U.S. Department of Health and Human Services Office of Inspector General ACF Should Improve Oversight of Head Start To Better Protect Children's Safety Suzanne Murrin Deputy Inspector General for Evaluation and Inspections September 2022, OEI-BL-19-00560 U.S. Department of Health and Human Services Office of Inspector General Report in Brief September 2022, OEI-BL-19-00560 Why OIG Did This Review ACF Should Improve Oversight of Head Start The Office of Head Start (OHS)-part of the Department of Health and To Better Protect Children's Safety Human Services' Administration for Children and Families (ACF)- What OIG Found oversees Head Start grant recipients Approximately one in four Head Key Takeaway to ensure their compliance with Start grant recipients received an ACF is not aware of all incidents in program standards, including adverse finding from ACF for child standards addressing children's which children in the care of a abuse, lack of supervision, or Head Start center are abused, left safety. The Office of Inspector General (OIG) initiated this review to unauthorized release between unsupervised, or released to an determine the extent to which October 2015 and May 2020. These unauthorized person, impeding recipients received adverse findings adverse findings encompassed ACF's efforts to protect children's from ACF for violating program 1,029 individual incidents. safety. standards that prohibit child abuse, leaving a child unsupervised (lack of Additionally, Head Start grant supervision), or releasing a child to an recipients did not promptly self-report all incidents of child abuse, lack of unauthorized person (unauthorized supervision, and unauthorized release as required. Of recipients with one or release); assess ACF's oversight of more adverse findings from ACF in these three categories, 24 percent also how recipients identify, address, and received an adverse finding for failure to promptly report these incidents. prevent such incidents; and identify Further, using data from 2 States for 2017 through 2019, OIG identified 130 opportunities to better protect additional incidents that occurred in Head Start centers but of which ACF children. was not aware. ACF noted that recipients are not required to report How OIG Did This Review incidents in which the victim is not a child funded by Head Start, even if the We reviewed data from ACF incident occurs in a blended classroom that includes Head Start enrollees. monitoring of Head Start grant Head Start grant recipients most often addressed incidents of child abuse, recipients to identify adverse findings for child abuse, lack of supervision, lack of supervision, and unauthorized release at Head Start centers through unauthorized release, and failure to a combination of disciplinary action, administrative improvements, and report significant incidents. We also training. Additionally, ACF uses monitoring data to target recipients for reviewed narratives from ACF training and technical assistance on these issues. Some OHS regional offices monitoring reports to determine the have also developed data systems and procedures to better track and number and nature of incidents respond to incidents that threaten children's safety. associated with each adverse finding. Additionally, we obtained data from What OIG Recommends and How the Agency Responded child care licensing agencies and To ensure that ACF is aware of and can appropriately respond to all child protective services agencies in incidents in which children in the care of a Head Start center are abused, left Florida and Texas to determine whether ACF was aware of all unsupervised, or released to an unauthorized person, we recommend that incidents that those State agencies ACF (1) improve Head Start grant recipients' self-reporting of incidents of had identified. Finally, we child abuse, lack of supervision, and unauthorized release through better interviewed OHS staff and reviewed guidance and stronger consequences for failure to report; (2) extend the internal and public documents to reporting requirement to include incidents in blended classrooms in which better understand ACF's processes the victim is not a Head Start-funded child; (3) improve data-sharing with and criteria for monitoring, incident States about incidents in Head Start centers; and (4) disseminate information response, and adverse finding about innovative practices that OHS regional offices have developed to determinations related to child better identify and prevent incidents that threaten children's safety. ACF abuse, lack of supervision, and unauthorized release. concurred with all four of our recommendations. TABLE OF CONTENTS BACKGROUND................................................................................................................................................ 1 FINDINGS.........................................................................................................................................................8 Approximately one in four Head Start grant recipients received an adverse finding from ACF for child abuse, lack of supervision, or unauthorized release between October 2015 and May 2020 ...................... 8 Head Start grant recipients did not promptly report all incidents of child abuse, lack of supervision, and unauthorized release as required ............................................................................................................................11 Head Start grant recipients most often addressed incidents of child abuse, lack of supervision, or unauthorized release through a combination of disciplinary action, administrative improvements, and training........................................................................................................................................................................................15 ACF uses monitoring data to target training and technical assistance regarding children's safety in Head Start centers ..................................................................................................................................................................17 CONCLUSION AND RECOMMENDATIONS........................................................................................... 19 Improve Head Start grant recipients' self-reporting of incidents of child abuse, lack of supervision, and unauthorized release through better guidance and stronger consequences for failure to report 19 Extend the reporting requirement to include incidents of child abuse, lack of supervision, and unauthorized release in blended classrooms in which the victim is not a Head Start-funded child......20 Improve data-sharing with States about incidents of child abuse, lack of supervision, and unauthorized release in Head Start centers..................................................................................................................20 Disseminate information about innovative practices that OHS regional offices have developed to better identify and prevent incidents that threaten children's safety.................................................................21 AGENCY COMMENTS AND OIG RESPONSE ........................................................................................ 22 DETAILED METHODOLOGY...................................................................................................................... 23 APPENDICES................................................................................................................................................. 29 Appendix A: Head Start Program Standards Included in OIG's Review............................................................29 Appendix B: Agency Comments.......................................................................................................................................33 ACKNOWLEDGMENTS AND CONTACT ................................................................................................ 39 Acknowledgments ..................................................................................................................................................................39 Contact........................................................................................................................................................................................39 ABOUT THE OFFICE OF INSPECTOR GENERAL.................................................................................... 40 ENDNOTES .................................................................................................................................................... 41 BACKGROUND OBJECTIVES 1. To determine the extent to which Head Start grant recipients received adverse findings from ACF for violating program standards that prohibit child abuse, leaving a child unsupervised, and releasing a child to an unauthorized person 2. To assess ACF oversight of how Head Start grant recipients identify, address, and prevent incidents of child abuse, leaving a child unsupervised, and releasing a child to an unauthorized person Head Start grant recipients ("recipients") are funded by the Department of Health and Human Services (HHS) to provide early childhood education and social services to infants and young children. To ensure the well-being of children receiving Head Start services, all recipients must meet Federal requirements for quality and safety. The Office of Head Start (OHS) within the HHS Administration for Children and Families (ACF) monitors recipients to ensure their compliance with all program standards. OHS also provides training and technical assistance to support recipients' compliance. OIG initiated this review to assess ACF oversight of recipients to prevent, identify, and respond to incidents of child abuse, leaving a child unsupervised (lack of supervision), and releasing a child to an unauthorized person (unauthorized release). Head Start: Program Overview Head Start is a nationwide grant program to provide early childhood education and social services to children, birth to age five, whose families meet income or other eligibility requirements. With a fiscal year (FY) 2022 budget of $11 billion, ACF awards funds to approximately 1,600 recipients, including public agencies, for-profit organizations, nonprofit organizations, school systems, and Tribal governments. Head Start programs served approximately 861,000 children in FY 2021. 1 Recipients typically provide services in child care centers or school settings. Some recipients are very large, operating multiple facilities, while other recipients operate only one facility. Some recipients operate "blended classrooms," i.e., classrooms that include children funded through Head Start as well as children funded through other sources, such as State programs. Additionally, some recipients serve families through a home visiting and family child care model. In this report, we use the term "Head Start center" to refer to any location or program serving children enrolled in Head Start. In total, recipients operate approximately 20,000 individual Head Start centers. ACF Should Improve Oversight of Head Start To Better Protect Children's Safety OEI-BL-19-00560 Background | 1 Head Start Program Standards Addressing Child Safety Recipients must comply with Federal regulations containing detailed standards for Head Start program governance, operations, and financial and administrative requirements. 2 These standards also require that recipients establish and enforce a system of health and safety practices that ensure children are kept safe at all times. In this review, we focus on the following subset of standards that affect children's safety: • Standards that prohibit child abuse, including physical abuse; corporal punishment; other prohibited disciplinary practices such as binding or tying; emotional abuse; and verbal abuse;3 • Standards that prohibit leaving any child alone or unsupervised; 4 • Standards that prohibit release of a child to an unauthorized person. 5 The Head Start program standards also require recipients to report to ACF "immediately or as soon as practicable" any significant incident that affects Head Start program participants' health and safety, including any incident regarding staff or volunteer compliance with laws addressing child abuse and neglect. Recipients must also comply with State, Tribal, and local reporting laws regarding child abuse and neglect. 6 Please see Appendix A for a complete list of regulatory citations for all Head Start program standards included in our review. Oversight of Head Start Grant Recipients ACF oversees recipients to ensure that they meet all grant requirements. Many recipients operate multiple centers; however, ACF awards grant funding and provides oversight to grant recipients, not individual centers. Federal law requires that each recipient receive a full review immediately after completion of the recipient's first year and at least once during each 3-year period. 7 In addition to these Federal monitoring reviews, staff in 12 OHS regional offices maintain ongoing contact with recipients throughout the grant cycle to gather information, assess performance, and facilitate training and technical assistance. Federal Monitoring All recipients undergo multiple types of regularly scheduled Federal monitoring reviews. These reviews address operations across all of a recipient's program sites, but do not necessarily include an onsite visit to each individual Head Start center operated by the recipient. Recipients may also be subject to additional reviews initiated as needed to address performance issues brought to ACF's attention outside of the scheduled review cycle or to ensure that previously identified performance concerns have been corrected. Typically, some reviews are conducted through phone calls and document reviews, while others are conducted onsite. However, ACF ACF Should Improve Oversight of Head Start To Better Protect Children's Safety OEI-BL-19-00560 Background | 2 suspended all in-person reviews in March 2020 as a result of the COVID-19 pandemic. ACF resumed onsite reviews in January 2022. Adverse Findings ACF uses data collected during monitoring reviews to determine whether recipients comply with all Head Start program standards. Monitoring reviews may result in the following adverse findings issued to the recipient: • Deficiency findings indicate that a recipient exhibits systemic or substantial noncompliance with significant State or Federal requirements. 8 Examples include failure to adhere to program standards regarding threats to children's health or safety or the misuse of Head Start funds. • Noncompliance findings indicate that a recipient is out of compliance with a Federal or State requirement, but not to a level that constitutes a deficiency. 9 Correction of Adverse Findings After receiving an adverse finding, the recipient works with the appropriate OHS regional office to identify appropriate steps to correct the finding. Recipients with either a deficiency or noncompliance finding must correct these findings within the time period designated by ACF (typically 30 to 120 days, depending on the nature of the finding; however, ACF may require a shorter timeline, including immediate correction of a deficiency, under certain circumstances). If the finding is a deficiency, the recipient may also be required to submit to ACF a Quality Improvement Plan. 10 ACF conducts followup reviews to confirm that recipients have completed all actions necessary to correct adverse findings. If a noncompliance is not corrected in the allotted time, it is reclassified as a deficiency. 11 Designation Renewal System In 2011, ACF established the Designation Renewal System (DRS) to improve the quality of Head Start program services by increasing competition for grant funding. Recipients are typically funded for a 5-year project period. Under the DRS, at the conclusion of the project period, a recipient is non-competitively renewed for funding if it does not meet any of seven conditions specified in regulation. These conditions include measures that reflect poor classroom quality, problems with financial management, significant noncompliance with Head Start standards, and other factors that signal concerns about the recipient's performance. 12 A recipient that meets any of the seven conditions must participate in open competition for funding renewal, a process also known as "recompetition." When the DRS was initially implemented, one of the seven conditions that triggered recompetition was receipt of a single deficiency finding at any point in the project period. In late 2020, that condition was revised, and a recipient is now required to compete for renewal if it receives two or more deficiencies during the project period (or if it meets any of the remaining six conditions that trigger recompetition). ACF Should Improve Oversight of Head Start To Better Protect Children's Safety OEI-BL-19-00560 Background | 3 Suspension or Termination of Funding Regulations at 45 CFR § 1304 detail reasons for which ACF may suspend or terminate a recipient's funding, which include failure to timely correct one or more deficiency findings. 13 Typically, a recipient receives notice prior to suspension or termination; however, ACF may issue an emergency suspension without advance notice for certain reasons, including harm to participants' health and safety. Termination decisions are subject to appeal through the Departmental Appeals Board. Recipients may also submit written materials or request a meeting with ACF about why the recipient believes financial assistance should not be suspended. Training and Technical Assistance ACF provides training and technical assistance to recipients to support delivery of quality services and compliance with program standards. At the national level, OHS and the ACF Office of Child Care jointly operate four training and technical assistance centers, each with a specific focus (e.g., program management and fiscal operations, early childhood health and wellness, etc.). Additionally, OHS regional offices work with training and technical assistance contractors to provide direct assistance to individual recipients. State Oversight of Head Start Programs Most Head Start grant recipients are licensed child care providers in their respective States and are subject to State child care licensure standards as well as Head Start program standards. Some Head Start programs are therefore monitored by State- level agencies to ensure compliance with State standards. Additionally, State Child Protective Services (CPS) agencies receive and investigate reports of child abuse and neglect in child care settings, including Head Start centers. Head Start State Collaboration Offices As directed by the Head Start Act, ACF has established Head Start State Collaboration Offices. 14 The purpose of these offices is to facilitate ongoing collaboration between the Office of Head Start and State and local entities to support activities related to the delivery of Head Start services, such as building early childhood systems and improving access to comprehensive services for children from low-income families. Related OIG Work OIG regularly conducts audits of Head Start grant recipients and periodically conducts broader reviews of programmatic issues. In 2016, OIG recommended improvements to the DRS to more effectively identify lower-performing recipients that should participate in open competition for funding renewal. In 2011, OIG reviewed 24 recipients (selected in collaboration with ACF because the recipients were considered most at risk for noncompliance with health and safety standards) and found that none complied fully with health and safety requirements. ACF Should Improve Oversight of Head Start To Better Protect Children's Safety OEI-BL-19-00560 Background | 4 Methodology Data Sources This review combined data from the following sources: Federal Monitoring Data. We reviewed OHS data on recipients' monitoring review results from October 1, 2015, through May 11, 2020 (i.e., from the beginning of fiscal year 2016-when the current Head Start performance standards went into effect-to the start of this study's data collection). These data included identifying information for each recipient; the dates of each review; regulatory citations for each adverse finding; the status of each finding (e.g. active, corrected, etc.); and narrative fields detailing the reasons for the review results and corrective actions taken. Program Information Report (PIR) Data. We used publicly available PIR data for the 2018-2019 program year (the most recent available at the time of analysis) to obtain a list of recipients and identify their location, program type, and other characteristics. OHS Center List. We requested from OHS the names and addresses of each individual Head Start center operated by recipients. State Data. We requested the following data from the appropriate State agencies in Texas and Florida: • State CPS Data. 15 We requested all substantiated abuse and neglect reports for children in the care of a child care facility for calendar years 2017 through 2019. • State Child Care Monitoring Data. 16 We requested child care monitoring data for all licensed child care facilities for calendar years 2017 through 2019. We requested State data for calendar years 2017 through 2019 because this range provided a 3-year subset of the broader period of Federal monitoring data in our review. Interviews With Selected OHS Staff. We conducted a panel interview with OHS central office staff and five panel interviews with staff from selected OHS regional offices (Regions 1, 4, 6, 11, and 12). We also submitted questions in writing to OHS and received written responses. OHS Procedural and Training Documents. We requested from OHS documentation on written policies, procedures, and training materials related to identifying and addressing incidents of child abuse, lack of supervision, or unauthorized release; reporting requirements; and determining whether a violation will be classified as a deficiency or a noncompliance. OHS Administrative Action Records. We requested from OHS documentation about recipients that were subject to suspension, termination, and/or recompetition for renewal of funding under the DRS from October 1, 2015, through January 6, 2022. ACF Should Improve Oversight of Head Start To Better Protect Children's Safety OEI-BL-19-00560 Background | 5 We requested administrative actions for this date range to ensure that we captured all actions taken during the period of monitoring data we reviewed (October 1, 2015, through May 11, 2020) as well as actions taken more recently, given that the administrative processes involved may take months or years to complete. Data Analysis Analysis To Determine the Extent of Adverse Findings To determine the extent to which recipients received adverse findings from ACF for violating Head Start standards related to child abuse, lack of supervision, or unauthorized release, we analyzed Federal monitoring data to identify recipients that received noncompliance or deficiency findings for violating these standards, as well as findings for failure to report such incidents, from October 1, 2015, through May 11, 2020. (Please see Appendix A for a list of all Head Start program standards included in our review.) We also reviewed narratives from monitoring reports to determine the number and type of individual incidents associated with each adverse finding. We analyzed these data to determine the number and percentage of recipients with adverse findings for child abuse, lack of supervision, unauthorized release, and failure to report, as well as the total number of each type of incident within those categories. Analysis To Assess ACF Oversight To better understand ACF oversight processes and procedures, we analyzed interview and written responses from OHS regional and central office staff. We also reviewed policy, procedural, and guidance documents to identify processes and criteria for monitoring, incident response, and adverse finding determinations. To determine the extent to which all incidents are reported to ACF, we matched child care licensing and CPS data from Texas and Florida with Federal monitoring data. For each incident observed in State data, we determined whether the incident was also reflected in Federal monitoring data. We provided to OHS a list of incidents we identified that had not been previously reported to ACF. To determine which corrective actions recipients took, we analyzed OHS monitoring data to identify actions taken by recipients to resolve adverse findings for child abuse, lack of supervision, or unauthorized release. We then calculated the percentage of incidents addressed through different types of actions: disciplinary actions against staff, administrative improvements, and staff training. We also reviewed data on recompetition, suspension, and termination to determine the extent to which recipients were subject to these ACF administrative actions. Limitations The number of incidents of child abuse, lack of supervision, or unauthorized release that occurred in Head Start centers in Texas and Florida may be higher than the number OIG identified. OIG linked State and ACF data by geocoding child care center addresses. If a center that appeared in the State data had a matching address in the ACF Should Improve Oversight of Head Start To Better Protect Children's Safety OEI-BL-19-00560 Background | 6 ACF data, and was confirmed by our manual review of the center name and other information, we identified it as a Head Start center. This method could have failed to identify a Head Start center in the State data if, for example, the addresses were slightly different, outdated, or the State used post office box addresses. Additionally, we did not independently verify the evidence that ACF collected from recipients to confirm that they had completed all corrective actions timely. Standards We conducted this study in accordance with the Quality Standards for Inspection and Evaluation issued by the Council of the Inspectors General on Integrity and Efficiency. ACF Should Improve Oversight of Head Start To Better Protect Children's Safety OEI-BL-19-00560 Background | 7 FINDINGS Approximately one in four Head Start grant recipients received an adverse finding from ACF for child abuse, lack of supervision, or unauthorized release between October 2015 and May 2020 From our review of Federal monitoring data, we found that 438 out of 1,611 recipients received one or more adverse findings (noncompliance or deficiency) from ACF for child abuse, lack of supervision, or unauthorized release between October 2015 and May 2020. ACF issues an adverse finding when it determines that a recipient has violated one or more Head Start program standards. One adverse finding may be associated with multiple incidents, which may have occurred at one or more centers that the recipient operates. For example, an adverse finding for lack of supervision could be based on a monitoring report describing children left unsupervised on multiple dates. 17 Our review of monitoring report narratives associated with these adverse findings identified 1,029 individual incidents. Exhibit 1: During the period of review, 27 percent of Head Start grant recipients received an adverse finding for child abuse, lack of supervision, or unauthorized release. Source: OIG analysis of Federal Head Start monitoring data, 2022 Among Head Start grant recipients with one or more adverse findings in these three categories, the most common citation was for lack of supervision Among recipients with one or more adverse findings for child abuse, lack of supervision, or unauthorized release, the most common citation was for lack of supervision. Six percent of recipients received adverse findings from ACF in more than one of these three categories. ACF Should Improve Oversight of Head Start To Better Protect Children's Safety OEI-BL-19-00560 Findings | 8 Nineteen percent of Head Start grant recipients were cited for lack of supervision Nearly one in five recipients received an adverse finding for lack of supervision between October 2015 and May 2020. Our review of monitoring report narratives associated with these findings identified 533 individual incidents in which a child was left unsupervised. Exhibit 2: Examples of incidents involving lack of supervision "[A] 4-year-old child left the [school] building and was unsupervised for approximately 5 to 10 minutes, during which time the child ran into the street." "[The] contracted bus driver left the child unattended on the bus for an undetermined period of time during very cold weather. The parent discovered discoloration on the child's feet later that day. An examination by a physician the following day indicated evidence of frostbite." Source: OIG analysis of Federal Head Start monitoring reports, 2022 The most frequent type of supervision incident, with 358 incidents, involved children left alone on the grounds of a Head Start facility, such as in a classroom, in a bathroom, or on a playground. We also identified 95 incidents of children left alone outside a facility, such as in a parking lot or on a nearby street. Finally, we identified 80 incidents of children left alone on a bus. Twelve percent of Head Start grant recipients were cited for child abuse Between October 2015 and May 2020, 12 percent of recipients received an adverse finding for child abuse. Our review of monitoring report narratives associated with these findings identified 454 separate incidents (i.e., unique child-date combinations) of abuse. A single incident of child abuse sometimes included multiple forms of abuse (for example, both physical and verbal abuse). Exhibit 3: Examples of incidents involving child abuse "[T]he Head Start Director… observed the teacher dragging the child across the floor on three separate occasions and placing the child's cot in the office and turning off the light. When the child got off his cot, the teacher roughly pushed him into one of the children's cubbies and stood in front of the cubby to prevent him from getting out." "[P]arents came forward to report concerns of the teacher and teacher assistant using verbal abuse by calling the children demeaning nicknames, including chancho-pig, in Spanish-and mustache girl.... [C]hildren described a stick that was used to scare and hit them and the children expressed fear of the staff returning." Source: OIG analysis of Federal Head Start monitoring reports, 2022 ACF Should Improve Oversight of Head Start To Better Protect Children's Safety OEI-BL-19-00560 Findings | 9 Of the 454 incidents of child abuse described in Federal monitoring reports: • 374 incidents included physical abuse or corporal punishment, defined as hitting, spanking, shaking, slapping, twisting, pulling, squeezing, or biting a child; • 102 incidents included emotional or verbal abuse, defined as any form of public or private humiliation; rejecting, terrorizing, extended ignoring, or corrupting a child; or using profane, sarcastic, threatening, or derogatory language or remarks about a child or a child's family; • 54 incidents included prohibited disciplinary practices, defined as using isolation; binding or tying; taping a child's mouth; using or withholding food as a punishment or reward; using toilet learning or training methods that punish, demean, or humiliate a child; or using physical activity or outdoor time as a punishment or reward; and • 8 incidents included sexual abuse. Two percent of Head Start grant recipients were cited for unauthorized release Few recipients had adverse findings for unauthorized release. We identified 42 incidents between October 2015 and May 2020 in which a child was released to an unauthorized person. Of these, 25 incidents involved children released from a bus to an unauthorized person and 17 involved children released from a Head Start facility to an unauthorized person. Exhibit 4: Examples of incidents involving unauthorized release "A Head Start child was released to an unauthorized adult… and the location of the child was undetermined for approximately 1.5 hours." "[A] 4-year-old child.… was dropped off at the wrong bus stop and not released to an authorized adult. The child was then left unsupervised for approximately 7 minutes." Source: OIG analysis of Federal Head Start monitoring reports, 2022. Most adverse findings for child abuse, lack of supervision, or unauthorized release were classified as deficiencies, which carry greater consequences than findings of noncompliance Of all adverse findings for child abuse, lack of supervision, and unauthorized release during the period of OIG's review, 65 percent were cited as deficiencies and 35 percent were cited as noncompliances. ACF provided internal documents that describe its criteria for determining whether these types of citations will be classified as a deficiency or a noncompliance. These criteria include the severity of an abuse incident; the number of minutes a child was left alone; whether an unauthorized release involved a known or unknown individual; the number of incidents involved; and other factors. Although both types of adverse findings must be corrected, ACF Should Improve Oversight of Head Start To Better Protect Children's Safety OEI-BL-19-00560 Findings | 10 recipients who receive a deficiency finding may be required to complete a Quality Improvement Plan and will be terminated if they fail to correct the deficiency. Additionally, unlike a noncompliance, a deficiency finding may contribute to a recipient being required to recompete for funding renewal. If a noncompliance is not corrected in the allotted time, it is reclassified as a deficiency. 18 During the period of review, most Head Start grant recipients did not have repeated adverse findings for child abuse, lack of supervision, or unauthorized release Between October 2015 and May 2020, 110 recipients received adverse findings for child abuse, lack of supervision, or unauthorized release in 2 or more fiscal years. This represents 25 percent of recipients that were ever cited for violating these Head Start standards during the period of review and 7 percent of all recipients. In addition, 18 recipients were cited for any of these 3 categories of violations during 3 or more fiscal years during the period of review. Head Start grant recipients did not promptly report all incidents of child abuse, lack of supervision, and unauthorized release as required Recipients are required to report to ACF "immediately or as soon as practicable…any significant incidents affecting the health and safety of program participants." 19 Additionally, all staff at Head Start centers are considered mandated reporters, i.e., they are legally obligated to report suspected child abuse or neglect to the appropriate State agency. 20 We found that despite these requirements, recipients did not promptly report all incidents of child abuse, lack of supervision, and unauthorized release. Of Head Start grant recipients with one or more adverse findings for child abuse, lack of supervision, or unauthorized release, 24 percent also received an adverse finding for failure to promptly report these incidents We found that nearly one quarter of recipients that were cited for child abuse, lack of supervision, or unauthorized release were also cited for failure to properly report these incidents. Recipients are in violation of Head Start program standards if they do not report an incident as soon as practicable or if they fail to report the incident at all. From our review of monitoring report narratives associated with these adverse findings, we identified 202 individual incidents that were not promptly reported to the proper authorities as required. Of these, 78 percent had not been appropriately reported to ACF, and 22 percent had not been appropriately reported to State or local authorities. Many of these incidents violated the Head Start standards because the recipient did not report the incident "as soon as practicable," which OHS central office ACF Should Improve Oversight of Head Start To Better Protect Children's Safety OEI-BL-19-00560 Findings | 11 staff explained is generally interpreted as 3 to 5 days. However, in nearly one-third of cases, the recipient failed to report the incident at all. Exhibit 5: Examples of incidents of failure to report "[A]n incident occurred in which a child was injured and required stitches due to a lack of proper supervision in the bathroom. The Regional Office was not notified of the incident by the grantee but was informed of the incident by an anonymous complaint." "A parent reported allegations of a teacher restraining a 4-year-old child… The Head Start Director was informed of the allegation, but expressed that she was unaware of the need to report the incident since the child was not injured." Source: OIG analysis of Federal Head Start monitoring reports, 2022. ACF staff explained that they learn of most incidents of child abuse, lack of supervision, and unauthorized release through recipient self-reporting. However, incidents are also reported through other sources, enabling ACF to identify recipients' failure to self-report. For example, ACF operates a hotline that parents and community members can use to alert ACF to possible child abuse or other safety concerns. ACF staff stated that they also learn of nonreported incidents during monitoring reviews and through media reports regarding allegations of abuse. In reviewing narratives associated with adverse findings for reporting failures, we found that some incidents were discovered when ACF became aware of State child care licensing violations or CPS reports involving Head Start recipients, although the ACF staff we interviewed did not characterize this as a routine source of information. OIG identified 130 additional incidents of child abuse, lack of supervision, and unauthorized release in Head Start centers in 2 States that were never reported to ACF Federal monitoring data reflect only the incidents of which ACF is aware. To determine whether additional incidents occurred that were never reported to ACF, OIG reviewed data from Texas and Florida State agencies that oversee child care settings, including Head Start centers. Specifically, we obtained data on child care licensing violations and substantiated reports of abuse and neglect from these two States' child care licensing and CPS agencies, respectively. We then reviewed these State records to identify incidents of child abuse, lack of supervision, or unauthorized release that occurred in Head Start centers from January 1, 2017, through December 31, 2019, and then reviewed Federal monitoring data to determine whether these incidents were also present in the Federal data. For any incident that did not appear in the Federal data, we consulted ACF to confirm whether ACF was aware of the incident. From our analysis, at least 130 incidents of child abuse, lack of supervision, or unauthorized release occurred in Head Start centers in Florida and Texas that were never reported to ACF. In Florida, we identified 41 incidents; of these, 16 involved child abuse, 23 involved lack of supervision, and 2 involved unauthorized release. In ACF Should Improve Oversight of Head Start To Better Protect Children's Safety OEI-BL-19-00560 Findings | 12 Texas, we identified 89 incidents; of these, 82 involved child abuse, 6 involved lack of supervision, 21 and 1 involved unauthorized release. OIG's figures represent the minimum number of incidents in these two States of which ACF was unaware; the true number may be higher. Specifically, because of challenges identifying Head Start centers in the State records through address matching, and our exclusion of all Texas licensing violations involving supervision, 22 it is likely that we did not identify every incident in a Head Start center in these two States. Without knowledge of these incidents, ACF may have been limited in its ability to ensure the safety and quality of care that children received in Head Start centers. The incidents we identified in State data that recipients had not reported to ACF included physical and sexual abuse; severe and humiliating punishments; and injury resulting from lack of supervision. Furthermore, we identified 22 recipients who received multiple State citations or reports but for which the Federal monitoring data had no record of any related performance issues. Head Start grant recipients are not required to report to ACF incidents in which the victim is not a child funded by Head Start OIG shared with ACF information about the incidents we identified, and ACF reported that it is determining the appropriate followup actions. ACF staff also noted that some Head Start centers operate blended classrooms, which include children funded through Head Start as well as children funded through other sources. ACF staff explained that recipients are not required to report to ACF significant health and safety incidents that affect only children who are not funded by Head Start. This is true even if the incident occurs in a blended classroom that also includes Head Start enrollees. 23 ACF does not routinely communicate with State agencies about incidents in Head Start centers involving child abuse, lack of supervision, or unauthorized release OHS central office staff reported that ACF does not routinely communicate with State child care licensing or CPS agencies about incidents of child abuse, lack of supervision, and unauthorized release in Head Start centers. Four of the five OHS regional offices that OIG interviewed similarly reported no routine communication with these State agencies. One OHS regional office reported more frequent communication, noting that its office obtains information about maltreatment and neglect incidents directly from State agencies, when necessary, by working with its Head Start State Collaboration Office Director. ACF Should Improve Oversight of Head Start To Better Protect Children's Safety OEI-BL-19-00560 Findings | 13 Head Start grant recipients' failure to report child abuse, lack of supervision, and unauthorized release may reflect confusion about reporting requirements as well as weaker consequences for violating these requirements Head Start grant recipients may not fully understand the reporting requirements The Head Start standards require that recipients report to ACF "immediately or as soon as practicable" any significant incidents affecting Head Start participants' "health and safety." ACF central office staff explained to OIG that generally, this requirement is interpreted to mean that an incident should be reported within 3 to 5 days of its occurrence. In 2015, ACF issued written guidance to recipients stating that incidents of unsupervised children should be reported to the regional office within 3 days of the incident. 24 However, ACF has not issued written guidance to recipients to similarly clarify the reporting timeframe for incidents in which a child was abused or released to an unauthorized person. Similarly, ACF has not issued written guidance to recipients providing details or examples of the types of incidents they should report. 25 In some cases, recipients with adverse findings for failure to report told ACF that they had misunderstood the requirements, including confusion about the reporting timeframe and the types of incidents that must be reported. Additionally, some monitoring narratives described recipients that reported incidents to appropriate State or local authorities but failed to report the same incidents to ACF. Other narratives noted that the recipient attributed its failure to report to the fact that it had already taken steps internally to address the incident. Exhibit 6: Examples of explanations that recipients provided to ACF to explain their failure to report incidents as required Failure To Report a Child Left Unsupervised: "The grantee informed the Program Specialist it was waiting to collect all supporting documents before reporting." Failure To Report Child Abuse: "[T]he Head Start Director reported she was not aware that she needed to make a report to the Office of Head Start." Failure To Report Unauthorized Release of a Child: "The grantee stated they did not believe this was an incident that required reporting." Source: OIG analysis of Federal Head Start monitoring reports, 2022. Consequences for failing to report an incident are usually less severe than the consequences for the incident itself Of all adverse findings for failure to report during the period of OIG's review, 87 percent were classified as noncompliances and 13 percent were classified as deficiencies. Failure-to-report violations were often cited as a noncompliance even when the originating incident was itself cited as a deficiency. ACF informed OIG that ACF Should Improve Oversight of Head Start To Better Protect Children's Safety OEI-BL-19-00560 Findings | 14 it does not have written criteria for classifying a failure-to-report finding as a deficiency or a noncompliance. ACF also stated that failure to report is typically classified as a noncompliance but that such citations may be elevated to a deficiency on a case-by-case basis, based on the severity of the incident or length of time it took the recipient to report the incident. Until late 2020, under the DRS, a single deficiency finding within a recipient's 5-year project period would force the recipient to recompete for grant renewal. ACF staff suggested that the DRS may discourage recipients from reporting, because the recompetition process is onerous and could result in the recipient failing to be renewed. Staff noted that in late 2020, the DRS criteria were altered such that two deficiencies (rather than just one) trigger recompetition and expressed that this change might improve reporting. However, under the current DRS criteria, and given typically lesser consequences for failure to report, recipients still have an incentive to withhold information about significant incidents. Recompetition for funding renewal is triggered by two deficiencies. If a recipient does not report an incident of (for example) leaving a child unsupervised, it may never come to ACF's attention, and the recipient will avoid the possibility of receiving a deficiency for that incident. On the other hand, if ACF learns of the incident through other means, the recipient may receive a deficiency finding for the supervision incident-but will likely receive only a noncompliance finding for failure to report. In this scenario, the recipient is not significantly worse off than if it had properly reported the incident to begin with, because the additional noncompliance finding for failure to report does not trigger recompetition. Head Start grant recipients most often addressed incidents of child abuse, lack of supervision, or unauthorized release through a combination of disciplinary action, administrative improvements, and training Following an adverse finding for child abuse, lack of supervision, or unauthorized release, the recipient works with the appropriate OHS regional office to determine the necessary steps to correct the finding. This may include developing a corrective action plan or quality improvement plan, depending on the finding. 26 ACF staff reported to OIG that corrective actions should not be limited to staff termination but should also address the root cause of the incident to reduce the likelihood of similar incidents in the future. For 81 percent of all incidents of child abuse, lack of supervision, or unauthorized release in our review, recipients took disciplinary action against the staff member(s) involved, including written warnings, administrative leave, suspension, or terminating employment. Staff were terminated or resigned in response to 75 percent of incidents of child abuse, 41 percent of incidents of lack of supervision, and 52 percent of incidents of unauthorized release. ACF staff noted that recipients are responsible ACF Should Improve Oversight of Head Start To Better Protect Children's Safety OEI-BL-19-00560 Findings | 15 for developing their own human resources and standards of conduct policies, including determining when termination or other personnel actions are appropriate. Recipients also commonly implemented administrative improvements and staff training as part of their corrective actions to address incidents of child abuse, lack of supervision, or unauthorized release. For example, recipients reported revising their operating procedures by creating clearer and more prescriptive supervision and standards of conduct policies. Recipients also responded to incidents by increasing classroom monitoring through in-person visits or video cameras, as well as by making facility improvements, such as installing alarms and physical barriers to prevent children from leaving the facility without supervision. Finally, recipients reported providing staff training on current and new policies, including instituting mandatory trainings. Exhibit 7: Examples of corrective actions that Head Start grant recipients took to address incidents of child abuse and lack of supervision Incident: Child Abuse Incident: Leaving a Child Unsupervised A child was lying on the floor while the A child exited the center without shoes. teacher dragged him by one arm down The child walked through the parking lot the hallway. and onto the sidewalk where he was picked up and returned to the center by a passerby in a car. The child had a history of leaving unsupervised. Removed teacher from service on day of Provided intensive guidance and retraining for incident, terminated 1 day later teachers involved Conducted announced and unannounced visits Updated the Child Health History form to by supervisors, education staff, and mental include more information on each child's health providers behaviors Installed video cameras Installed alarms on classroom doors Implemented new child supports such as Seated Family Advocates at main doors social-emotional skill building activities, fidget Installed whiteboards in each classroom to be manipulatives, soft seats to help strengthen updated when each child arrives/departs core and focus, and motion bottles for self- soothing Directed center management to monitor classrooms monthly, conduct teacher-coach Implemented new staff supports such as classroom observations twice monthly, and coaching on child communication strategies conduct unannounced classroom observations Held training for staff on standards of conduct, twice annually; arranged for State licensing child abuse and neglect, Right Response consultant to make unannounced site visits Toolkit, Child-Adult Relationship Enhancement, Created Active Supervision and Accountability mental health, and dealing with challenging checklists used by staff behaviors Created training video library for staff Source: OIG analysis of Federal Head Start monitoring data, 2022. ACF Should Improve Oversight of Head Start To Better Protect Children's Safety OEI-BL-19-00560 Findings | 16 ACF conducts followup reviews to ensure that recipients complete their corrective actions within the specified timeframe. In the case of an adverse finding for child abuse, lack of supervision, or unauthorized release, a recipient typically must correct the finding within 30 to 120 days. 27 ACF reported that its followup reviews include reviewing documentation of corrective actions (e.g., policies and procedures, training logs); when necessary, ACF may also conduct site visits and interviews. From our review of monitoring data, OIG did not identify any adverse findings for child abuse, lack of supervision, or unauthorized release that recipients had failed to correct through ACF's corrective action process. 28 ACF requires a recipient to recompete for funding renewal if it receives two or more deficiency findings, even if the deficiencies are corrected. Recompetition is a demanding process that requires the recipient to demonstrate that it remains the best-qualified entity to provide Head Start services. From October 1, 2015, through January 6, 2022, ACF required 459 recipients to recompete for funding renewal. Of the 302 recipients with at least one deficiency finding for child abuse, lack of supervision, or unauthorized release during the period of OIG's review, 68 percent had recompeted for funding renewal as of January 6, 2022. In contrast, ACF has rarely used its suspension and termination authorities. From October 1, 2015, through January 6, 2022, ACF suspended six recipients and terminated seven recipients for any reason. 29 ACF uses monitoring data to target training and technical assistance regarding children's safety in Head Start centers ACF provides recipients with training and technical assistance to support compliance with Head Start standards. Both OHS central office and regional office staff reported using monitoring data to identify training needs, including training about child abuse, supervision, and safe release of children. For example, staff from one OHS regional office explained that, based on the health and safety incidents they observed among recipients in their region, the office had developed a training program involving monthly check-ins with recipients to discuss challenges related to children's safety and identify systemic improvements to better protect children. Another OHS regional office described initiating Statewide trainings on child maltreatment after noticing an uptick in incidents. ACF also uses monitoring data to prioritize technical assistance for recipients with adverse findings. Some OHS regional offices have developed data systems and procedures to better track and respond to incidents of child abuse, lack of supervision, and unauthorized release ACF does not centrally track individual incidents of child abuse, lack of supervision, or unauthorized release. Instead, ACF tracks violations of Head Start standards at the ACF Should Improve Oversight of Head Start To Better Protect Children's Safety OEI-BL-19-00560 Findings | 17 finding level: during a review, a recipient receives a single adverse finding for each standard it is found to have violated. A single adverse finding may be based on one or many individual incidents related to the same standard. Information about individual incidents associated with an adverse finding is included in the monitoring report narrative. In reviewing narratives associated with adverse findings for child abuse, lack of supervision, and unauthorized release, we found that most findings were based on a single incident. Overall, the number of incidents involved in a single adverse finding ranged from 1 to 21 incidents. We also noted variation in narratives' format and level of detail. Exhibit 8: Example of a Head Start monitoring report narrative describing multiple incidents "[T]here were four maltreatment incidents between June 2018 and October 2018… Incident One: On June 12, 2018, a teacher grabbed a 21-month-old child by the arm and shoved him against the wall, injuring the toddler. Incident Two: On June 12, 2018, a teacher grabbed a 23-month-old child by the arm while she was cleaning up after painting. There was water on the sink and floor, causing the child to slip when she was grabbed by the teacher. The child hit her head on the sink, causing an injury. Incident Three: On June 28, 2018, a teacher slapped a 21-month-old child in the face after repeatedly asking the child to stop hitting others… Incident Four: On October 2, 2018, a student teacher observed a teacher firmly grab a child's wrist… and told him to wash his hands. When the child was washing his hands and appeared to be playing in the water, the teacher grabbed his arm, swung him around, and swatted him on the behind. The child began crying." Source: OIG analysis of Federal Head Start monitoring data, 2022. Staff from three OHS regional offices indicated that the current monitoring data meets their oversight needs without tracking individual incidents. However, two of the five OHS regional offices we interviewed reported that they had developed supplementary internal data systems to track individual incidents within their region. Staff reported that these supplementary tracking systems helped them examine trends in the nature and number of significant incidents and develop actions to support recipients' efforts to ensure children's safety. Some OHS regional offices also reported that they had created additional procedures to better address these types of incidents within their region. For example, one regional office established a dedicated team that meets weekly to discuss significant incidents. Another regional office developed a standardized list of questions to gather information about incidents; staff reported that this tool aided them in collecting consistent data about incidents that threaten children's safety. ACF Should Improve Oversight of Head Start To Better Protect Children's Safety OEI-BL-19-00560 Findings | 18 CONCLUSION AND RECOMMENDATIONS All Head Start grant recipients are required to adhere to Head Start standards, which include standards that prohibit child abuse, leaving a child unsupervised, and releasing a child to an unauthorized person. To identify violations of these standards, ACF relies primarily on recipients' self-reporting of significant incidents. However, OIG found that recipients do not always report these incidents to ACF as required. In some cases, ACF only learns that a child in a Head Start center was abused, left unsupervised, or released to an unauthorized person long after the incident occurs. In other cases, ACF never learns of an incident at all. Our review identified at least 130 incidents of child abuse, lack of supervision, and unauthorized release over 3 years in just 2 States that had never been reported to ACF. Accurate information about incidents that threaten children's safety is critical for effective oversight. ACF reported that it uses monitoring data to identify training needs and target technical assistance to Head Start grant recipients to better prevent child abuse, lack of supervision, and unauthorized release. ACF also reported that when it does learn of an incident of this nature, it works closely with the recipient to identify and carry out a comprehensive set of corrective actions intended to address the root cause of the incident. However, given OIG's findings, ACF's ability to effectively implement these approaches is limited by its lack of complete and timely information. To better protect children's safety in Head Start centers, we recommend that ACF: Improve Head Start grant recipients' self-reporting of incidents of child abuse, lack of supervision, and unauthorized release through better guidance and stronger consequences for failure to report ACF should provide recipients with more detailed guidance about the requirement to report significant incidents that threaten children's health and safety, which includes incidents of child abuse, lack of supervision, and unauthorized release. In particular, ACF should more clearly specify the reporting timeframe and provide examples of the types of incidents that must be reported. ACF should then provide additional training and technical assistance to ensure that recipients clearly understand the requirement. Head Start programs may also benefit from regularly recurring training on these important requirements. ACF should also impose stronger consequences for failure to report. Currently, citations for failure to report are typically classified as a noncompliance rather than as a deficiency. ACF should establish specific criteria for circumstances under which it ACF Should Improve Oversight of Head Start To Better Protect Children's Safety OEI-BL-19-00560 Conclusion and Recommendations | 19 will classify failure to report as a deficiency, just as it has established similar criteria for citations involving child abuse, lack of supervision, and unauthorized release. For example, ACF should consider classifying failure to report as a deficiency when multiple incidents were not reported or when the unreported incident itself results in a deficiency finding. Given that a recipient must recompete for funding if it receives two deficiencies, this change could encourage recipients to report incidents rather than risk an additional deficiency finding by withholding information that ACF could later discover. Extend the reporting requirement to include incidents of child abuse, lack of supervision, and unauthorized release in blended classrooms in which the victim is not a Head Start-funded child Some recipients operate blended classrooms that serve children funded by Head Start together with children funded through other sources. Currently, a recipient is not required to report an otherwise reportable incident if the victim is not a Head Start- funded child-even though an incident of child abuse, lack of supervision, or unauthorized release clearly reflects a danger to Head Start-funded children in the same blended classroom. ACF's oversight approach relies in part on information about the number and nature of incidents that have occurred at Head Start centers operated by each recipient. Without complete information about all incidents, ACF may erroneously believe that a problem is more limited in scope than is in fact the case-or may be unaware that a safety problem exists at all. To protect children's safety, ACF should take the necessary steps (such as issuing new program guidance or revising regulations, depending on its assessment of its authorities) to seek full transparency regarding incidents of child abuse, lack of supervision, and unauthorized release that occur in any Head Start center, regardless of whether the individual victim is a Head Start enrollee. This additional reporting would better inform ACF's determinations about whether and what training, technical assistance, or other actions are necessary to prevent future incidents and ensure the recipient's compliance with these critical Head Start standards. Improve data-sharing with States about incidents of child abuse, lack of supervision, and unauthorized release in Head Start centers OIG was able to identify incidents of child abuse, lack of supervision, and unauthorized release of which ACF was not aware by reviewing data from child care licensing and CPS agencies in two States. This suggests that State data could be a useful source of information to help ACF identify incidents that recipients fail to self- report. As noted above, such information is critical to support effective ACF oversight of recipients. ACF Should Improve Oversight of Head Start To Better Protect Children's Safety OEI-BL-19-00560 Conclusion and Recommendations | 20 To improve its access to information about children's safety, ACF should develop data-sharing agreements with States regarding incidents of child abuse, lack of supervision, and unauthorized release at Head Start centers. ACF should consider leveraging the Head Start State Collaboration Offices for this purpose. OIG notes that information-sharing need not include child victims' personally identifying information; the Head Start center location, date, and general nature of the incident is sufficient to enable ACF to follow up with the recipient. Additionally, many States make child care licensing reports publicly available online; ACF should explore methods to systematically retrieve and search these materials as they pertain to Head Start centers. Disseminate information about innovative practices that OHS regional offices have developed to better identify and prevent incidents that threaten children's safety We found that some OHS regional offices have implemented innovative methods to track and respond to incidents of child abuse, lack of supervision, and unauthorized release in Head Start centers in their regions. For example, some OHS regional offices reported practices such as creating a supplemental data system to better track incidents at the child level and establishing a dedicated response team that meets routinely to discuss incidents in Head Start centers within the region. ACF should foster information-sharing across OHS regional offices so that these and other innovative practices can be shared and exchanged. ACF Should Improve Oversight of Head Start To Better Protect Children's Safety OEI-BL-19-00560 Conclusion and Recommendations | 21 AGENCY COMMENTS AND OIG RESPONSE ACF concurred with all four of our recommendations. ACF stated that it will take both immediate and long-term steps to improve child safety in the Head Start program. ACF provided examples of how it will improve oversight and monitoring of Head Start grant recipients to better address incidents of child safety. Regarding our first recommendation, ACF described steps to improve recipients' reporting of incidents of child abuse, lack of supervision, and unauthorized release. For example, ACF stated that in August 2022, the agency provided recipients with letters of guidance and webinars about Head Start reporting requirements. In addition, ACF stated that it will provide ongoing training and technical assistance resources for these requirements. Regarding our second recommendation, ACF agreed that its current protocol leaves OHS unaware of important safety information. ACF stated that it does not currently have authority to fully implement OIG's recommendation but will explore options to extend its reporting requirements. ACF also stated that while exploring those options, it will take immediate steps to increase training and guidance to clarify reporting requirements as they apply to blended classrooms, such as providing guidance that the reporting requirement applies to incidents involving Head Start staff, contractors, and volunteers in all Head Start and Early Head Start settings, including blended classrooms. Regarding our third recommendation, ACF stated that it will take a tiered approach to improving data-sharing with States. As a first step, ACF will examine publicly available State child care licensing reports in the 2022–2023 program year. ACF will also explore additional ways to get information from States about child safety incidents, including creating data-sharing plans with States. Regarding our fourth recommendation, ACF described steps to disseminate information about innovative child safety practices across Head Start regional offices. For example, ACF stated that program managers from the regional offices currently meet on a weekly basis to collaborate and share information. ACF also stated that it will consider more formal structures for sharing innovations across regional offices, such as analyzing trends and technical assistance activities. OIG appreciates ACF's commitment to protect children from incidents of child abuse, lack of supervision, and unauthorized release while in the care of a Head Start center. For the full text of ACF's comments, see Appendix B. ACF Should Improve Oversight of Head Start To Better Protect Children's Safety OEI-BL-00-00560 Agency Comments and OIG Response | 22 DETAILED METHODOLOGY To determine the extent to which Head Start grant recipients received adverse findings from ACF for violating program standards that prohibit child abuse, leaving a child unsupervised, and releasing a child to an unauthorized person, and to assess ACF oversight of recipients' compliance with these standards, we used mixed methods to analyze Federal monitoring data, State monitoring data, Federal administrative action data, interviews, written responses, and documents. Data Sources This review combined data from the following sources: Federal Monitoring Data. We reviewed OHS data on recipients' monitoring review results from October 1, 2015, through May 11, 2020 (i.e., from the beginning of fiscal year 2016-the year the current Head Start performance standards went into effect- to the start of this study's data collection). These data included identifying information for each recipient (e.g., name, address, etc.); the dates of each review; regulatory citations for each adverse finding identified during the review; the level of adverse finding (deficiency or noncompliance); the status of each finding (new, active, corrected, or not corrected); and narrative fields detailing the incidents and reasons for the review results and corrective actions taken by the recipient. Program Information Report (PIR) Data. Recipients report PIR data annually. We used PIR data for the 2018-2019 program year (the most recent available at the time of analysis) to obtain a list of recipients and each recipient's location, program type, number of children the recipient is funded to serve, and ACF regional affiliation. We identified 6 recipients with adverse findings during the period of review who were not on the 2018-2019 PIR recipient list; we removed these 6 recipients from our analysis. OHS Center List. We requested from OHS the names and addresses of each Head Start center operated by recipients. State Data. To gain further insight into the extent to which ACF reliably receives information about all incidents of child abuse, lack of supervision, and unauthorized release, OIG reviewed data from two States' agencies that oversee child care settings, including Head Start centers. We purposively selected two States, Texas and Florida, based on representation of multiple types of Head Start programs, multiple ACF administrative regions (OHS regional offices 4 and 6), urban and rural populations, and ability to provide data. We requested the following from each State: • State CPS Data. From the two selected States, we requested all substantiated abuse and neglect reports for children in the care of a child care facility for calendar years 2017 through 2019. These data included the child care facility's name and address, the date of the report, and descriptive information about ACF Should Improve Oversight of Head Start To Better Protect Children's Safety OEI-BL-19-00560 Detailed Methodology | 23 the abuse or neglect incident. We did not request or receive identifying information about individual children. • State Child Care Monitoring Data. From the two selected States, we requested child care monitoring data for all licensed child care facilities for calendar years 2017 through 2019. These data included the child care facility's name and address, the date of the State's review, the licensing violation, and descriptive information about the violation. We did not request or receive identifying information about individual children. We requested State data for calendar years 2017 through 2019 because this range provided a 3-year subset of the broader period of Federal monitoring data in our review, ensuring that we would be able to identify incidents in the State data that were reported to ACF either before or after they were reported to the State. Interviews with Selected OHS Staff. We conducted a panel interview with OHS central office staff and five panel interviews with staff from selected OHS regional offices (Regions 1, 4, 6, 11, and 12). We purposively selected these five OHS regional offices to capture variation in key regional and programmatic features (e.g., geographic location, program types) and to ensure we met with regional offices responsible for the largest number of children potentially at risk of harm (e.g., overall number of children served and overall number of incidents of child abuse, lack of supervision, and unauthorized release). Specifically, Regions 4 (serving Kentucky, Tennessee, North Carolina, South Carolina, Georgia, Alabama, Mississippi, and Florida) and 6 (serving New Mexico, Oklahoma, Arkansas, Louisiana, and Texas) serve large numbers of children and had the highest number of incidents during the period of review. Region 1 (serving Maine, New Hampshire, Vermont, Massachusetts, Rhode Island, and Connecticut) was among the smallest regions, serving a smaller number of children. Regions 11 and 12 oversee Head Start programs nationwide that serve children and families in American Indian and Alaska Native communities and migrant and seasonal communities, respectively. We also submitted questions in writing to OHS and received written responses. OHS Procedural and Training Documents. We requested from OHS documentation on written policies, procedures, and training materials related to identifying and addressing incidents of child abuse, lack of supervision, and unauthorized release; reporting requirements; and determining whether a violation will be classified as a deficiency or a noncompliance. OHS Administrative Action Records. We requested from OHS documentation about recipients that were subject to suspension, termination, and/or recompetition for renewal of funding under the DRS from October 1, 2015, through January 6, 2022. ACF Should Improve Oversight of Head Start To Better Protect Children's Safety OEI-BL-19-00560 Detailed Methodology | 24 Data Analysis Identifying and Categorizing Head Start Grant Recipients We analyzed PIR data to obtain a list of Head Start recipients. To do so, we defined a "recipient" as an organization operating in a unique State that oversees one or more individual Head Start center(s) and received one or more grants directly from ACF to provide Head Start services. We removed all recipients with no funded enrollment for the 2018-2019 enrollment year. Through this process, we identified 1,611 unique recipients; collectively, these recipients operate approximately 20,000 Head Start centers. We also used PIR data fields to identify the type(s) of Head Start programs each recipient operated (i.e., Head Start; Early Head Start; AI/AN Head Start; and Migrant and Seasonal Head Start) 30 and the recipient's ACF regional affiliation (1 through 12). Analysis of Monitoring Data We identified regulatory citations in both the 2009 and 2016 Head Start program standards that are related to child abuse, lack of supervision, unauthorized release, or failure to report significant incidents. (Although the current standards went into effect in 2016, ACF sometimes cited to the 2009 version of the standards during the early part of our review period; the 2009 and 2016 standards are not substantively different with regard to the types of incidents included in our review.) Please see Appendix A for a detailed list of all Head Start program standards included in our review and how they align with these categories. To determine the extent to which recipients received adverse findings from ACF for violating these Head Start standards, we analyzed Federal monitoring data for October 1, 2015, through May 11, 2020. This dataset lists all adverse findings associated with each recipient. We identified and summed the number of unique adverse findings (noncompliance or deficiency) for each of the four citation categories in our review (i.e., child abuse, lack of supervision, unauthorized release, and failure to report). To confirm the accuracy of the citation, we also reviewed the monitoring report narrative associated with each adverse finding. Then, we calculated the number and percentage of adverse findings in each of the four citation categories, as well as the number and percentage of recipients with such findings. We also analyzed review dates to identify recipients with child abuse, lack of supervision, or unauthorized release findings in separate fiscal years. To further categorize these violations, we conducted qualitative analysis of each report narrative to identify and classify each child abuse, lack of supervision, unauthorized release, and failure-to-report incident by its subcategory. The incidents were categorized as follows: • physical abuse and/or corporal punishment, emotional and/or verbal abuse, other prohibited disciplinary practices, or sexual abuse; ACF Should Improve Oversight of Head Start To Better Protect Children's Safety OEI-BL-19-00560 Detailed Methodology | 25 • child left unsupervised on the grounds of the Head Start facility, outside the Head Start facility, or on a bus; • release of a child to an unauthorized person from the facility or from a bus; and • failure to report an incident to ACF or to State or local authorities as required, including not reporting the incident "as soon as practicable" or not reporting the incident at all. Then, we identified the number of individual incidents associated with the finding. We defined an "incident" as a unique child-date combination. For example, a child left unsupervised on four separate dates was treated as four incidents, and three children subjected to child abuse on the same date was treated as three incidents. We summed these counts to identify the number and percentage of incidents related to each category (e.g., child abuse) and subcategory (e.g., physical abuse and/or corporal punishment), as well as the number and percentage of recipients with such incidents. To determine the status of adverse findings, we manually matched each citation's record with its followup record where necessary (due to a change in ACF process partway through the dataset, some records included followup action information in the same entry as the original citation, whereas others had separate followup records). We determined the number of adverse findings listed as "new" or "active" (i.e., currently in the corrective action process); "corrected"; or "not corrected." We also calculated the number and percentage of individual incidents associated with adverse findings of each status. To describe corrective actions that recipients took to address incidents of child abuse, lack of supervision, or unauthorized release, we conducted qualitative analysis of each report narrative to identify and categorize actions as follows: • disciplinary actions against staff, including written warnings; administrative leave; suspension; and termination of staff; • administrative improvements, including revised policies and procedures; updated monitoring procedures (e.g., in-person or video classroom monitoring); and facility improvements (e.g., installation of alarms, installation of physical barriers to prevent children from leaving the facility without supervision); and • staff training. We calculated the percentage of incidents addressed through each category of corrective action. We also calculated the percentage of incidents for which the involved staff member was terminated or resigned. ACF Should Improve Oversight of Head Start To Better Protect Children's Safety OEI-BL-19-00560 Detailed Methodology | 26 Analysis of Data on ACF Administrative Actions To determine the extent to which ACF used administrative authorities, we requested and reviewed data on recipients' termination, suspension, and recompetition from October 1, 2015 through January 6, 2022. We requested administrative actions for this date range to ensure that we captured all actions taken during the period of monitoring data we reviewed (October 1, 2015 through May 11, 2020) as well as actions taken more recently, given that the administrative processes involved may take months or years to complete. We summed the number of recipients that were terminated, suspended, and/or required to recompete for funding renewal under the DRS. We also compared our list of recipients with deficiency findings for child abuse, lack of supervision, or unauthorized release to the recompetition list to calculate the percentage of such recipients who had been required to recompete. Analysis of State Data To determine whether all incidents of child abuse, lack of supervision, and unauthorized release are reported to ACF, we requested data on child care licensing violations and substantiated CPS reports from Texas and Florida for calendar years 2017 through 2019. To match the State and Federal monitoring data, we identified Head Start centers by geocoding addresses in both the State data and the OHS Center List (which contains all center names and addresses, as well as their associated recipient organizations). We further confirmed all matches through manual review of the center names. Next, we identified incidents in these State datasets that described child abuse, lack of supervision, or unauthorized release. To do so, we reviewed the citation descriptions and narrative fields to identify those that reflect actions that are prohibited under the Head Start standards included in our review (child abuse, lack of supervision, or unauthorized release). We collected additional procedural and regulatory information from the State agencies where needed to make accurate determinations. We then manually compared the date and description of each incident in the State data to all incidents associated with the recipient in the Federal monitoring data. We summed the number of incidents of each type (child abuse, lack of supervision, unauthorized release) that were present in a State dataset but were not present in the Federal monitoring data, as well as the number of recipients with such incidents. We provided to OHS a list of incidents we identified that had not been previously reported to ACF. ACF Should Improve Oversight of Head Start To Better Protect Children's Safety OEI-BL-19-00560 Detailed Methodology | 27 Interviews and Written Responses We conducted a thematic qualitative analysis of interview responses from our panel interviews with staff from OHS central office and five OHS regional offices. We also analyzed OHS written responses to questions. Document Review We reviewed policy, procedural, and guidance documents to identify processes and criteria for monitoring, incident response, and adverse finding determinations. ACF Should Improve Oversight of Head Start To Better Protect Children's Safety OEI-BL-19-00560 Detailed Methodology | 28 APPENDICES Appendix A: Head Start Program Standards Included in OIG's Review This table lists Head Start program standards that ACF cited in one or more adverse findings involving child abuse, leaving a child unsupervised, or releasing a child to an unauthorized person from October 1, 2015, through May 11, 2020. The current Head Start program standards went into effect in September 2016; however, ACF sometimes cited to the 2009 version of the standards during the early part of our review period. Type of Violation Head Start Program Standard Head Start Program Standard Text Regulatory Citation 45 CFR § 1304.52(i)(1)(iv) [S]taff, consultants, and volunteers… will use Child Abuse (October 2009) positive methods of child guidance and will not engage in corporal punishment, emotional or physical abuse, or humiliation. In addition, they will not employ methods of discipline that involve isolation, the use of food as punishment or reward, or the denial of basic needs 45 CFR §§ 1302.90(c)(1)(ii)(A-I) Ensure staff, consultants, contractors, and (September 2016) volunteers do not maltreat or endanger the health or safety of children, including, at a minimum, that staff must not: (A) Use corporal punishment; (B) Use isolation to discipline a child; (C) Bind or tie a child to restrict movement or tape a child's mouth; (D) Use or withhold food as a punishment or reward; (E) Use toilet learning/training methods that punish, demean, or humiliate a child; (F) Use any form of emotional abuse, including public or private humiliation, rejecting, terrorizing, extended ignoring, or corrupting a child; (G) Physically abuse a child; (H) Use any form of verbal abuse, including profane, sarcastic language, threats, or derogatory remarks about the child or child's family; or (I) Use physical activity or outdoor time as a punishment or reward ACF Should Improve Oversight of Head Start To Better Protect Children's Safety OEI-BL-19-00560 Appendix A | 29 Type of Violation Head Start Program Standard Head Start Program Standard Text Regulatory Citation 45 CFR § 1304.52(i)(1)(iii) No child will be left alone or unsupervised while Child Left (October 2009) under their [i.e., staff, consultants, and volunteers] Unsupervised1 care 45 CFR § 1306.35(b)(2)(iv) Children are supervised at all times. Providers (October 2009) must have systems for assuring the safety of any child not within view for any period (e.g. the provider needs to use the bathroom or an infant is napping in one room while toddlers play in another room) 45 CFR § 1310.10(g) Each agency must ensure that children are only (October 2009) released to a parent or legal guardian, or other individual identified in writing by the parent or legal guardian. This regulation applies when children are not transported and are picked up from the classroom, as well as when they are dropped off by a vehicle. Agencies must maintain lists of the persons, including alternates in case of emergency, and up-to-date child rosters must be maintained at all times to ensure that no child is left behind, either at the classroom or on the vehicle at the end of the route 45 CFR § 1302.90(c)(1)(v) Ensure no child is left alone or unsupervised by (September 2016) staff, consultants, contractors, or volunteers while under their care 45 CFR § 1302.47(b)(5)(iii) All staff and consultants must follow appropriate (September 2016) practices to keep children safe during all activities, including… [a]ppropriate indoor and outdoor supervision of children at all times 45 CFR §1303.72(a)(3) A program must ensure… [u]p-to-date child (September 2016) rosters and lists of the adults each child is authorized to be released to, including alternates in case of emergency, are maintained and no child is left behind, either at the classroom or on the vehicle at the end of the route 45 CFR §§ 1302.47(a)-(b)(4)(i)(I)2 All staff with regular child contact have… training (September 2016) in… [a]ppropriate precautions in transporting children 45 CFR § 1303.72(a)(4) A program must ensure… [w]ith the exception of (September 2016) transportation services to children served under a home-based option, there is at least one bus monitor on board at all times, with additional bus monitors provided as necessary 1 ACF cited 45 CFR § 1310.10(g) and 45 CFR § 1303.72(a)(3) in adverse findings for both child supervision and unauthorized release incidents. 2 ACF cited 45 CFR §§ 1302.47(a)-(b)(4)(i)(I) in an adverse finding for leaving a child alone and unsupervised in a vehicle. ACF Should Improve Oversight of Head Start To Better Protect Children's Safety OEI-BL-19-00560 Appendix A | 30 Type of Violation Head Start Program Standard Head Start Program Standard Text Regulatory Citation 45 CFR § 1310.10(g) Each agency must ensure that children are only Child Released To an (October 2009) released to a parent or legal guardian, or other Unauthorized Person3 individual identified in writing by the parent or legal guardian. This regulation applies when children are not transported and are picked up from the classroom, as well as when they are dropped off by a vehicle. Agencies must maintain lists of the persons, including alternates in case of emergency, and up-to-date child rosters must be maintained at all times to ensure that no child is left behind, either at the classroom or on the vehicle at the end of the route 45 CFR § 1302.47(b)(5)(iv) All staff and consultants follow appropriate (September 2016) practices to keep children safe during all activities, including… [o]nly releasing children to an authorized adult 45 CFR § 1302.47(b)(7)(v) Programs establish, follow, and practice, as (September 2016) appropriate, procedures for… [m]aintaining procedures and systems to ensure children are only released to an authorized adult 45 CFR § 1303.72(a)(3) A program must ensure… [u]p-to-date child (September 2016) rosters and lists of the adults each child is authorized to be released to, including alternates in case of emergency, are maintained and no child is left behind, either at the classroom or on the vehicle at the end of the route 3 ACF cited 45 CFR § 1310.10(g) and 45 CFR § 1303.72(a)(3) in adverse findings for both child supervision and unauthorized release incidents. ACF Should Improve Oversight of Head Start To Better Protect Children's Safety OEI-BL-19-00560 Appendix A | 31 Type of Violation Head Start Program Standard Head Start Program Standard Text Regulatory Citation 45 CFR § 1302.47(b)(5)(i) All staff and consultants follow appropriate practices Failure to Report a (September 2016) to keep children safe during all activities, Significant Incident including…[r]eporting of suspected or known child Affecting Children's abuse and neglect, including that staff comply with Health and Safety applicable federal, state, local, and tribal laws 45 CFR § 1302.102(d)(1)(ii)(A) A program must submit… [r]eports, as appropriate, (September 2016) to the responsible HHS official immediately or as soon as practicable, related to any significant incidents affecting the health and safety of program participants, circumstances affecting the financial viability of the program, breaches of personally identifiable information, or program involvement in legal proceedings, any matter for which notification or a report to state, tribal, or local authorities is required by applicable law, including at a minimum: (A) Any reports regarding agency staff or volunteer compliance with federal, state, tribal, or local laws addressing child abuse and neglect or laws governing sex offenders ACF Should Improve Oversight of Head Start To Better Protect Children's Safety OEI-BL-19-00560 Appendix A | 32 Appendix B: Agency Comments Office of the Assistant Secretary | 330 C Street, S.W., Suite 4034 Washington, D.C. 20201 | www.acf.hhs.gov September 9, 2022 Ms. Suzanne Murrin Deputy Inspector General U.S. Department of Health and Human Services 330 Independence Avenue, SW Washington, DC 20201 Dear Ms. Murrin: The Administration for Children and Families (ACF) appreciates the opportunity to respond to the Office of Inspector General's (OIG) draft report titled "ACF Should Improve Oversight of Head Start To Better Protect Children's Safety" (OEI-BL-19-00560). Protecting the health and safety of children is central to ACF's mission across programs. ACF places the utmost priority on child health and safety and appreciates OIG's work to identify actions that will improve the Head Start program and better ensure children's well- being. Program rules are designed to support child safety, and the Office of Head Start (OHS) expects that any incident will be carefully assessed by the grant recipient and appropriate actions will be taken swiftly. At the same time, ACF is committed to continuous quality improvement to enhance oversight of Head Start programs and support grant recipients in preventing incidents that jeopardize children's safety. We appreciate OIG utilizing OHS' robust data set in its report in order to identify areas we can build on our rigorous process and make additional improvement to our system. ACF is committed to implementing the recommendations proposed by OIG as soon as possible and identifying additional opportunities to improve OHS' approach to protecting young children enrolled in Head Start. All child safety incidents are unacceptable. During the 5 year period covered in the report (2015 to 2020), Head Start's 1,600 grant recipients operated more than 20,000 centers and cumulatively served 5 million children. The OIG report identified 1,029 incidents involving lack of supervision, release to an unauthorized adult, and child abuse (defined broadly to include child abuse, corporal punishment, derogatory language, or discipline practices prohibited in Head Start). OIG finds that during the 5 year time period reviewed, 27 percent of Head Start grant recipients had an adverse finding in one of these categories. In alignment with these findings, further analysis by OHS shows that fewer than one in 1,000 children were involved in an incident involving lack of supervision, release to an unauthorized adult, or inappropriate discipline or child abuse. In addition, less than 5 percent of Head Start centers experienced an incident involving lack of supervision, release to an unauthorized adult, or inappropriate discipline or child abuse. In response to the incidents referenced in the report, local programs took contemporaneous, immediate, and appropriate actions in the vast majority of cases. ACF Should Improve Oversight of Head Start To Better Protect Children's Safety OEI-BL-19-00560 Appendix B | 33 Deputy Inspector General Murrin – Page 2 OHS supports and oversees Head Start grantees that adhere to Head Start program standards, which set a high bar for quality early childhood programs and are research-based, as mandated by Congress in the Head Start Act. Head Start Program Performance Standards are comprehensive and include standards with regard to health, safety, and adult-child interactions. Head Start staff, contractors, and volunteers are required to abide by a code of conduct informed by child development experts that includes using positive, developmentally appropriate strategies to support children's well-being and prohibits inappropriate discipline or harsh language. All grant recipients must follow these standards and undergo a rigorous federal monitoring process every 3 years. Grantees found to be out of compliance with the Head Start Program Performance Standards must take corrective action. Serious infractions result in a finding(s) of "deficiency," and multiple deficiencies require grant recipients to participate in a competitive grant process in order to retain funding. OHS proactively works with grant recipients through technical assistance efforts and provides resources to support the mitigation of health and safety incidents. The OIG report found that only 7 percent of Head Start grant recipients had adverse child safety findings for multiple years. OHS believes this finding reflects the need to continue to strengthen and build on existing technical assistance and training to prevent incidents from occurring and addressing them immediately if they do. Since the time period covered by the report concluded, specifically in November 2021, OHS launched a new tool that supports Head Start in preventing and reporting child abuse and adhering to the Head Start standards of conduct, called iLookOut. The iLookOut tool is an interactive, online professional development course for mandated reporters of child abuse. OHS offers iLookOut training specifically for early childhood staff and providers to promote a deeper understanding of how to protect children from abuse. This includes people who work as staff, consultants, and contractors or volunteer in Head Start programs, child care centers, and other early childhood education settings. This evidence-based, online course covers a variety of topics critical to keeping children safe from abuse. Additionally, OHS is scheduled to release iLookOut in Spanish in November 2022. Grant recipients also have access to technical assistance resources such as podcasts, webinars, and toolkits to support their understanding of how to reduce incidents of lack of supervision, inappropriate discipline, and unauthorized release. OHS will continue to update and make resources more accessible to grant recipients and their staff. For example, OHS has tools to provide practical support to programs in complying with the Head Start Program Performance Standards related to child supervision, and OHS and its national technical assistance centers work closely with programs to implement a culture of safety and environments where every program staff member understands their role and responsibilities in preventing childhood injuries. However, the OIG report makes clear OHS must take additional immediate and long-term steps to improve child safety. The following are ACF's responses to the report's four recommendations to OHS. ACF Should Improve Oversight of Head Start To Better Protect Children's Safety OEI-BL-19-00560 Appendix B | 34 Deputy Inspector General Murrin – Page 3 OIG Recommendation 1: Improve Head Start grant recipients' self-reporting of child abuse, lack of supervision, and unauthorized release incidents through better guidance and stronger consequences for failure to report. ACF Response: ACF concurs with this recommendation. Currently, ACF requires Early Head Start and Head Start grant recipients, under 45 CFR 1302.102(d)(ii)(A), to report immediately or as soon as practicable, any significant incidents affecting the safety of program participants and staff or volunteer violations of federal, state, tribal, or local laws addressing child abuse and neglect or governing sex offenders. Most Head Start grant recipients report as required. Seventy-six percent of grant recipients cited for violations of Head Start health and safety standards during the timeframe of the study reported incidents in a timely manner. Consequences for violation of the reporting requirement vary based on the severity of the incident affecting the program participant or the failure of the grant recipient to report immediately or as soon as practicable. In most cases, grant recipients who fail to report health and safety incidents within 7 to 10 days of an incident are determined to have a noncompliance, and the grant recipient must correct the violation within 120 days. If a health and safety incident is more severe, or the grant recipient fails to report for a longer period of time, OHS issues a deficiency determination, which typically must be corrected sooner and could lead to re-competition or termination of the grant. OHS has recently taken the following additional steps to further improve self-reporting of incidents by grant recipients: • In August 2022, OHS issued a Dear Colleague Letter to align with the beginning of the typical program year outlining Early Head Start and Head Start reporting requirements, examples of what to report, and emphasizing the importance of mandated reporting and child safety. • In August 2022, OHS provided a health and safety webinar for programs to reinforce additional reporting guidance and online professional development tools, including the Head Start Health Services Competencies and modules that address safety practices. • In September 2022, OHS will issue an Information Memorandum on the current Early Head Start and Head Start reporting requirements that will further define when child safety incidents must be reported and to whom, including examples of what types of incidents to report. In addition, OHS will: • Develop more robust strategies to communicate the consequences for grant recipients of not reporting, or not reporting immediately, significant incidents affecting the health and safety of children to OHS; • Clearly communicate the criteria for when delays in reporting are classified as an area of non-compliance or deficiency, while also preserving some discretion based on the severity of the incident and context; ACF Should Improve Oversight of Head Start To Better Protect Children's Safety OEI-BL-19-00560 Appendix B | 35 Deputy Inspector General Murrin – Page 4 • Provide regular training on reporting requirements, especially targeting new Head Start program directors; • Include more information and guidance on reporting requirements in OHS' annual monitoring information memorandum and webinar; and • Highlight Training and Technical Assistance (TTA) resources and information available to Early Head Start and Head Start grant recipients regarding reporting and preventing child health and safety on the Early Childhood Knowledge & Learning Center website. OIG Recommendation 2: Extend the reporting requirement to include child abuse, lack of supervision, and unauthorized release incidents in blended classrooms in which the victim is not a Head Start-funded child. ACF Response: ACF concurs with this recommendation. Head Start staff are already mandated under state, local, or tribal law to report all suspected child abuse to state or tribal officials. However, if the behavior involves a child who is not a Head Start program participant, they are not required to report the incident to OHS, but to the appropriate state or tribal agency. Conversely, most incidents involving lack of supervision or unauthorized release of a child are typically not reported to state or tribal child welfare officials because, while they violate Head Start Performance Standards and should be reported to OHS, they may not meet a state or tribal reporting requirement. ACF agrees with OIG that this current protocol means OHS is left unaware of important information about the safety of children in Head Start programs and changes to procedures are warranted. Though ACF does not currently have authority to fully implement this recommendation, ACF agrees with the need to explore options to extend its requirement about grantees reporting to OHS incidents that compromise the safety of any child in their care. While exploring additional options, ACF will take immediate steps to increase training and guidance, including to: • Provide guidance that the requirement to report applies to incidents involving Head Start staff, contractors, and volunteers in all Head Start and Early Head Start settings including blended classrooms; • Enhance efforts to encourage Head Start to utilize the iLookOut tool to train Head Start staff; • Provide additional guidance to all Early Head Start and Head Start recipients on mandated reporting; • Identify OHS legal authority in this area and identify available ways to improve guidance on mandated reporting involving children in blended classrooms; • Increase OHS TTA related to mandated reporting to local authorities; and • Create Early Head Start and Head Start resources to mitigate safety incidents, to be available on ACF websites for use by early childhood education agencies, including those operating blended classrooms. ACF Should Improve Oversight of Head Start To Better Protect Children's Safety OEI-BL-19-00560 Appendix B | 36 Deputy Inspector General Murrin – Page 5 OIG Recommendation 3: Improve data-sharing with states about incidents of child abuse, lack of supervision, and unauthorized release in Head Start centers. ACF Response: ACF concurs with this recommendation. OHS will take a tiered approach to addressing this recommendation. As an immediate first step, OHS will incorporate a review of publicly available state child care licensing reports as part of its monitoring reviews in the 2022 to 2023 program year. OHS will review publicly available licensing findings from the grant period to inform Head Start monitoring reviews and to determine if there were any incidents of child safety reported to the state licensing agencies that were not reported to OHS. Simultaneously, OHS will explore additional methods for obtaining data on incidents reported via state licensing and other child protective services, such as through data sharing plans with states, partnerships with other ACF program offices, and through Head Start State Collaboration Offices. Finally, OHS will consider ways to clarify requirements related to concurrently reporting information to both the state and federal government. OIG Recommendation 4: Disseminate information about innovative practices that OHS regional offices have developed to better identify and prevent incidents that threaten children's safety. ACF Response: ACF concurs with this recommendation. OHS regional offices are critical to the effective implementation and monitoring of Early Head Start and Head Start programs across the country. Historically, OHS regional offices have shared knowledge, skills, and abilities across the 12 OHS regions. OHS regional program managers (RPMs) meet weekly to share ideas, trends, and concerns. RPMs also meet in person at least quarterly to share innovations. Regional offices are consulted on the development and innovation of strategies to mitigate child health and safety risks, and many of those strategies are the cornerstones of the OHS monitoring system. Sharing innovations is an intrigal part of the OHS structure, and ideas generated are disseminated through multiple channels. ACF will consider more formal structures for collecting and disseminating information across regions including new innovations and strategies to identify and prevent incidents that threaten children's safety. Examples of structures and resources being considered, include: • Analyzing regional TTA Activity reports in the areas of child health and safety, sharing trends and innovative practices, and summarizing by region planned yearly regional professional development related to child incidents and reporting; • Documenting the innovative strategies of the five OHS regional offices interviewed by OIG for its report; • Documenting additional strategies from the remaining seven OHS regional offices not interviewed by OIG for its report; and • Using regional innovations to improve the identification of child health and safety issues across regions. ACF Should Improve Oversight of Head Start To Better Protect Children's Safety OEI-BL-19-00560 Appendix B | 37 Deputy Inspector General Murrin – Page 6 ACF is proud to be the home of Head Start, which over the course of its 57-year history has upheld rigorous national standards and reporting requirements to promote the well-being of the children participating in Early Head Start and Head Start-funded programs. We appreciate OIG for the review of this critical national program and for identifying areas that should be strengthened. Again, thank you for the opportunity to review and comment on this draft report. Please direct any follow-up inquiries to Scott Logan, Office of Legislative Affairs and Budget, Administration for Children and Families, (202) 401-4529. Sincerely, January Contreras Assistant Secretary Administration for Children and Families U.S. Department of Health and Human Services ACF Should Improve Oversight of Head Start To Better Protect Children's Safety OEI-BL-19-00560 Appendix B | 38 ACKNOWLEDGMENTS AND CONTACT Acknowledgments Jennifer Hutnich served as the team leader for this study. Others in the Office of Evaluation and Inspections who conducted the study include Suriya Karim, Tori Lawson, and Anne Leong. Office of Management and Policy staff who provided support include Jason Lee. Office of Evaluation and Inspections headquarters staff who provided support include Kevin Manley and Christine Moritz. This report was prepared under the direction of Dave Tawes, Regional Inspector General for Evaluation and Inspections in the Baltimore regional office, and Louise Schoggen, Assistant Regional Inspector General and Heather Barton, Deputy Regional Inspector General. Contact To obtain additional information concerning this report, contact the Office of Public Affairs at Public.Affairs@oig.hhs.gov. OIG reports and other information can be found on the OIG website at oig.hhs.gov. Office of Inspector General U.S. Department of Health and Human Services 330 Independence Avenue, SW Washington, DC 20201 ACF Should Improve Oversight of Head Start To Better Protect Children's Safety OEI-BL-19-00560 Acknowledgments and Contact | 39 ABOUT THE OFFICE OF INSPECTOR GENERAL The mission of the Office of Inspector General (OIG), as mandated by Public Law 95-452, as amended, is to protect the integrity of the Department of Health and Human Services (HHS) programs, as well as the health and welfare of beneficiaries served by those programs. This statutory mission is carried out through a nationwide network of audits, investigations, and inspections conducted by the following operating components: The Office of Audit Services (OAS) provides auditing services for HHS, either by conducting audits with its own audit resources or by overseeing audit work done by others. Audits examine the performance of HHS programs and/or its recipients and contractors in carrying out their respective responsibilities and are intended to provide independent assessments of HHS programs and operations. These audits help reduce waste, abuse, and mismanagement and promote economy and efficiency throughout HHS. The Office of Evaluation and Inspections (OEI) conducts national evaluations to provide HHS, Congress, and the public with timely, useful, and reliable information on significant issues. These evaluations focus on preventing fraud, waste, or abuse and promoting economy, efficiency, and effectiveness of departmental programs. To promote impact, OEI reports also present practical recommendations for improving program operations. The Office of Investigations (OI) conducts criminal, civil, and administrative investigations of fraud and misconduct related to HHS programs, operations, and beneficiaries. With investigators working in all 50 States and the District of Columbia, OI utilizes its resources by actively coordinating with the Department of Justice and other Federal, State, and local law enforcement authorities. The investigative efforts of OI often lead to criminal convictions, administrative sanctions, and/or civil monetary penalties. The Office of Counsel to the Inspector General (OCIG) provides general legal services to OIG, rendering advice and opinions on HHS programs and operations and providing all legal support for OIG's internal operations. OCIG represents OIG in all civil and administrative fraud and abuse cases involving HHS programs, including False Claims Act, program exclusion, and civil monetary penalty cases. In connection with these cases, OCIG also negotiates and monitors corporate integrity agreements. OCIG renders advisory opinions, issues compliance program guidance, publishes fraud alerts, and provides other guidance to the health care industry concerning the anti-kickback statute and other OIG enforcement authorities. ACF Should Improve Oversight of Head Start To Better Protect Children's Safety OEI-BL-19-00560 About OIG | 40 ENDNOTES 1 Recipients operate one or more of the following program types: (1) Head Start programs, which serve children ages 3 to 5; (2) Early Head Start programs, which serve infants and toddlers up to age 3, as well as pregnant women; (3) Migrant and Seasonal Head Start programs, which serve children and families in the migrant and seasonal worker community; and (4) American Indian-Alaska Native Head Start programs, which serve children and families in the American Indian and Alaska Native communities. 2 The Head Start program standards were revised and reissued in 2016; provisions related to child abuse, lack of supervision, and unauthorized release did not substantively change. We cite the 2016 standards in this report, because they were in effect during our study period. For the complete current Head Start program standards, see 45 CFR Chapter XIII. 3 45 CFR § 1302.47(b)(5)(v); 45 CFR § 1302.90(c)(1)(ii). 4 45 CFR § 1302.47(b)(5)(iii); 45 CFR § 1302.90(c)(1)(v). 5 45 CFR § 1302.47(b)(5)(iv). 6 45 CFR § 1302.102(d)(1)(ii); 45 CFR § 1302.47 (c). 7 42 U.S.C. § 9836a(c)(1). 8 42 U.S.C. § 9832(2). 9 45 CFR § 1304.2(a). 10 42 U.S.C. § 9836a(d)(2)(A). 11 42 U.S.C. § 9832(2)(C). 12 The seven DRS conditions are as follows: (1) Two or more deficiencies, defined as a systemic or substantial material failure of an agency in an area of performance; (2) Score below a competitive threshold in one or more Classroom Assessment Scoring System (CLASS®) domains; (3) Two or more audit findings of material weakness or questioned costs associated with Head Start funds, or a "going concern" finding (i.e., a determination that the grantee is at risk of financial failure); (4) Failure to establish and take steps to achieve school readiness goals; (5) License revocation; (6) Suspension by OHS; and (7) Debarment by another federal or state agency or disqualification from the Child and Adult Care Food Program. See 45 CFR § 1304.11. 13 The complete list of reasons ACF may terminate Head Start funding to a recipient is as follows: (i) The grantee is no longer financially viable; (ii) The grantee has lost the requisite legal status or permits; (iii) The grantee has failed to timely correct one or more deficiencies as defined in the Act; (iv) The grantee has failed to comply with eligibility requirements; (v) The grantee has failed to comply with the Head Start grants administration or fiscal requirements set forth in 45 CFR part 1303; (vi) The grantee has failed to comply with requirements in the Act; (vii) The grantee is debarred from receiving federal grants or contracts; or (viii) The grantee has failed to abide by any other terms and conditions of its award of financial assistance, or any other applicable laws, regulations, or other applicable federal or state requirements or policies. See 45 CFR § 1304.5. 14 42 U.S.C. § 9837b. 15 We requested data from the Florida Department of Children and Families-Office of Child Welfare and the Texas Department of Family and Protective Services. In this report, we refer to these generally as CPS agencies and CPS data. 16 We requested data from the Florida Department of Children and Families-Office of Child Care Regulation and the Texas Health and Human Services Commission-Child Care Regulation/Regulatory Services Division. In this report, we refer to these generally as child care licensing agencies and child care monitoring data. ACF Should Improve Oversight of Head Start To Better Protect Children's Safety OEI-BL-19-00560 Endnotes | 41 17 We defined an "incident" as a unique child-date combination. For example, a child subjected to child abuse on two separate dates was treated as two incidents, and three children subjected to child abuse on the same date was treated as three incidents. 18 42 U.S.C. § 9832(2)(C). 19 45 CFR § 1302.102(d). 20 ACF, Information Memorandum: Mandated Reporting of Child Abuse and Neglect, ACF-IM-HS-15-04, September 18, 2015. 21 In addition to 6 incidents of children left unsupervised that were reported to the Texas CPS agency but were not reported to ACF, our review of Texas child care licensing data also identified 102 licensing violations related to child supervision that were not in the ACF monitoring database. However, the Texas licensing violation data did not clearly distinguish between incidents involving children who are left completely unsupervised-incidents included in this report-and other types of supervision events (e.g., child/caregiver ratio, room capacity limitations) that are not addressed in this report. Therefore, we did not include them in the total number of incidents not reported to ACF. 22 See Endnote 20. 23 ACF was not able to provide a list of recipients or centers that operate blended classrooms. OIG was therefore unable to determine whether any of the incidents that we identified in State data, but which were not present in Federal data, had occurred in blended classrooms and thus could have involved a child who was not funded through Head Start. 24 ACF, Supervision and Transitions, Information Memorandum ACF-IM-HS-15-05, Sep. 18, 2015. Accessed at https://eclkc.ohs.acf.hhs.gov/policy/im/acf-im-hs-15-05 on May 12, 2022. 25 In January 2018, the Acting Director of the Office of Head Start wrote a letter to recipients cautioning against over-reporting of "practices, and/or behaviors that need improvement but do not harm or endanger children, versus reportable practices or behaviors that harm or endanger children." The letter does not describe or give examples of incidents that do or do not warrant reporting, instead directing that recipients "work with management, governing bodies, Health Advisory Committees, mental health consultants, and local or state licensing agencies" to identify reportable incidents. 26 Deficiencies with corrective action periods over 90 days must submit a quality improvement plan (QIP) to ACF for approval. (See 42 U.S.C. § 9836a(e)(2).) However, ACF staff reported that QIP's are rarely used because, in practice, ACF requires most of the types of incidents included in this report to be corrected within 30 to 45 days. 27 42 U.S.C. §§ 9836a(e)(1)(B) and (e)(2)(A)(ii). The Head Start Act requires all deficiencies to be corrected within one year, but ACF may require some deficiencies to be corrected immediately or within 90 days depending on the nature and magnitude of the deficiency and whether it threatens the health and safety of staff or program participants. In practice, ACF staff reported that the corrective action period for the incidents included in this report usually range between 30 to 120 days. 28 All adverse findings for child abuse, lack of supervision, and unauthorized release in our review had either been corrected or were still listed as "new" or "active" in ACF's database; ACF informed OIG that a "new" or "active" status indicates that the finding is still in the corrective action process. One adverse finding was excluded from our corrective status analysis because of anomalous data; specifically, the primary record showed the status as "not corrected," but multiple incomplete followup records with missing data were listed for the same recipient. 29 This number does not include terminations for which the termination was reversed by the Departmental Appeals Board or the appeal was in progress at the time of our data collection. 30 See Endnote 1 for information on the different types of Head Start programs. ACF Should Improve Oversight of Head Start To Better Protect Children's Safety OEI-BL-19-00560 Endnotes | 42