^b
≥20.0	0.00 (0.00 to 0.01)
Total population unemployed, %
0.0-4.3	0.16 (0.10 to 0.23)^b
4.4-4.8	0.15 (0.10 to 0.19)^b
4.9-6.0	0.35 (0.31 to 0.40)^b
6.1-7.1	0.26 (0.21 to 0.31)^b
≥7.2	0.33 (0.26 to 0.40)^b
Median annual household income, $
26 472-33 162	0.00 (–0.01 to 0.01)
33 163-39 286	0.08 (0.06 to 0.10)^b
39 287-47 124	0.40 (0.35 to 0.45)^b
≥47 125	0.77 (0.65 to 0.90)^b
Female-headed households, %
0.0-19.9	0.74 (0.65 to 0.83)^b
20.0-39.9	0.52 (0.43 to 0.61)^b
Residents with private or public health insurance, %
0.0-79.9	0.05 (0.04 to 0.07)^b
≥80.0	1.20 (1.04 to 1.37)^b
College-educated residents, %
0.0-39.9	1.20 (1.04 to 1.35)^b
≥40.0	0.06 (0.04 to 0.08)^b

Each estimate was obtained by fitting stratified segmented regression models, including report date, indicator for statewide stay-at-home order, and an interaction between the date and the indicator. The estimate is represented by the coefficient of the interaction term. The estimate obtained for nonsevere cases, for example, indicates that the rate of average hotline calls increased by 1.07 calls per day after the stay-at-home order began.

The change in rate for that category is significant from 0 at the α = .05 level.

Discussion

We used data from a southern state in the United States to examine changes in the trend in hotline calls before and during the COVID-19–associated stay-at-home order across various age groups, allegation severity, case severity, and socioeconomic characteristics. We found that despite an immediate decrease in the average number of hotline calls on the day of the stay-at-home order, the rate of change after the order increased across all socioeconomic groups and case and allegation severity. The rates at which reporting increased varied by county-level education, health insurance coverage, median annual household income, and unemployment.

Many studies are consistent with ours in that they show an immediate decrease in the number of child maltreatment reports after COVID-19–related stay-at-home orders at the city and state levels in the United States. Across the United States, studies indicated a decrease in child maltreatment reporting during the early stages of COVID-19.^59,60 In Los Angeles, the number of child abuse and neglect reports made to the Los Angeles Police Department decreased by 8.0% following stay-at-home orders.⁴⁸ Similar stay-at-home policies in New York City led to 28.8%, 51.1%, and 46.6% fewer reports of child maltreatment than expected in March, April, and May 2020, respectively, when compared with 2019.⁴⁷ In California, the number of child protection reports dropped 14.3% after stay-at-home orders, although the proportion alleging domestic violence increased.⁵⁰ After statewide stay-at-home orders in Colorado, referrals and responses of child maltreatment dropped by 27.9% as compared with 2019, with 68.7% and 39.2% drops coming from school professionals and daycare providers, respectively.³⁵ The number of daily reports fell from 43.5 in 2019 to 31.6 in 2020 following stay-at-home orders in Michigan.⁵⁴ Decreases in reports were also seen in states near the state that we studied. The number of reported allegations of child maltreatment was 27% lower than expected in March and April 2020 in Florida.⁴⁹ Reports in Georgia decreased by 58% during 2020 as compared with earlier years, driven by fewer reports made by educational personnel.⁵³ Lastly, in North Carolina the rate of child maltreatment reporting fell from 2716 per 100 000 population in 2019 to 2547 in 2020.⁵² In the United States, reports made by education or medical personnel make up almost one-third of all child maltreatment reporting³¹; therefore, we believe that the immediate separation of children and these mandatory reporters, among all others, likely led to the decrease in reporting following stay-at-home orders.

Although the number of child maltreatment reports tended to decrease after COVID-19 stay-at-home orders, previous studies showed that parental experiences associated with higher risk for child maltreatment increased during the pandemic,⁶¹ consistent with our results of increased rates of reporting following stay-at-home orders. Domestic violence as a whole and substance use also increased during the pandemic, with police calls indicating that domestic violence increased from 7.5% to 12.0% on average as compared with before the pandemic in major US cities.^62,63 Mohler et al⁶⁴ reported that 911 calls for domestic violence increased in Los Angeles and Indianapolis during the pandemic. In addition, family violence increased among families with a history of violence in rural and suburban areas of North Carolina,⁶⁵ and violence-related trauma increased by 55% in Connecticut communities.⁶⁶ Finally, substance use and drug overdoses in the United States increased during the COVID-19 pandemic.⁶⁷

Strengths and Limitations

Our study had several strengths. First, we analyzed the impact of the stay-at-home order on the number of hotline calls by using the interrupted time-series method, a widely used method in evaluating the effect of policies with a clearly defined start date,⁶⁸ as was the case in our context. Second, we examined the number and rate of change in hotline calls, which gives us a better understanding of the immediate and evolving consequences of COVID-19–related policies. By September 30, 2020, the COVID-19–related stay-at-home order was still in place in this southern state; therefore, trends reported from our analysis were not affected by the policy ending. Third, our analysis incorporated county-level socioeconomic variables. Child maltreatment reporting has been shown to differ among socioeconomic groups,²⁸ and our study quantified some of those differences in the context of COVID-19. To our knowledge, no other study has examined possible disparities in child maltreatment reporting across socioeconomic groups.

Our study also had several limitations. First, the data used in this analysis did not include information on child race or reporter type. Differences in reporting of child maltreatment across various racial groups^54,58 and reporter type^51,54 have been shown. These variables were not available in our data but could have provided further insight into these differences during the pandemic. Second, we incorporated socioeconomic information into our analysis, although we included these continuous values as county-level categorical variables. Third, while we did include time as a variable in our models, possible autocorrelation of error terms may still be present, although we did not adjust for this further. Fourth, to ensure that our results were easily interpretable, we did not include a model that adjusted for all possible covariates, although unadjusted models may have led to biased estimates. Fifth, we modeled trends of hotline calls linearly, although the relationship between time since the stay-at-home order and allegations may be more complex than a linear relationship. Lastly, we used a single source of data on child maltreatment allegations from a single southern state up to September 30, 2020; therefore, our results should be considered within such context. A more complete picture of how child maltreatment has been affected by COVID-19 across the United States requires understanding our results alongside all other studies on this topic.

Conclusion

We found that the stay-at-home order related to the COVID-19 pandemic was associated with an immediate decrease in child maltreatment allegations in a single southern state. We also noted a subsequent increase in the rate of change in child maltreatment allegations across most age groups, allegation and case severity, and county-level socioeconomic characteristics. Immediate declines in the average number of allegations are likely linked to the physical separation of mandatory reporters of child maltreatment in professional settings and children. The increases in the day-to-day rates of change in allegations may be associated with the increase of domestic violence and/or parental substance use during the COVID-19 pandemic. Evaluating these trends is important for policy makers and practitioners to understand how policies enforced during the pandemic influence child maltreatment reporting and how these policies may affect reporting differently across socioeconomic groups.

Supplemental Material

sj-docx-1-phr-10.1177_00333549241245846 – Supplemental material for Reporting of Child Maltreatment During the COVID-19 Pandemic in a Southern State in the United States

Supplemental material, sj-docx-1-phr-10.1177_00333549241245846 for Reporting of Child Maltreatment During the COVID-19 Pandemic in a Southern State in the United States by Julia C. Thome, Kathy Gracey, Richard A. Epstein, Michael J. Cull, Tarah Kuhn and Rameela Raman in Public Health Reports

Footnotes

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

ORCID iDs

Julia C. Thome, BA

Rameela Raman, PhD

Supplemental Material

Supplemental material for this article is available online. The authors have provided these supplemental materials to give readers additional information about their work. These materials have not been edited or formatted by Public Health Reports’s scientific editors and, thus, may not conform to the guidelines of the AMA Manual of Style, 11th Edition.

References

Petersen

Joseph

Feit

, eds. New Directions in Child Abuse and Neglect Research. National Academies Press; 2014.

Gilbert

Widom

Browne

Fergusson

Webb

Janson

Burden and consequences of child maltreatment in high-income countries. Lancet. 2009;373(9657):68-81. doi:10.1016/S0140-6736(08)61706-7

Belsky

Child maltreatment: an ecological integration. Am Psychol. 1980;35(4):320-335. doi:10.1037/0003-066X.35.4.320

Peterson

Florence

Klevens

The economic burden of child maltreatment in the United States, 2015. Child Abuse Negl. 2018;86:178-183. doi:10.1016/j.chiabu.2018.09.018

Butchart

Harvey

Mian

Fürniss

Kahane

Preventing Child Maltreatment: A Guide to Taking Action and Generating Evidence. World Health Organization; 2006.

Belsky

Etiology of child maltreatment: a developmental–ecological analysis. Psychol Bull. 1993;114(3):413-434. doi:10.1037/0033-2909.114.3.413

National Research Council. Understanding Child Abuse and Neglect. National Academies Press; 1993.

Ben-Arieh

Localities, social services and child abuse: the role of community characteristics in social services allocation and child abuse reporting. Child Youth Serv Rev. 2010;32(4):536-543. doi:10.1016/j.childyouth.2009.11.006

Coulton

Korbin

Neighborhoods and child maltreatment: a multi-level study. Child Abuse Negl. 1999;23:1019-1040. doi:10.1016/s0145-23134(99)00076-9

10.

Ernst

JS.

Community-level factors and child maltreatment in a suburban county. Soc Work Res. 2001;25(3):133-142.

11.

Fong

Child welfare involvement and contexts of poverty: the role of parental adversities, social networks, and social services. Child Youth Serv Rev. 2017;72:5-13. doi:10.1016/j.childyouth.2016.10.011

12.

McLoyd

VC.

The impact of economic hardship on Black families and children: psychological distress, parenting, and socioemotional development. Child Dev. 1990;61(2):311-346. doi:10.1111/j.1467-8624.1990.tb02781.x

13.

Eckenrode

Smith

McCarthy

Dineen

Income inequality and child maltreatment in the United States. Pediatrics. 2014;133(3):454-461. doi:10.1542/peds.2013-1707

14.

Gracia

López-Quílez

Marco

Lila

Mapping child maltreatment risk: a 12-year spatio-temporal analysis of neighborhood influences. Int J Health Geogr. 2017;16(1):38. doi:10.1186/s12942-017-0111-y

15.

Gracia

López-Quílez

Marco

Lila

Neighborhood characteristics and violence behind closed doors: the spatial overlap of child maltreatment and intimate partner violence. PLoS One. 2018;13(6):e0198684. doi:10.1371/journal.pone.0198684

16.

Kim

Drake

Child maltreatment risk as a function of poverty and race/ethnicity in the USA. Int J Epidemiol. 2018;47(3):780-787. doi:10.1093/ije/dyx280

17.

Lindo

Schaller

Hansen

Caution! Men not at work: gender-specific labor market conditions and child maltreatment. J Public Econ. 2018;163:77-98. doi:10.1016/j.jpubeco.2018.04.007

18.

Maguire-Jack

Multilevel investigation into the community context of child maltreatment. J Aggress Maltreat Trauma. 2014;23(3):229-248. doi:10.1080/10926771.2014.881950

19.

Merritt

DH.

Child abuse potential: correlates with child maltreatment rates and structural measures of neighborhoods. Child Youth Serv Rev. 2009;31(8):927-934. doi:10.1016/j.childyouth.2009.04.009

20.

Spearly

Lauderdale

Community characteristics and ethnicity in the prediction of child maltreatment rates. Child Abuse Negl. 1983;7:91-105. doi:10.1016/0145-2134(83)90036-4

21.

Lotspeich

Jarrett

Epstein

, et al. Incidence and neighborhood-level determinants of child welfare involvement. Child Abuse Negl. 2020;109:104767. doi:10.1016/j.chiabu.2020.104767

22.

Cherry

Wang

The link between male employment and child maltreatment in the US, 2000-2012. Child Youth Serv Rev. 2016;66:117-122. doi:10.1016/j.childyouth.2016.05.008

23.

Paxson

Waldfogel

Parental resources and child abuse and neglect. Am Econ Rev. 1999;89(2):239-244. doi:10.1257/aer.89.2.239

24.

Paxson

Waldfogel

Work, welfare, and child maltreatment. J Labor Econ. 2003;20(3):435-474. doi:10.3386/w7343

25.

Weissman

Jogerst

Dawson

JD.

Community characteristics associated with child abuse in Iowa. Child Abuse Negl. 2003;27:1145-1159. doi:10.1016/j.chiabu.2003.09.002

26.

Child Abuse Prevention and Treatment Act, S 1191, Pub L No 93-247 (1974).

27.

McDaniel

In the eye of the beholder: the role of reporters in bringing families to the attention of child protective services. Child Youth Serv Rev. 2006;28(3):306-324. doi:10.1016/j.childyouth.2005.04.010

28.

Wolf

Baiocchi

Argüello

Child maltreatment reporting in the general population: examining the roles of community, collective efficacy, and adverse childhood experiences. Child Abuse Negl. 2018;82:201-209. doi:10.1016/j.chiabu.2018.06.003

29.

Ashton

The effect of personal characteristics on reporting child maltreatment. Child Abuse Negl. 2004;28(9):985-997. doi:10.1016/j.chiabu.2004.03.012

30.

Zellman

GL.

The impact of case characteristics on child abuse reporting decisions. Child Abuse Negl. 1992;16(1):57-74. doi:10.1016/0145-2134(92)90008-F

31.

Children’s Bureau. Child Maltreatment 2019. US Department of Health and Human Services; 2021.

32.

Koh

Geller

VanderWeele

TJ.

Deaths from COVID-19. JAMA. 2021;325(2):133-134. doi:10.1001/jama.2020.25381

33.

Nicola

Alsafi

Sohrabi

, et al. The socio-economic implications of the coronavirus pandemic (COVID-19): a review. Int J Surg. 2020;78:185-193. doi:10.1016/j.ijsu.2020.04.018

34.

Parolin

Unemployment and child health during COVID-19 in the USA. Lancet Public Health. 2020;5(10):e521-e522. doi:10.1016/S2468-2667(20)30207-3

35.

Brown

Doom

Lechuga-Peña

Watamura

Koppels

Stress and parenting during the global COVID-19 pandemic. Child Abuse Negl. 2020;110(pt 2):104699. doi:10.1016/j.chiabu.2020.104699

36.

Stith

Liu

Davies

, et al. Risk factors in child maltreatment: a meta-analytic review of the literature. Aggres Violent Behav. 2009;14(1):13-29. doi:10.1016/j.avb.2006.03.006

37.

Banerjee

Rai

Social isolation in COVID-19: the impact of loneliness. Int J Soc Psychiatry. 2020;66(6):525-527. doi:10.1177/0020764020922269

38.

Carrington

Morley

Warren

, et al. The impact of COVID-19 pandemic on Australian domestic and family violence services and their clients. Aust J Soc Issues. 2021;56(4):539-558. doi:10.1002/ajs4.183

39.

Platt

Guedert

Coelho

EBS

. Violence against children and adolescents: notification and alert in times of pandemic. Rev Paul Pediatr. 2020;39:e2020267. doi:10.1590/1984-0462/2021/39/2020267

40.

Nakamura

Children’s health and well-being during the COVID-19 pandemic in Japan. Asian Soc Work J. 2021;6(2):article 2. doi:10.47405/aswj.v6i2.164

41.

Fouché

Theron

LC.

Child protection and resilience in the face of COVID-19 in South Africa: a rapid review of C-19 legislation. Child Abuse Negl. 2020;110(pt 2):104710. doi:10.1016/j.chiabu.2020.104710

42.

Rengasamy

Long

Rees

Davies

Hildebrandt

Payne

Impact of COVID-19 lockdown: domestic and child abuse in Bridgend. Child Abuse Negl. 2022;130(pt 1):105386. doi:10.1016/j.chiabu.2021.105386

43.

Massiot

Launay

Fleury

, et al. Impact of COVID-19 pandemic on child abuse and neglect: a cross-sectional study in a French Child Advocacy Center. Child Abuse Negl. 2022;130(pt 1):105443. doi:10.1016/j.chiabu.2021.105443

44.

Cabrera-Hernández

Padilla-Romo

. Hidden Violence: How COVID-19 School Closures Reduced the Reporting of Child Maltreatment. University of Tennessee Department of Economics; 2020. Accessed August 31, 2022. https://ideas.repec.org/p/ten/wpaper/2020-02.html

45.

Kim

YE.

Unemployment and child maltreatment during the COVID-19 pandemic in the Republic of Korea. Child Abuse Negl. 2022;130(pt 1):105474. doi:10.1016/j.chiabu.2021.105474

46.

Petrowski

Cappa

Pereira

Mason

Daban

RA.

Violence against children during COVID-19: assessing and understanding change in use of helplines. Child Abuse Negl. 2021;116(pt 2):104757. doi:10.1016/j.chiabu.2020.104757

47.

Rapoport

Reisert

Schoeman

Adesman

Reporting of child maltreatment during the SARS-CoV-2 pandemic in New York City from March to May 2020. Child Abuse Negl. 2021;116(pt 2):104719. doi:10.1016/j.chiabu.2020.104719

48.

Barboza

Schiamberg

Pachl

A spatiotemporal analysis of the impact of COVID-19 on child abuse and neglect in the city of Los Angeles, California. Child Abuse Negl. 2021;116(pt 2):104740. doi:10.1016/j.chiabu.2020.104740

49.

Baron

Goldstein

Wallace

CT.

Suffering in silence: how COVID-19 school closures inhibit the reporting of child maltreatment. J Public Econ. 2020;190:104258. doi:10.1016/j.jpubeco.2020.104258

50.

Rebbe

Lyons

Webster

Putnam-Hornstein

Domestic violence alleged in California child maltreatment reports during the COVID-19 pandemic. J Fam Violence. 2022;37(7):1041-1048. doi:10.1007/s10896-021-00344-8

51.

Brown

Orsi

Chen

PCB

Everson

Fluke

The impact of the COVID-19 pandemic on child protection system referrals and responses in Colorado, USA. Child Maltreat. 2022;27(1):3-11. doi:10.1177/10775595211012476

52.

Arthur

Schiro

Tumin

, et al. Shelter in place and an alarming increase in penetrating trauma in children and concerning decrease in child abuse. Am Surg. 2023;89(12):5386-5390. doi:10.1177/00031348221148361

53.

Bullinger

Boy

Feely

, et al. Home, but left alone: time at home and child abuse and neglect during COVID-19. J Fam Issues. 2023;44(2):338-362. doi:10.1177/0192513X211048474

54.

Park

Victor

Perron

Ryan

Braun

Persistent racial disproportionality in investigated and substantiated child maltreatment reports: trend analysis before and during the COVID-19 pandemic (2019-2020). J Public Child Welf. 2024;18(1):1-20. doi:10.1080/15548732.2022.2153960

55.

Abrams

Greenhawt

Shaker

Pinto

Sinha

Singer

The COVID-19 pandemic: adverse effects on the social determinants of health in children and families. Ann Allergy Asthma Immunol. 2022;128(1):19-25. doi:10.1016/j.anai.2021.10.022

56.

US Census Bureau. American Community Survey, ACS 5-year estimates detailed tables. 2020. Accessed July 30, 2021. https://data.census.gov/table?d=ACS%205-Year%20Estimates%20Detailed%20Tables

57.

Krieger

Chen

Waterman

Rehkopf

Subramanian

SV.

Race/ethnicity, gender, and monitoring socioeconomic gradients in health: a comparison of area-based socioeconomic measures—the Public Health Disparities Geocoding Project. Am J Public Health. 2003;93(10):1655-1671. doi:10.2105/ajph.93.10.1655

58.

Drake

Jolley

Lanier

Fluke

Barth

Jonson-Reid

Racial bias in child protection? A comparison of competing explanations using national data. Pediatrics. 2011;127(3):471-478. doi:10.1542/peds.2010-1710

59.

Cappa

Jijon

COVID-19 and violence against children: a review of early studies. Child Abuse Negl. 2021;116(pt 2):105053. doi:10.1016/j.chiabu.2021.105053

60.

Jonson-Reid

Drake

Cobetto

Ocampo

. Child abuse prevention month in the context of COVID-19. April 14, 2020. Accessed October 27, 2022. https://cicm.wustl.edu/child-abuse-prevention-month-in-the-context-of-covid-19

61.

Rodriguez

Lee

Ward

DF.

The perfect storm: hidden risk of child maltreatment during the COVID-19 pandemic. Child Maltreat. 2021;26(2):139-151. doi:10.1177/1077559520982066

62.

McCrary

Sanga

The impact of the coronavirus lockdown on domestic violence. Am Law Econ Rev. 2021;23(1):137-163. doi:10.1093/aler/ahab003

63.

Leslie

Wilson

Sheltering in place and domestic violence: evidence from calls for service during COVID-19. J Public Econ. 2020;189:104241. doi:10.1016/j.jpubeco.2020.104241

64.

Mohler

Bertozzi

Carter

, et al. Impact of social distancing during COVID-19 pandemic on crime in Los Angeles and Indianapolis. J Crim Justice. 2020;68:101692. doi:10.1016/j.jcrimjus.2020.101692

65.

Machlin

Gruhn

Miller

, et al. Predictors of family violence in North Carolina following initial COVID-19 stay-at-home orders. Child Abuse Negl. 2022;130(pt 1):105376. doi:10.1016/j.chiabu.2021.105376

66.

O’Neill

Dodington

Gawel

, et al. The effect of the COVID-19 pandemic on community violence in Connecticut. Am J Surg. 2023;225(4):775-780. doi:10.1016/j.amjsurg.2022.10.004

67.

National Institute on Drug Abuse. COVID-19 and substance use. 2022. Accessed October 24, 2022. https://nida.nih.gov/research-topics/comorbidity/covid-19-substance-use

68.

Bernal

Cummins

Gasparrini

Interrupted time series regression for the evaluation of public health interventions: a tutorial. Int J Epidemiol. 2017;46(1):348-355. doi:10.1093/ije/dyw098

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.10 MB