Free Order - District Court of Connecticut - Connecticut

Case 2:89-cv-00859-AHN

Document 539-2

Filed 10/01/2007

Page 1 of 46

Juan F. v Rell 2006 Comprehensive Targeted Review Executive Summary September 24, 2007

Table of Contents
Section Introduction Acknowledgements Collective Findings Monitor's Comprehensive Qualitative Report ­ Findings at a Glance Outcome Measure 1: Commencement of Investigations Outcome Measure 2: Completion of Investigations Outcome Measure 3: (Treatment Plans) & Outcome Measure 15 (Needs Met) Outcome Measure 4: Search for Relatives Outcome Measure 5: Repeat Maltreatment of Children Outcome Measure 6: Maltreatment in Out-of-Home Care Outcome Measure 7: Reunification Outcome Measure 8: Adoption Outcome Measure 9: Transfer of Guardianship Outcome Measure 10: Sibling Placement Outcome Measure 11: Re-Entry into DCF Custody Outcome Measure 12: Multiple Placements Outcome Measure 13: Foster Parent Training Outcome Measure 14: Placement within Licensed Capacity Outcome Measure 16: Out-of-Home Visitation Outcome Measure 17: In-Home Visitation Outcome Measure 18: Caseload Standards Outcome Measure 19: Reduction in the Number of Children Placed in Residential Care Outcome Measure 20: Discharge Measures Outcome Measure 21: Discharge of Mentally Ill or Retarded Children Outcome Measure 22: Multi-Disciplinary Examinations (MDE's) Page 2 3 4 6 8 8 11 17 19 21 23 26 28 29 31 34 36 34 34 38 41 42 43 43 45

1

Case 2:89-cv-00859-AHN

Document 539-2

Filed 10/01/2007

Page 2 of 46

Juan F. v Rell 2006 Comprehensive Targeted Review Executive Summary September 24, 2007

Juan F. v Rell 2006 Comprehensive Targeted Review Executive Summary Introduction The Revised Juan F. Exit Plan requires that the "Court Monitor will conduct case reviews to produce annual reports documenting the Department of Children and Families' performance and progress toward achieving the Outcome Measures defined within this Exit Plan, except Outcome Measures 3 and 15 for which there will be separate quarterly case record reviews. The annual report concerning 2006 data will include a synopsis of the quantitative data provided by DCF in the third and fourth quarters of the calendar year of 2006 as well as the quantitative and qualitative findings from the research questions documented in the attached addendum document. . . . Subsequent annual reports will include a synopsis of the quantitative data provided by DCF as well as the quantitative and qualitative findings for those subsequent calendar years from the research questions documented in the attached addendum document." On March 23, 2006 the Court Monitor submitted an overall methodological outline for the Comprehensive Annual Review to the Juan F. parties. Initially, Outcome Measure 5, Outcome Measure 6, Outcome Measure 12, Outcome Measure 16 and Outcome Measure 17 were slated for only quantitative analysis given the verified accuracy of data. However, upon reconsidering the reporting requirements, and to provide the most informative reporting to the parties, the Monitor added these measures to the qualitative reporting schedule. The Court Monitor's Office then began preparations to conduct the reviews with the input of all parties and with the assistance of the Technical Advisory Committee (TAC). The TAC was identified as a component in the Revised Monitoring Order dated October 12, 2005. Each of the measures was approached systematically with the development of protocols and methodologies which were submitted to each party with sufficient opportunity to comment and contribute to the development of the review process. To accommodate fiscal and resource restraints within the allotted timeframes, changes were made to the sampling populations. On December 12, 2006, the Court Monitor proposed changes to the methodology of the 2006-2007 Comprehensive Targeted Case Review. The reductions in sampling methodology were identified as follows: · Outcome Measures 1 and 2 - Change sample size from ~238 to a sample of 100 · Outcome Measure 7 - Change sample size from ~160 to a sample of 80 · Outcome Measure 11 - Change sample size from ~191 to a sample of 68 · Outcome Measure 13 - Change sample size from ~237 to a sample of 70 · Outcome Measure 17 - Change sample size from ~249 to a sample of 100 All parties considered the suggested samples sufficiently robust for evaluating the qualitative aspects of each of the measures. All quantitative data for these identified Outcome Measures was reviewed on numerous occasions, with query logic verified and corrections implemented as required. Those measures that presented with persistent data quality issues over the course of time since the Juan F. Revised Exit Plan was approved on July 11, 2006, were subject to review at more stringent confidence ratios and margin of errors. As the sample size and confidence level to review only qualitative data is not detailed in the Juan F. v Rell Revised Exit Plan and

2

Case 2:89-cv-00859-AHN

Document 539-2

Filed 10/01/2007

Page 3 of 46

Juan F. v Rell 2006 Comprehensive Targeted Review Executive Summary September 24, 2007

there is no controlling language regarding qualitative reviews, a Court Order was not required related to these methodological changes. This unprecedented review of approximately 2,500 cases provided a unique opportunity to analyze the Department's progress with respect to the Outcome Measures and consider the expectations that we place on DCF as an organization. This review confirms that tremendous progress has been made and sustained for many of the 22 Outcome Measures, but important work remains to ensure that children permanency, well-being and service needs are effectively addressed. Acknowledgements The Court Monitor would like to acknowledge the continued collaborative efforts of the Juan F. parties in developing this review process and the consultations provided by the TAC. Specifically I wish to acknowledge the following individuals who contributed to the development, training, review, data collection, analysis and drafting of this document:
DCF Personnel Kathy Acosta Debra Collins Liz Cyr Janet Gonzalez Juliann Harris Marcy Hogan Doug Howard Allon Kalisher Stanley Kasanowski Wanda Ladson Maxine McIntyre Jorge Martinez Fred North Barbara O'Connell Linda Raitt, RN Kim Somaroo-Rodriguez Sandra Tapia Isabel Turmeque Joan Twiggs Rynep VanEldik Maribel Vazquez Lynette Warner DCF Central Office Administration Lou Ando Rudy Brooks Darlene Dunbar Susan Hamilton Karl Kemper Brian Mattiello Technical Advisory Committee Claire Anderson Linda Arnold Dick Matt Judith Meltzer

Monitor's Staff and Contracted Reviewers Mary Corcoran Charlene Fleming Tom Gallese Mary Ann Hartmann Susan Marks Roberts Melodie Peet Joni Beth Roderick Michelle Turco

Other Stakeholders Mark Horwitz CAFAP Juan F. Attorneys Stephen Frederick, Esq. Marcia Lowry, Esq. Ira Lustbader, Esq. Jessica Polansky, Esq. Ann Rubin, Esq.

A number of overarching findings emerged from the targeted reviews conducted over the last six months. The findings confirm the sustained improvements in many areas of case practice that have been highlighted during the course of our quarterly reports. The reviews offered insight into areas where progress has not been adequate, or quality of the work requires continued improvement. Each of the targeted reviews presents a series of findings that pertain to the specific focus of that Outcome Measure. These findings are narrated within each corresponding chapter within this document. In addition, there are a number of overarching findings that span many, if not all of the measures. These are presented as collective findings below: 3

Case 2:89-cv-00859-AHN

Document 539-2

Filed 10/01/2007

Page 4 of 46

Juan F. v Rell 2006 Comprehensive Targeted Review Executive Summary September 24, 2007

Collective Findings · The results from the targeted review of approximately 2,500 cases confirm the tremendous progress and improvement in fundamental areas of case practice that has occurred over the course of the past three years. The Department has elevated its practice in key areas such as: o Visitation contact o Timely permanency outcomes for children o Provision of Multidisciplinary Exams (MDE) for children, o Timeliness of investigations o Increased emphasis on kinship resource searches, o Reduction of residential care placements, particularly those out of state o Improved educational/vocational outcomes for youth discharged after age 18 · Despite impressive improvements referenced above, the Department has not been successful in two significant and fundamental areas of practice: Treatment Planning and Needs Met. · Appropriate treatment planning remains areas of concern. The findings indicate that the Department is performing considerably lower than the benchmarks set within the Juan F. Revised Exit Plan. In spite of increased training, resources and numerous initiatives to focus the work, treatment plans are regularly developed without the full participation of the active case participants and often lack: · Clear focused goals for the child in placement and/or family · Inclusive action steps for the active case participants, providers and DCF, · Identification of progress related to the goals set in place during the prior six months This is not to say that treatment planning has not shown some improvement. Several areas of the treatment plan assessment have made gradual progress since implementation of the format currently in use. Further, progress has been noted due to the statewide implementation of the Family Conference (FC) model which places the focus on appropriate engagement and empowerment of families. While there is evidence that this process now has a foothold within many area offices, much more work is required to achieve system-wide competency and consistent utilization of this process. System gridlock exists in the treatment and placement service array creating a barrier to meeting children and families' needs. Discharge delays routinely occur at emergency departments (ED), group homes, residential treatment facilities, SAFE homes, and STAR/Shelter programs. There is a tremendous need for all levels of foster care and adoptive resources. Wait lists for in-home services and outpatient mental or behavioral health services exist for most area offices. The permanency outcome measures (OM7, OM8, and OM9) have been maintained at or above the Exit Plan requirement for three consecutive quarters. However, while the Department has greatly improved its performance in achieving timely permanency, these measures are predicated on exit cohorts. Of those children that remain in the system, many have the non-preferred goal "Another Planned Living Arrangement" (APPLA) and 4

·

·

Case 2:89-cv-00859-AHN

Document 539-2

Filed 10/01/2007

Page 5 of 46

Juan F. v Rell 2006 Comprehensive Targeted Review Executive Summary September 24, 2007

are languishing in the system with insufficient focus on development of connections to lifelong resources, and inadequate life skills training and discharge planning. More focused attention and actions must be taken on behalf of these children. · Given the success of the Department in achieving fundamental quantitative changes required by the Exit Plan benchmarks, attention must now focus on improving the quality of efforts. The targeted review findings suggest numerous opportunities for continued development and improvement. Continuous quality improvement must be emphasized at all levels of the Department. Each targeted review identified the Social Work Supervisor (SWS) as a critical component to producing positive outcomes for children and families. Competent guidance and support is required to navigate the multiple tasks required on the path to permanency and well-being. In every area measured, reviewers identified accurate assessment, discussion of risk, development of action steps, and follow through within the supervisory process as keys to successful outcomes. Appropriate use of the Area Resource Group (ARG) and external consultants in situations of substance abuse, mental health issues and domestic violence remains inconsistent across the cases sampled. There were many opportunities for ARG consultations that were not utilized - resulting in less robust assessment of risk and safety and less effective treatment planning. The culmination of these review efforts provides confirmation that the automated reports produced by the Department for the Exit Plan Outcome Measures accurately represent the levels of performance. Given the repeated confirmation and verification of the quantitative results for the majority of the measures, attention must now be focused on developing qualitative methodologies for analyzing system strengths and opportunities for development. The Court Monitor's future reviews will focus on targeted areas of concern related to meeting the needs of children and families. The overwhelming expectations and complexities facing DCF staff were noted by reviewers throughout this process. The current workload, especially for the SW and SWS level staff must be analyzed and considered in the context of producing positive outcomes. Increasing expectations to meet multiple, overlapping, focused and unfocused, funded and unfunded mandates are compounded by layers of responsibility and new labor intensive processes. An unmanageable workload will result in inefficient efforts by DCF on behalf of children and families. Left unchecked, this will create a situation that will threaten the progress that has been achieved.

·

·

·

·

The following chart provides an overview of the timeframes and populations from which all samples were drawn, as well as our qualitative findings for that measure. Further, the Monitor's Juan F. v. Rell Exit Plan Quarterly Report: April 1, 2007 ­ June 30, 2007 includes a table which provides the longitudinal data for each of the fourteen quarters that have been reported to date.

5

Case 2:89-cv-00859-AHN

Document 539-2

Filed 10/01/2007

Page 6 of 46

Juan F. v Rell 2006 Comprehensive Targeted Review Executive Summary September 24, 2007

Monitor's Comprehensive Qualitative Report ­ Findings at a Glance
Outcome Measure OM1: Investigation Commencement OM2: Investigation Completion OM3: Treatment Plans OM4: Search for Relatives OM5: Repeat Maltreatment OM6: Maltreatment in Out-ofHome (OOH) Care OM7: Reunification OM8: Adoption OM9: Transfer of Guardianship OM10: Sibling Placement Requirement 90% 85% 90% 85% 7% 2% 60% 32% 70% 95% Baseline N/A 73.7% N/A 58% 9.3% 1.2% 57.8% 12.5% 60.5% 57.0% Universe All reports accepted at the Hotline during the period of Oct 1, 2006 ­ Oct 31, 2006 (N= 2,387) All reports accepted at the Hotline during the period of Oct 1, 2006 ­ Oct 31, 2006 (N =2,387) 2nd Qtr 2007 ACR schedule All children that entered DCF custody during the period of April 1, 2006 through June 30, 2006. (N=737) All children that experienced substantiated abuse or neglect during the month of July 2006. (N= 876) All children that were in out of home placement during the 3rd Qtr 2006. (N=6,688) All children discharged via adoption from DCF custody during the 4th Qtr 2006 (N=131) All children discharged from care to reunification during the 4th qtr 2007 (N=327) All children that were discharged from DCF custody during the quarter of Oct 1, 2006 through Dec 31, 2006. (N=106) New sibling entries and changes in sibling placement within full population of all siblings in placement during 4th Qtr 2006. (N=2,333 children) All children that were discharged from DCF custody during the quarter of July 1, 2005 through Sept 30, 2005 (excluding Voluntary Service Placements) (N=639) All children in DCF custody for at least 30 days in the 4th Qtr 2006 (excluding voluntary, interstate and probate cases (N=5,808) Sample 100 reports 100 reports 76 cases 196 children 134 children: (67 w/ repeat and 67 w/ no repeat) 27 children: (12 w/maltreatment and 15 w/no maltreatment) 71 Children 80 Children 80 Children 604 Children (all identified siblings with change in placement or new entry) 113 Children (55 children w/ re-entry date, and 58 children w/ no re-entry date) 256 Children Findings 97.0% 95.0% 30.3% 94.4% 7.6% 0.18% 64.8% 27.5% 76.3% 85.5%

OM11: Re-Entry into DCF Custody

7%

6.9%

8.6%

OM12: Multiple Placements

85%

N/A

95.3%

6

Case 2:89-cv-00859-AHN

Document 539-2

Filed 10/01/2007

Page 7 of 46

Juan F. v Rell 2006 Comprehensive Targeted Review Executive Summary September 24, 2007

Monitor's Comprehensive Qualitative Report ­ Findings at a Glance
Outcome Measure OM13: Foster Parent Training OM14: Placement within Licensed Capacity OM15: Children and Families' Needs Met OM16: OOH Visitation Requirement 100% 96% Baseline N/A 94.9% Universe All licensed foster homes available to the Department on December 31, 2006. (N=3,723) All children in DCF custody for at least 30 days in the 4th Qtr 2006 (excluding voluntary, interstate and probate cases.) .(N=5,808) 2nd Qtr 2007 ACR schedule All children in DCF custody for at least 30 days in the 4th Qtr 2006 (excluding voluntary, interstate and probate cases). (N=5,808) All in-home treatment cases open 30 days or more as of December 31, 2006 (excluding probate, interstate and voluntary cases). (N=737) All open cases during 4th qtr 2006: 17,622 cases distributed among 1,340 workers. (including 170 trainees) All children in placement on Dec 31, 2006. All youth, 18 years of age or older, discharged from care 4th Qtr, 2006. Excluding from Juvenile Justice, Interstate, Probate, and Voluntary cases opened for the sole purpose of making monetary payments on behalf of the youth. (N=70) All youth, 18 years of age or older, discharged from care 4th Qtr, 2006 and who require discharge to DMHAS or DMR. Excludes Juvenile Justice, Interstate, Probate, and Voluntary cases opened for the sole purpose of making monetary payments on behalf of the youth. (N=29) All Juan F. children entering care for the first time and reaching 30 days in placement during the 4th quarter. (N=388) Sample 75 Foster Homes 202 Children Findings Qualitative Findings do not translate to a score. 90.1%

80% 85% (Month) 100% (Qtr) 85%

N/A 72.0% 87.0% N/A

76 cases 256 Children

51.3% 99.4% (Month) 99.2% (Qtr) 89.2%

OM17: In-Home Visitation

250 Families

OM18: Caseload Standards OM19: Residential Reduction OM20: Discharge Measures

100% 11% 85%

69.2% 13.5% 61.0%

N/A N/A 70 Youth

100.0% 11.0% 100.0%

OM21: Discharge to Department of Mental Health and Addiction Services (DMHAS)/Department of Mental Retardation (DMR) OM22: Multi Disciplinary Exams (MDE)

100%

N/A

29 Youth

97.0%

85%

5.6%

21 children identified as noncompliant with OM22

96.9%

7

Case 2:89-cv-00859-AHN

Document 539-2

Filed 10/01/2007

Page 8 of 46

Juan F. v Rell 2006 Comprehensive Targeted Review Executive Summary September 24, 2007

Outcome Measure 1: Commencement of Investigations and Outcome Measure 2: Completion of Investigations Overview The Juan F. v Rell Revised Exit Plan requires two performance benchmarks related to child protective service investigations. They are Outcome Measure 1: Commencement of Investigations and Outcome Measure 2: Completion of Investigations. The requirements of these two measures are defined as follows: Outcome Measure 1: Commencement of Investigations requires that, "DCF shall assure that at least 90% of all reports of children alleged to be abused, or neglected, shall be prioritized, assigned and the investigation shall commence within the timeframes specified below. If the report of child abuse or neglect is determined by the DCF Hotline to be... A. A situation in which failure to respond immediately could result in the death of, or serious injury to a child , then the response time for commencing and investigation is the same calendar day Hotline accepts the report. B. A non-life threatening situation that is severe enough to warrant a 24 hour response to secure the safety of the child and to access the appropriate and available witnesses, then the response time for commencing an investigation is 24 hours. C. A non-life threatening situations that, because of the age or condition of the child, the response time for commencing an investigation is 72 hours." In definitions within the Juan F. v Rell Revised Exit Plan document, the "commencement of the investigation" occurs when the DCF investigator attempts to make face-to-face contact with the parent or person responsible for the child's care, and or with the child(ren)". The "attempt at faceto-face contact" is made "when the investigator visit the home, school or other setting in an effort to interview the child(ren) and family members regarding the allegation of abuse or neglect." (See DCF Policy 34-4) The Department has reported achievement of this measure every quarter since the Fourth Quarter 2004. This review found 97.0% of the case sample (n=100) met Outcome Measure 1 as the response time achieved within the investigation was within the priority established for response at Hotline. This is consistent with the Department's reported performance for this measure. Outcome Measure 2: Completion of the Investigation requires that "at least 85% of all reports of alleged child maltreatment accepted by the DCF Hotline shall have their investigations completed within 45 calendar days of acceptance by the Hotline. Guidelines within the Exit Plan document indicate that completion of the investigation occurs only after the DCF SWS verifies the determination of substantiation or non-substantiation and enters that in LINK via the SWS approval of the investigation. The Department has reported achievement of Outcome Measure 2 consistently since the fourth quarter 2004. Our quantitative findings indicate that 95.0% of the reports accepted for investigation within our sample had investigations completed within 45 days of the date of acceptance at Hotline. Of those four cases not achieving the benchmark performance, the

8

Case 2:89-cv-00859-AHN

Document 539-2

Filed 10/01/2007

Page 9 of 46

Juan F. v Rell 2006 Comprehensive Targeted Review Executive Summary September 24, 2007

completion was documented at 51, 53, 54 and 55 days from date of acceptance. This is consistent with the Department's LINK reporting for this measure. Review Findings and Trends: The reviewers identified that 69% of the cases within the sample incorporated a majority of the elements of the risk assessment component measured, but that only 43% incorporated all expected policy elements. The latter issue may be one of disparity of standards for documentation which vary slightly office by office rather than poor practice. Overall rankings of the quality of the investigations resulted in 49.0% at a very good or optimal level, 40% at a marginal level, 10% at a poor level and 1.0% scored as adverse. There are data elements of the protocol left blank in many cases. Most frequently this was related to descriptive data elements, services, and notification. In some cases, this deficit was overcome by text within the investigation protocol. In others it was not. While there are variations between area offices, the differences are more apparent when the Special Investigation Unit (SIU) conducts investigations. This unit appears to have different practice guidelines regarding the data elements that are filled out within the investigation protocol document. While 97.0% of the sample commenced investigations within the specified timeframe, actual contact with alleged child victims was 4.6 days. Investigators were most successful with the Same Day priority response, in that 92.9% of those cases had documented contact with the alleged victim on the same date of the acceptance. The 24 hour response and 72 hour response cases achieved contact with the alleged victim within the priority response designation in 54.1% and 47.0% respectively. In 94.0% of the sample there was documented contact with the child victim. In the six cases where there was no contact with the alleged victim, reviewers felt that circumstances preventing contact in all but one case were well documented and reasonable. In 97% of the sample, there was documented contact with the alleged perpetrator(s). In the three situations where there was no contact with the alleged perpetrator, the circumstances preventing contact were well documented and reasonable. Use of the Area Resource Group (ARG) in situations of substance abuse, mental health issues and domestic violence remains inconsistent across the investigations sampled. Reviewers felt that there were many opportunities for ARG consultations that were not utilized - resulting in less robust assessment of risk and safety. Specifically, reviewers noted 17 cases for which they felt ARG consultation was required, and not included within the process. As a result, the risk assessments did not appear to be comprehensive and accurate. Key background and collateral contacts were not always documented. Fourteen percent of the cases did not have criminal background checks documented. Twenty percent did not document medical collateral contacts. Fifteen percent of the cases failed to incorporate interviews with other adult residents in the home where the alleged abuse took place.

9

Case 2:89-cv-00859-AHN

Document 539-2

Filed 10/01/2007

Page 10 of 46

Juan F. v Rell 2006 Comprehensive Targeted Review Executive Summary September 24, 2007

In 31.0% of the sample, service needs were identified by the Investigator but no follow up is documented upon disposition of the case. This potentially leaves a family that is not transferred to Ongoing Services no better informed or connected with resources available in the community then when they came to the Department's attention. In all, 88.7% of the reports with one specific incident date failed to identify that incident date in the appropriate data field. This can be a source of error when reporting on repeat maltreatment as the date of incident and report are often not the same. The majority of reports in this sample came from mandated reporters outside of the Department. In 74.7% of the cases reported by mandated reporters there was no indication of letters of notification to the reporter regarding the determination of abuse/neglect. Notification letters or contact with the alleged perpetrator to advise them of the investigation findings is poorly documented. In all, 42% of the sample lacked this documentation. For more details regarding each of these areas of measurement see the analysis within this chapter. Methodology The Monitor's Office created the tool for measurement of Outcome Measure 1 and Outcome Measure 2 with input of all parties in the fall of 2006. The parties agreed with a randomly selected sample of 100 cases. On October 2, 2006 the Monitor's Office requested the universe of all reports of abuse or neglect accepted at the Hotline during the period of October 1, 2006 through October 31, 2006. This allowed an opportunity to conduct a review verifying LINK data reporting related to Outcome Measure 1 and Outcome Measure 2 as well as to provide a universe from which to sample and review more qualitative practice elements regarding investigative practices. The Department provided data which identified 2,387 accepted, unduplicated reports during the quarter. A sample of 100 was selected from that universe. For full details of sampling see the Juan F. v Rell 2006 Comprehensive Targeted Review. The LINK record review was conducted over the first and second quarters of 2007. DCF Court Monitor review staff and DCF personnel assisted in the data collection efforts. A pilot test was conducted and necessary changes resulted to improve validity and reliability of scoring.

10

Case 2:89-cv-00859-AHN

Document 539-2

Filed 10/01/2007

Page 11 of 46

Juan F. v Rell 2006 Comprehensive Targeted Review Executive Summary September 24, 2007

Monitor's Office Case Review for Outcome Measure 3 and Outcome Measure 15 Overview The Juan F. v Rell Revised Exit Plan and subsequent stipulated agreement reached by the parties and court ordered on July 11, 2006, requires the Monitor's Office to conduct a series of quarterly case reviews to monitor Outcome Measure 3 (Treatment Planning) and Outcome Measure 15 (Needs Met). The implementation of this review began with a pilot sample of 35 cases during the third quarter 2006 and has been conducted quarterly since that time with samples of approximately 70-75 cases. For the Third Quarter 2007 the sample included 76 cases. Outcome Measure 3 requires that, "....in at least 90% of the cases, except probate, interstate and subsidy only cases, appropriate treatment plans shall be developed as set forth in the "DCF Court Monitor's 2006 Protocol for Outcome Measures 3 and 15" dated June 29, 2006 and the accompanying "Directional Guide for OM3 and OM15 Reviews" dated June 29, 2006." Review Findings and Trends The fourth quarter case review data for Outcome Measure 3 indicates that the Department attained the level of "Appropriate Treatment Plan" in 23 of the 76-case sample or 30.3%. · This reduction in the treatment plan scores reflects the overarching issue raised by reviewers ­ the need for greater supervisory oversight in the development of and approval of plans. Currently, plans are often approved that clearly are not in accordance with expected practice. Recent changes in the Administrative Case Review (ACR) SWS role in relation to reviewing plans with an eye to Outcome Measure 3, as well as fulfilling the role of the third party, objective reviewer may improve scores in coming quarters. However, reviewers have noted many times in the past that the documentation provided by the ACR SWS often addresses action steps, goals, and services, yet fails to be incorporated prior to the approval of the plan. The Ongoing SWS bears a great deal of the responsibility for improving the level of performance for this measure. · Treatment plans continue to lack inclusive action steps, identified goals for the upcoming six months, identification of services, and the input of providers (other than with foster parents ­ who are being engaged in discussion with regularity). 98.7% of the cases sampled had a plan less than 7 months old at the point of review. Three of the plans not passing (3.9%) did not have SWS approval. All three of these plans had one or more sections with less than a "very good" rating and would have been deemed inappropriate regardless of approval status. With respect to accommodating the primary language of clients, 94.7% of the cases had documentation that families' language needs were met (4 cases did not have documentation of meeting language needs). In-Home cases (n=22) appear to be most problematic in regards to inclusive and appropriate treatment planning. In all only 18.2% of the In-Home Cases were deemed as having appropriate treatment plans. Child in Placement cases with CPS designation had the greatest percentage of appropriate treatment planning for the sample set, with 69.6% of those cases deemed appropriate. Voluntary Service Child in Placement cases were considered appropriate in 60% of the cases reviewed.

·

·

11

Case 2:89-cv-00859-AHN

Document 539-2

Filed 10/01/2007

Page 12 of 46

Juan F. v Rell 2006 Comprehensive Targeted Review Executive Summary September 24, 2007

·

There have been recent changes to the definitions of the APPLA goals, and changes in expectations related to assignment of these less permanent scores. It may be that this transition along with the confusion of implementing this new policy is impacting the scoring for both treatment plans and needs met related to permanency. There was documented engagement of foster parents in discussions around treatment planning in 94.4% of the cases sampled. This is the highest rate to date. Mothers (73.8%), and identified support/kin (66.7%) are the next participants most frequently involved in discussions. Attendance rates at the ACRs and Family Conferences remain problematic for most case participants. Reviewers noted a failure to invite adolescents and fathers, and the overall lack of engagement with both children's and parents' attorneys.

·

As a result of discussion with area offices regarding the in-home cases reviewed, the Monitor has determined a need for a methodological change for the future reviews. Area Offices were identifying cases that they felt had appropriate treatment planning but that were not passing as a result of poor or missing LINK documentation or information known to the SW or SWS that was not represented in the LINK narrative but was of importance to how the treatment planning efforts unfolded. While we feel that scoring has given an accurate accounting of the treatment planning documents and records we reviewed, beginning with the Third Quarter 2007 reviews, we have made a change in approach. Many in-home cases do not have a family conference that typically provides an opportunity for the reviewer to clarify issues or obtain first hand knowledge. Therefore, for all in-home cases where there is no opportunity to attend the family conference and speak with the SWS or SW in person, we will include a brief phone interview with the SWS or SW. Outcome Measure 15 requires that, "....at least 80% of all families and children shall have all their medical, dental, mental health and other service needs met as set forth in the "DCF Court Monitor's 2006 Protocol for Outcome Measures 3 and 15 dated June 29, 2006, and the accompanying `Directional Guide for OM3 and OM15 Reviews dated June 29, 2006." Review Findings and Trends The case review data indicates that the Department attained the designation of "Needs Met" in 51.3% of the 76-case sample. · Nine of the 11 categories measured within Outcome Measure 15 showed some level of improvement over the prior quarter. The Department shows promising practices in legal action, safety of children in placement, attending to medical needs, and recruitment efforts for the prior period. Most problematic are provision of timely dental services, mental health, behavioral health, and substance abuse services. The permanency goals of children appear to have an impact on the ability of the Department to attain a level of practice and service provision that meets the identified needs of children and families. Children with non-preferred treatment planning permanency goals more frequently do not have their identified needs met. APPLA: Permanent Non-Relative Foster Care, APPLA: Other and Long Term Foster Care with a Relative are respectively 33.3%, 42.9% and 33.3% compliant. Preferred goals are compliant at rates greater than 55.0%.

·

12

Case 2:89-cv-00859-AHN

Document 539-2

Filed 10/01/2007

Page 13 of 46

Juan F. v Rell 2006 Comprehensive Targeted Review Executive Summary September 24, 2007

·

·

·

Structured Decision Making (SDM) training is now completed across the state area offices. The practice shows great promise if DCF maintains the integrity of the scoring protocols' definitions. There is an expected learning curve as SW and SWS work with the assessment tools in establishing both permanency and services needs. Of the needs which remained unmet from the prior planning period, mental health services continued to be most prevalent. Most frequently, the barrier documented was "client refusal"; however, reviewers noted that these circumstances often did not have documented engagement efforts such as client contact by SWS or ARG, or collaborative efforts with active providers to engage clients in needed services. 68.4% of the cases reviewed had clear documentation of a service need that should have been carried over to the treatment plan reviewed for this quarter, but was not.

For more details regarding each of these areas of measurement see the analysis within this chapter. Methodology: The Juan F. v Rell Revised Exit Plan and subsequent stipulated agreement reached by the parties and court ordered on July 11, 2006, requires the Monitor's Office to conduct a series of quarterly case reviews to monitor Outcome Measure 3 (Treatment Planning) and Outcome Measure 15 (Needs Met). The implementation of this review began with a pilot sample of 35 cases during the Third Quarter 2006. During the Second Quarter 2007, the Monitor's Office reviewed a total of 76 cases1 . Methodology will change with the Third Quarter 2007 reviews in that 50 cases will be reviewed by individual reviewers. This was necessitated by changes in DCF staff assignments and additional monitoring activities resulting from the Juan F. v Rell Action Plan. This quarter's 76 case sample was stratified based upon the distribution of area office caseload on March 1, 2007. The sample incorporated both in-home and out-of-home cases based on the overall statewide percentage reflected at the point that each sample is determined. For full details of sampling see the Juan F. v Rell 2006 Comprehensive Targeted Review. This is the last quarter where the methodology included pairing of DCF staff with Monitor's Review staff. Within the course of seven to twelve hours, each case was subjected to the following methodology. 1. A review of the Case LINK Record documentation for each sample case concentrating on the most recent six months. This includes narratives, treatment planning documentation, investigation protocols, and the provider narratives for any foster care provider during the last six-month period. 2. Attendance/Observation at the Treatment Planning Conference (TPC)/Administrative Case Review (ACR) or Family Conference (FC)2. 3. A subsequent review of the final approved plan is conducted fourteen to twenty days following the date identified within the TPC/ACR/FC schedule from which the sample was
The Exit Plan required a total of 70 cases be reviewed. Due to rounding and ensuring that each area office had representation of both in-home and out of home case assignments, a total of 76 cases were selected. 2 Attendance at the family conference is included where possible. In many cases, while there is a treatment plan due, there is not a family conference scheduled during the quarter we are reviewing. To compensate for this, the monitoring of in-home cases includes hard copy documentation from any family conference held within the six month period leading up to the treatment plan due date.
1

13

Case 2:89-cv-00859-AHN

Document 539-2

Filed 10/01/2007

Page 14 of 46

Juan F. v Rell 2006 Comprehensive Targeted Review Executive Summary September 24, 2007

drawn. Each reviewer completes an individual assessment of the treatment plan and needs met outcome measures and fills out the scoring forms for each. 4. A final meeting with the assigned teammate is held to jointly arrive at the final scores for each section and overall scoring for OM3 and 15. Individual scoring and joint scoring forms are then submitted to the Monitor. (This step may change as determined appropriate by the DCF Court Monitor after evaluation of the process, feedback from review staff and fiscal/staffing considerations.) Although the criterion for scoring requires consistency in definition and process to ensure validity, no two treatment plans will look alike. Each case has unique circumstances that must be factored into the decision-making process. Each reviewer has been provided with direction to evaluate the facts of the case in relationship to the standards and considerations and have a solid basis for justifying the scoring. In situations where agreement cannot be reached, the team requests that the supervisor become a third voice on those areas of concern. They present their opinions and findings and the supervisor determines the appropriate score to reflect the level of performance for the specific item(s) and assists them in the overall determination of compliance for OM3 and OM15. If the team indicates that there are areas that do not attain the "very good" or "optimal" level, yet consensus is the overall score should be "an appropriate treatment plan" or "needs met" the team clearly outlines their reasoning for such a determination and it is reviewed by the Court Monitor for approval of an override exception. These cases are also forwarded to the Technical Advisory Committee (TAC) for review. During the Fourth Quarter, there were 19 such cases submitted for consideration/assistance of supervisory oversight. Of the 19 cases, seven resulted in the approval of an override to allow one or the other measure to achieve a passing score. These cases will be identified in the overall scoring tables later in this document.

14

Case 2:89-cv-00859-AHN

Document 539-2

Filed 10/01/2007

Page 15 of 46

Juan F. v Rell 2006 Comprehensive Targeted Review Executive Summary September 24, 2007

Table 9: Measurements of Treatment Plan OM 3 ­ Number and Percent of Rank Scores for All Cases Across All Categories of OM3
Category I.1 Reason for DCF Involvement I.2. Identifying Information I.3. Strengths/Needs/Other Issues I.4. Present Situation and Assessment to Date of Review II.1 Determining the Goals/Objectives II.2. Progress3 II.3 Action Steps to Achieving Goals Identified II.4 Planning for Permanency Optimal "5" 42 (55.3%) 3 (3.9%) 17 (22.4%) 17 (22.4%) 15 (19.7%) 19 (25.0%) 3 (3.9%) 24 (31.6%) Very Good "4" 30 (39.5%) 57 (75.0%) 37 (48.7%) 38 (50.0%) 26 (34.2%) 30 (39.5%) 27 (35.5%) 34 (44.7%) Marginal "3" 4 (5.3%) 15 (19.7%) 22 (28.9%) 20 (26.3%) 29 (38.2%) 22 (28.9%) 37 (48.7%) 14 (18.4%) Poor "2" 0 (0%) 1 (1.3%) 0 (0%) 1 (1.3%) 6 (7.9%) 2 (2.6%) 8 (10.5%) 2 (2.6%) Adverse/Absent "1" 0 (0%) 0 (0%) 0 (0%) 0 (0%) 0 (0%) 1 (1.3%) 1 (1.3%) 2 (2.6%)

Table 10: Measurements of Treatment Plan OM 3 ­ Number and Percent of Rank Scores for Out of Home (CIP) Cases Across All Categories of OM3
Category I.1 Reason for DCF Involvement I.2. Identifying Information I.3. Strengths/Needs/Other Issues I.4. Present Situation and Assessment to Date of Review II.1 Determining the Goals/Objectives II.2. Progress4 II.3 Action Steps to Achieving Goals Identified II.4 Planning for Permanency Optimal "5" 25 (46.3%) 1 (1.9%) 13 (24.1%) 16 (29.6%) 8 (14.8%) 13 (25.0%) 2 (3.7%) 14 (25.9%) Very Good "4" 26 (48.1%) 40 (74.1%) 26 (48.1%) 23 (42.6%) 19 (35.2%) 24 (46.2%) 23 (42.6%) 24 (44.4%) Marginal "3" 3 (5.5%) 12 (22.2%) 15 (27.8%) 14 (25.9%) 23 (42.5%) 13 (25.0%) 23 (42.6%) 13 (24.1%) Poor "2" 0 (0%) 1 (1.9%) 0(0%) 1 (1.9%) 4 (7.4%) 1 (1.9%) 6 (11.1%) 2 (3.7%) Adverse/Absent "1" 0 (0%) 0 (0%) 0 (0%) 0 (0%) 0 (0%) 1 (1.9%) 0 (0%) 1 (1.9%)

Table 11: Measurements of Treatment Plan OM 3 ­ Number and Percent of Rank Scores for In-Home Family Cases Across All Categories of OM3
Category I.1 Reason for DCF Involvement I.2. Identifying Information I.3. Strengths/Needs/Other Issues I.4. Present Situation and Assessment to Date of Review II.1 Determining the Goals/Objectives II.2. Progress II.3 Action Steps to Achieving Goals Identified II.4 Planning for Permanency Optimal "5" 17 (77.3%) 2 (9.1%) 4 (18.2%) 1 (4.5%) 7 (31.8%) 6 (27.3%) 1 (4.5%) 10 (45.5%) Very Good "4" 4 (18.2%) 17 (77.3%) 11 (50.0%) 15 (68.2%) 7 (31.8%) 6 (27.3%) 4 (18.2%) 10 (45.5%) Marginal "3" 1 (1.3%) 3 (13.6%) 7 (31.8%) 6 (27.3%) 6 (27.3%) 9 (40.9%) 14 (63.6%) 1 (4.5%) Poor "2" 0 (0%) 0 (0%) 0 (0%) 0 (0%) 2 (9.1%) 1 (4.5%) 2 (9.1%) 0 (0%) Adverse/Absent "1" 0 (0%) 0 (0%) 0 (0%) 0 (0%) 0 (0%) 0 (0%) 1 (4.5%) 1 (4.5%)

3 4

Excludes two cases that were newly opened ­ ranked as N/A- too early to note progress (2.6%). Excludes two cases that were newly opened ­ ranked as N/A-too early to note progress (3.7%).

15

Case 2:89-cv-00859-AHN

Document 539-2

Filed 10/01/2007

Page 16 of 46

Juan F. v Rell 2006 Comprehensive Targeted Review Executive Summary September 24, 2007

Table 13: Measurements of Treatment Plan OM 15 ­ Percentage of Rank Scores Attained Across All Categories5
Category # Ranked Optimal "5" # Ranked Very Good "4" # Ranked Marginal "3" # Ranked Poor "2" # Ranked Adverse/Absent "1" N/A To Case

I.1 Safety ­ In Home I.2 Safety ­ Children in Placement II.1 Securing the Permanent Placement ­ Action Plan for the Next Six Months II.2. DCF Case Management ­ Legal Action to Achieve the Permanency Goal During the Prior Six Months II.3 DCF Case Management ­ Recruitment for Placement Providers to achieve the Permanency Goal during the Prior Six Months II.4 DCF Case Management ­ Contracting or Providing Services to achieve the Permanency Goal during the Prior Six Months III.1 Medical Needs III.2 Dental Needs III.3 Mental Health, Behavioral and Substance Abuse Services IV.1 Child's Current Placement IV.2 Educational Needs

8 (30.8% 28 (48.3%) 25 (43.1%) 54 (71.1%) 38 (60.3%)

11 (42.3%) 24 (41.4%) 18 (31.0%) 20 (26.3%) 14 (22.2%)

6 (23.1%) 5 (8.6%) 14 (24.1%) 1 (1.3%) 11 (17.5%)

1 (3.8%) 1 (1.7%) 1 (1.7%) 1 (1.3%) 0 (0%)

0 (0%) 0 (0%) 0 (0%) 0 (0%) 0 (0%)

50 18 18 0 13

29 (38.7%)

27 (36.0%)

19 (25.3%)

0 (0%)

0 (0%)

1

35 (46.1%) 43 (56.6%) 19 (27.5%) 25 (44.6%) 25 (42.4%)

28 (36.8%) 11 (14.5%) 30 (43.5%) 18 (32.1%) 19 (32.2%)

10 (13.2%) 15 (19.7%) 15 (21.7%) 13 (23.2%) 10 (16.9%)

1 (1.3%) 3 (3.9%) 5 (7.2%) 0 (0%) 4 (6.8%)

2 (2.6%) 4 (5.3%) 0 (0%) 0 (0%) 1 (1.7%)

0 0 7 20 17

Percentages are based on applicable cases for the individual measure. Those cases marked N/A are excluded from the denominator in each row's calculation of percentage. At the point of sampling, the total number identified for the in-home sample was 23 cases. However, a number of cases had both in-home and out of home status at some point during the six month period of review.

5

16

Case 2:89-cv-00859-AHN

Document 539-2

Filed 10/01/2007

Page 17 of 46

Juan F. v Rell 2006 Comprehensive Targeted Review Executive Summary September 24, 2007

Outcome Measure 4 ­ Search for Relatives Overview This review of Outcome Measure 4 ­ Search for Relatives, is a qualitative review that will supplement the quarterly data provided by DCF and verified by the Court Monitor, regarding the Department's efforts to conduct thorough searches for possible familial and social resources for placement or support of a child who requires out of home placement. Outcome Measure 4 requires that DCF comply and sustain the following level of practice: If a child(ren) must be removed from his/her home, DCF shall conduct and document a search for maternal and paternal relatives, extended formal or informal networks, friends of the child or family, former foster parents, or other persons known to the child. The search period shall extend through the first six months following removal from home. The search shall be conducted and documented in at least 85% of the cases. The Department has reported compliance with the search for relatives measure in each quarter reported to date from the second quarter 2005 through the third quarter 2006. Our review is consistent with the Department's reporting of Outcome Measure 4, in that 94.4% of the children who were in our sample had a documented relative search narrative entered within the LINK record. Review Findings and Trends As shown below, all but one of the area offices has attained compliance with the measure defined as an entry in the appropriate narrative form regarding relative search. Table 17: Compliance with OM 4 within Each of the Area Office Locations
Area Office Bridgeport Danbury Greater New Haven Hartford Manchester Meriden Middletown New Britain New Haven Metro Norwalk Norwich Stamford Torrington Waterbury Willimantic Grand Total Sample 9 3 12 26 25 5 4 22 26 4 11 4 4 27 14 196 % of Compliance with Measure 100.0% 100.0% 100.0% 80.8% 96.0% 100.0% 100.0% 95.5% 96.2% 100.0% 100.0% 100.0% 100.0% 88.9% 100.0% 94.4%

Relative search entries are most often noted in investigation or early after the transfer of the cases with little or no subsequent follow up. While the measure is met, concerted activity appears 17

Case 2:89-cv-00859-AHN

Document 539-2

Filed 10/01/2007

Page 18 of 46

Juan F. v Rell 2006 Comprehensive Targeted Review Executive Summary September 24, 2007

minimal in a number of cases with regard to follow up on all potential resources identified. In some cases this is due to placement with a resource early in the process. The failure to thoroughly pursue potential resources impacts identification of placement opportunities in the event of disruption. Even with these limitations, results yielded are impressive. There were 75 foster placements, 5 respite resources and 43 visiting resources identified within the 196 case sample. In situations where documentation offered insight, the following barriers or issues were identified: · There is considerable improvement in the consistency of conducting at least one preliminary relative search in comparison with prior reviews. In all, 168 (85.7%) of the cases identified at least one possible relative or known party to be researched. Within these 168 cases, 146 had indication of contact attempted with that resource, and 135 documented interest of the identified resource to be considered for placement or visitation. · Relative search was too often a once and done entry. In some instances, the narrative entry briefly indicates "no relative resources identified by mother". There is little discussion documented, or follow up once a rapport has been established to revisit the issue with the parent, guardian, other known relatives or children. Reviewers identified 21.4% of the cases as having resources identified, but not pursued. · Parents sometimes refuse to provide information when first approached and documented follow up at a later date is not readily evident. · Children and adolescents are often not included in the dialogue to identify relative resources. Only 68.6% of the cases documented any conversation with a child about possible resource. · Unknown fathers often present challenges and delays in resource identification. · In many cases relatives often had many of the same CPS, criminal backgrounds, and mental health or substance abuse issues and were therefore are not viable options. · When placements with relatives are made early in the investigation or Ongoing Services case, further exploration of additional relatives is frequently not documented to prepare for the possible disruption. This lack of planning inhibits timely consideration of alternate placement options which could have been researched and assessed for appropriateness during the initial six month period. · A portion of the children or adolescents were in a high level of care due to mental health or physical needs. The review found that given this fact, the search was not conducted immediately or was postponed in light of lengthy treatment process. This inappropriate deferral of research inhibits the timely identification of visiting resources and preparation to potentially prepare and meaningfully plan the discharge. · Paternal relatives are pursued with less frequency than maternal relatives. · Situations of relatives coming forward and then rescinding the offer to be a resource were noted. Delays in placement with viable relatives can result from these interruptions in planning. · Interstate compact issues cause delays in assessment of resources. · Criminal and CPS background checks are sometimes not documented in conjunction with placements. There were 54 instances within this sample that failed to document the

18

Case 2:89-cv-00859-AHN

Document 539-2

Filed 10/01/2007

Page 19 of 46

Juan F. v Rell 2006 Comprehensive Targeted Review Executive Summary September 24, 2007

· ·

criminal background check. It is assumed they are completed by FASU as part of the process, but documentation is not consistently incorporated into the LINK record. Documentation indicates that prior foster parents are not often considered as possible resource. Appropriately, non-custodial parents are given preferential option to become resource. However, it appears in several situations that resource searches are not concurrently pursued when this option is being explored.

Methodology The Monitor's Office requested the DCF provide the universe of all children that entered DCF custody during the period of April 1, 2006 through June 30, 2006. This request was fulfilled with the Department's submittal of an Excel Database including 737 children. Sampling methodology required a sample at a 95% confidence level (+/-6%). This resulted in the need to identify 196 children for the sample. For full details of sampling see the Juan F. v Rell 2006 Comprehensive Targeted Review. The LINK record review was conducted in the second quarter of 2007. This allowed for a sixmonth window of practice upon which to measure the level of performance in regard to "Search for Relatives." Court Monitor review staff and DCF personnel assisted in the data collection efforts. A pilot test was conducted and necessary changes and training resulted to improve validity and reliability of scoring.

Outcome Measure 5 Case Review: Repeat Maltreatment of Children Overview The DCF Court Monitor's Office is required by the Revised Juan F. v Rell Exit Plan to conduct a series of reviews on the 22 outcome measures resulting in a full status report in April 2007. This review, Outcome Measure 5 Case Review: Repeat Maltreatment of Children is a qualitative review that will supplement the quarterly data provided by DCF and verified by the Court Monitor, regarding instances of repeat maltreatment. Outcome Measure 5: Repeat Maltreatment of Children requires that DCF comply and sustain the following level of practice related to re-entry: "No more than 7% of the children who are victims of substantiated maltreatment during any six month period shall be the substantiated victim of additional maltreatment during any subsequent six month period. This outcome shall begin to be measured within the six month period beginning January 1, 2004." Review Findings and Trends The Court Monitor's findings indicate performance at a rate of 7.6%. This is highly compatible to the Department's automated reports. Our analysis included the full universe of children with repeat maltreatment (Children with Repeat Maltreatment (N=67) and a like number of Children in Care with No Repeat Maltreatment (n=67). The review results showed no major factors of significance distinguishing cases with repeat maltreatment from those that did not have repeat maltreatment. There were, however, several noteworthy trends and qualitative information that resulted from the review: 19

Case 2:89-cv-00859-AHN

Document 539-2

Filed 10/01/2007

Page 20 of 46

Juan F. v Rell 2006 Comprehensive Targeted Review Executive Summary September 24, 2007

·

·

·

· ·

· ·

·

Documentation of risk and safety assessments within the supervisory conferences or in the LINK entries for visits with children and families were often lacking. In 39% of the cases open during the interim period between substantiations the SW and SWS failed to document identification of risks. In many situations, red flags are identified, but were not incorporated into supervisory discussions. Within the sample set, 74.5% of the cases were deemed to have adequate risk assessment activity in the period following the July substantiation. Of those cases scored as adequate by reviewers, the rate of repeat maltreatment is 50.6%. Of those deemed inadequate, the rate of repeat maltreatment was higher, at 80%. Reviewers identified 14 cases where children in the sample set were reunified yet information documented within the case record raising concerns related to the reunification being rushed, or premature. Thirteen of these 14 children were in the repeat maltreatment group. None of the instances of repeat maltreatment documented in this sample were at the hands of an out of home care provider. All occurred within the home or community setting. The most frequently cited form of repeat abuse or neglect is "physical neglect", often the result of parents' substance abuse or domestic violence episodes. In all, 41 of the 67 cases documenting repeat maltreatment had the same or very similar allegations raised as identified in the allegation substantiated in July 2006. A total of 34.3% of all perpetrators in the sample had three or more substantiations in the last 12-month period. Determining the level of impact of service provision as a factor in repeat maltreatment presents unique challenges. Engagement opens families up to additional scrutiny by mandated reporters, and at the same time offers opportunities for growth or rehabilitation. There are also factors beyond program attendance that are related to clients' ability to apply or comprehend the intent and lessons provided. These are not captured through this record review process but could be well served by a methodology incorporating interviews of stakeholders in each case sampled. The rate of collaboration between the DCF ISW and the Ongoing Services SW was quite high, with 92.6% of the cases sampled documenting some level of contact during the investigation.

Methodology The Monitor's Office requested the DCF provide the universe of all children that experienced substantiated abuse or neglect during July 2006. On March 28, 2007, the Department provided a universe from the Performance Outcomes for Children (POC) Outcome Measure 5 database of 876 children identified as victims of substantiated abuse or neglect. Of that number, a system query identified 69 children, or 7.9% of that population as having a repeat incident of substantiation within 6 months of the first incident. The Monitor's Office initiated a process to review those 69 cases as well as 69 cases in which the system did not identify a subsequent episode. Upon initial review of the 138 cases, our office found cause to eliminate two cases due to episodes with incidents outside of the period of review. To be equitable, we also eliminated two cases in which there was no subsequent episode. The resulting sample therefore includes 134 cases with equal number of repeat maltreatment situations and no repeat maltreatment.

20

Case 2:89-cv-00859-AHN

Document 539-2

Filed 10/01/2007

Page 21 of 46

Juan F. v Rell 2006 Comprehensive Targeted Review Executive Summary September 24, 2007

The DCF Court Monitor review staff and DCF personnel assisted in the data collection efforts. A pilot test was conducted to ensure issues of reliability and validity can be addressed prior to initiating the full review. Minor edits resulted to both the tool and directional guide as a result of this process.

Outcome Measure 6 ­ Maltreatment in Out of Home Care Overview The DCF Court Monitor's Office is required by the Revised Juan F. v Rell Exit Plan to conduct an Outcome Measure 6 Case Review: Maltreatment of Children in DCF Care. This is a qualitative review that will supplement the quarterly data provided by DCF and verified by the Court Monitor, regarding the instances of maltreatment while in DCF custody. Outcome Measure 6: Maltreatment of Children in DCF Care requires that DCF comply and sustain the following level of practice related to re-entry: "No more than 2% of the children in out of home care on or after January 1 2004 shall be the victims of substantiated maltreatment by substitute caregivers while in out of home care." Review Findings and Trends The Court Monitor's findings indicate compliance was achieved at a rate of 0.18%. This is highly comparable to the Department's automated reports. An analysis of the two subsets within the population reviewed (Children Maltreated in Out of Home Care (N=12) and a like number of Children in Care with No Substantiated Maltreatment (n=15) did not identify any major factors of significance distinguishing those cases with repeat maltreatment, from those with not repeat maltreatment. As the maltreatment rate of 0.18% suggests, there are positive notes regarding case practice. The reviewers indicated that the majority of cases had appropriate action taken by the Ongoing SW in the period of time leading up to the report of alleged abuse or neglect (whether substantiated or not). In one case, the reviewer noted that the SW proactively used supervision, documented assessment, documented the resulting plan to address "red flags", and initiated consultation within the Department and among providers as necessary to prevent the abuse of the child. Issues were addressed, and the child was prevented from experiencing maltreatment while in outof-home placement. Worker-Child visitation appears to have a positive impact on the rate of maltreatment in care. Across the two subsets, visitation met or exceeded monthly standards in 24 cases (88.8%). Within the 12 cases of substantiated abuse or neglect there were additional regulatory violations cited against the provider. None of the cases reviewed in either cohort had stand alone regulatory violations. Reviewers noted that none of the six DCF foster care providers that were substantiated, had documentation of updates to their annual support plans. Throughout this process, reviewers observed four trends related to maltreatment of children in out of home care. 21

Case 2:89-cv-00859-AHN

Document 539-2

Filed 10/01/2007

Page 22 of 46

Juan F. v Rell 2006 Comprehensive Targeted Review Executive Summary September 24, 2007

·

· ·

·

The first issue was the lack of coordinated communication between the DCF Ongoing SW, FASU SWs and service providers. In the three month period leading up to the report to the Hotline and during the period shortly thereafter, documented contacts were often limited to narrative lacking collaborative assessment. In only 50% of the 12 cases was there documentation that the Foster and Adoptive Services Unit (FASU) or Program Review and Evaluation Unit (PREU) was informed by the SW or ISW of the incident. The second trend relates to workers not assessing or even minimizing the "red flags" that they often document within their own narratives. The third trend is related to effective use of supervision. While supervisory conferences are documented in most cases, the actual use of this time to create action steps and follow-up on prior concerns is not often documented as it relates to "red flags". The three month period under review had an average of 3.44 supervisory sessions documented. Frequency of supervision in itself, however, was not felt to be a driving factor in the rate of maltreatment. It was the content of discussion that was significant. In 14 of the cases in which risks were identified by the SW there was no documentation that these issues were raised to the SWS, and 100% had maltreatment substantiations. And finally, the reviewers noted that Case Aides' documentation of transport and visitation are routinely not addressed by the SW or SWS. Situations detailed by the case aides are not recognized or discussed at the supervisory level in spite of the noted risks and safety. The role of the Case Aide in contributing to the overall risk assessment ongoing during the life of the case needs to be examined.

For more details regarding each of these areas of measurement see the analysis within this chapter. Methodology The Monitor's Office requested the DCF provide the universe of all children that were in out of home placement during the third quarter 2006. The Department had some difficulty in provision of the data for this request which resulted in delay. This necessitated changing the period of review to the fourth quarter, 2006. The request was fulfilled with the submission of an Excel Database including 6,688 children (excluding committed delinquent, children in placement as result of ICO, and Probate cases). DCF provided a universe of 14 children in out of home placement identified as victims of substantiated abuse or neglect and a total population of 6,688 children in care over the period. Per the automated reporting, 0.2% of the population in care during the quarter was victims of substantiated abuse or neglect. This is correct given the logic implemented for reporting which substitutes the reporting date if no incident date is identified. However, upon initial review of the identified cases, two cases actually had an incident of abuse/neglect with an incident date that preceded the start of the period (but were incorrectly reported during the period). Therefore the rate of maltreatment while in care during the fourth quarter was revised to 0.18%. In order to provide a basis for comprehensive analysis of this outcome measure, the sample included all 12 children who had an incident in the fourth quarter resulting in substantiation and an additional 15 cases randomly selected from the statewide population of children in care, with

22

Case 2:89-cv-00859-AHN

Document 539-2

Filed 10/01/2007

Page 23 of 46

Juan F. v Rell 2006 Comprehensive Targeted Review Executive Summary September 24, 2007

or without an investigation undertaken during the period..6 This was done to allow a review of case practice issues that may impact the success or barriers of DCF in meeting OM 6. Due to the low number of maltreatment in care cases upon which this measure can be assessed, we caution against the use of this data for anything other than qualitative or descriptive purposes. Statistical significance should not be determined from one quarter's performance. A LINK record review was conducted during April-May 2007. DCF Court Monitor review staff and two DCF personnel assisted in the data collection efforts. A pilot test was conducted d