Audit Report on the Administration for Children’s Services’ Monitoring of the Safety of Children in Foster Care

Audit Impact

Summary of Findings

The audit found that the New York City Administration for Children’s Services’ (ACS) mechanisms for monitoring the safety of children in foster care during the period reviewed were generally adequate.

The audit found that incidents of neglect and abuse have been trending downward in recent years, from 4% of children in foster care with substantiated cases in 2020 to 3.1% in 2023, but that many children in foster care still suffer neglect and/or abuse. ACS data shows that between Fiscal Years 2020 and 2023, there were more than 2,000 cases of substantiated neglect and/or abuse involving more than 1,600 children.[1] Over 76% of incidents occurred while children were visiting with their birth families, and roughly a quarter occurred while the child was in the physical custody of a foster care family (whe                                                                                                                           ther kinship or non-relative) or some other agent employed by a foster care agency.

In some instances, decisions concerning visitation and supervision are not made by ACS or the foster care agencies but by Family Court. The audit found that ACS does not track the aggregate number of incidents that occurred during court-mandated versus non-court-mandated visitations with birth families, nor does ACS track incidents that occur during supervised versus unsupervised visits. Compiling and monitoring such information would help ACS assess both the extent to which incidents occur during non-court-mandated unsupervised visitations and the effectiveness of the agency’s efforts to reduce their occurrence. ACS should evaluate the effectiveness of decisions that are within its direct control; evaluation could also inform future ACS decisions and its recommendations to Family Court.

Although ACS’ monitoring mechanisms were found adequate in most areas of the foster care system, auditors found that ACS did not conduct site visits at the sampled residential care facilities in a timely manner, as called for in its policies. For four sampled providers that had been issued safety and risk alerts, ACS allowed an average of 10 months to elapse between visits—far exceeding the target timeframe of at least one visit every six months. Foster children in residential care account for less than 10% of the foster care population.

Intended Benefits

This audit identified areas in which ACS could improve its monitoring of the safety of children in foster care and steps it could take to better ensure the safety of the foster children under its care.

Introduction

Background

The mission of ACS is to protect and promote the safety and well-being of New York City’s children and families. As part of this effort, the agency facilitates foster care for children when needed. Foster care is intended to provide safe and stable out-of-home care for children until they can be either safely returned home or placed in another permanent arrangement. Children in foster care include those who were removed by order of the Family Court due to neglect or abuse or based on a determination that risk of such harm exists. Foster care placement may also take place when primary caregivers have died or are otherwise unable to care for them.

ACS relies on contracted providers to operate four types of foster care programs. These include regular Family Foster Care (FFC), Treatment Family Foster Care (TFFC), Specialized Family Foster Care (SFFC), and Residential Services.

FFCs are placements in which foster parents provide care to children in family-type home settings . A relative or other person who has a significant relationship with the child’s family may also be certified to be a foster parent (known as kinship foster parents).

A TFFC program serves children/youth up to age 21 who have moderate to severe behavioral or emotional issues and can be supported in foster family settings for the short term.

Children in SFFC have special medical needs and/or severe intellectual/developmental disabilities that require higher and more consistent levels of caregiving. SFFC children are often in situations that demand immediate intervention.

Residential Services offer intensive clinical services for youth (12 years and older ) in a group home setting. These may include, but are not limited to, youth whose current clinical,medical, or other needs cannot be met safely and adequately in a family setting.

Foster Care Provider Responsibilities

ACS contracts with foster care providers (providers) to place and oversee children in appropriate foster care settings. Providers are required to create and maintain written service plans that include the specific steps and services needed to meet the needs of each child/youth in their care. All case records (including progress notes) are maintained in Connections (CNNX), the electronic recordkeeping system managed by the New York State Office of Children and Family Services (OCFS), the oversight agency for ACS.

Providers are responsible for recruiting foster parents and training them to provide safe, nurturing home environments. Providers also employ caseworkers who are required to conduct regular visits (casework contacts) to the foster homes where children are placed to assess the child’s safety and well-being. During these contacts, caseworkers conduct safety assessments, which include determining whether there are any safety hazards in the foster home or any visible signs of abuse, including signs of mental/emotional trauma; evaluating service needs of children and their families; and verifying whether the required services are being provided (e.g., medical, therapy, and tutoring).

As stated in the New York State Department of Social Services Rules and Regulations , a minimum of three casework contacts must be conducted during the first 30 days of placement—two with the child and one with the foster parent. [2] After that, casework contacts are held at least once a month, with at least two contacts made at the child’s foster care placement location every 90 days.

Providers also oversee visitation between foster children and their birth parents. New York State Law recognizes the importance of maintaining and strengthening children’s connections with their birth families while in foster care. According to the Family Court Act § 1030-c, parents are to be granted reasonable and regularly scheduled visitation unless the court finds that the child’s life or health would be endangered. In addition, a parent’s visitation with a child who is in foster care is to be unsupervised unless supervision is warranted due to the foster care agency’s concerns relating to the safety of the child. Providers are required to regularly assess the necessity for supervision, and providers’ attorneys make a request to the family court for an increase or decrease to the level of supervision as appropriate. [3]

If a provider identifies any safety concerns during a child’s court-ordered visitation, the provider works with ACS attorneys within its Family Court Legal Services (FCLS) Division to decide whether to make a motion to change a visiting plan. FCLS and the provider present the case-specific details and efforts to work within the existing plan and any alternative plans. It is at the Family Court’s discretion that modifications are made to the existing visitation plan. FCLS can decide whether to appeal the Family Court’s denial of a requested change in visits.

ACS’ Monitoring of Providers

ACS’ Division of Policy, Planning and Measurement (DPPM) is responsible for managing and overseeing policy development, new program development, and quality improvement through ACS’ contract provider agencies. The Division of Family Permanency Services (FPS) is responsible for ensuring that quality services are provided to children in foster care by offering contract provider agency staff with technical assistance and services support.

The DPPM has two units―the Provider Agency Measurement System (PAMS) and the Agency Program Assistance (APA)―which assist. The PAMS unit conducts biannual, statistically random reviews of foster children cases managed by every foster care provider, including a review of casework contacts performed by the foster care providers’ caseworkers. The APA unit is responsible for monitoring the performance of foster care provider agencies, including provider-operated residential care facilities. This is done by examining case records, performing site assessments at the facilities where children have been placed, and conducting interviews with children and program staff.

According to ACS’ COVID-19 emergency guidance (issued March 31, 2020, and revised July 14, 2020), video conferences and phone calls were accepted in lieu of face-to-face casework contacts. However, particularly high-risk cases were still expected to have in-person contacts. All COVID-related polices ended in June 2023.

ACS’ Protocol for Allegations of Neglect and/or Abuse of Children within the Foster Care System

New York State Family Court Act defines neglect and abuse as follows:

• Neglect occurs when a child’s physical, mental, or emotional condition has been impaired or is in imminent danger as a result of the parent’s (or authorized caregiver’s) failure to provide a minimum degree of care, such as food, clothing, etc.
• Abuse occurs when a child experiences serious injury, is in danger of serious injury, or experiences a sex offense through the conduct of the parent (or authorized caregiver).

A child named in an allegation may either be the subject of neglect or abuse or be present in the same household as the subject at the time that the alleged incident occurred. The Statewide Central Register (SCR) of Child Abuse and Maltreatment operates a 24-hour hotline to receive complaints of possible neglect and abuse and is responsible for screening and assigning reports of suspected child neglect and abuse to local jurisdictions for investigation. Allegations of neglect or abuse are entered by SCR into CNNX.

Alleged incidents that occur either during a supervised or unsupervised visit or trial discharge, involving the child’s birth family or a close family member or friend, are designated as “familial.”[4] Alleged incidents that occur away from the child’s birth family are categorized accordingly (e.g., “foster care,” “day care”).[5]

Allegations that fall under ACS’ jurisdiction (within the five boroughs) are investigated by its Division of Child Protection (DCP) .[6] Investigative steps and timeframes to which ACS must adhere are detailed in state procedures and various ACS and OFCS directives, including the ACS Division of Child Protective Services Casework Practice Requirements Manual. Once an allegation of child maltreatment, including neglect and/or abuse, has been reported, a team consisting of a Child Protective Specialist manager, supervisor, and investigator conducts the investigation. The investigative team must conduct a thorough assessment of the safety risk level of every child in the household.

The investigation must begin within 24 hours of receipt of the report. Within this 24-hour period, the investigative team must have face-to-face or telephone contact with the subject(s) and other person(s) named in the report, the source of the report, and/or other persons in a position to provide information about whether the child may be in immediate danger of serious harm. In addition, a criminal history check must be conducted on all family members within the household. All details of each investigation must be documented by the investigation team within the progress notes section of CNNX.

New York State mandates require that the investigation be concluded, and a final determination be made of the allegation, within 60 days of receipt of the report. If the investigation reveals that “some credible evidence” of child neglect and/or abuse exists, the report is deemed “indicated” (or substantiated).

When safety deficiencies involving a child in traditional (non-relative) or kinship foster care are identified, the Office of Special Investigations (OSI) investigative team may make corrective action plan (CAP) recommendations to a foster care provider.[7] However, a CAP will not be issued in certain circumstances, such as a determination that there are no serious safety factors and/or risk issues requiring on-going assessment, a recommendation to close the foster home, or when preventive services, court-ordered services and/or court-ordered supervision have not been deemed necessary.

The provider has up to 20 days to respond to and implement the CAP, unless the determination requires immediate implementation, such as in cases requiring the removal of a child from the foster home.

According to the Fiscal Year 2023 Mayor’s Management Report (MMR), there has been a steady decline in the average number of children in ACS’ foster care system, going from 8,341 children in FY2019 to 6,728 children in FY2023—a decrease of 19.3%. According to ACS’ data on foster care investigations, for the period covering FY2020 through FY2023, there were a total of 2,154 substantiated instances of neglect and/or abuse involving 1,641 children in foster care. A review of the data shows a slight decrease in the total percentage of substantiated incidents of neglect and/or abuse involving children in foster care, dropping from 4% to 3.1% from FY2020 to FY2023.

Objective

The objective of the audit was to determine whether ACS adequately monitors the safety of children in foster care.

Discussion of Audit Results with ACS

The matters covered in this report were discussed with ACS officials during and at the conclusion of this audit. An Exit Conference Summary was sent to ACS and discussed with ACS officials at an exit conference held on August 20, 2024. On October 4, 2024, we submitted a Draft Report to ACS with a request for written comments. We received a response from ACS on October 21, 2024. In its response, ACS generally agreed with four recommendations (#1, #3, #4, and #5) and disagreed with one recommendation (#2).

ACS’ written comments have been fully considered and, where relevant, changes and comments have been added to the report.

The full text of ACS’ response is included as an addendum to this report.

Detailed Findings

The audit found that the mechanisms used by ACS to monitor the safety of children in foster care are generally adequate, with the exception of oversight of residential care facilities, where less than 10% of the foster children placed in ACS’ care resided in Fiscal Year 2023. The audit also found that an alarming number of children in foster care continue to suffer substantiated cases of neglect and/or abuse (also referred to generally throughout this report as maltreatment). These do not always occur in circumstances within the direct control of ACS.

The audit found that ACS reviews foster care case records on schedule, issues safety and risk alerts to foster care providers when necessary, and assesses caseworker contacts with foster children. A sample review of 172 case record reviews conducted by the PAMS unit found that ACS satisfactorily identified and addressed safety risks for children receiving foster care from contracted providers and followed up to ensure that providers addressed deficiencies. The PAMS reviews also assessed whether providers conducted in-person visits to foster homes. A detailed examination of 30 of the 172 PAMS case record reviews found that caseworkers made 96% of the required number of contacts with foster children (e.g., in-person visits to foster homes) during the year, meeting federally mandated requirements.

However, the audit found a serious deficiency in the oversight over residential care facilities. APA did not conduct site visits at the sampled residential care facilities in a timely manner. For four sampled providers that had been issued safety and risk alerts by PAMS reviewers between March 2021 and December 2023, ACS’ APA unit allowed an average of nine months to elapse between visits—far exceeding the target timeframe of at least one visit every six months . Failure to conduct the required in-person visits at residential facilities increases the risk that health and safety issues or program inefficiencies are not identified and corrected in a timely manner.

In addition, the data obtained from ACS shows that between FYs 2020 and 2023 there were 2,154 cases of substantiated neglect and/or abuse, involving more than 1,600 children (some children have more than one case ). A majority of these were neglect cases—76.4% of these substantiated incidents involved neglect only; 21.3% involved both neglect and abuse; and 2.3% involved abuse alone. An analysis of substantiated incidents of neglect and abuse cases found that 24% of such incidents—508 in total—occurred while the child was in the physical custody of a foster care family or some other agent (e.g., residential facility) employed by a foster care agency. One incident occurred at day care, away from the child’s birth family. The remaining 1,645 (76%) incidents were familial in nature, meaning they occurred while children were visiting with a birth parent, family member, or a person with a significant relationship with the child, after being placed in foster care. In some instances, decisions concerning visitation and supervision are made by Family Court rather than ACS.

The audit reviewed ACS’ processes for addressing and tracking incidents of neglect and/or abuse and found that ACS does not compile or monitor in the aggregate whether incidents occurred during non-court-mandated visitations with birth families or during supervised or unsupervised visits. Compiling and monitoring such information would help ACS assess both the extent to which incidents occur during non-court-mandated unsupervised visitations and the effectiveness of the agency’s efforts to reduce the rate of such occurrences. This information could in turn inform related internal policies and future decisions.

Similarly, ACS does not aggregate data for incidents that occur during court-ordered unsupervised visitations and trial discharges. While ACS does not have control over such decisions, it has access to this data through OSI. Tracking and evaluating this information could help ACS identify trends in the frequency of incidents. Evaluating these should inform future recommendations by ACS to Family Court.

2,154 Cases of Substantiated Neglect and Abuse Incidents between 2020–2023

There was a total of 2,154 incidents of neglect and/or abuse involving 1,641 children in foster care during the four-year period from FY2020 to FY2023. Although there has been a slight decrease in the average percentage of children in foster care who have been the subject of substantiated incidents of neglect and/or abuse each year (dropping from 4% in FY2020 to 3.1% in FY2023), 341 children in foster care were still maltreated in FY2023.

Table I below provides a breakdown of the substantiated incidents among the above-mentioned categories.

Table I: Breakdown of Substantiated Instances of Neglect and/or Abuse of Children in Foster Care during Fiscal Years 2020 through 2023

*Figures include children who resided for at least one day in foster care during the year.

**This column does not total because some children were included in the figures for more than one year.

A substantiated incident does not automatically require that children be removed from their foster care settings: only those children who cannot remain safely are removed, as in the case of substantiated abuse incidents. In cases of substantiated neglect, a child may remain at a foster home if investigators determine that their health, well-being, and safety are not at immediate risk. For example, if a child is not receiving adequate medical or mental health services, OSI may direct the foster care provider to furnish such services without removing the child.

As shown in the table above, more than three quarters of the substantiated incidents occurred while the children were visiting with family members (i.e., familial). According to ACS data, 1,646 (76.4%) of the 2,154 substantiated incidents involved neglect only; 458 (21.3%) involved both neglect and abuse; and 50 (2.3%) involved abuse alone.

Of the 508 cases where foster parents were accused, 370 (72.8%) involved neglect only, 12 (2.4%) involved abuse only, and 126 (24.8%) involved both neglect and abuse. These percentage breakdowns were relatively consistent over the four-year period, though the overall percentage of children involved in such instances trended downward along with the total number of children in foster care.[8]

Incidents of Neglect and Abuse Involving Foster Children Trending Downward but Persistent

ACS prepares a Strategic Management Report (SMR) each quarter that is presented in its monthly monitoring meeting with OCFS. This document reports on the agency’s efforts in the areas of child safety, prevention services, youth justice, early care and education, and foster care, and provides various statistics on a yearly basis.

A review of this data shows downward trends in some key metrics, including the number of children impacted by substantiated investigations and in repeated substantiated investigations within a single year. One such indicator is the number of children per every 100,000 days of care provided to all children. According to ACS, this measure was established by the federal government. (In Calendar Year 2023, ACS provided over 2.3 million days of care to more than 6,000 children per day on average.) According to the March 2024 SMR, the number of children who were maltreated while in foster care at the time of the incidents has significantly decreased in recent years, going from 21.7 children per 100,000 care days in CY2020 to 12.8 children in CY2023. Regarding children who were maltreated by a foster parent, the numbers went from 7.7 children in CY2020 to 2.9 children in CY2023.

This apparent improvement is not, however, reflected in overall percentages of children suffering confirmed instances of neglect and/or abuse. Overall numbers remain within a 2% margin over this four-year period, and as reported in the March 2024 SMR, more than 300 children in foster care were still impacted by neglect and/or abuse during CY2023. In 70 cases (23%), this maltreatment was by foster parents directly overseen by ACS and its providers.

At the exit conference for this audit, ACS stated that the agency implemented several initiatives in the period following the audit (after 2023) that are intended to reduce the rate of incidents involving foster children, including:[9]

• Case Conferences with key stakeholders to discuss concerns/issues found during the investigation (started in FY2024).

• A special program to reduce neglect called Trauma Informed Parenting Program (TRIPP) to strengthen foster parents’ capacity to support children who have experienced neglect and abuse.
• New contracts (implemented on July 1, 2023) that integrate regular and therapeutic family foster care into a new model called Enhanced Family Foster Care (EFFC). EFFC recognizes that all children in care have experienced some level of trauma, that their needs may change over time, and that their strengths must be supported so they can reach their full potential.

Officials also referenced various ongoing efforts, including (1) a focus on kinship foster home placements; (2) increasing parent advocacy support for birth parents; and (3) working with providers to strengthen recruitment and support of foster parents. Auditors are unable to assess the extent to which these efforts contributed to the declining rate of neglect and abuse incidents for children in foster care.

ACS Cannot Identify Incidents That Occur During Visitations under Provider Control

While ACS tracks several key indicators pertaining to the types of substantiated incidents and those responsible for these incidents, the agency does not track data that would allow it to better identify neglect and/or abuse that occurs under circumstances that are primarily controlled by ACS and foster care agencies—namely, incidents involving birth families during visitations that were not ordered by the court and were unsupervised.[10] Compiling and monitoring such information would help ACS assess whether appropriate decisions were made to allow such visits. These could also inform related internal policies.

As indicated earlier, approximately one-quarter of neglect and abuse incidents occurred while the children were in the foster family placement, which is under the direct purview of the foster care providers who report to ACS. In addition, more than 75% of the incidents occurred during visits, trial discharges, or kinship foster homes with the birth family that were scheduled by and at the discretion of foster care providers.

Auditors attempted to determine the extent to which substantiated familial incidents occurred during court-ordered visitations and trial discharges. Although this information is recorded by OSI in the files it compiles for each investigation, ACS does not track or aggregate this information. ACS also does not track the aggregate number of incidents that occur during supervised versus unsupervised visitations.

Consequently, ACS does not track the number of familial incidents of abuse that occurred during non-court-mandated visits and whether they were supervised or unsupervised, and therefore does not know what proportion of all familial incidents occur during such incidents. ACS is not currently evaluating the impact of its decisions―and those of its providers―to allow discretionary visitation and whether such visitation is supervised or unsupervised. This is a missed opportunity to more effectively evaluate its efforts to safeguard children in its care. This information would help ACS identify familial incidents that occur during non-court-ordered visits with the birth families and track the effectiveness of strategies to reduce the number of such incidents.

High Number of Substantiated Neglect and/or Abuse Incidents Involved Birth Families

As part of a review of 172 sampled cases , auditors found that 10 children were named in 17 substantiated cases of neglect and/or abuse that occurred during FYs 2021, 2022, and 2023. Of the 10 children, four were named in two or more incidents.

A review of ACS’ Division of Child Protection investigation data showed that all 17 incidents were familial —meaning that a member of the child’s birth family perpetrated the neglect or abuse during visits, trial discharges or while in kinship foster homes. Of the 10 children, one moved, so their case and the responsibility for subsequent follow-up was transferred to another county outside of ACS’ jurisdiction. For the remaining 15 incidents (involving nine children), the investigative records indicate that none resulted in physical injury to the foster children. Auditors reviewed the measures taken by ACS and the responsible foster care agencies in response to those incidents.

In only one case do the investigative notes indicate that a request was made to suspend the foster child’s visitation with the birth families. The court denied the request. Rather, the investigative notes for these cases indicate that foster care providers attempted to work with birth families by providing additional support services and increasing oversight of these families.

In one case, the investigation determined that the birth parent neglected the child during a court-ordered trial discharge by failing to provide the minimum standard of care, including lack of adequate food, shelter, and clothing. OSI determined in a subsequent incident that the birth parent—who had by then been granted custody —abused the child by using excessive corporal punishment (spanking). The report does not indicate that any injuries were sustained during the incident.

To address these issues, OSI issued a CAP recommending that the foster care agency closely monitor the birth parent through announced and unannounced visits to the home; assist the birth parent in re-engaging in therapeutic services; provide babysitting services; and provide social service funds for food, medical care, and rent. The foster care agency, through prevention services provided by ACS designed to keep children with birth families, continued to monitor the home on a weekly basis.

In another case, the investigation determined that the birth parent abused their child during a routine weekend visitation (i.e., one not ordered by the court ). The birth mother used excessive corporal punishment (forcefully hitting the child on its bottom multiple times). This report also notes that no injuries were sustained during the incident.

To address this incident, OSI issued a CAP recommending that the birth mother be referred to specialized training for children with special needs, and that the foster child be provided individual therapy. The provider agency was then required to follow all recommendations from the therapeutic evaluation.

ACS officials stated that under certain circumstances, ACS or foster care agencies have no say in whether or not a child is allowed to visit their birth family. For example, a judge can order a foster child to be temporarily returned to their birth family on a trial discharge at any stage of the child protective proceeding/court process. A judge can also order a child to be released to their birth family with supervision by ACS if the judge believes that the child will not be in imminent danger if returned to family care.

Overall, the audit found that the measures taken by foster care agencies appeared to be targeted to address the circumstances that led to the neglect and abuse incidents. However, auditors did not have access to the detailed histories of the children and families involved and the audit is unable to offer opinions regarding the adequacy of those measures.

Requisite Site Visits Not Consistently Performed at Sampled Residential Foster Care Facilities

APA is required to conduct site visits to all contracted child welfare residential facilities to ensure they are following ACS Quality Assurance Standards and New York State foster care regulations.

In FY2023, approximately 9% of the children in foster care lived in residential foster care facilities. ACS procedures require APA to visit sites every six months (or approximately every 180 days).  During these visits, the monitors conduct walkthroughs to review ongoing provider activities and interview both staff and children to get their perspectives on the residential programs. All rooms at the facility must be observed to assess for safety hazards, including rooms that youth do not access (i.e., basements, boiler rooms, kitchens, etc.).

APA safety alerts can be generated from site visits if there is a child-specific safety concern identified. The basis of an alert could be, for example, a child saying that they are being bullied by another child or staffer, or if they are not being given prescribed medication. Program staff are required to address immediate safety concerns and repairs as needed.

The auditors sampled four foster care providers that had received safety and risk alerts from PAMS relating to their residential facilities and found that APA did not make the requisite number of site visits to residential facilities during FYs 2021 through 2023 (as of December 2023).^[11] For these facilities, the time between visits consistently exceeded the 180-day target.

On average, the providers conducted visits every 282 days (around nine months). The number of days between visits ranged from 153 days (five months) to 636 days (21 months). For the latter, APA visited a male residential facility in the Bronx on February 3, 2022, but did not return until November 1, 2023. ACS officials stated that all APA site visits were halted to launch the labor-intensive process of implementing new contracts and closing programs during the winter of 2022–2023.

According to documentation summarizing the results of the visits, there were no APA visits at any of the four providers’ residential care facilities (a total of 14 facilities ) for a full year, covering the period September 17, 2022, through September 28, 2023. As of December 31, 2023, four of the 14 facilities (one from each of the four sampled providers) had at least one in-person visit in November or December 2023.

No child safety alerts had been identified during the last visits conducted in FYs 2021 and 2022.[12] Notes from the reviews indicated that children interviewed by APA stated they felt safe at the facilities . APA detailed certain environmental conditions that needed correction (e.g., a foster care provider did not conduct a fire drill; peeling paint on door frames), but none that required safety alerts. Nonetheless, failure to conduct periodic visits increases the risk that unsafe conditions or instances of noncompliance with APA Residential Care Monitor Visit Protocols will not be identified or addressed in a timely manner.

Due to COVID concerns, APA only performed virtual visits of the audit’s selected foster care agencies during FY2021. In FY2022, APA provided all residential care programs the option of having their visits conducted virtually (i.e., via MS Teams or Zoom ) rather than in-person, depending on the degree of severity of COVID-related conditions at the sites/campuses. During that year, APA increased its overall number of residential visits (including both in-person and virtual visits).

Of the four sampled providers, ACS stated that only one (Children’s Village) opted out of in-person visits. According to ACS, APA and Children’s Village discussed the feasibility of conducting in-person visits and, in the interest of keeping youth and staff safe, agreed to the virtual option based on the increased number of COVID cases on the Children’s Village campus and to comply with Mayoral directives and Centers for Disease Control’s guidelines. However, ACS stated that this was discussed during a phone call and as there is no documentation of this agreement, auditors were unable to verify this statement. The auditors also note that APA had still not performed any in-person visits to Children’s Village through the first quarter of FY2023 (October 2023). By this time, the entire City work force had returned to the office and in-person shopping and dining had resumed for a full year.

Periodic in-person visits are essential to better assess these facilities and ensure that children are living in safe conditions. Following the exit conference, ACS officials stated that the Residential Care & Permanency Planning Unit (RCPPU), a unit within ACS’ Division of Family Permanency Services (FPS) conducted weekly site visits to all residential programs (including Children’s Village).^[13] According to the agency, the unit met with the programs to discuss concerns and provide support for youth and the program staff, as needed.

ACS provided the auditors with the dates and internal emails showing the results of the RCPPU visits. A review of the internal emails revealed that the scope and nature of the visits performed by RCPPU did not mirror the duties typically performed by APA, which include observing the rooms at the facility for safety hazards and generating alerts when child-specific safety concerns are identified. Rather, the RCCPU visits were more geared towards casework visits and discussions with facility staff and foster children regarding discharge planning and permanency.[14]

Very Few Visits to Residential Facilities Were Unannounced

Through FY2023, APA residential site visit protocols stated that APA visits may be announced or unannounced (i.e., with or without advance notice). According to ACS, visits to review physical conditions and conduct interviews with staff and youth are announced. Unannounced visits are generally performed on a case-by-case basis to follow up on previously identified safety and program issues.

Unannounced visits are more likely to provide an accurate picture of the conditions at residential sites and reduce the opportunity for staff at those sites to mask deficiencies. Such visits are critical to ensuring that children in foster care are safe and receiving appropriate care.

However, during the two-year period covered in the auditors’ review of 26 residential visits conducted at the four sampled facilities, only one was unannounced . The sole purpose of this site visit (made to the Jewish Child Care Association, located in Pleasantville, NY) was to observe and identify safety-related concerns pertaining to food supply management, meal preparation, and food distribution operations (youth had previously expressed dissatisfaction with the food preparation and options).

ACS officials informed the auditors that the agency modified its policy, requiring APA to conduct at least one announced and one unannounced visit to residential facilities each year (effective July 3, 2023). ACS officials stated that APA staff plan to conduct announced site visits during the first round of the fiscal year (July 2023 to January 2024) and unannounced visits during the second round (March 2024 to June 2024).

PAMS Biannual Case Reviews Completed in a Satisfactory Manner

According to the PAMS Case Record Review Instrument Guidelines, PAMS reviewers are required to conduct reviews of sampled foster care cases for each contracted foster care provider agency twice a year.[15] Reviewers must use the prescribed case record review instrument (a standardized list of questions and response categories) and answer each question accurately based on the documentation in CNNX case records.

The auditors reviewed PAMS responses to a randomly selected stratified sample of 172 of the 1,654 case file reviews completed in FY2021. The sample review focused on the 25 categories included in the review instrument that were devoted to child safety, which include questions related to whether: (1) appropriate actions and/or services were provided in response to the child’s service needs; (2) the case record adequately documents that the child has been placed in a safe and healthy setting; and (3) the child was safe at the time of the review. (All categories are listed in Appendix I.)

The auditors found that the reviewers generally followed guideline instructions by performing safety reviews of foster care case records and answering all safety-related review instrument questions for each case.^[16] Sufficient evidence was found showing that the reviewers conducted systematic reviews of the case records and that their responses were generally consistent and thorough. Subsequent PAMS reviews were conducted for 16 of the 172 sampled cases over the next two fiscal years (nine in FY2022 and seven in FY2023).[17] Auditors again identified no deficiencies in the PAMS reviews.

To determine whether providers took steps to mitigate safety concerns, auditors performed an in-depth review of 30 case files in which PAMS reviewers identified safety issues but determined that safety alerts were not warranted. For each of these cases, auditors determined that there was evidence that providers put steps in place to protect the child, as indicated.

Safety and Risk Alerts Issued as Needed and Appropriately Followed Up

According to the PAMS Case Record Review Instrument Guidelines, a “Safety Alert” must be generated and issued immediately to the foster care provider if a reviewer determines that there is a behavior, condition, or circumstance that potentially places a child in immediate or impending danger or causes serious harm. Safety Alerts should also be issued if the current safety of the child cannot be determined. When a PAMS reviewer identifies no immediate danger but a likelihood that a child may be abused or maltreated in the future, an “Unattended Risk Alert” should be issued.

Upon receipt of an alert, foster care providers are expected to develop safety plans when warranted and intervene with the family within 48 hours to resolve the issue. Providers are required to document in CNNX the actions taken within two business days. The APA Unit follows up on all PAMS-issued alerts. Between March 2020 and June 2021, ACS issued new guidance to providers stating that they were not required to provide responses to risk alerts unless specifically asked to do so by APA, though providers were still required to respond to all safety alerts. ACS issued this directive as part of an effort to focus on immediate safety issues, while adjusting to unforeseen challenges presented by the COVID-19 pandemic.

Auditors’ analysis of the PAMS reviews found that reviewers were unable to determine whether children were safe in 15 of the 172 sampled cases due to a lack of substantive or consecutive contact with the child throughout the review period and/or the identification of an issue that posed an immediate risk concern. In these 15 cases, ACS issued five risk alerts (covering five cases) and 14 safety alerts (covering 10 cases) to foster care providers as a result of PAMS reviews.[18]

In the five risk alerts, PAMS reviewers cited the foster care providers for not providing essential services, such as medical services. In the 14 safety alerts, PAMS reviewers cited the foster care agencies because reviewers were not able to assess the safety of the children for a variety of reasons (e.g., providers lacked contact with children, children were absent without leave).

Of the 14 safety alerts that required a response, the auditors found that the foster care providers responded to 13 of the alerts within two days. For the remaining safety alert, which was issued because of an incomplete Family Assessment Service Plan (FASP), this was due to a delay in document entry that was completed after the PAMS review was completed.[19]

Based on evidence received from the providers, APA subsequently determined that the safety alerts were adequately resolved. The auditors found that the responses and documentation submitted by the providers supported APA’s determination.

Sample-Based Review Found Foster Care Providers Made 95% of Required Caseworker Contacts with Foster Children

According to the PAMS Case Record Review Instrument Guidelines, a PAMS reviewer is required to review and assess a caseworker’s contacts and visits to assess a foster child’s safety, as part of the safety review. In accordance with Federal Social Security Act § 424, each state is required to make 95% of the visits that would occur during the year if each child were visited once every month while in foster care.

There are two key types of contacts used to assess a child’s safety:

• Casework contacts with child, which are defined as individual or group face-to-face contacts between the caseworker and the child.^[20] This includes documented face-to-face contacts, which are used to implement the service task in the family and child’s service plan; and
• Casework contacts with foster parent/childcare worker, which are defined as face-to-face contacts by the caseworker with the person or people immediately responsible for day-to-day care.

PAMS reviewers are required to determine whether the progress notes recorded in CNNX reflect a meaningful observation focused on assessing the behaviors, conditions, and/or circumstances relevant to the contact type, and whether credit should be given for a caseworker’s participation in a specific contact by assessing the quality of the contacts.

In a random sample of 30 foster care cases , auditors reviewed the PAMS-completed assessments and compared them to the progress notes and contact summary reports.[21] Auditors found the PAMS reviews to be thorough, comprehensive, and sufficient to determine the level of interaction and observation during the contacts made by caseworkers.

For the 30 sampled cases, the caseworkers were required to make a total of 342 contacts—178 with the foster children and 164 with the foster parent/childcare worker. Upon review of the case files, PAMS reviewers determined that caseworkers met the minimum criteria in 326 contacts (95%).^[22] Auditors examined the progress notes extracted from CNNX and determined that there was sufficient evidence that these contacts took place.

For the remaining 16 required contacts (5%)—eight with the foster children and eight with the foster parent/childcare workers—PAMS reviews determined there was lack of evidence in the progress notes that the contacts were conducted. However, the PAMS reviews determined that the children were safe at the time of the review because contacts were made prior to, or during the reviews.

To address the concerns for the previously missed contacts, PAMS reviews issued past practice risk alerts to inform them of the policy. Providers are then required to explain the deficiencies. In addition, the providers’ missed contacts put them at risk of falling below the 95% requirement for the review year.

Recommendations

To address the abovementioned findings, the auditors propose that ACS should:

1. Make all reasonable efforts to collect, aggregate, and analyze data that would allow the agency to track substantiated incidents that occur during unsupervised visitations, court-ordered visitations, and trial discharges to better identify and track the occurrence of incidents over which ACS and foster care agencies have a greater degree of control and to evaluate outcomes from Family Court decisions. Use related data and analysis to identify risk factors to inform future decisions and recommendations.
   ACS Response: ACS agreed with this recommendation.

2. Require any parent found to engage in neglect or abuse during a visitation undergo mandatory counseling to address underlying factors (e.g., substance abuse, anger management) that are a contributing cause of the behavior leading to the maltreatment.
   ACS Response: ACS disagreed with this recommendation, stating that, “service recommendations and other outcomes of any substantiated allegation of maltreatment are highly individualized […].”

   Auditor Comment: The auditors understand the need for individualized responses but reiterate the need to ensure parents receive counseling when involved in substantiated instances of abuse or neglect during visitation.

3. Identify the factors contributing to the decrease in the rate of substantiated instances of neglect or abuse and, where feasible, take steps to continue that trend.
   ACS Response: ACS agreed with this recommendation.

4. Ensure that the required number of residential site visits are conducted at all residential care facilities, and that those visits are done in-person and at reasonable intervals.
   ACS Response: ACS agreed with this recommendation.

5. Perform unannounced visits to all foster care agency-run residential care facilities on a regular basis to ensure that the children at those facilities continue to receive appropriate care.
   ACS Response: ACS agreed with this recommendation.

Recommendations Follow-up

Follow-up will be conducted periodically to determine the implementation status of each recommendation contained in this report. Agency reported status updates are included in the Audit Recommendations Tracker available here: https://comptroller.nyc.gov/services/for-the-public/audit/audit-recommendations-tracker/

Scope and Methodology

We conducted this performance audit in accordance with Generally Accepted Government Auditing Standards (GAGAS). GAGAS requires that we plan and perform the audit to obtain sufficient, appropriate evidence to provide a reasonable basis for our findings and conclusions based on our audit objectives. We believe that the evidence obtained provides a reasonable basis for our findings and conclusions within the context of our audit objective(s). This audit was conducted in accordance with the audit responsibilities of the City Comptroller as set forth in Chapter 5, §93, of the New York City Charter. The scope of this audit was July 1, 2020 through June 30, 2023.

To obtain an understanding of the responsibilities and regulations relating to ACS’s monitoring of the safety of children in foster care, the auditors reviewed and used the following documents as their audit criteria:

• New York State Office of Children and Family Services (OCFS) Child Protective Services Manual;
• ACS Foster Care Quality Assurance Standards 2011;
• Title 18 New York Codes, Rules and Regulations (NYCRR) part 441.21 – Casework Contacts; NYCCR part 428.5 Progress Notes; NYCCR part 428.6 Family Assessments and Service Plans;
• Social Security Act § 422;
• ACS PAMS Foster Care Reviewer Guidelines for Fiscal Years 2021 and 2022;
• ACS Performance Monitoring Guidance Update – March 31, 2020;
• ACS Emergency Guidance for Foster Care Providers: Casework Contacts, Family Time and Family Team Conferences, Revised June 4, 2021;
• APA Residential Care Monitor Site Visit Protocols for Fiscal Years 2021 through 2023;
• APA Interim Residential Care Virtual Site Visit Protocol, Updated March 31, 2021;
• ACS Division of Child Protection Casework Practice Requirements Manual; and
• New York State Child Protective Services Manual .

To gain an understanding of how ACS monitors foster care agencies to ensure the safety of the foster children, the auditors evaluated the agency safety mechanisms in place and interviewed ACS officials and staff including:

• Associate Commissioner of ACS’s Office of Policy, Contracts and Monitoring within the Division of Policy Planning and Measurement (DPPM);[23]
• PAMS Director;
• APA Manager; and
• APA Lead Monitor.

In addition, the auditors met with the President and CEO of The Children’s Village foster care agency to obtain a foster care provider’s point of view of the foster care process.

Additionally, the auditors interviewed the Associate Commissioner and Executive Director from ACS’ Office of Research and Analytics (ORA), the unit tasked with calculating the sample size of eligible foster children from the general population to be reviewed by PAMS annually, to obtain an understanding of the foster children population and sample size determination process.

To obtain an understanding of the NYS OCFS CNNX system (New York State’s case management records system used by foster care agencies as their system of record to document each foster child’s case and related activities), and ACS’ review (including PAMS), the auditors conducted a walkthrough of the system led by the CNNX Senior Business Manager under OCFS.

The auditors obtained electronic data records on a population consisting of 1,654 Fiscal Year 2021 foster care cases that were reviewed by PAMS ; 1,374 Office of Special Investigations (OSI) substantiated cases of child abuse covering 1,573 incidents while children were in foster care from July 1, 2020 through July 30, 2023.

The auditors conducted various data reliability tests of the dataset to check for questionable entries (including duplicates, blank fields, and inconsistent information). In addition, to determine whether the Fiscal Year 2021 PAMS reviews were performed and complete, the auditors first stratified the population of 1,654 of cases reviewed by program type and randomly selected 172 cases for review, using the RAT-STATS statistical sampling software. The 172 case files raw data was evaluated for indications that all 25 PAMS Safety Case Record Review Questions were addressed, including whether controlling interventions and safety plans, if applicable, were evaluated. The auditors also reviewed the responses and determined whether: (1) appropriate actions and/or services were provided in response to the child service needs; (2) the case record document adequately documented the child was placed in a safe and healthy setting; and (3) the child was safe at the time of the review.

From the auditors’ analysis of PAMS reviewer answers to the review questions, 15 cases were identified in which risk or safety alerts were required to be issued. For these 15 cases, the auditors reviewed documentation to ascertain whether: (1) ACS appropriately notified the foster care agencies of the identified risk and safety alerts; (2) there were the required timely response to the alerts by the foster care agencies; (3) there was adequate APA follow-up to ensure that the foster care agencies addressed the risk and safety issues; and (4) any of the identifies issues re-occurred during subsequent PAMS reviews, if applicable.

To test the accuracy and thoroughness of the PAMS Fiscal Year 2021 Safety and Risk Assessment reviews, auditors randomly selected 30 cases for in-depth analysis. The auditors attempted to corroborate PAMS reviewers’ responses to the safety record questions with their own review using the children’s case file information, including the documented FASPs and Case Progress Notes (recorded in CNNX) for each of the 30 cases. Further, to determine whether casework contacts were held with the children and foster parents, the auditors reviewed and verified PAMS recorded case work contacts documented in each of the Case Progress Notes and the Contact Summary report.[24]

To determine the status of the 172 Fiscal Year 2021 PAMS reviewed foster cases as of November 20, 2023, the auditors obtained information on each case including whether : (1) the child’s case was discharged; (2) the child was returned to birth parents or adopted; or (3) the child’s case was still active. The auditors analyzed data on 45 sampled cases (the 15 cases with alerts and 30 cases in which auditors performed in-depth reviews). From these 45 cases, there were 16 cases that had subsequent PAMS reviews in 2022 and 2023. To evaluate the accuracy of PAMS case record reviews and to determine whether additional alerts were issued by PAMS reviewers, the auditors reviewed responses to all 25 PAMS Safety Case Record Review Questions for all 16 cases. Furthermore, to ascertain whether casework contact requirements were satisfied with the children and foster parents, the auditors evaluated the required case contact raw data received from ACS.

To determine whether the APA unit conducted the required in-person and/or virtual site visits to residential facilities during the period from Fiscal Years 2021 through 2023, auditors selected 4 of 14 residential foster care providers based on PAMS-identified alerts issued to those providers during Fiscal Year 2021. The auditors analyzed APA Residential Site Visit Summary Reports that APA monitors drafted to document the site visits.

To determine whether any of the Fiscal Years 2021 through 2023 OSI substantiated child abuse incidents were associated with any of the sampled 172 PAMS-reviewed cases performed in Fiscal Year 2021, the auditors compared the data for both populations and identified 10 foster care cases involving 17 incidents. The auditors obtained detailed information on these 17 incidents to determine what actions were taken on the part of ACS and the foster care agencies to resolve incidents and ensure that children involved were safe.

The results of the above tests, while not projectable to their respective populations whenever a sample was used, provided a reasonable basis for the auditors to evaluate whether ACS adequately monitored the children in its foster care program.

Appendix I

See attachment.

Addendum

See attachment.

---

Endnotes

[1] During Fiscal Years 2020 through 2023, 76.4% of the more than 2,000 cases of substantiated incidents involved neglect only.

[2] Title 18 § 441.21

[3] According to New York Social Services Law § 384-a, when a foster care agency determines to terminate or limit visiting rights between a parent or guardian and the child is voluntarily placed in foster care, parental visitation may not be terminated or limited except by court order in a proceeding in which the parent or guardian was a party. Visitation is to continue until a court order is obtained, except in cases of imminent danger to the child’s life, health, and safety. In such cases, the authorized agency may terminate or limit the parent’s visitation without a court order. On the same day visitation is terminated or limited, the foster care agency must commence a court action. If action is already before the court, ACS shall seek an order of the court as if the child had been taken into protective custody. The above-referenced provisions do not apply where the parent or guardian had agreed to such limitation in the voluntary placement agreement.

[4] As a general rule, a child should not be trial or final discharged without first having experienced successful overnight and weekend visits with the parent over a period of time.

[5] The OSI data provided by ACS regarding alleged and substantiated incidents of neglect and/or abuse did not distinguish between kinship and traditional (non-relative) foster care placements. The instances of neglect and/or abuse for foster children placed with kin are included in the foster care numbers, not in the familial numbers.

[6] DCP is comprised of several offices including: (1) the Office of Special Investigations, which investigates allegations where the foster parent is the respondent in the allegation , and (2) the Child Protective Borough Offices, which investigates allegations concerning the child’s birth family.

[7] Kinship foster care placements are overseen and monitored by foster care agencies as with traditional (non-relative) foster care placements.

[8] In its response to the Draft Report, ACS argued that the audit misrepresents the number of children who were maltreated while in foster care. However, a comparison of the figures from this report with those in the ACS response show the difference to be negligible, with the latter showing a higher number of incidents (2,258) and number of children involved in those incidents (1,972). The figures used in this report represent a summary of the detailed data (e.g., listing of individual incidents, type of maltreatment) provided by ACS during the course of the audit. Conversely, ACS provides no support for the summary data contained in its response, so the auditors are unable to independently corroborate those figures.

[9] These were implemented too recently to assess during the current audit.

[10]   In its response to the Draft Report, ACS disagreed with this finding, stating that the details about cases are fully documented and understood by personnel involved in the respective cases. However, this information is not captured in the aggregate data and can only be ascertained through independent reviews of individual case records.

[11] A total of 14 foster care providers operated residential care facilities during the audit scope period. The four sampled foster care providers that were issued alerts based on PAMS reviews were as follows: Children’s Village, Good Shepherd Services (GSS), Jewish Child Care Association (JCCA), and St. John’s Residence.

[12] Although no safety alerts were issued at these site visits, APA monitors did identify physical site safety concerns such as fire extinguishers that were missing or needed to be replaced, peeling paint, and blocked exits that pose potential safety risks in case of an emergency.

[13] The RCPPU aims to provide technical assistance to residential providers in support of timely permanency planning, response to social environment and effective transitions.

[14] In its response to the Draft Report, ACS disagreed with this finding, stating that the APA semi-annual residential site visits are not required by law or state regulations. However, they are required according to ACS’ written procedures, which served as the basis of the audit’s criteria and analysis.

[15] There are two rounds of reviews each year: Round 1 is conducted from September to February, and Round 2 is conducted from February/March through June.

[16] The 172 cases were selected based on a stratified population of cases by program type. There were 126 program types, including Family Foster Care, Residential Care, and Treatment Family Foster Care.

[17] From the 172 sampled cases, auditors selected for review 45 cases of which they either performed in-depth reviews (30 cases) or had safety alerts (15 cases). There were 16 of the 45 cases that had subsequent PAMS reviews either in Fiscal Year 2022 or 2023.

[18] PAMS reviewers sent safety alerts and risk alerts to the following foster care providers: The Children’s Village, St. John’s Residence for Boys, Children’s Aid Society, SCO Family Services, Jewish Child Care Association, Good Shepherd Services, Abbott House, Graham Windham, Seaman’s Society, Little Flower Children Family Services of New York.

[19] The FASP is the primary tool for documentation of all information and casework activity related to an ongoing child welfare services case in NYS. It provides a uniform framework for gathering and documenting assessment information, supporting and recording decisions, and developing and approving plans to address a family’s most significant child welfare issues, needs, and concerns.

[20] Starting in March 2020, video conferences and phone calls were allowed by OCFS as a substitute for in-person, face-to-face contacts because of the COVID pandemic. As of June 30, 2023, only in-person contacts are allowed.

[21] The contact summary reports are used by the reviewer to summarize the number of casework contacts required, made or missed for the review period.

[22] Sampled case contacts consisted of contacts between caseworker and child and caseworker and foster parents.

[23] DPPM includes the PAMS and APA units.

[24] The Case Summary Report summarizes all the contacts for each case under review. It includes the number of required contacts completed, the credited contacts, the non-credited contacts and those that were attempted for the period under review. The report also captures the month(s) contacts were missed and the number of days between missed contacts.