Aliza Ali¹, Audrey Nath^2*

¹University of the Incarnate Word School of Osteopathic Medicine

²University of Texas Medical Branch

*Corresponding Author: Audrey Nath, University of Texas Medical Branch, aunath@utmb.edu

Introduction
There are known racial disparities in the child protective services involvement with children in the United States [1]. Child protective services (CPS) contact is more likely for Black than white newborn babies [9]. Medical personnel are four times more likely to report Black and Hispanic children to CPS than other children [2]. The risk for CPS investigation Black children were highest in a study that analyzed CPS investigations in 20 most populous counties [3].

Children from poor neighborhoods are five times more likely to be reported to CPS than those from prosperous neighborhoods [2]; however, Black families experience CPS involvement even at higher income levels [12].

In addition to there being racial disparities in CPS involvement with families, there are also differences in rates of separation and child removal. A study of over 26 million child investigations in the National Child Abuse and Neglect Data Systems and Adoption and Foster Care Analysis and Reporting System showed that Black, Native and multiracial children are more likely to be separated from their families following a CPS investigation [5]. The current study aims to determine if the same trends exist in the state of Texas, specifically, if there are differences in the rates of removal of children by the Texas Department of Family and Protective Services (TDFPS) by racial category.

Methods
Institutional Review Board (IRB) exemption
As per the United States Department of Health and Human Services Title 45 Part 46 (Protection of Human Subjects, 2018), research involving a publicly-available dataset does not require IRB review, given that 1) the data is publicly-available and 2) the data is deidentified. Both of these criteria are fulfilled by the datasets used in this study.

Datasets
The data used in this study was a part of the publicly-available Texas Open Data Portal website (https://data.texas.gov/). A search query was performed for the category of TDFPS, resulting in numbers of de-identified children whose families were under TDFPS investigation and were removed from their home, by race/ethnicity. There were 5 categories for race/ethnicity: Black White, Asian, Hispanic and Native American. The fiscal years included in this analysis were 2013-2022.

The total numbers of Black, White, Asian/Pacific Islander, Hispanic and American Indian/Alaska Native children under the age of 19 years in Texas during this time frame were estimated based on publicly available 2021 population data [13].

Statistical Analysis
The percentage of the Texas child population that was removed by TDFPS, by each race/ethnic category, was calculated using Microsoft Excel (Microsoft Corporation, 2018, Microsoft Excel.

Retrieved from https://office.microsoft.com/excel).

The 5 percentages of Texas children removed by TDFPS (Black, White, Asian, Hispanic and Native American) were then compared using the k-proportions test, where the null hypothesis was that the proportion of each race/ethnicity’s Texas child population that was removed by TDFPS would be equal. Significance was set as P<0.05. The k-proportions test was performed with a Chi-squared approach (degrees of freedom = 3, alpha = 0.05) using XLSTAT (XLSTAT, 2007, Statistical Software for Excel. https://www.xlstat.com).

Results
Demographics
There were a total of 161,172 Texas children that were removed from their family home by TDFPS from 2013-2021. There was a total of 8,065,123 children in Texas in 2021. Both of these totals are broken down by race/ethnicity in Table 1.

Statistical Analysis
The percentage of Texas children in 2013-2021 who were removed from their family home by TDFPS out of the total number of children by race/ethnic group are as follows: 3.3% of Black children, 2.2% of White children, 0.2% of Asian children, 1.8% of Hispanic children and 0.7% of Native American children.

To determine if these percentages of removed children by race/ethnicity were significantly different from each other, a k- proportions test was performed with a Chi-squared approach to compare the 5 percentages. The observed chi-square value was 8190 (p<0.0001).

Table 1. Left-most data column: Number of children removed from home in Texas from 2013-2021 by TDFPS, by race/ethnicity and total.

Middle data column: Number of children (under age 19 years) in Texas in 2021, as an estimate for the number of children in Texas, by race/ethnicity and total.

Right-most data column: The percentage of children removed from home in Texas from 2013-2021 by TDFPS out of the total number of children in Texas by race/ethnicity.

Discussion
In this study of children in Texas removed from their family home by TDFPS from 2013-2021, the percentages of children removed from their homes was not uniform across racial/ethnic lines. Over 3% of the estimated total Black children in Texas were removed from their homes by TDFPS, while 2.2% of White children, 1.8% of Hispanic children, 0.7% of Native American children and 0.2% of Asian children were removed from their homes by TDFPS. This finding of a disparity in child removals by race/ethnicity is supported by studies done in other parts of the United States. A study that analyzed CPS decision making using data from National Child Abuse and Neglect Data Systems and Adoption and Foster Care Analysis and Reporting System found that Black, Native American, and multiracial children were more likely than non-Black Hispanic children to be placed out of home. Additionally, higher poverty rates were linked to increased incidence of unsubstantiated child protective services child abuse and neglect reporting against families [5]; these unsubstantiated investigations may also result in biased and unnecessary removals.

Our finding in this study is also in line with literature of greater state family services involvement with Black families. Black children were more than twice as likely as White children to enter into foster care before the age of 5 years [8].

One caveat with our analysis is that there were far fewer children removed from Asian and Native American homes (e.g. hundreds) than Black, White or Hispanic homes (e.g. tens of thousands) in Texas in this time period. As a result, while we were able to calculate a percentage of Asian and Native American children removed from the home, these proportions are under-powered and may not reflect true removal rates. Indeed, this issue of lower numbers of Asian and Native American patients in medical research studies has been seen across disciplines. Less than 1% of National Institutes of Health (NIH) funding is directed toward Asian health research [6]. A meta- analysis that included 83 studies with a total of 12,027 patients showed that only 1.27% identified as Asian while 0.12% identified as Native American [10]. Further research into these racial/ethnic groups is needed to better characterize trends within these groups. Additionally, there may be different rates of child protective services involvement within families of different Asian countries of origin, which are not reflected in statistics which put all Asian countries of origin and ethnicities into a single group [4].

Limitations and Future Directions
A major limitation with the analysis in this study is that the total number of children’s families under TDFPS investigation by race/ethnicity was not provided (e.g. the number of children who were removed and not removed by TDFPS). This information would have allowed for us to determine a percentage removed out of a total number of investigations, by race/ethnicity. Instead, we used total numbers of children of racial/ethnic categories in Texas as determined by the US Census as a denominator to determine percentages of the total racial/ethnic child population that was removed. Future research should allow us to determine if the bias is present in the investigation process and the removal process, or only the removal process.

References

1. Berkman E, Brown E, Scott M, Adiele A (2022) Racism in child welfare: Ethical considerations of harm. Bioethics. 36(3): 298- 304.
2. Cort NA, Cerulli C, He H (2010) Investigating health disparities and disproportionality in child maltreatment reporting: 2002- 2006. J Public Health Manag Pract. 16(4): 329-36.  
3. Edwards F, Wakefield S, Healy K, Wildeman C (2021) Contact with Child Protective Services is pervasive but unequally distributed by race and ethnicity in large US counties. Proc Natl Acad Sci U S A. 118(30): e2106272118.
4. Finno-Velasquez M, Palmer L, Prindle J, Tam CC, Putnam- Hornstein E (2017) A birth cohort study of Asian and Pacific Islander children reported for abuse or neglect by maternal nativity and ethnic origin. Child Abuse Negl. 72: 54-65.
5. Maguire-Jack K, Font SA, Dillard R (2020) Child protective services decision-making: The role of children's race and county factors. Am J Orthopsychiatry. 90(1): 48-62.
6. Obra JK, Lin B, Đoàn LN, Palaniappan L, Srinivasan M (2021) Achieving Equity in Asian American Healthcare: Critical Issues and Solutions. J Asian Health. 1(1): e202103.
7. Protection of Human Subjects, 45 C.F.R. §46 (2018).
8. Putnam-Hornstein E, Needell B, King B, Johnson-Motoyama M (2013) Racial and ethnic disparities: a population-based examination of risk factors for involvement with child protective services. Child Abuse Negl. 37(1): 33-46.
9. Roberts SC, Nuru-Jeter A (2012) Universal screening for alcohol and drug use and racial disparities in child protective services reporting. J Behav Health Serv Res. 39(1): 3-16.
10. Spielman DB, Liebowitz A, Kelebeyev S, Smith TL, McKinney K, et al. (2021) Race in Rhinology Clinical Trials: A Decade of Disparity. Laryngoscope. 131(8): 1722-1728.
11. Taylor VM, Ko LK, Hwang JH, Sin MK, Inadomi JM (2014) Gastric cancer in asian american populations: a neglected health disparity. Asian Pac J Cancer Prev. 15(24): 10565-71.
12. Thomas MMC, Waldfogel J, Williams OF (2023) Inequities in Child Protective Services Contact Between Black and White Children. Child Maltreat. 28(1): 42-54.
13. US Census Bureau. Population estimates based on bridged race categories released by the National Center for Health Statistics.