[Senate Prints 115-18]
[From the U.S. Government Publishing Office]

  115th Congress  }                                   {   S. Prt
                          COMMITTEE PRINT
  1st  Session    }                                   {   115-18


                     AN EXAMINATION OF FOSTER CARE

                         IN THE UNITED STATES

                     AND THE USE OF PRIVATIZATION


                      Prepared by the Staff of the

                          COMMITTEE ON FINANCE
                          UNITED STATES SENATE

                        Orrin G. Hatch, Chairman
                       Ron Wyden, Ranking Member

                              OCTOBER 2017

            Printed for the use of the Committee on Finance

                      AND THE USE OF PRIVATIZATION

115th Congress     }                                      {     S. Prt
 1st Session       }         COMMITTEE PRINT              {     115-18


                     AN EXAMINATION OF FOSTER CARE

                         IN THE UNITED STATES

                     AND THE USE OF PRIVATIZATION


                      Prepared by the Staff of the

                          COMMITTEE ON FINANCE
                          UNITED STATES SENATE

                        Orrin G. Hatch, Chairman
                       Ron Wyden, Ranking Member


                              OCTOBER 2017

            Printed for the use of the Committee on Finance
 26-354 PDF               WASHINGTON : 2017       
 For sale by the Superintendent of Documents, U.S. Government Publishing Office,
Internet:bookstore.gpo.gov. Phone:toll free (866)512-1800;DC area (202)512-1800
  Fax:(202) 512-2104 Mail:Stop IDCC,Washington,DC 20402-001                

                          COMMITTEE ON FINANCE

                     ORRIN G. HATCH, Utah, Chairman

CHARLES E. GRASSLEY, Iowa            RON WYDEN, Oregon
MIKE CRAPO, Idaho                    DEBBIE STABENOW, Michigan
PAT ROBERTS, Kansas                  MARIA CANTWELL, Washington
MICHAEL B. ENZI, Wyoming             BILL NELSON, Florida
JOHN CORNYN, Texas                   ROBERT MENENDEZ, New Jersey
JOHN THUNE, South Dakota             THOMAS R. CARPER, Delaware
RICHARD BURR, North Carolina         BENJAMIN L. CARDIN, Maryland
JOHNNY ISAKSON, Georgia              SHERROD BROWN, Ohio
ROB PORTMAN, Ohio                    MICHAEL F. BENNET, Colorado
PATRICK J. TOOMEY, Pennsylvania      ROBERT P. CASEY, Jr., Pennsylvania
DEAN HELLER, Nevada                  MARK R. WARNER, Virginia
TIM SCOTT, South Carolina            CLAIRE McCASKILL, Missouri

                     A. Jay Khosla, Staff Director

              Joshua Sheinkman, Democratic Staff Director

                          INVESTIGATIVE STAFF

Senate Finance Committee Majority    Senate Finance Committee Minority 
Staff                                Staff
KIMBERLY BRANDT, Chief Oversight     IAN M. NICHOLSON, Investigator
Counsel                              EMILY DOUGLAS, SRCD/AAAS 
DONALD ABBOTT, U.S. Secret Service   Congressional Fellow
Detailee                             DAVID M. BERICK, Chief 
BECKY SHIPP, Health and Human        Investigator
Services Policy Advisor              LAURA BERNTSEN, Chief Human 
JOHN A. CARLO, Senior Oversight      Services Advisor
Counsel                              DOUGLASS V. CALIDAS, Legislative 

                            C O N T E N T S



Executive Summary................................................     1

I. Introduction..................................................     2

II. Overview of the Problem and Justification for the Committee 
  Investigation..................................................     3

    A. Child Abuse and Neglect and Foster Care...................     3

    B. Federal Financing of Foster Care Services.................     3

    C. Recent Committee History on Foster Care and Related Issues     4

    D. Initial Media Reports About For-Profit Foster Care and 
      Children's Deaths..........................................     5

    E. MENTOR and Private Foster Care Agencies...................     6

    F. Therapeutic Foster Care...................................     8

III. The Committee's Investigation and Surveys of the States: 
  Foster Care Services and Performance...........................     8

    A. The 50-State Overview Letter and Request..................     8

    B. The 5-State In-Depth Letter and Request...................     9

    C. U.S. HHS Child and Family Services Reviews................    10

    D. The MENTOR Letter and Request.............................    10

IV. Findings From the Committee's Investigations and Surveys.....    11

    A. Use of Private Child Welfare Services.....................    11

    B. Background Checks.........................................    13

    C. Child Welfare Workforce Operations and Concerns...........    14

    D. Physical and Mental Well-Being of Children in Foster Care.    16

    E. Failure to Identify and Respond to Risk to Children.......    17

    F. Actions When Maltreatment Is Substantiated in Foster Homes    18

    G. MENTOR Incident Reports...................................    20

    H. MENTOR Mortality Report...................................    23

    I. Financial Settlements From MENTOR.........................    26

V. Oversight of Child Welfare Services and Protecting Vulnerable 
  Children.......................................................    27

VI. Conclusions..................................................    29

VII. Recommendations.............................................    31

    Recommendations for States and Tribes........................    31

    Recommendations for the Department of Health and Human 
      Services (HHS).............................................    31

    Recommendations for Congress.................................    32

Appendix A--50-State Overview Letter.............................    35

Appendix B--Responses to 50-State Overview Letter................    38

    Exhibit 1--Response From Alabama, Department of Human 
      Resources..................................................    39

    Exhibit 2--Response From Alaska, Department of Health and 
      Social Services............................................    47

    Exhibit 3--Response From Arkansas, Office of the Governor....    51

    Exhibit 4--Response From California, Health and Human 
      Services Agency, Department of Social Services.............    55

    Exhibit 5--Response From Colorado, Office of the Governor....    62

    Exhibit 6--Response From Connecticut, Department of Children 
      and Families...............................................    66

    Exhibit 7--Response From Delaware, Department of Services for 
      Children, Youth, and Their Families........................    69

    Exhibit 8--Response From Guam, Department of Public Health 
      and Social Services........................................    74

    Exhibit 9--Response From Hawaii, Office of the Governor......    79

    Exhibit 10--Response From Illinois, Department of Children 
      and Family Services........................................    82

    Exhibit 11--Response From Indiana, Department of Child 
      Services...................................................    88

    Exhibit 12--Response From Iowa, Department of Human Services.    93

    Exhibit 13--Response From Kansas, Department for Children and 
      Families...................................................    97

    Exhibit 14--Response From Kentucky, Cabinet for Health and 
      Family Services............................................   101

    Exhibit 15--Response From Maryland, Department of Human 
      Resources..................................................   107

    Exhibit 16--Response From Massachusetts, Office of the 
      Governor...................................................   114

    Exhibit 17--Response From Minnesota, Department of Human 
      Services...................................................   120

    Exhibit 18--Response From Nebraska, Office of the Governor...   126

    Exhibit 19--Response From New Hampshire, Department of Health 
      and Human Services.........................................   130

    Exhibit 20--Response From New Jersey, Department of Children 
      and Families...............................................   134

    Exhibit 21--Response From New Mexico.........................   139

    Exhibit 22--Response From New York, Office of Children and 
      Family Services............................................   153

    Exhibit 23--Response From North Dakota.......................   158

    Exhibit 24--Response From Oklahoma, Department of Human 
      Services...................................................   162

    Exhibit 25--Response From Oregon, Office of the Governor.....   166

    Exhibit 26--Response From Pennsylvania, Department of Human 
      Services...................................................   172

    Exhibit 27--Response From South Dakota, Office of the 
      Governor and the Department of Social Services.............   179

    Exhibit 28--Response From Tennessee, Department of Children's 
      Services...................................................   184

    Exhibit 29--Response From Texas, Department of Family and 
      Protective Services........................................   189

    Exhibit 30--Response From Utah, Department of Human Services.   201

    Exhibit 31--Response From Washington, Department of Social 
      and Health Services........................................   206

    Exhibit 32--Response From West Virginia, Department of Health 
      and Human Resources........................................   209

    Exhibit 33--Response From Wisconsin, Department of Children 
      and Families...............................................   213

    Exhibit 34--Response From Wyoming, Department of Family 
      Services...................................................   219

Appendix C--5-State In-Depth Letters.............................   222

    Exhibit 1--Letter to Georgia, Department of Community Health.   223

    Exhibit 2--Letter to Illinois, Department of Children and 
      Family Services............................................   228

    Exhibit 3--Letter to Massachusetts, Department of Children 
      and Families...............................................   233

    Exhibit 4--Letter to Maryland, Department of Human Resources.   238

    Exhibit 5--Letter to Texas, Department of Family and 
      Protective Services........................................   243

Appendix D--Congressional Research Service Memorandum on the U.S. 
  HHS Children and Family Services Reviews.......................   248

Appendix E--Letter to The MENTOR Network.........................   275

Appendix F--MENTOR Level 4 Incident Reports......................   282

Appendix G--MENTOR Mortality Report and Backup Data..............   606

    Exhibit 1--Foster Care Mortality Analysis....................   607

    Exhibit 2--Backup Data and Explanation of Mortality Analysis.   613

                     AN EXAMINATION OF FOSTER CARE

                          IN THE UNITED STATES

                      AND THE USE OF PRIVATIZATION

                           EXECUTIVE SUMMARY

    Foster care placements for children who are victims of 
abuse and neglect have historically been managed by a 
combination of private and public resources. However, the need 
for specialized foster care services and a shortage of foster 
care homes in recent years has led to the privatization of many 
core foster care services. Today, both non-profit and for-
profit private agencies contract with and provide foster care 
services on behalf of State agencies. In 2015, 671,000 children 
in the United States were provided out-of-home foster care 
services. There are no official statistics on what proportion 
of these children received contracted foster care, case 
management, or other services. State child welfare agencies 
report they have procedures in place to monitor child welfare 
providers' performance and outcomes. But this investigation 
conducted by the bipartisan staff of the U.S. Senate Finance 
Committee shows that these policies are not always followed; 
exceptions are made, waivers are granted, profits are 
prioritized over children's well-being, and sometimes those 
charged with keeping children safe look the other way. High 
turnover among staff sometimes makes it impossible to develop 
case plans to ensure that children are ``on-track.'' Foster 
parents with questionable backgrounds, who lack the skills to 
provide care to vulnerable children, are given licenses to 
parent challenging children, and these children are then 
inadequately monitored. The outcome of this investigation shows 
that the child welfare system does not always protect children. 
The data collection and oversight structures at both the State 
and Federal levels make it difficult and sometimes impossible 
to monitor the operations of the child welfare system, as well 
as its private contractors.
    A recent bout of national media attention concerning 
questionable behavior by private for-profit agencies, abuse and 
neglect by foster parents working for those providers, and in 
some instances abuse and neglect which caused children's 
deaths, led the Finance Committee to investigate this issue. As 
the Finance Committee has primary jurisdiction over Federal 
child welfare and foster care funding and policy (largely 
through the Social Security Act), the Committee launched an 
investigation in April 2015 to examine the privatization of 
foster care services. One specific private company, The MENTOR 
Network, one of the largest for-profit providers of foster care 
services in the United States, was used as a case study to 
highlight the problems that exist with the privatization of 
human services. This report documents the findings of this 
investigation and reveals problems with child welfare 
contracting practices as well as public agency oversight of 
such contracts and services.
    The investigation was conducted by collecting information 
from public child welfare agencies across the Nation concerning 
their general policies and practices, including how they 
contract with and monitor private agencies. The Committee also 
gathered information from The MENTOR Network, specifically, by 
reviewing incident reports about the deaths of children in the 
company's care, an internal ``mortality report,'' legal 
settlements, case notes, foster parent applications, and other 
related documentation.
    The Committee staff concluded that children who are under 
the legal authority of their State, yet receive services from 
private for-profit agencies, have been abused, neglected, and 
denied services. The very agencies charged with and paid to 
keep foster children safe too often failed to provide even the 
most basic protections, or to take steps to prevent the 
occurrence of tragedies. In MENTOR's case in particular, 
investigations into fatalities were never followed up after the 
fact; autopsy reports which were pending years ago were 
excluded from files; and the vast majority of children who died 
were not the subject of internal investigations, even when 
their deaths were unexpected. The MENTOR Network issued a 
report which falsely claimed that its death rates are in line 
with national death rates and the rates of death among all 
children in the foster care system. Moreover, families of these 
and other victims of inadequate care have received millions of 
dollars in financial settlements, significant enough for The 
MENTOR Network to receive less favorable terms from its 
    As the role of private for-profit and non-profit providers 
of foster care services has grown, oversight of these entities 
by State agencies--as well as Federal oversight of the States--
has been inadequate. The Finance Committee staff has made 
recommendations to HHS, the States, and to Congress addressing 
these shortcomings.

                            I. INTRODUCTION

    The privatization of foster care, and specifically for-
profit foster care, has been a growing trend in the delivery of 
child welfare services over the past few decades.\1\ Recent 
national media attention concerning questionable performance by 
these private agencies, including abuse and neglect by foster 
parents working for these agencies, and in some instances abuse 
and neglect which led to children's deaths, led the Senate 
Finance Committee (hereinafter, the ``Committee'' or ``SFC'') 
to investigate these issues.
    \1\ Flaherty, C., Collins-Camargo, C. and Lee, E., ``Privatization 
of child welfare services: Lessons learned from experienced states 
regarding site readiness assessment and planning.'' Children and Youth 
Services Review, Vol. 30, No. 7, pp. 809-820, http://
    The Committee has jurisdiction over Federal child welfare 
and foster care funding and policy in the United States, and 
thus has a responsibility in ensuring children receive the most 
suitable placements to appropriately support their healthy 
development. Chairman Hatch and Ranking Member Wyden launched 
an investigation in April 2015 to examine the privatization of 
foster care services within the context of the larger child 
welfare system. One specific private company, The MENTOR 
Network (hereinafter, ``MENTOR''), was used as a case study in 
order to highlight some of the problems that exist with the 
privatization of human services in general. At the time the 
Committee initiated the investigation, MENTOR reported it was 
the ``leading provider of home- and community-based health and 
human services to must-serve individuals and families.'' \2\ 
MENTOR continues to make this claim today.\3\ This report 
documents the findings of this investigation and reveals 
problems with child welfare contracting practices as well as 
public agency oversight of such contracts and services.
    \2\ Civitas Solutions, Inc., U.S. Securities and Exchange 
Commission, 2014 10-K filing for the fiscal year ending September 30, 
2014, https://www.sec.gov/Archives/edgar/data/1608638/
    \3\ Civitas Solutions, Inc., U.S. Securities and Exchange 
Commission, 2017 10-Q filing for the quarterly period ending March 31, 
2017, https://www.sec.gov/Archives/edgar/data/1608638/


               A. Child Abuse and Neglect and Foster Care

    According to the National Child Abuse and Neglect Data 
System (NCANDS), in 2015 there were 683,000 children who were 
victims of abuse or neglect in the United States, representing 
a rate of 9.2 victims per 1,000 U.S. children.\4\ In instances 
where children are abused or neglected and cannot safely remain 
at home or with relatives, they are placed in foster care. 
According to the Adoption and Foster Care Analysis and 
Reporting System (AFCARS), 671,000 children were served by the 
foster care system in 2015 either because they were already in 
foster care, or because they newly entered foster care that 
year.\5\ When children are placed in foster care, they are most 
often placed in one of three settings: 
nonrelative foster care (45%), relative/kinship care (30%), or 
institutions/group homes (14%).\6\ Foster care placements can 
occur either through a child's State or public child welfare 
agency, or through private entities that contract with public 
child welfare agencies to find placements for children. These 
private organizations can be non-profit or for-profit agencies. 
The private agencies that were the focus of this report 
(specifically MENTOR) provide non-relative foster care for 
children outside of institutional settings.
    \4\ U.S. Department of Health and Human Services, Administration 
for Children and Families, Children's Bureau, ``Child Maltreatment 
2015,'' Report, January 2017, https://www.acf.hhs.gov/sites/default/
    \5\ U.S. Department of Health and Human Services, Administration 
for Children and Families, Children's Bureau, ``Trends in Foster Care 
and Adoption,'' Published: June 30, 2016, last reviewed: March 13, 
2017, https://www.acf.hhs.gov/cb/resource/trends-in-foster-care-and-
    \6\ U.S. Department of Health and Human Services, Administration 
for Children and Families, Children's Bureau, ``Foster Care Statistics 
2015,'' Report, pp. 2-4, March 2017, https://www.childwelfare.gov/

              B. Federal Financing of Foster Care Services

    The Committee has jurisdiction over many areas of public 
finance including the Internal Revenue Code, major health-care 
programs such as Medicare and Medicaid, and Social Security. 
Federal child welfare policy is largely guided by the Social 
Security Act, originally established in 1935.\7\ Foster care 
services are partly funded through titles IV-B and IV-E of the 
Social Security Act. In addition, services and supports for 
children and their families, including foster care, can be 
funded through title XX and title IV-A of the Social Security 
Act. Federal assessment and monitoring of State child welfare 
systems are also covered by title IV-E. Thus, potential misuse 
or mismanagement of these funds to place children in foster 
homes where they may potentially be unsafe is of keen interest 
to the Committee.
    \7\ U.S. Social Security Administration, Social Security Act of 
1935, https://www.ssa.gov/history/35act.html.

              C. Recent Committee History on Foster Care 
                           and Related Issues

    The Committee and its members have a long history of 
working to improve the State and Federal child welfare systems. 
For decades, the child welfare advocacy and provider 
communities, as well as families and children impacted by the 
system, have recognized that the government is no substitute 
for a family when it comes to raising children. Frequent news 
stories highlighting traumatic experiences children in foster 
care sometimes face have led Congress to take steps to improve 
the system in two key ways: first, to do more to ensure that 
foster care is an intervention used only when in the best 
interest of the child; and second, to ensure that when foster 
care is necessary, it is of the highest possible quality and 
promotes normalcy.
    In the 114th Congress, the Committee held several hearings 
and roundtable discussions related to the child welfare system, 
its incentives, and its funding structure. Specifically, in May 
2015, the Committee held a hearing entitled ``No Place to Grow 
Up: How to Safely Reduce Reliance on Foster Care Group Homes.'' 
\8\ The purpose of the hearing was to examine how Congress can 
best address the challenges facing foster children and protect 
them from the unfit environment and risk of sex trafficking 
found in group homes. In August 2015, the Committee held a 
hearing entitled, ``A Way Back Home: Preserving Families and 
Reducing the Need for Foster Care.'' \9\ Its purpose was to 
explore safe alternatives to foster care and better understand 
the interventions, services, and funding mechanisms States and 
Tribes are using--or would like to use--to help keep families 
and children safely together. And in February 2016, the 
Committee held a hearing entitled ``Examining the Opioid 
Epidemic: Challenges and Opportunities.'' \10\ The purpose of 
this hearing was to examine the opioid abuse epidemic and its 
effect on the health and child welfare systems, as well as to 
consider solutions. Committee members heard testimony detailing 
the links between rising opioid use and fatalities and the 
corresponding strain on State foster care systems.
    \8\ U.S. Senate Finance Committee, ``No Place to Grow Up: How to 
Safely Reduce Reliance on Foster Care Group Homes,'' Committee Hearing, 
114th Congress, May 19, 2015, https://www.finance.senate.gov/imo/media/
    \9\ U.S. Senate Finance Committee, ``A Way Back Home: Preserving 
Families and Reducing the Need for Foster Care,'' Committee Hearing, 
114th Congress, August 4, 2015, https://www.
    \10\ U.S. Senate Finance Committee, ``Examining the Opioid 
Epidemic: Challenges and Opportunities,'' Committee Hearing, 114th 
Congress, February 23, 2016, https://www.finance.
    As a result of these hearings and working closely with 
stakeholders from the foster care provider community, State 
groups, advocates, and current and former foster youth, the 
chairman and ranking member developed a bipartisan proposal, 
``The Family First Act.'' Working with House Ways and Means 
Committee Chairman Brady and Ranking Member Levin, Chairman 
Hatch and Ranking Member Wyden introduced the bipartisan/
bicameral ``Family First Prevention Services Act of 2016'' 
(H.R. 5456/S. 3065).\11\ The House of Representatives passed 
H.R. 5456 by voice vote on June 21, 2016.\12\ The goal of this 
legislation was to increase the availability of prevention 
services so that more children can stay at home with their 
families and avoid the trauma associated with foster care when 
safely possible. The legislation also aimed to reduce the 
unnecessary use of congregate care and group homes.
    \11\ S. 3065: ``Family First Prevention Services Act of 2016,'' 
114th Congress, https://www.congress.gov/bill/114th-congress/senate-
bill/3065; see also: Kelly, John, ``Massive child welfare Finance bill 
planned for 2016,'' The Chronicle of Social Change, December 18, 2015, 
    \12\ H.R. 5456: ``Family First Prevention Services Act of 2016, 
Actions/Overview,'' 114th Congress, https://www.congress.gov/bill/
    While the legislation ultimately was not enacted into law, 
the Committee continues to work to advance the goals of the 
legislation. The findings of this investigation underscore the 
importance of holding States accountable for their child 
welfare system management and outcomes, as well as providing 
States with the tools necessary to improve their capacity to 
both prevent foster care (when it is possible to do so without 
jeopardizing the safety of children) and to ensure that 
children who enter foster care are safe from harm.

       D. Initial Media Reports About For-Profit Foster Care and 
                           Children's Deaths

    In 2015, BuzzFeed News \13\ and Mother Jones \14\ reported 
similar stories concerning the private, for-profit foster care 
company, MENTOR. The reports provided evidence of a company 
that prioritizes profits over children's well-being; a company 
that skirted corners when screening foster parents, that 
increased social workers' caseloads, that hired unlicensed 
workers, and whose primary mission was to ``fill beds'' in 
order to increase company profits. According to these reports, 
children were placed in homes with individuals who had been 
convicted of kidnapping and other serious crimes, with parents 
who had substance abuse problems, and in homes where caretakers 
had previous ``failed'' foster care placements. Some children 
were deprived of emotional care, were sexually abused, and even 
beaten to death in their foster homes. The news articles 
included both allegations of wrong-doing and information that 
was substantiated through criminal trials and lawsuit 
settlements to families of the children who had been harmed.
    \13\ Roston, Aram and Singer-Vine, Jeremy, ``Fostering Profits,'' 
BuzzFeed News, February 20, 2015, https://www.buzzfeed.com/aramroston/
    \14\ Joseph, Brian, ``The Brief Life and Private Death of 
Alexandria Hill,'' Mother Jones, October 26, 2015, http://
    The Committee received numerous questions and expressions 
of concern from the public as a result of these news accounts. 
The chairman and ranking member also felt strongly that the 
allegations in the press accounts deserved a more thorough 
examination. As a result, the chairman and ranking member 
directed the majority and minority oversight teams to 
investigate the issue of privatization within the foster care 
system using MENTOR as a case study, as it is among the largest 
providers of private foster care services in the United 
States.\15\, \16\ Because the Committee did not 
conduct an in-depth investigation of other providers, direct 
comparisons cannot be made with other private providers. 
However, it did collect State-level data on the performance of 
other for-profit and non-profit providers, and it is also 
notable that MENTOR is by no means alone when it comes to 
negative attention and questionable practices. For example, the 
following headlines demonstrate that contracting practices, 
provider quality, and inadequate oversight issues are rampant 
across States and providers:
    \15\ McBeath, Bowen, Collins-Camargo, Crystal, and Chuang, 
Emmeline, ``Portrait of Private Agencies in the Child Welfare System: 
Principal Results From the National Survey of Private Child and Family 
Serving Agencies,'' National Quality Improvement Center on the 
Privatization of Child Welfare Services, September 2011, http://
    \16\ Civitas Solutions, Inc., 2014 10-K filing (September 30, 
2014). See Footnote 2.

         L``Foster care scandal deepens: `Every single staff 
        person has a criminal record' '' (The Oregonian, 
        January 9, 2016).\17\
    \17\ Theriault, Dennis C., ``Foster care scandal deepens: `Every 
single staff person has a criminal record,' '' The Oregonian, January 
9, 2016, http://www.oregonlive.com/politics/index.ssf/2016/01/
         L``Federal Judge: Texas Foster Care System Violates 
        Children's Rights'' (The Texas Tribune, December 17, 
    \18\ Walters, Edgar and Ramshaw, Emily, ``Federal Judge: Texas 
Foster Care System Violates Children's Rights,'' The Texas Tribune, 
December 17, 2015, https://www.texastribune.org/2015/12/17/judge-
         L``Suit Alleges Widespread Deficiencies in South 
        Carolina Foster Care'' (The New York Times, January 12, 
    \19\ Blinder, Alan, ``Suit Alleges Widespread Deficiencies in South 
Carolina Foster Care,'' The New York Times, January 12, 2015, https://
         L``Report Finds `Blatant Lack of Oversight By DCF' In 
        Licensing of Foster Home Where Toddler Died'' (WBUR 
        News, October 1, 2015).\20\
    \20\ Conway, Abby Elizabeth, ``Report Finds `Blatant Lack of 
Oversight by DCF' in Licensing of Foster Home Where Toddler Died,'' 
WBUR News, October 1, 2015, http://www.wbur.org/news/2015/10/01/auburn-
         L``State Must Step up on Foster Care Deaths'' (The 
        Courier, June 11, 2017).\21\
    \21\ Hines, Doug, ``State Must Step up on Foster Care Deaths,'' The 
Courier, June 11, 2017, http://wcfcourier.com/opinion/editorial/state-
         L``Minnesota Faces Penalties for Failed Placements of 
        Foster Children'' (The Star Tribune, February 10, 
    \22\ Serres, Chris, ``Minnesota faces penalties for failed 
placements of foster children,'' The Star Tribune, February 10, 2014, 

               E. MENTOR and Private Foster Care Agencies

    MENTOR, headquartered in Boston, Massachusetts, is a for-
profit service agency that describes itself as a ``national 
network of local health and human services providers in 35 
States offering an array of quality, community-based services 
to adults and children[. . .].'' \23\ It is owned by Civitas 
Solutions, Inc., a publicly traded company. Civitas is majority 
owned (approximately 68%) by Vestar Capital Partners and 
management investors.\24\ Public investors hold roughly 32% of 
the company according to the company's filings with the 
Securities and Exchange Commission (SEC).\25\ The group 
emphasizes its work with higher-risk youth in foster care, 
particularly those with intellectual or developmental 
disabilities, or who are medically fragile.
    \23\ ``The MENTOR Network.'' Available on the company's website, 
    \24\ On its own, Vestar owns 53% of the company's shares according 
to the Civitas Solutions, Inc. U.S. Securities and Exchange Commission 
2016 10-K filing for the fiscal year ending September 30, 2016, https:/
    \25\ Civitas Solutions, Inc., 2014 10-K filing (September 30, 
2014). See Footnote 2.
    According to information reported in MENTOR's SEC filings 
for 2014, it was the leading provider of human services to 
29,100 clients in 36 States during that year--12,600 in 
residential settings and 16,500 in non-residential 
settings.\26\ With regard to the foster care population, which 
is the focus of the Committee's investigation, MENTOR served 
10,300 at-risk children, adolescents, and their families in 18 
different States in 2014. By way of comparison, according to a 
2011 national survey of non-profit and for-profit private child 
welfare agencies conducted by the National Quality Improvement 
Center on the Privatization of Child Welfare Services, only 13 
child and family-serving agencies, or 3%, provided services in 
more than one State.\27\ This survey also showed that 
nationally the largest private agency budgets range from $17 
million to $140 million. Again, for comparison, in 2014 MENTOR 
reported to the SEC that its gross revenue for serving at-risk 
youth was $203 million and that its net revenue for this same 
population was about $198 million.\28\ Private child welfare 
agencies across the country largely rely on public government 
contracts in order to provide services to children and 
families. In 2011, half of the surveyed child and family-
serving agencies reported that almost 100% of their revenue 
came from public contracts.\29\ This is also the case for 
    \26\ The Committee began its investigation in 2015, which is why 
2014 SEC information is reported here. See Footnote 2.
    \27\ McBeath, B., Collins-Camargo, C., and Chuang, E. See Footnote 
    \28\ Civitas Solutions, Inc., 2014 10-K filing (September 30, 
2014). See Footnote 2.
    \29\ McBeath, B., Collins-Camargo, C., and Chuang, E. See Footnote 
    \30\ Statement by Civitas Solutions: ``We derive approximately 90% 
of our revenue from contracts with state and local government agencies, 
and a substantial portion of this revenue is state-funded with federal 
Medicaid matching dollars,'' 2014 10-K filing, p. 16. See Footnote 2.
    The Committee focused on MENTOR's work as a provider of 
foster care services, since it was one of the largest providers 
of those services nationally. At the time the Committee began 
its investigation, MENTOR provided foster care services to 
thousands of children who are involved with their State's child 
welfare system. As recently as 2015, MENTOR provided foster 
care services to children in 15 different States.\31\ Since the 
Committee launched its investigation, MENTOR has withdrawn from 
a number of States. During FY 2015, MENTOR discontinued at-risk 
youth services in the States of Florida, Louisiana, Indiana, 
North Carolina, and Texas.\32\
    \31\ Roston, A. and Singer-Vine, J. See Footnote 13.
    \32\ Civitas Solutions, Inc., 10-Q filing for the period ending 
March 31, 2017, p. 20. See Footnote 3. (Note: Illinois terminated its 
contract with AHS/MENTOR on July 1, 2015.)

                       F. Therapeutic Foster Care

    In representations to the Committee, MENTOR claims to 
largely serve high-risk children classified as in need of 
therapeutic foster care (TFC) because they are medically 
complex or fragile. There is no uniform definition of TFC in 
the field or in statute, but the Foster Family-based Treatment 
Association describes it as ``a clinical intervention, which 
includes placement in specifically trained foster parent homes, 
for youth in foster care with severe mental, emotional, or 
behavioral health needs. This includes medically fragile or 
developmentally delayed youth whose physical and emotional 
health needs require more intensive clinical and medical 
intervention than can be accommodated in traditional foster 
care.'' \33\
    \33\ Boyd, Laura W., ``Therapeutic Foster Care: Exceptional Care 
for Complex, Trauma-Impacted Youth in Foster Care,'' State Policy 
Advocacy and Reform Center, Report, July 2013, https://
    Many States claim reimbursements from Medicaid for 
components of TFC services. A 2015 report by the Medicaid and 
CHIP Payment and Access Commission noted that 3% of child 
Medicaid enrollees receive TFC services.\34\ States may also 
claim reimbursement under title IV-E for some of the costs 
associated with TFC. There is significant variation across 
States and providers both with regard to eligibility for and 
the provision of services related to TFC. A study that was 
commissioned and funded by MENTOR showed that 17.3% of U.S. 
children in foster care were in TFC-level placements.\35\ 
MENTOR reports that 75% of its caseload is comprised of TFC-
level placements.\36\ In its 2014 SEC filings, MENTOR reported 
billing Medicaid for the provision of at-risk youth 
    \34\ Medicaid and CHIP Payment and Access Commission (MACPAC), 
``Report to Congress on Medicaid and CHIP,'' Report, June 2015, https:/
    \35\ Gonyea, J.G., Bachman, S.S., Rajabiun, S., Springwater, J.S., 
Tobias, C.R., Hirschi, M. and Little, F., ``The 50 State Chartbook on 
Foster Care.'' Retrieved March 28, 2017. As originally cited on p. 5 of 
the MENTOR Mortality Report in Appendix G, Exhibit 1, http://
    \36\ The MENTOR Network, ``Backup Data and Explanation of Mortality 
Analysis Final,'' March 2016. Supplied by MENTOR to the Finance 
Committee. See Appendix G, Exhibit 2.
    \37\ Statements by Civitas Solutions: ``We derive approximately 90% 
of our revenue from contracts with state and local government agencies, 
and a substantial portion of this revenue is state-funded with federal 
Medicaid matching dollars,'' p. 16; ``We derive revenues for our I/DD 
and ARY services and a significant portion of our SRS services from 
Medicaid programs,'' p. 9. See Footnote 2.


              A. The 50-State Overview Letter and Request

    In April 2015, the Committee requested information from all 
50 governors regarding their States' privatization of child 
welfare and/or foster care services.\38\ For example, the 
Committee asked each State to describe its process used to 
select and contract with private agencies providing child 
welfare services as well as the process used to inspect the 
safety of foster care settings in which children are placed. 
The Committee also asked States to outline how they investigate 
and respond to allegations and substantiations of maltreatment 
when a child is in out-of-home care. Thirty-three (33) States 
responded to that request with information prepared by their 
child welfare administrators (see Appendix B for State 
responses to the 50-State Overview Letter).\39\
    \38\ See Appendix A for the complete 50-State Overview Letter.
    \39\ The States that responded: Alabama, Alaska, Arkansas, 
California, Colorado, Connecticut, Delaware, Guam, Hawaii, Illinois, 
Indiana, Iowa, Kansas, Kentucky, Maryland, Massachusetts, Minnesota, 
Nebraska, New Hampshire, New Jersey, New Mexico, New York, North 
Dakota, Oklahoma, Oregon, Pennsylvania, South Dakota, Tennessee, Texas, 
Utah, Washington, West Virginia, Wisconsin, and Wyoming.

               B. The 5-State In-Depth Letter and Request

    In March 2016, the Committee sent in-depth inquiries to 
five States regarding their child welfare operations in order 
to obtain more information about MENTOR and its affiliates as 
well as other for-profit and non-profit providers in each 
State.\40\ The request for additional information was sent to 
the directors of the State child welfare agencies in Georgia, 
Illinois, Maryland, Massachusetts, and Texas. These States were 
selected because they were served by MENTOR and highlighted in 
news accounts that documented serious allegations of 
mismanagement of services and mistreatment of children served 
by private foster care agencies. As of today, MENTOR still 
provides foster care services in Georgia, Maryland, and 
Massachusetts; it no longer provides those services in Illinois 
or Texas. The primary goal of the request for additional 
information was to compare performance indicators of the public 
agencies and private agencies providing foster care services. 
The Committee sought information related to standard 
performance measures for foster care using the Child and Family 
Services Reviews' (CFSRs) performance metrics as a basis (see 
subsection C below). Among the many questions directed to the 
State agencies, SFC staff focused on the following information 
from these select States:
    \40\ See Appendix C for copies of the 5-State In-Depth letters.

         LPhysical and behavioral subgroups (special needs, 
        physically disabled, infants, etc.);
         LMaltreatment during a foster care episode;
         LRate of maltreatment in foster care;
         LPermanency outcomes (reunification, adoption, 
         LPhysical and mental health screenings of children in 
        foster care;
         LChildren receiving monthly caseworker visits;
         LAverage caseload for each caseworker employed by the 
        contractor; and
         LTotal cost to the State under the contract.

    Four out of five States complied with the Committee's 
request for this detailed information. Despite repeated contact 
with the Commonwealth of Massachusetts, its public child 
welfare agency never complied with official requests from the 
Committee to provide the requested in-depth information.\41\
    \41\ For purposes of clarification, Massachusetts did respond to 
the initial 50-State Letter to provide overview information about their 
State, but did not submit any documentation to the Committee in 
response to the 5-State In-Depth Letter.
    In addition to the performance metrics, the Committee also 
asked these five States for copies of any rankings or reviews 
of contractors, as well as performance and investigative 
reports of MENTOR, particularly in the case of reports relating 
to the death, sexual abuse, or injury to a child while in the 
care or custody of foster parents recruited or employed by 
MENTOR. Each of the four responding States complied with this 
request. In addition, Illinois provided Committee staff an 
extensive ``Statewide Specialized Foster Care Review'' focused 
on Alliance Human Services/Illinois MENTOR, which was conducted 
prior to the State terminating services with the company.\42\ 
(The Committee's initial official requests for information from 
State child welfare agencies are in Appendices A and C.)
    \42\ Illinois Department of Children and Family Services, Division 
of Quality Assurance and Research. ``Alliance/Illinois MENTOR Statewide 
Specialized Foster Care Review,'' Report, August 8, 2014, Print.

             C. U.S. HHS Child and Family Services Reviews

    The Children's Bureau in U.S. Department of Health and 
Human Services (HHS) conducts Child and Family Services Reviews 
(CFSRs), which are periodic reviews of State child welfare 
systems, to achieve three goals: (1) ensure conformity with 
Federal child welfare requirements; (2) determine what is 
actually happening to children and families as they are engaged 
in child welfare services; and (3) assist States in helping 
children and families achieve positive outcomes. The first CFSR 
round began in 2001. HHS is currently conducting the third 
round of CFSRs between 2015 and 2018. In October 2016, SFC 
staff asked the Congressional Research Service (CRS) to provide 
a compilation and ranking of State-level data and indicators 
from AFCARS that are used in the CFSR assessments (see Appendix 
D).\43\ The CRS analysis provided the Committee with national 
data concerning the performance of all States in the country, 
including those States that did not respond directly to the 
Committee's requests.
    \43\ Stoltzfus, Emilie, Memorandum prepared by the Congressional 
Research Service: ``Statewide data indicators used in the Child and 
Family Services Review (CFSR),'' October 27, 2016. Available in 
Appendix D.

                    D. The MENTOR Letter and Request

    The Committee sent its first letter to Bruce Nardella, the 
President and CEO of MENTOR, in June 2015. This initial letter 
and correspondence requested information about the company's 
structure, performance, and standards. Specifically, the 
Committee requested the total number of children served 
nationwide; copies of assessments and performance reviews 
conducted on MENTOR; average caseloads of MENTOR caseworkers; 
processes for investigating, vetting, and training potential 
foster parents; details surrounding the use of bonuses for 
placing children; processes for handling allegations of 
misconduct against foster caregivers; copies of settlement 
agreements entered into by MENTOR since 2005; total funding 
received from States; nondisclosure/confidentiality clauses; 
and critical incident reports. Additionally, because MENTOR 
operates under different names in different States, questions 
regarding its corporate structure, affiliates, and related 
organizations were asked as well.\44\ (The Committee's official 
request for information from MENTOR is available in Appendix 
    \44\ In Illinois, for example, MENTOR operated under the name 
Alliance Human Services.


                A. Use of Private Child Welfare Services

    One of the first goals of the Finance Committee's 
investigation was to determine the extent to which States use 
or rely on contracted child welfare services. Information 
obtained by the Committee from the initial 50-State Letter 
shows that of the 33 States that responded, 31 use private 
agencies to provide services to children in foster care and 16 
of these States contract with for-profit and non-profit 
providers. The nature of contracted services provided by the 
entities described in responses to the 50-State Letter varied 
considerably. Twenty States volunteered that services from 
private providers are targeted toward the specialized 
population of youth needing TFC. With the exception of two 
States, administrators were adamant that they were obligated to 
provide oversight--and that they provided this oversight--of 
all foster care placements. According to the responses, private 
agencies might recruit, screen, train, and provide case 
management services to foster families, but the public agencies 
were responsible for approving all placements and for ensuring 
that children were living in safe conditions. For example, 
Texas wrote that children are placed in homes that are 
``directly overseen by child protective services.'' \45\ 
Delaware wrote that the State ``retain[s] . . . legal and case 
management responsibilities for meeting the needs of all 
children in foster care, whether they are placed in a [public] 
foster home or private provider home. . . .'' \46\
    \45\ Response from Texas to the 50-State Overview Letter, Appendix 
B, Exhibit 29.
    \46\ Response from Delaware to the 50-State Overview Letter.
    Other States reported inconsistent information. For 
example, Massachusetts reported that the public agency handles 
96% of placements for the almost 11,000 children in the State 
who are in out-of-home care. Cases that involve a conflict of 
interest with the child welfare agency (for example, employees 
who are the subject of maltreatment allegations), adoptions, or 
unaccompanied refugees who are minors are handled by private 
contract agencies. Nevertheless, in that same response, 
Massachusetts also reported statistics showing that roughly 35% 
of its foster care caseload is managed by a contracted 
agency.\47\ Similarly, Maryland reported ``we contract with 
private providers for placement services only,'' but then went 
on to say ``100% of Maryland foster youth are placed by the 
public agency.'' \48\ Oregon listed two county-run shelters and 
a Youth Villages facility as ``for-profit'' entities when other 
sources identified these entities as non-profits.\49\ News 
reports out of Oregon also show how the non-profit/for-profit 
distinction can be abused, and even non-profit entities can be 
used for financial gain. For example, the director of the now-
shuttered Oregon foster care provider ``Give Us This Day'' was 
accused of using three non-profit organizations to buy property 
for personal use ($100,000), remodel and furnish her home 
($213,000), and pay for trips, meals, clothes, and beauty 
expenses including cosmetic surgery ($249,800).\50\
    \47\ Massachusetts, Department of Children and Families, Letter to 
SFC, June 10, 2015, p. 1. The full response is listed in Appendix B, 
Exhibit 16.
    \48\ Maryland, Department of Human Resources, Letter to SFC, July 
20, 2015, p. 1. See full response listed in Appendix B, Exhibit 15.
    \49\ United States District Court, District of Oregon, Portland 
Division, United States of America v. Mary Holden Ayala, 3:16-CR-00495-
HZ; https://www.justice.gov/usao-or/press-release/file/965436/download; 
Lincoln County, Oregon Juvenile Shelter website; http://
us/juvenile/page/shelter; Douglas County, Oregon Juvenile services 
website, http://www.co.
    \50\ Theriault, Dennis C., ``Oregon accuses foster care provider of 
`plundering' $2 million in state funds,'' The Oregonian, October 15, 
2015. Updated: October 16, 2015, http://www.oregonlive.
    When asked what types of services private agencies provide, 
21 States indicated they were used for case management, even if 
this task was shared or duplicative of services provided by the 
public agency. Twenty-eight States indicated that private 
agencies provide support, services, or training to foster 
families. All States have licensing standards, but only six 
(California, Kansas, Kentucky, Illinois, Tennessee, and Texas) 
reported that they require all of the agencies that contract 
with the State to be accredited. In addition, some States, such 
as Illinois, use benchmarks that private agencies are expected 
to meet, such as an annual permanency rate of 40%. The business 
model is to reward top-performing agencies with ``a greater 
share of new, incoming foster cases.'' \51\
    \51\ Illinois DCFS Letter to SFC, December 15, 2015, pp. 2-3. Full 
response is listed in Appendix B, Exhibit 10.
    MENTOR is one of the largest contractors providing foster 
care services in many of the States the Committee staff 
examined. Information provided by Texas in response to the 5-
State In-Depth Letter showed that when combining the number of 
children served from all MENTOR jurisdictions, it ranked either 
5th or 6th in total size among all Texas private child welfare 
agencies from 2010-2013, before Texas stopped contracting with 
the company. It was always the largest for-profit provider. In 
Maryland, for each of the years the Committee staff reviewed 
in-depth information, MENTOR always had the highest number of 
children receiving contracted services among all providers.
    With regard to the financing of private child welfare 
services, information from the four States that responded to 
the 5-State In-Depth Letter shows that between 2010 and 2015, 
these States spent between $63 million and $291 million 
annually on private child welfare services. Roughly 20% of 
these expenditures went to pay for-profit agencies for 
services. This means these States paid between $18 million and 
$50 million annually to companies that profited from children 
and families involved with the foster care system. Using MENTOR 
as an example, in 2015 MENTOR Maryland was paid an average of 
$47,542 per foster child, for a total approaching $16 million. 
The contractor with the highest annual rate per child in 2015 
in this one State was paid an average of $69,242 per child.

                          B. Background Checks

    All of the States that responded to the Committee's 50-
State Overview Letter described, in various levels of detail, 
how they assess children's safety in out-of-home placements and 
specifically among potential or current foster parents. 
Sometimes there were variations in public versus private 
settings, but all described a process that involves some 
version of a State criminal background check, a national or 
Federal background check, and a check into their own State's 
child abuse and neglect registry--and sometimes in registries 
of other States where potential foster parents have recently 
lived. Some States volunteered information showing or 
suggesting that positive findings of criminal activity or a 
history of maltreating children do not automatically close a 
door to family foster care. For example, California wrote: 
``For persons with criminal convictions, the Department of 
Justice provides the county child welfare agency with the 
record information report [which is reviewed], to determine 
whether the crimes are those for which an exemption may be 
granted.'' \52\ These kinds of exemptions and waivers turned up 
in the materials reviewed by Committee staff concerning 
MENTOR's operations in other States.
    \52\ Response from California to 50-State Overview Letter. See 
Appendix B, Exhibit 4.
    In response to inquiries from the Committee, MENTOR wrote: 
``as a part of our commitment to quality, MENTOR entities 
conduct criminal and non-criminal background checks on 
prospective foster parents in accordance with local/State 
requirements and regulations. This has always been part of our 
practices. Not only do background checks support our rigorous 
vetting efforts to find the most qualified, caring foster 
parents, they are also a requirement of the States and 
referring agencies with whom we partner.'' Yet, several news 
accounts tell a different story.\53\, \54\ Most 
notably is one media account which notes that MENTOR placed 
children in a home with a household member who had previously 
been convicted of aggravated kidnapping and robbery when she 
kidnapped a pregnant convenience store employee.\55\ Similarly, 
the Committee staff determined that MENTOR is often out of 
compliance with its own guidelines, as well as State 
guidelines, with respect to conducting background checks of 
those who care for foster children or those who are routinely 
in homes where foster children are placed. In the State of 
Texas, case records showed that on four separate occasions in 
an 8-month period, MENTOR was instructed to conduct background 
checks on a frequent visitor to a foster home. That check was 
never completed.\56\
    \53\ Joseph, Brian. See Footnote 14.
    \54\ Ansari, Talal, and Roston, Aram, ``Parent at Nation's Leading 
For-profit Foster Care Firm Facing Murder Charges,'' BuzzFeed News, 
February 23, 2016, https://www.buzzfeed.com/talalansari/parent-at-
    \55\ Roston, A. and Singer-Vine, J. See Footnote 13.
    \56\ MENTOR bates numbers 0001822-1823.
    Even when MENTOR met background screening guidelines, it 
sometimes waived the outcomes of the findings. For example, in 
the Committee staff's investigation of documents from MENTOR 
Maryland, the Committee staff found that the husband of a 
foster parent, who was later convicted of sexually abusing 
foster children in their home, had been the subject of four 
previous abuse allegations. The MENTOR worker marked in 
handwriting on the criminal background search results, ``Not 
Mentor [sic] parent,'' presumably indicating that the husband's 
criminal history was irrelevant because the foster mother was 
the primary caretaker. Similarly, MENTOR Texas noted in 
documentation provided to the Committee that in one case 
potential foster parents reported that they or family members 
had been convicted of a crime, but no further information was 
provided. The exact quote is: ``Yes, no description given.'' 
\57\ These individuals went on to become MENTOR foster parents, 
and the foster mother killed a 3-year-old foster child by blunt 
force trauma to the head.\58\
    \57\ Notes from MENTOR Texas Documents: Clemon and Sherill Small 
    \58\ Joseph, Brian. See Footnote 14.
    In another case reviewed by Committee staff, during the 
vetting process for one particular set of MENTOR foster parents 
in Maryland, the foster father indicated that he had a 
preference for ``white, male children.'' This was not seen as a 
red flag to the agency. Male children who were subsequently 
placed with this MENTOR foster parent were sexually abused. In 
fact, media accounts report that children who disclosed abuse 
were not believed at first and were instead sent back to live 
with their abusive foster father. Eventually a foster child was 
believed and the case moved through the legal system.\59\
    \59\ Roston, A. and Singer-Vine, J. See Footnote 13.

                 C. Child Welfare Workforce Operations
                              and Concerns

    Caseworker Turnover. The child welfare field consistently 
has high rates of turnover among its workforce. The national 
range of turnover among child welfare workers is 30-40% 
annually.\60\ When there is movement among staff, children are 
often served by many different caseworkers, which can make it 
difficult for children to form relationships with their 
caseworkers, for caseworkers to put together a treatment plan 
for children, and for cases to be adequately 
monitored.\61\, \62\ Many records and news accounts 
referenced the high turnover rate among staff at MENTOR; it was 
of keen interest to the Committee staff to examine this at the 
State and provider level.
    \60\ U.S. General Accounting Office, ``Child Welfare: HHS Could 
Play a Greater Role in Helping Child Welfare Agencies Recruit and 
Retain Staff,'' Report to Congressional Requesters, No. GAO-03-357, 
March 2003, http://www.gao.gov/new.items/d03357.pdf.
    \61\ Illinois DCFS Report. See Footnote 42.
    \62\ Garner, Bryan R., Hunter, Brooke D., Modisette, Kathryn D., 
Ihnes, Pamela C., and Godley, Susan H., ``Treatment staff turnover in 
organizations implementing evidence-based practices: turnover rates and 
their association with client outcomes,'' Journal of Substance Abuse 
Treatment, March 2012, pp. 134-142, https://www.ncbi.nlm.nih.gov/pmc/
articles/PMC3268938/. See also U.S. Department of Health and Human 
Services, Substance Abuse and Mental Health Services Administration, 
``Report to Congress on the Nation's Substance Abuse and Mental Health 
Workforce Issues,'' January 24, 2013, https://www.google.com/
    Illinois reported that MENTOR had trouble maintaining 
qualified staff. Its overall turnover rate for caseworkers was 
actually consistent with national norms, but the turnover rate 
for therapists working with MENTOR children was 44% in 2012. At 
one Illinois site, the turnover rate for therapists was 80% in 
2013. In this same year, the turnover rate for MENTOR child 
welfare program directors in Illinois was an astounding 82%. In 
its own periodic reviews of MENTOR, the State of Georgia also 
noted concerns about staff turnover. In the Macon, Georgia 
jurisdiction, MENTOR staff turnover reached 83%.
    Caseload Size. Another workforce issue is the size of 
caseloads that child welfare professionals carry. Despite the 
field's attention to caseload size in the child welfare 
profession, Committee staff found it difficult to obtain this 
information from the States that were investigated. Three 
States that responded to the 5-State In-Depth Letter do not 
maintain information on the average caseload per caseworker 
employed by each private contractor. Georgia indicated that it 
tracks this information, but it was not easily accessible to 
the State or the Committee, stating, ``There is no 
comprehensive database that collects average caseload 
numbers.'' Apparently caseload size is one of many indicators 
that is recorded on paper files during periodic onsite reviews 
of private foster homes and other similar agencies. This 
information is never transmitted into an electronic database. 
To further complicate matters, Georgia also insisted that it 
does not use ``caseworkers,'' per se. Instead, the State uses 
the term ``case support workers,'' a distinction which almost 
prevented SFC staff from obtaining any information about the 
caseload sizes in that State.
    National standards for special needs children or children 
requiring TFC services indicate that workers should not have 
more than 10-12 cases per worker.\63\ Illinois contracted with 
MENTOR to provide TFC services at the ratio of 10 cases per 
worker. The State learned that MENTOR was not in compliance 
with this standard. In some instances the ratio was as high as 
14.5 to 1.\64\ Even though Illinois required specific levels of 
case management, it did not maintain a data reporting system 
that would ensure these levels were met and monitored. Maryland 
indicated that, should caseloads exceed what is contracted, a 
corrective action plan would be put in place.
    \63\ Hughes, Sean and Lay, Suzanne, ``Direct Service Workers' 
Recommendation for Child Welfare Financing and System Reform,'' Child 
Welfare League of America, January 2012.
    \64\ Illinois DCFS Report. See Footnote 42.
    Communication. Communication within the company was also 
noted to be a problem. When the State of Illinois was 
conducting its own annual review and subsequent comprehensive 
investigation of MENTOR, the Illinois team had to provide 
copies of the two previous reports from the Illinois Office of 
the Inspector General to MENTOR company officials, as they were 
not aware of the prior investigations completed by the State 
about their company. Nor were company officials aware of 
commitments made by previous MENTOR officers as a result of 
these reports. In these reports, Illinois child welfare 
administrators noted that ``interagency communication issues 
were evident at the onset of the review.''
    Multiple Violations. Maryland conducted periodic reviews of 
MENTOR from 2010-2015. Each review noted any specific violation 
of regulations. Committee staff examined 22 quarterly reviews 
that were conducted of MENTOR Maryland. In only two of those 
quarters were no violations noted. Common and repeated 
violations included missing documentation from employee hiring, 
missing foster parent case files and child client case files, 
concerns around board management and oversight of MENTOR 
operations, and licensing and staffing issues. In many cases, 
foster parents did not document the required hours of annual 
training, nor did children's case files document all medical 
and psychiatric exams. FBI clearance checks were also 
incomplete in some cases. These reviews, over a period of 6 
years, also noted frequent changes in the MENTOR Maryland 
Board's composition and a high rate of staff turnover at 
    Recruitment Bonuses. MENTOR provides incentives for 
recruiting new foster parents to the company. Specifically, the 
company provides financial incentives for employees who recruit 
new foster parents that result in the placement of a child 
through MENTOR. Employees who oversee recruitment efforts in 
each State are called ``recruitment managers.'' Between 2012 
and 2014, MENTOR reported paying an average of $92,000 each 
year in bonuses to employees for foster parent recruitment, 
which is about what a State would pay annually to take care of 
two children who were placed with MENTOR families. According to 
MENTOR, this is an average of $3,800 per recipient. In 
addition, MENTOR also provides incentive payments to foster 
parents who recruit other foster parents to have a child placed 
in their home through MENTOR. The financial bonus for this 
action is $250 for each new foster parent/foster home. 
According to MENTOR records, in 2014 and the first 8 months of 
2015, foster parents received bonuses totaling $126,000 for the 
recruitment of new foster parents, which, using MENTOR's own 
numbers, would represent about 500 new foster parents. There is 
evidence that similar bonuses or incentives are used by State 
and other private foster care agencies as well.\65\ While 
Committee staff were unable to document a direct impact on 
child care resulting from these bonuses, further investigation 
regarding the use of bonuses and incentives may be warranted.
    \65\ For example, in Kentucky State foster parents receive bonuses 
ranging from $100-$250 for each ``resource home'' that is successfully 
recruited. State of Kentucky, Department for Community Based Services, 
``Chapter 12.2.3: Recruitment bonus,'' Standards of Practice Online 
Manual; http://manuals.sp.chfs.ky.gov/chapter12/22/Pages/
1223RecruitmentBonus.aspx. Also, AdoptUSKids, which is a project of the 
Children's Bureau within HHS, recommends using recruitment incentives 
for both staff and existing foster/adoptive parents to increase the 
pool of potential placements for children. See McKenzie Consulting, 
Inc., ``Practitioner's guide: Getting more parents for children from 
your recruitment efforts,'' AdoptUSKids, https://www.

                  D. Physical and Mental Well-Being of
                        Children in Foster Care

    SFC staff sought to determine the physical and mental well-
being of children in foster care and if this varied by provider 
type. According to Federal policy, all States must develop a 
plan for the oversight and coordination of health-care services 
for children who are in foster care. This plan must involve the 
State's Medicaid agency and input from health-care and child 
welfare experts. One of the most basic elements of this plan is 
to determine the timeline under which children will have an 
initial health-care screening upon entering foster care. The 
American Academy of Pediatrics recommends that children and 
youth receive comprehensive health-care screenings within 30 
days of entering foster care.\66\ States vary considerably in 
their own timelines, ranging from requiring health screenings 
within 24 hours to 30 days after a child enters foster care. 
Health-care screenings are one of the items on which States are 
assessed in the periodic CFSRs, but a 2015 review of children's 
health-care needs and services by the Inspector General of HHS 
showed that one-third of children in foster care did not 
receive one of their health-care screenings. Further, one-
quarter of the children received their health-care screening 
    \66\ American Academy of Pediatrics, ``Fostering health: Health 
care for children and adolescents in foster care, 2nd edition,'' 2005, 
Report, p. 22, https://www.aap.org/en-us/advocacy-and-policy/aap-
    \67\ U.S. Department of Health and Human Services, Office of 
Inspector General, ``Not all children in foster care who were enrolled 
in Medicaid received required health screenings,'' Report, OEI-07-13-
00460, March 2015, https://oig.hhs.gov/oei/reports/oei-07-13-00460.pdf.
    In the 5-State In-Depth Letter, States were asked to 
determine what portion of children in public State agencies and 
private agencies had a full physical and mental health 
assessment within 60 days of entering foster care. According to 
the Illinois DCFS review of MENTOR, children being served by 
Illinois MENTOR did not have their physical or mental health 
needs met in a timely manner. Mental health assessments in 
Illinois are to be completed within 30 days of contact with a 
therapist, a standard that MENTOR met for only 60% of cases 
reviewed by the State. Foster children served by MENTOR in 
Illinois waited an average of 122 days before having contact 
with a psychotherapist. A treatment plan is to be established 
within 45 days of a mental health assessment, a standard that 
was met for only 73% of MENTOR cases reviewed by the State. 
Finally, treatment plans are to be updated every 5 months, a 
standard which MENTOR met only 52% of the time. By way of 
comparison, Georgia transmitted data on physical and mental 
health assessments reporting compliance rates for assessments 
within 60 days of placement. The performance in this area was 
very poor for both MENTOR and the entire State, at 12% and 11%, 
    In other instances, MENTOR ranked better than the State 
averages in its compliance with the 60-day mark. In Texas, the 
overall rate for the State was 73%, but for MENTOR it was 84%. 
A similar situation was true in Maryland, where overall only 
60% of children in the State were seen by a health provider for 
a physical and mental health exam within 60 days of entering 
foster care, but MENTOR met this mark 69% of the time.

                 E. Failure to Identify and Respond to
                            Risk to Children

    The documents that were provided by MENTOR and in responses 
to the 5-State In-Depth Letter showed that public agencies and 
MENTOR repeatedly failed to identify and respond to the risk 
that was presented to children in out-of-home care.
    During the 2015-2016 reviews that Georgia conducted with 
its contracted service providers, the MENTOR jurisdiction in 
Athens reported 41 ``significant events,'' which included four 
child protection investigations (each unsubstantiated), two 
suicidal/homicidal threats, one ``child-on-child sexual 
event,'' and two ``child-to-child physical confrontations.'' 
Despite these reports, the State gave this particular MENTOR 
jurisdiction an overall qualitative safety score of 93%. Yet, 
of the two caregiver homes randomly selected for the 
jurisdiction's annual review, the second home presented 
environmental risks to the foster children. The case notes 
reviewed by Committee staff read: ``the provider stored the 
garden tools, rake, lawn mower and the bottom of a water cooler 
in the living room.''
    In this same review, one of the ``well-being strengths'' 
listed is the ``documentation of the younger children's 
academic needs being met,'' despite the fact the review also 
noted the following:

         LThe caregiver reported the two youth have no 
        ambition, motivation, or life goals;
         LThe youth refuse to attend school. . . .

    Similarly, under the category of ``Well-being Areas Needing 
Improvement,'' the review noted two youth failing in school, 
not making adequate progress, and not receiving tutoring or 
academic support, in addition to a lack of documentation 
explaining more than five unexcused absences.
    A MENTOR foster parent in Texas killed one of her foster 
children.\68\ Case notes reviewed by Committee staff indicate 
there were clear warning signs that the safety and well-being 
of children in her care were compromised before the fatal 
abuse, but Texas MENTOR failed to see the risk to the children 
placed with her and ultimately did not protect the children 
being served. MENTOR described the foster mother and her 
partner as ``mature, responsible, healthy individuals capable 
of meeting the needs of a child placed in their care.'' Yet the 
records also show that the foster mother reported being 
overwhelmed and uncertain if she could care for foster 
children. Children placed in her care were removed with 
``negative outcomes,'' placements in the home ``failed,'' and 
Early Childhood Intervention staff felt that children should 
not be in this particular foster home. Further, the foster 
parents were investigated by the Texas Department of Family and 
Protective Services for concerns about children in their foster 
home. Records from MENTOR report that the children in this home 
had bruises and the foster mother reported that the children 
would make false allegations against her and her partner. Texas 
MENTOR did not terminate their license, but instead reinforced 
its commitment to working with this family with case records 
stating that MENTOR staff ``agree that this family should 
continue to work as foster parents. . . . We will be decreasing 
the number of children the family is licensed to care [for] in 
efforts to ensure the family remains a Mentor [sic] family.''
    \68\ Joseph, Brian. See Footnote 14.

                    F. Actions When Maltreatment Is
                     Substantiated in Foster Homes

    All of the MENTOR children who were highlighted in media 
accounts and were maltreated by their foster parents were 
living in very high-risk situations. (As previously noted, in 
one MENTOR home, children had disclosed their maltreatment, but 
their disclosures were dismissed by those in a position to take 
action. The preceding section noted the presence of bruises on 
children and the investigation of maltreatment in one MENTOR 
home where a child was killed.\69\) As a result, in the 50-
State Letter, SFC asked States about their procedures for when 
maltreatment is substantiated in a foster home. In such a 
situation, this would mean it was confirmed, founded, or 
substantiated that a foster child was being abused or neglected 
in his or her foster placement. According to data presented in 
a report to Congress by the U.S. Department of Health and Human 
Services, the State median of maltreatment among children who 
were in foster care from 2010-2013 was 0.35%.\70\ Missouri, 
Wyoming, and Virginia had the lowest rates, as determined by 
the 2015 NCANDS and AFCARS datasets; the highest rates were in 
New York, Iowa, and Massachusetts.\71\
    \69\ Roston, A. and Singer-Vine, J. See Footnote 13.
    \70\ U.S. Department of Health and Human Services, Administration 
for Children and Families, Children's Bureau, ``Child Welfare Outcomes 
2010-2013--Report to Congress,'' Report, February 1, 2016, https://
    \71\ Stoltzfus, Emilie, Memorandum prepared by the Congressional 
Research Service. See Footnote 43.
    Of the 33 States that responded to the 50-State Letter, 
only 9 indicated that substantiation for abuse or neglect in a 
foster home would unequivocally result in the revocation of a 
foster home license: Arkansas, Delaware, Indiana, Kansas, 
Nebraska, New Hampshire, South Dakota, Tennessee, and 
Wisconsin. All of the other responding States report that 
license revocation is one possible outcome when maltreatment is 
    Many of the States compared the rates of re-victimization 
of children in foster care based on their foster home setting: 
public, private non-profit, or private for-profit. There was no 
evidence that children who were in privatized foster care 
settings were more likely to be re-victimized than children in 
publicly run foster homes.
    Some of the information provided by the States about 
maltreatment in foster care raised serious concerns. 
Massachusetts reported the number and percentage of 
substantiated child maltreatment episodes as a share of the 
total foster care population. Of the 44,240 children in 
substitute care during Federal fiscal years 2010-2014, 
Massachusetts reported 739 instances of maltreatment in foster 
care, which is 1.67% of all children in care. \72\ At face 
value, this might appear to be a relatively low rate, but as 
previously noted, it is well above the State median of 0.35%. 
Further, data used in the third round of the CFSRs shows that 
Massachusetts had the highest rate of child re-victimization in 
the Nation \73\ and it steadily climbed between 2010 and 
    \72\ See Appendix B, Exhibit 16--Response From Massachusetts, 
Office of the Governor.
    \73\ Stoltzfus, Emilie, Memorandum prepared by the Congressional 
Research Service. See Footnote 43.
    \74\ ``Child Welfare Outcomes 2010-2013--Report to Congress.'' See 
Footnote 70.
    In response to the 5-State In-Depth Letter, Maryland 
reported that it does not track the occurrence of maltreatment 
in foster care by provider or by type of provider, which means 
that the performance of individual contractors and type of 
contractor is not monitored in this way. Texas reported that 
``serious instances of confirmed abuse and neglect cases result 
in licensure revocation.'' But, it only reported one instance 
of that happening between 2010 and 2014, despite the fact that 
the State also provided the Committee with data showing there 
were 295 instances of confirmed maltreatment among children in 
foster care in that same time period. \75\ Further 
documentation provided by Texas shows that between 2010 and 
2015 about 7% of youth in foster care had an episode of 
maltreatment against them substantiated; the rate was 8% for 
Texas MENTOR.\76\
    \75\ See Appendix B, Exhibit 29--Response From Texas, Department of 
Family and Protective Services.
    \76\ Maltreatment in foster care was for many years measured as a 
percentage of children in foster care who had a substantiated or 
indicated report of maltreatment where the perpetrator was coded as the 
child's foster care provider (i.e., the perpetrator was the child's 
foster parent or a staff member at a group home or institution where 
the foster child was placed). The State median data cited above, for 
example, uses that metric. Further, the data provided by Massachusetts 
and Texas on the number of children maltreated while in foster care 
from FY2010-FY2014 appears to use a similar if not identical metric. By 
contrast, when it separately reported a percentage of children in Texas 
foster care who were maltreated, the State appears to have made this 
calculation based on all reports of maltreatment of children in foster 
care, without regard to the perpetrator of the abuse or neglect. HHS 
has begun to move its measurement of maltreatment of children in foster 
care to include maltreatment without regard to who is the perpetrator. 
However, the new HHS calculation, which measures incidents of 
maltreatment for every 100,000 days of foster care provided by the 
State, also takes certain steps to ensure that reports of maltreatment 
for children in foster care do not unintentionally capture those 
reports that were responsible for bringing a child in to foster care.
    There is also a troubling example from Massachusetts that 
was captured by investigative journalists where the State 
changed its determination of whether an infant died from child 
abuse or neglect.\77\ A 2-month-old infant died in a 
Massachusetts foster home that was run by MENTOR. Initially the 
case was ruled by the State as a death related to neglect 
because of unsafe sleeping conditions. There is a provision in 
Federal law which requires States to release information about 
abuse and neglect-related deaths to the public.\78\ When 
journalists used this provision to press the Massachusetts 
Department of Children and Families to release information 
about the infant's death, the substantiation of neglect was 
reversed by the State--2.5 years after the death. Instead, the 
State ruled that the death was not related to neglect. This 
meant that Massachusetts was no longer required to release 
information about the circumstances relating to and causes of 
the infant's death. Accordingly, this information and the 
record can remain sealed from the public.
    \77\ Roston, Aram, ``In an unmarked grave, a baby's untold story,'' 
BuzzFeed News, June 18, 2015, https://www.buzzfeed.com/aramroston/in-
    \78\ Child Abuse Prevention and Treatment Act, ``2.1A.4, Assurances 
and requirements, Access to child abuse and neglect information, Public 
disclosures.'' The full statute can be found at https://

                       G. MENTOR Incident Reports

    The Committee staff requested that MENTOR submit all of its 
highest-level incident reports from FY 2005 to FY 2014 for 
review. This would allow SFC staff to investigate the most 
serious cases where children died or were seriously harmed. 
MENTOR submitted a total of 98 ``level 4'' incident reports 
which capture the agency's most serious incidents of injury, 
assault, abuse, or other similar events; 86 involved the death 
of a child.\79\ The other cases involved psychiatric 
admissions, allegations of sexual assault perpetrated against 
foster children, allegations of sexual assault committed by 
foster children, and accidents or injuries that happened to 
foster children. Table 1 shows that about half of the reports 
(45%) involved a child with a behavioral health concern and 40% 
involved a child who was medically complex. Of the cases that 
involved a death, almost three-quarters (73%, or 62 cases) of 
the deaths were listed as ``unexpected,'' which is a check box 
on MENTOR's incident report form.
    \79\ See Appendix F for a sample of the level IV incident reports 
provided by MENTOR. The company provided Committee staff all death-
related level IV incident reports for children in foster care under 
MENTOR for FY2005-FY2014, with the exception of one incident report 
that could not be located by the company. In addition, Committee staff 
reviewed non-death-related level IV incident reports provided to the 
Committee for those years, in addition to death-related and non-death-
related incident reports for FY2015. The Committee staff considered all 
of these reports in its analysis. Committee staff cannot draw 
conclusions about this full time frame given that the most recent 
reports were provided after the Committee staff's analysis and have not 
been fully reviewed.

              Table 1. Summary of Incident Reports Reviewed
               Area of Assessment                     No.       Percent
Service Category (Indicated by MENTOR)
Behavioral health                                        44         45%
Blank                                                     2          2%
Juvenile Justice                                          3          3%
Medically Complex                                        39         40%
Missing Incident Report                                   1          1%
Mentally Retarded/Dev. Delay                              3          3%
Other                                                     6          6%
    Total                                                98        100%
Death Was Expected? *
Yes                                                      23         27%
No                                                       62         73%
    Total **                                             85        100%
Was an Internal Investigation Launched?
Yes                                                      13         13%
No                                                       84         87%
    Total                                                97        100%
* Calculation only includes death cases.
** One incident report for a death was not provided to the Senate
  Finance Committee.

    Table 2 shows that of the deaths that were unexpected, an 
internal investigation was launched only 21% of the time (13 
cases--set bold in Table 2), which suggests MENTOR does not 
seek opportunities to learn from unexpected critical incidents. 
The child welfare profession,\80\ along with many other 
professions including law enforcement,\81\ health care,\82\ and 
transportation,\83\ is moving in the direction of increasing 
transparency and trying to learn from crises. In many cases and 
jurisdictions, this includes systematic reviews of incidents 
that result in unexpected deaths. The Committee staff also 
determined that in at least nine of the incidents, there were 
financial settlements paid to families of the victims.
    \80\ See the National Center for the Review and Prevention of Child 
Deaths, https://www.ncfrp.org/. See also: Hochstadt, N.J., ``Child 
death review teams: a vital component of child protection,'' Child 
Welfare, 2006 July-August, 85(4): 653-70, https://www.ncbi.nlm.nih.gov/
    \81\ See ``Critical Incident Review Library'' at The Police 
Foundation, https://www.
    \82\ Wald, Heidi and Shojania, Kaveh G., U.S. Department of Health 
and Human Services, ``Chapter 4: Incident reporting,'' Agency for 
Healthcare Research Quality--Archive. Retrieved March 28, 2017, https:/
    \83\ National Transportation Safety Board, ``History of the 
National Transportation Safety Board.'' Retrieved March 28, 2017, 

     Table 2. Internal Investigations and Death Expected/Unexpected
                                      Death Expected?
    Internal Investigation?    ----------------------------     Total
                                     No            Yes
No                                 49 (79%)     23 (100%)            72
Yes                                13 (21%)        0 (0%)            13
    Total                         62 (100%)     23 (100%)            85

    At the most basic level, MENTOR's incident reports have 
typos, errors, inconsistencies, and missing information. More 
concerning instances include inaccurate information and 
diagnostically implausible conditions. For example:

         LMENTOR's incident reports are incomplete. For 
        example, several reports mention that an internal 
        investigation is underway, but the outcome is never 
        indicated. Similarly, other reports note that an 
        investigation by law enforcement is underway, but there 
        was never any follow-up information available from the 
        incident reports to indicate the outcomes of these 
         LOne MENTOR incident report was missing. In the list 
        of incident reports that was presented to Committee 
        staff, one incident report was not provided. In fact, 
        the company was unable to locate the document. Yet, 
        documentation attached to that case ID indicated that 
        the outcome of the case was serious enough to warrant a 
        settlement from MENTOR with the family.
         LMENTOR's incident reports include information that 
        is diagnostically inaccurate. For example, one report 
        documenting the death of a 2 month-old infant described 
        the deceased as being ``oppositional.'' At best, this 
        was an error. At worst, it was an actual (although 
        implausible) diagnosis, since the conditions leading to 
        a diagnosis of ``oppositional defiant disorder'' need 
        to persist for a minimum of 6 months before a diagnosis 
        can be made.\84\ Even then, it is developmentally 
        inappropriate to give an infant this kind of diagnosis. 
        The average age of onset for oppositional defiant 
        disorders is between ages 5-15.\85\ Similarly, in 
        another case, a 4-month-old infant who also died was 
        described as having ``behavioral health'' problems when 
        the field widely recognizes that most serious mental/
        behavioral health conditions are not diagnosed until 
        adolescence or early adulthood. Even when more 
        childhood-based conditions are treated (such as 
        attention deficit/hyper-activity disorders) the average 
        age of onset is 4-11 years old.\86\
    \84\ American Academy of Child and Adolescent Psychiatry, 
``Oppositional defiant disorder: A guide for families by the American 
Academy of Child and Adolescent Psychiatry,'' Report, 2009, https://
    \85\ Kessler, Ronald C., Amminger, G. Paul, Aguilar-Gaxiola, 
Sergio, Alonso, Jordi, Lee, Sing, and Ustun, T. Bedirhan, ``Age of 
onset of mental disorders: A review of recent literature,'' Current 
Opinion in Psychiatry, Vol. 20, No. 4, 2007, https://
    \86\ Ibid.
         LMENTOR's incident reports contain information that 
        conflicts with media accounts of incident. One incident 
        report documents the death of a 4-year-old child and 
        states that the child died from cardiac arrest. Media 
        accounts of that incident (which were discovered by the 
        Committee staff based on the State and age of the 
        deceased, as well as the date of death) indicate that 
        the foster mother of the child was convicted of second 
        degree manslaughter in the death of the child. MENTOR 
        also reached a legal settlement with this family.

                       H. MENTOR Mortality Report

    In response to the national attention concerning children 
who died on MENTOR's watch and the investigation by SFC staff, 
MENTOR conducted its own analysis of children who died in the 
company's care. This MENTOR ``mortality report'' was completed 
through a contract held with a research center at a public 
university in the company's home State (see Appendix G, 
Exhibits 1 and 2).\87\ In this analysis, MENTOR concludes that 
its child death rates are in keeping with the rates of deaths 
among foster children and among the youth population (in 
general) at the national level. As discussed further in this 
section, these conclusions are inaccurate and they appear to 
misrepresent the experiences of children who are served by 
MENTOR. The company did not have the report independently 
validated. When SFC staff inquired about having the report 
``peer reviewed'' by independent researchers with expertise in 
child maltreatment, MENTOR indicated that this would only be 
possible with the company's approval.
    \87\ Note: On October 5, 2017, MENTOR provided Committee staff with 
an updated mortality analysis, which reflected data through August 
2017. To the extent Committee staff feels this data meaningfully 
changes the analysis in the Committee Print, Committee staff will make 
this information available on the Committee website in the future.
    Committee staff found this report to be inaccurate and 
misleading. The report used unequal points of comparison 
between deaths that occurred under MENTOR's watch and national 
rates of the deaths of foster children, according to AFCARS. 
MENTOR's report included a comparison of its own annual death 
rate based on the total number of children in its care each 
year, with the national, annual death rate of foster children. 
This national rate is based on the number of children who were 
in care on the single date of September 30th, which is when 
annual counts are taken. The result is the appearance of the 
national death rate being much higher than MENTOR's death rate, 
when in fact the opposite is true, as explained below.
    In addition to these false conclusions, MENTOR's methods 
and standards of analysis are not consistent with the field, 
nor are they employed by HHS or the Centers for Disease Control 
and Prevention (CDC). For example, the report stated the rate 
of child deaths per 100 live children, as opposed to 100,000 
live children, which is what is used by HHS \88\ and the 
    \88\ U.S. Department of Health and Human Services, Administration 
for Children and Families, Administration on Children, Youth, and 
Families, Children's Bureau, ``Child Maltreatment 2015,'' Report, 
January 19, 2017, https://www.acf.hhs.gov/sites/default/files/cb/
    \89\ Kochanek, K.D., Murphy, S.L., Xu, J., and Tejada-Vera, B., 
U.S. Department of Health and Human Services, Centers for Disease 
Control and Prevention, National Center for Health Statistics, National 
Vital Statistics System, ``Deaths: Final Data for 2014.'' National 
Vital Statistics Reports, Vol. 65, No. 4. June 30, 2016, https://
    Using the numbers provided in its own report, MENTOR's 
average rate of death for 2010-2014 is .074 per 100 foster 
children served (or as the field would express it, 74 per 
100,000 children served), as compared with national rates of 
.052 per 100 foster children served (or 52 per 100,000 children 
served). Yet, MENTOR concludes: ``The MENTOR Network serves 
significantly more children and youth with heightened risk 
factors relative to others in foster care, and sustains child 
mortality rates that are comparable with national norms'' (bold 
emphasis in the original document). In fact, MENTOR's death 
rate among foster children is 42% higher than the national 
    Figure 1 demonstrates MENTOR's misleading display of 
information. The red comment box points to MENTOR's inclusion 
of a national death rate that is based on point-in-time counts 
of the number of children in foster care on a single day. Using 
this number in the denominator makes it look like the death 
rate among children in foster care is higher than is actually 
the case. The Committee staff concluded that this information 
is not a valid comparison. The black comment box points to the 
national death rate that is based on the total numbers of 
children served annually in foster care, which provides a more 
accurate estimate of the death rate among foster children. The 
blue comment box points to the death rate among foster children 
who are being served by MENTOR. This is also based on the 
number of children served annually by MENTOR. This chart 
clearly shows that the blue bar and rate, which captures 
MENTOR's death rate, is higher than the black bar and rate, 
which captures the comparable national death rate of children 
who are in foster care.

    Further, in its mortality report, MENTOR also states: 
``Mortality rate in TMN [The MENTOR Network] foster care 
programs has been very similar to, and more recently equal to 
or better than, national norms.'' To substantiate this, MENTOR 
provided a line chart of death rates for 2009-2014, with data 
points that are too small to decipher and no numbers. The chart 
appears intended to capture the national death rate among those 
aged 0-22. Regardless, this chart is also misleading. Mortality 
information collected by Committee staff from the CDC Wide-
ranging Online Data for Epidemiologic Research (WONDER) 
database shows that the national death rate in the United 
States between 2009 and 2014 among those aged 0-22 is 54.7 per 
100,000 in the population.\90\ That means that the MENTOR death 
rate (74 per 100,000) among youth aged 0-22 is 35% higher than 
the national average.
    \90\ For more information about the WONDER database, visit its 
website https://wonder.cdc.gov/.
    Furthermore, MENTOR's graphical analyses only focused on 5 
years of data, 2009-2014, but in its report, MENTOR included 10 
years of data. Figure 2 displays the full 10 years of rates of 
children's deaths per 100,000 for 2005-2014. There are only 2 
years during which MENTOR's death rates were at or below the 
foster care population and national averages.\91\
    \91\ The national data in Figure 2 is for youth ages 0-22, which is 
the same age range that MENTOR used in their mortality report.

    Finally, in the MENTOR conclusion cited above in this 
section, the company highlights its contention that it serves 
``significantly more children and youth with heightened risk 
factors'' and therefore a higher death rate should be expected 
among its served population. However, as noted earlier in the 
report, when the Committee staff examined MENTOR's incident 
reports, it found that of the cases involving a death, nearly 
three-quarters of those cases (73%) indicated that MENTOR 
itself concluded the death was ``unexpected.'' In short, 
MENTOR's own incident reports do not support the conclusion 
that MENTOR's fatality rate is attributable to these heightened 
risk factors.

                  I. Financial Settlements From MENTOR

    The documentation provided by MENTOR allowed Committee 
staff to review over 20 financial settlement agreements 
resulting from alleged negligence and/or damages. However, this 
did not capture the full range of such settlements. According 
to documents reviewed by Committee staff in reference to 
Maryland dated between 2005 and 2015, there were 22 settlements 
of claims against MENTOR. Illinois submitted materials to the 
Committee indicating that there had been nine similar 
settlements in the State. In almost all of the MENTOR legal 
documents the plaintiffs and settlement amounts are redacted. 
However, the documents from Illinois cite a total of $19.5 
million in payouts. Given the numerous settlements made between 
MENTOR and dozens of parties, MENTOR has likely paid many 
millions of dollars in wrongful suit settlements. In fact, in 
its 2014 SEC filings, MENTOR wrote: ``Several years ago, we 
experienced a spike in claims filed against the Company, and we 
could face an increase in claims in the future. As a result of 
the prior increase in claims, we received less favorable 
insurance terms and have expensed greater amounts to fund 
potential claims.'' \92\
    \92\ Civitas Solutions, Inc., 2014 10-K Filing (September 30, 
2014), p. 13. See Footnote 2.


    The States that responded to the 50-State Letter were 
adamant about their oversight of children in their foster care 
systems, regardless of the nature of the children's placement. 
That said, the information provided to and reviewed by SFC 
staff describes systems that do not always keep children safe 
or allow their performance to be readily evaluated. In fact, 
one State--Massachusetts--failed to comply with a request from 
the Committee to submit more detailed information about its 
child welfare services. To further complicate matters, the 
complex and fragmented nature of the child welfare system makes 
it difficult for the Federal Government, and others, to monitor 
the operations and outcomes of children who are involved in the 
system. Some of the problems SFC staff encountered in trying to 
evaluate and compare the performance of States and providers 
through this review are as follows:

         LGeorgia does not specifically record data about 
        children needing TFC services, which means the State is 
        likely unable to track the well-being of this 
        subpopulation (and likely many others).
         LData from Illinois provided point-in-time 
        measurements for a single date on two key measures 
        (siblings placed in same living arrangement and average 
        caseload per caseworker), as opposed to over a period, 
        such as a fiscal year or multiple fiscal years. SFC 
        staff repeatedly raised this issue with Illinois staff 
        and requested a State average across all providers 
        which would have given the Committee at least some 
        benchmark for comparison purposes. The State was unable 
        to provide this information.
         LMaryland only provided the Committee with data on 
        private TFC placements, as opposed to data about their 
        entire child welfare population with subpopulations 
        that might have included those needing TFC placements.
         LNot all child welfare indicators are available 
        electronically. Some States collect information, 
        perform reviews, and maintain data in paper files that 
        are never entered into an electronic database or that 
        are never synthesized into a single report or review. 
        Without more systematic procedures in place, it is 
        almost impossible for States to have any meaningful 
        oversight over their own systems or the agencies that 
        provide contracted services for them.
         LThe field lacks a consistent language about child 
        welfare services, clients, and operations, which makes 
        it difficult to make comparisons between States and 
        between providers. For example:

                 LThere is no uniform definition of 
                what constitutes TFC. Some States use the term 
                ``treatment foster care,'' instead of 
                ``therapeutic foster care,'' although the 
                spirit of the definition, the needs of the 
                children, and the services provided would be 
                similar. Other common terms used include 
                ``special needs'' children or ``medically 
                fragile'' children.
                 LOne State does not use the term 
                ``caseworkers'' and would not provide client-
                to-caseworker ratios until SFC staff 
                established that this State called these 
                employees ``case support workers.'' MENTOR 
                calls its caseworkers ``child welfare 
                 LThe terms that States use to 
                refer to the private agencies with which they 
                contract to provide foster care services vary 
                as well. Some States refer to these agencies as 
                ``child placement agencies.'' Another common 
                term is ``contract agencies'' or ``foster care 

         LStates have varying definitions and 
        conceptualizations of what it is that private child 
        welfare agencies do in their States. Some States do not 
        use the term ``child placement agencies'' to refer to 
        contract agencies, while others do. This is because the 
        State has control over and responsibility for placing 
        children in foster homes. So when the Committee asked 
        States, ``What proportion of the children in foster 
        care in your State is placed by the public agency, not-
        for-profit providers, and for-profit providers,'' many 
        States indicated ``zero'' for anything outside of the 
        public agency because they maintain that all children 
        are placed by the State. Other States (which maintained 
        that the public agency places all children) still 
        provided rates of children placed by contracted 
        agencies. This meant it was impossible for SFC staff to 
        determine the proportion of children living in homes 
        that for all intents and purposes are run by private 
         LWith regard to substantiations of maltreatment in 
        foster care, the Committee asked States to indicate in 
        ``how many of these instances . . . were children 
        placed by: not-for-profit providers, for-profit 
        providers, and public providers?'' Some States 
        indicated ``zero'' for the first two categories, again 
        because they maintain that children in their State are 
        not placed by contracted agencies, they are only placed 
        by public agencies, even if the foster parents work for 
        and are managed by a contracted agency. Thus, the 
        information obtained from different States was not 
        always comparable.
         LOther States maintain that public employees provide 
        full case management services for the children in their 
        care, yet private agency workers set up appointments 
        for the clients, make arrangements for services, 
        provide transportation, provide support for foster 
        parents, and visit the children in their foster care 
        placements. According to the National Association for 
        Social Workers, the definition of case management 
        services is: ``A process to plan, seek, advocate for, 
        and monitor services from different social services or 
        health care organizations and staff on behalf of a 
        client.'' \93\ This seems consistent with the services 
        private agencies provide children who are in the 
        protective care and custody of their State, even if the 
        public agencies call it otherwise.
    \93\ National Association of Social Workers, ``NASW Standards for 
Social Work Case Management,'' Report, 2013, https://
         LThe Committee staff noted the various ways and 
        outcomes by which child welfare agencies and government 
        documents discuss the maltreatment of children involved 
        with child welfare agencies. CFSRs, which are 
        implemented by HHS and monitor State-level child 
        welfare outcomes, track the percentage of any 
        recurrence of maltreatment among all children in the 
        State. The standard: States should not be above 9.1%. 
        In 2013, 19 States failed to meet this benchmark.
         LMaltreatment of children in foster care is measured 
        as the number of children who were victims of 
        substantiated maltreatment per 100,000 days spent in 
        foster care. The national standard is set at 8.5 per 
        100,000 days. In 2013, 22 States exceeded this 
        standard. A document prepared by HHS for Congress 
        showed that the State median of maltreatment of 
        children in foster care is 0.35%. Meanwhile, individual 
        States submitted rates to SFC that ranged from 2% to 
        20%. These varying ways for measuring and reporting the 
        same construct make it difficult for regulators to 
        monitor outcomes and the well-being of children 
        involved with their State's child welfare system.
         LThe third round of CFSRs is being conducted between 
        2015 and 2018. In late 2016, HHS discovered an error in 
        the syntax the Department used to electronically gather 
        information from the States regarding their program 
        performance. Information is still being gathered, but 
        States that submitted data before the error was 
        discovered will not be assessed in the areas that were 
        affected by the syntax error. Thus, a major Federal 
        mechanism that is in place for monitoring child welfare 
        performance in the States is not fully functional and 
        means that for some States, there will be approximately 
        a 10-year gap on the Federal assessment of some child 
        welfare performance indicators.
         LNot all of the States responded to inquiries from 
        the Committee.

                 LSeventeen States failed to 
                respond to the 50-State Letter. Those States 
                were: Arizona, Florida, Georgia, Idaho, 
                Louisiana, Maine, Michigan, Mississippi, 
                Missouri, Montana, Nevada, North Carolina, 
                Ohio, Rhode Island, South Carolina, Vermont, 
                and Virginia.
                 LMassachusetts failed to respond 
                to the 5-State In-Depth Letter. Despite 
                repeated assurances from the Massachusetts 
                Department of Children and Family Services that 
                ``continuing efforts are being made to collect 
                and prepare the information'' with ``the full 
                intention to send a response to the 
                Committee,'' a response to the Committee's 
                questions was never received.

                            VI. CONCLUSIONS

    Despite the limitations on information supplied to the 
Committee, there is sufficient information from this single 
private agency and the States to show that children who are 
under the legal authority of their State are often ill-served. 
Some children served by MENTOR were abused, neglected, and 
denied services, and the very State agencies that have been 
charged with and paid to keep them safe have likely contributed 
to their suffering and deaths. As documented by MENTOR's 
incident reports, investigations about fatalities were never 
followed up, autopsy reports which were pending years ago are 
not part of case files, and the vast majority of children who 
died were not the subject of an internal investigation--even 
when their deaths were unexpected. MENTOR issued a report to 
the Committee which falsely claimed that its death rates are in 
keeping with national death rates and the rates of death among 
children in the foster care system. Add to this complicated 
narrative the fact that the families of MENTOR victims have 
received millions of dollars in financial settlements, 
significant enough for MENTOR to have received less favorable 
terms from its insurer in 2014. The Committee staff recognizes 
that a case study of one provider does not allow for direct 
comparisons with other individual providers, and thus, cannot 
draw conclusions regarding MENTOR's operations in relation to 
other contracted agencies, for-profit or otherwise. Regardless, 
information collected for this report shows that MENTOR 
repeatedly placed the health and well-being of children at 
    State agencies would likely counter concerns about their 
performance by noting the procedures they have in place to 
monitor private agencies. For example, Illinois notes that all 
licensed foster homes are physically inspected at least twice 
each year, with licenses valid for 4 years. Maryland states 
that it assesses all private TFC providers quarterly. This same 
State noted that providers can find themselves on the agency's 
``hotlist'' by not complying with contractual obligations or by 
committing license violations. In such situations, providers 
would not be able to take in new clients until a corrective 
action plan is generated and subsequently completed. Even 
though MENTOR repeatedly failed to meet all State licensing 
criteria in 20 of the 22 quarters reviewed that the Committee 
examined, MENTOR continues to operate in Maryland. Illinois 
described a similar regulatory approach, but MENTOR no longer 
operates in that State.
    State oversight guidelines are in place in almost every 
State in the Nation, either written into State statute or as 
part of agency policy. This investigation and precipitating 
media reports show that these policies are not always followed, 
exceptions are made, waivers are granted, and sometimes 
professionals serving children look the other way.
    The documents reviewed in this investigation show a system 
that does not always protect children. Profits are sometimes 
prioritized over children's safety and well-being. Turnover 
among staff sometimes makes it impossible to develop case plans 
to ensure that children are ``on-track'' and being monitored. 
Foster parents with questionable backgrounds who seemingly lack 
the skills to provide care to vulnerable children are given 
licenses to parent challenging children and then are often 
inadequately monitored. Further, the data and oversight 
structures at both the State and Federal level make it 
difficult and sometimes impossible to monitor the operations of 
the child welfare system itself, as well as its private 
contractors. Thus, the bipartisan Committee staff sets forth 
the following recommendations.

                          VII. RECOMMENDATIONS

                 Recommendations for States and Tribes

         LImprove outreach, customer service, and support 
        services for those interested in becoming foster 
        parents to attract and retain high-quality foster 
         LSupport enhanced oversight of foster families to 
        ensure robust background checks, home study 
        assessments, and ongoing placement oversight.
         LFrequently review performance of child welfare 
        service providers/contractors to ensure child safety, 
        permanency, and well-being standards are being met.
         LTrack child safety and well-being related outcomes 
        at the individual provider level, including whether 
        children served by specific providers have higher than 
        average needs (e.g., medically fragile, special needs, 
        or therapeutic foster care placement, etc.).
         LSet standards for maximum caseload size for child 
        welfare workers, which may include differentiated 
        standards based on variations in case type (e.g., 
        medically fragile children, children in therapeutic 
        foster care placements, etc.) or activity (e.g., 
        investigations of abuse or neglect, case planning for 
        children in foster care).
         LProvide greater funding for the training of front-
        end staff charged with making removal and placement 
        setting decisions for children entering foster care or 
        at risk of entry.
         LRevoke contracts from child welfare service 
        providers who are unable to demonstrate the capacity to 
        provide safe foster care placements for children.
         LProvide subsidized guardianship payments to 
        relatives willing and able to provide safe placements 
        for children who can no longer remain at home.
         LEnsure child death review teams are transparent, 
        timely, and well-staffed. Require the timely 
        publication of the results of child death reviews while 
        ensuring appropriate and robust privacy protection of 
        sensitive data.
         LMake placement setting decisions based on the 
        assessed strengths and needs of children entering 
        foster care using an age-appropriate, evidence-based, 
        validated, functional assessment tool to ensure 
        children receive the appropriate level of care in the 
        least restrictive, most family-like environment.
         LEstablish child welfare ombudsman offices through 
        which children in care, family members, child welfare 
        workers, foster parents, whistleblowers, and members of 
        the public at large can submit comments and concerns 
        about misconduct within the child welfare system.

             Recommendations for the Department of Health 
                        and Human Services (HHS)

         LWork to engage States, Congress, and the broader 
        child welfare community in understanding the purpose 
        and State-
        specific relevance of the CFSRs and ensure this process 
        contributes to meaningful improvement and reform.
         LSeek and provide clarification on how States and 
        Tribes are defining, using, and overseeing the delivery 
        of Therapeutic Foster Care (TFC) and establish a common 
        definition of TFC for the purposes of Medicaid and 
        title IV-E.
         LDevelop a uniform definition of ``child abuse and 
        neglect fatality'' and provide guidance related to 
        determining and reporting such fatalities and ensure 
        States and Tribes are using this new definition when 
        reporting data via the National Child Abuse and Neglect 
        Data System (NCANDS).
         LAid States in developing the means and mechanisms to 
        accurately collect provider-specific outcomes data, 
        consistent with the metrics and definitions associated 
        with AFCARS, NCANDS, and the CFSRs.
         LEstablish maximum caseload guidelines to promote 
        manageable caseload sizes for the child welfare 

                      Recommendations for Congress

         LSupport both funding and oversight for States and 
        Tribes to enhance foster parent recruitment and 
        retention activities to ensure robust background 
        checks, home studies, ongoing placement oversight, and 
        strong support services for foster parents.
         LSupport both funding and oversight for States and 
        Tribes to enhance caseworker recruitment and retention 
        activities to ensure child welfare caseworkers are both 
        prepared to enter the field and given the support 
        services necessary to carry out their jobs effectively.
         LAllow States and Tribes to use title IV-E funds to 
        support evidence-based services aimed at safely 
        preventing foster care entries.
         LConsider de-linking subsidized guardianship payments 
        from the Aid to Families with Dependent Children (AFDC) 
        income standard so that States and Tribes can receive a 
        Federal match on behalf of all children placed in 
        subsidized guardianship placements and promote equity 
        in the payment rate for kinship placements.
         LRequire all States to report to the National Child 
        Abuse and Neglect Data System (NCANDS) using standard 
        definitions and provide support for this data 
        collection and reporting.
         LConsider legislation creating an explicit private 
        right of action for children and youth in foster care 
        tied to components of the case plan and case review 
        requirements defined under section 475 of the Social 
        Security Act.\94\
    \94\ ``Compilation of the Social Security Laws,'' Social Security 
Act, section 475, https://www.ssa.gov/OP_Home/ssact/title04/0475.htm.
         LConsider statutory changes requiring HHS to assess 
        fiscal penalties on States for failing to meet CFSR 
        outcomes or system requirements and develop a penalty 
        reinvestment structure under which assessed penalties 
        must be used by the State to address the key identified 
        deficiencies (rather than be deposited into the Federal 
         LConsider amending section 479A of the Social 
        Security Act to require States to collect, and HHS to 
        audit, provider-
        specific child outcomes data in addition to State-
        specific data on outcomes such as: child fatalities, 
        maltreatment in care, recurrence of maltreatment within 
        6 months, exits from foster care by reason for the exit 
        (adoption or guardianship, reunification, 
        emancipation), time to reunification, re-entry rates, 
        and the average number of placements. Ensure this 
        performance data is available to the public and 
        considered by States or Tribes before making or 
        renewing a contract with the provider.
         LConsider prohibiting Federal title IV-E 
        reimbursements for providers who consistently perform 
        poorly on key safety, permanency, and well-being 
        indicators. Charge HHS with auditing States and 
        providers to determine which providers shall be 
        excluded from Federal title IV-E reimbursement.
         LRequire States to make their contracts with private 
        child welfare service providers publicly available and 
        include details on whether such providers are private 
        not-for-profit or private for-profit.



                               Appendix A






                               Appendix B





                               Appendix C




                               Appendix D





                               Appendix E






                               Appendix F






                               Appendix G