[ Main Page of Report | Contents of Report ]
Between March 1997 and December 1998, members of the NEFS study team conducted a series of telephone interviews and site visits to New Jersey's Division of Youth and Family Services (DYFS) in Trenton and to five of the seven NEFS study sites within the state to collect both descriptive and actual cost data information. This section describes the sources of cost data the study team found to be available in New Jersey and various methods for cost data collection. It will present what cost data are available on a per-case basis and offers options about how information on other non-child specific costs could be derived. Strengths and weaknesses associated with each approach are discussed.
The sources of cost data found in New Jersey include the state’s child expenditure ledger, contract management systems, case records, the administrative case management system, Medicaid, and information on the state’s indirect rates. Exhibit A-1 provides an overview by "crosswalking" each cost data element of interest with potential data source(s) in New Jersey.
| Service Category(5) | Cost Information Source | Availability of Case-specific Data |
|---|---|---|
Child Welfare Service Costs |
||
| Family Preservation Treatment and Administration | State Family Preservation Coordinator | Yes |
Foster Care |
||
| Maintenance and Administration | Child Ledger | Yes |
| Clothing | Child Ledger | Yes |
| Miscellaneous | Child Ledger | Yes |
DYFS Directly Provided Services |
||
| Investigations Assessments Counseling Case management |
There are no actual cost data in this area. Various means of estimating directly provided services are described below. | No - costs would be derived/estimated. |
Emergency Outlays |
||
| Food | Child Ledger | Yes |
| Shelter | Child Ledger | Yes |
| Transportation | Child Ledger | Yes |
| Indirect Public Agency Rate | DYFS Finance Administration | No - an average rate would be applied to all cases. |
External Service Costs - those funded by the agency and by external sources |
||
| Mental health Psychiatric exams Counseling | There are multiple sources depending on the funding stream including: child ledger, contract information system, case records, and Medicaid. | Case-specific data would be available if listed in child ledger or Medicaid database.(6) All other costs would be derived/estimated. |
| Substance Abuse Treatment | The majority of substance abuse cases are funded by the Department of Health or through Medicaid. | No, unless listed in Medicaid database. |
| Child day care | Contract Information System, Service Information System and case records | No, costs would be derived/estimated. |
| Parent Training | Contract Information System, Service Information System and case records | No, costs would be derived/estimated. |
| Homemaker Services | Contract Information System, Service Information System and case records | No, costs would be derived/estimated. |
| Respite Care | Contract Information System, Service Information System and case records | No, costs would be derived/estimated. |
The child ledger is the most useful source of cost data in New Jersey because it ties many child welfare service expenditures directly to individual children. As shown by Exhibit A-1, it provides information on foster care and related clothing costs, individual service costs including homemaking, psychological assessment, therapy, and medical expenses not reimbursable by Medicaid, transportation, and other services not generally available through existing contracts. It also includes the costs of emergency outlays to families given by the district offices. The State Administrator of the Office of Accounting reported that “conservatively, 85 percent of expenditures for contracted services received by the study population are accounted for in the child ledger.” The two major child welfare service costs not available on the child ledger are the costs of contracted services paid for by block grants or state funds, including family preservation, and services provided directly by the public agency.
The child ledger is produced separately for each county and is organized by the child’s last name. It contains the child's identifying number (“KC” number) found in case records and all automated case information systems, and date of birth. It contains the month and year of service delivery, provider, days/units of care, type of service and expenditure (a sample ledger entry follows this appendix). Ledgers are prepared monthly and biannual summaries are bound for data storage purposes. Each ledger also contains a summary line of all ledger expenditures made on behalf of the case since the case was opened in DYFS.(7) The Child Ledger draws on information contained in:
The Client Payment Accounting System (CPAS). An automated payment system that provides client specific cost information for all out-of-home board and clothing payments.(8)
The K100 Form. A manual payment system for child specific services (e.g. transportation, private therapist, child care) and partial month payments for out-of-home care.
District Office Bank Accounts. District offices have individual flexible funding accounts to support emergency needs of children and families. While service information is not consistently entered into the state's administrative database the Service Information System (SIS), because district offices can immediately issue a check, this cost information is available on a case-specific basis through the child ledger.
The ledger contains cost information on actual expenditures made to vendors. Therefore, information is only as up-to-date as the vendor’s own billing system. Reportedly, charges related to the CPAS and the K100 systems are entered into the child ledger within 30 days. Emergency expenses funded through the district offices are posted within 60 days only account for only about two percent of the total child ledger budget.
It is important to note that the child ledger is prepared by the Data Center within the umbrella agency that includes DYFS, the Department of Human Services. Therefore, the data are not controlled directly by child welfare. Further, the child ledger is not an online system; it is the product of matches between three payment systems and the state's administrative Service Information System, SIS. DYFS does not have access to the full database. Each month the Data Center prepares a printout for the Accounting Office in DYFS to review for accuracy. It does not appear that the Data Center maintains any of the historic matches on data files - historic information is only available in hard copy in Trenton or at the individual county offices. The only way to retrieve historic data for the study period is to manually review the hard-copy ledgers and copy the relevant pages.(9)
Pros and cons: The child ledger is the best single source for client specific cost data on a range of child welfare funded services and will reportedly account for the vast majority of contract service costs. The major drawback to this data source is its burden to study staff that will need to review all relevant ledgers and construct a new database from the data gathered.
A wide range of DYFS funded contracted services are not paid for on a child specific basis including family preservation. These services, known as cost reimbursement services, are funded on a unit basis; for example, DYFS might contract with a mental health center to provide 100 psychological assessments a year. Regional Offices maintain the database for these services, called the Contract Information System (CIS). Expenditures for contracted services such as family preservation, parent education and mental health are submitted monthly on a predetermined program “component” basis. In order to track payments and performance, CIS maintains information on:
From this information, for each contracted service a “unit rate” is calculated at six-month intervals based on either the number of cases that received services or the number of days (or other equivalent) of service provided.
To estimate costs for services funded on a cost reimbursement basis, the study team will need to gather child specific service utilization information from other sources discussed later in this plan and apply to each service, the unit cost from the Contract Information System. Specific steps would include:
Alternatively, (to get actual or true service costs) submit client specific service information to the service provider who must confirm service receipt, verify the source of payment and provide actual service costs.
Pros and cons: The Contract Information System includes a wide range of child welfare services and has useful information about unit rates that are time and provider specific. The major drawback to this approach is that the single best source of information on service utilization, the case records, will be labor intensive to retrieve and review (see next section for a description of case records). A review of the automated case management system is also recommended. In addition, since there is no single source that will identify all services a family received, it would be advisable to use multiple data sources (e.g., case records supplemented by service record printouts from the state's administrative system, and/or case worker interviews).
Members of the study team reviewed a small sample of case records in New Jersey’s Camden, Essex and Passaic Counties to look for documentation of caseworker activities and time dedicated to individual cases. Information was also sought on the utilization of contracted and other external services. Case records were reviewed from the time the case went through NEFS random assignment through the time of the site visit. To facilitate the identification of child welfare services funded by the agency, DYFS prepared county specific printouts of all standing agency contracts to match with services listed in the case records.(10)
Case records contain handwritten contact sheets listing the worker's ongoing interaction with the case, computer printouts from the agency's administrative database, and several forms that provide demographic and administrative information about the case. For a sample of a service-related case record form, see Appendix E.
Caseworker documentation of their ongoing contact with families seemed fairly thorough at each of the three sites. The number of entries on the contact forms after random assignment ranged from five (a case opened for two months) to 109 (a case opened for 13 months). The average number of entries across the nine records was 53. “Entries” are defined as any effort noted in the record by the DYFS worker on behalf of the sample family, for example, follow-up phone calls to the family or other concerned parties, home visits, correspondence that was recorded, etc. The detailed information on the contact sheets appears to supply a good deal of information about the types of directly provided services delivered to each family.
Client case records are located in the county office in which the family is based. For our sample reviews, District Administrators with the help of line workers collected all records and provided conference rooms to review the material. On average, each case record took team members two and a half hours to review and code (although this time might decrease as proficiency increases). The records are very detailed, with several pages of contact sheets, other forms and documents and computer printouts. Several of the forms and documents had duplicates in the files and often forms, computer printouts, and contact sheets were not in chronological order making the recording of services difficult.
To estimate the amount of time workers devoted to each case, the study team would conduct focus groups comprised of workers and supervisors to account for major bundles of activities that can account for the majority of casework time. Participants would be asked to define service units by identifying the 7-10 "bundles," or primary clusters of activities they engage in on a routine basis for different types of cases (e.g., those in foster care or those in ongoing protective services). For example, one "bundle" of activities might relate to court hearings. The bundle may include case review activities, meetings with the family and/or child to explain the court proceedings, completion of court papers, and actual time spent in court.
The focus group members would also be asked to assess whether existing case records document when these bundles of activities (service units) occurred. Determining service units that would likely be provided in different amounts to experimental and control cases would be especially important. Workers would also be asked to identify which service units would likely be used at different rates by the study sample. Using this information, case records would be reviewed using standardized data collection forms that list all activities developed by the focus groups.
A second set of focus group meetings would help estimate the average amount of time spent on providing each service unit. If time estimates vary widely, it would be necessary to further subdivide the service unit. For example, if workers indicate that court hearings may take as little as two hours and as much as two days, then it would be necessary to identify the types of hearing that take approximately two hours (e.g., initial custody hearings) versus those that take 16 hours (e.g., dispositional hearings). The result of this effort would be a list of key casework service units and the average number worker hours associated with each unit. Using data on the cost of an hour of worker time, a unit cost for each service unit would be established.
Pros and Cons: The benefit of this approach is that it will give the most accurate description of actual casework activities conducted for each family including both the number and types of activities performed for each case. The major drawbacks to this approach are that it will be labor intensive and costly to review all case records and will not be backed by data other than caseworker estimates.
While the case records contain valuable information on directly provided services, they contain little useful information about the utilization of external services. There are several problems with the records for this purpose. First, while contact forms would indicate that a worker made a service referral, it was rarely clear whether a family accepted the referral. There was rarely any follow-up in the contact sheets and the records contained only limited documentation about service receipt.(11)
A second problem with service information contained in the case records was that even if it became clear that the family accepted a service, there was no indication of how the service was funded. The funding source will likely impact the amount of expenditure. For example, in a number of cases reviewed, either the mother or a teen child went into a substance abuse program for an evaluation or to a hospital for a psychiatric assessment. In some cases, these are services that DYFS has under a cost reimbursement contract but agency officials warned that a family would also be encouraged to use Medicaid or their own private insurance provider. Without the necessary paperwork in the files, the only way to determine if DYFS paid for the service is to go the service provider themselves, because DYFS does not track individual client names for cost reimbursement contracts.
The third complication is that even if it is determined that a family accepted a service, and the service is funded through a cost reimbursement contract, there was rarely information on actual service utilization, that is, the intensity or length of service. While some services will be discrete activities, a psychiatric assessment for example, others, such as youth companions or psychiatric placements are ongoing. It will be necessary to either interview workers or families, or establish average length of service before we can use service unit costs to construct a total cost for each service.
Pros and cons: While often inconclusive, the case records still offer the most thorough source of information on receipt of external services. While labor intensive, all references to service referrals could be verified with the service providing agency to ensure service receipt, payment source and retrieve actual cost data. There are three drawbacks to this approach. First, case record reviews will be a heavy burden to study staff. Second, in order to verify the information gathered from the records, it would be advisable for study staff to submit client names and approximate dates of service to a wide range of service providers across the state. Third, it is unclear how willing the provider agencies will be to research and supply the requested cost information.
Alternatively, to address the second and third concerns, with the use of agency records, develop average or estimated service costs (based on average duration) for each service and apply them to the service referrals found in the case record.
New Jersey's automated administrative case registration and management system is called the Service Information System (SIS). All cases referred to DYFS for abuse or neglect, family problems, adoption services, or from juvenile probation are entered in this system.
The administrative data files can supply the number of referrals, the number of investigations of abuse and neglect, the number of times the case opened for service, the length of service (with opening date to closure dates), the number of foster and adoptive home placements, and the length of placement. In short, the administrative data files provide information on most of the major clusters of direct service activities that were provided to the case.
Information from the system's service file can also be used to supplement other sources of service information provided to a family or child. Services commonly listed in SIS include: foster care, homemaker services, parent training, respite care, health related services, psychological/therapeutic services, day care, day treatment, adoption, independent living, family preservation, camping, and clothing and food allowances. However, DYFS workers and administrators cautioned study staff on relying on SIS for service information, explaining that there is not a required data screen and there is no quality control. In addition, the service files do not contain any cost information.
Pros and cons: The SIS administrative data files are available to the study team members. These files were obtained from the state during the course of the study and were used in the program evaluation. These files contain information on the frequency of the major casework activities (e.g. number of investigations, number of foster care placements, etc.). SIS can provide the time reference needed to collect cost data from the child ledger since the service dates are on these files. This data would be useful to supplement other data sources described in this report.
As mentioned above, there is no thorough record maintained on the actual amount of time spent on individual cases by DYFS staff. To estimate these costs, two other data sources were reviewed: information from the state's Workload Standards Project and the Random Moment Survey. In addition to these two sources, two other cost estimation options are presented below.
A Workload Standards Project was implemented in New Jersey in 1992 to determine appropriate caseloads for case managers.(12) The new casework standards resulting from the project include a description of activities, frequency and duration of activities by case type (see Appendix F for examples of service descriptions). While available, the Director of Research and Analysis at DYFS noted that by 1997, workload standards were already out-of-date, largely due to the influx of substance abuse cases that are much more labor intensive and severely skew time spent per case.
Pros and cons: While information produced by this study should not be taken at face value, components of the analysis could be used for discussions with caseworkers and supervisors as a basis on which to build new estimates of direct service costs. For example, the study recommended that "placement/intake" activities should take 270 minutes (or four and a half hours), and is made up of several service components. This information could be used in the caseworker focus groups to devise discrete casework activities and establish activity duration.
The Random Moment Survey (RMS) is designed to allocate administrative costs associated with DYFS cases for claiming Title IV-E (foster care and adoption) reimbursement from the Federal government. The RMS provides a breakout of “observations” related to specific service categories delivered to specific children at one point in time by claiming categories (a sample Random Moment Survey is included in this appendix). Looking across all DYFS staff sampled, it provides a proportional account of activities (observations) taking place at that time.
Since the Random Moment Survey sorts discrete observed activities into cost categories, information from this source can be used to estimate the value of specific casework activities only when DYFS has the actual total number of those activities that took place over a year. A sample equation for determining investigations follows.
Where:
I: number of observations during RMS related to the Investigations category
O: total number of Observations during the RMS
Y: relative value of all investigations in a year (solve for this variable)
B: total actual annual budget for RMS population
I/O as Y/B
Using available information on “I’, “O” and “B”, once Y (the value of all investigations conducted over a year), has been solved for, divide Y by the actual number of investigations that took place that year to establish an average value for each investigation.
Pros and cons: No burden on state staff, but only produces average estimates that can be applied on case-specific basis. Also, this methodology will not work for less discrete activities in which the total number conducted over a given year is unknown (e.g., ongoing casework).
Through the use of focus groups with various types of workers, establish the average time devoted to different types of clients (for example, substance abusing), at different stages of DYFS involvement (screening, investigation, ongoing, placement). Establish both average number of contacts and length of time of each contact (home visits, office visits, court appearance, and paperwork). Information from the state’s Workload Standards Project might be applied. Using average salary levels, and total time (as measured by number of casework activities) devoted to a case, derive value of labor.
Pros and cons: Supervisors and workers appear willing to participate in these focus groups and this approach is much less labor intensive for study staff than reviewing individual case records. However, this option produces only average estimates and requires multiple assumptions of time requirements and client characteristics. These average estimates will be based on a composite of all families served or the most remembered anecdotal family and not necessarily the families studied in this evaluation.
d. Determine Value of Labor by Length of Stay.
A final alternative to establishing the value of directly provided services would be to disregard specific casework activities and determine casework cost by length of time a case remained opened in the child welfare agency. (As described above, this was the favored approach by the site.) Using administrative data on the length of time a case was opened, and monthly DYFS personnel costs related to individual DYFS units (e.g. Investigators, Ongoing CPS, Placement), create a weighted average monthly value for labor associated with each DYFS unit. Monthly unit labor costs would be the product of the agency's personnel expenditures for that unit divided by the size of the caseload served, divided by 12. Dependent on where the case was in the system, the study team could apply the average cost of labor to each month the case was opened.
Pros and cons: This option presents very limited burden to state staff, but only produces rough, average estimates contingent on length of stay. This method can be skewed when cases are left open erroneously without service and then artificially closed at some administratively appointed time. An additional problem with this approach is that it would not distinguish between important differences between and within the treatment and control groups in service intensity. To address this, it would be necessary to identify subgroups of cases that could be expected to receive differing service intensity and develop different cost multipliers.
Medicaid is used to fund a wide range of counseling and substance abuse services used by study families. In New Jersey, the Department of Medical Assistance maintains all Medicaid cost data. Information is available on a case-specific basis but due to staff shortages in the Department, information must be requested as much as two years prior to the time it is needed.
In order to access Medicaid expenditures from the Department of Medical Assistance (MA), the study team will need to gather the MA numbers for each study participant. In some cases these numbers will be included in the Service Information System along with other identifying information. In other cases the study team will need to request that the MA Systems Office conduct a data match using family names and dates of birth. The Director of this office reported that this could take up to two years given the backlog of data analysis in the office. Once the client MA numbers have been retrieved, the research team will need to submit a formal request to the department's oversight board specifying our data request and how we want the information delivered to us.
Pros and cons: Medicaid is used to fund a range of services provided to child welfare involved families such as mental health counseling, and substance abuse treatment. The major drawback to this data source will be the time involved in collecting the Medicaid Assistance number for each member of the study sample and the additional time required to retrieve the cost information.
Exhibit A-2 Sample New Jersey Child Ledger
Between April 1997 and early 1998, members of the study team conducted a series of telephone interviews and site visits to the Department of Children's Services (DCS) in Nashville and Shelby County (Memphis), Tennessee to review the state's financial administrative systems. The following summary of data availability describes the payment systems in place at that time. Since then, the state has been working to consolidate the systems described below. While state officials have informed the study team that the new system contains essentially all of the same information as the old, before conducting a full cost study in this site, a thorough update from the state would need to be conducted to verify data availability and means for accessing the information.
In addition to consolidating its financial systems, during the entire study period Tennessee was in the midst of transitioning to its new SACWIS administrative systems. Due to concerns expressed by state staff about the quality of data entered into the old administrative systems during the transition period from the former administrative system to the new one, in October 1999, members of the study team returned to Memphis to review case records on all sample families. The purpose of the case record review was to identify information on all child welfare activities (including allegations and information on foster care placements) and information on contracted services that might supplement information contained in the administrative database.
The following section describes the sources of cost data the study team found to be available in Tennessee. It begins with a description of the automated financial systems and then describes various approaches to gathering other cost information not tracked on a case-specific basis. Exhibit B-1 provides a crosswalk for each data element and the data source(s) that should be considered.
| Service Category(13) | Cost Information Source | Availability of Case-specific Data |
|---|---|---|
Child Welfare Service Costs |
||
| Family Preservation Treatment and Administration | State Family Preservation Coordinator | Yes |
Foster Care |
||
| Maintenance and Administration | DCS Automated Financial System(14) | Yes |
| Clothing | DCS Automated Financial System | Yes |
| Miscellaneous | DCS Automated Financial System | Yes |
DCS Directly Provided Services |
||
| Investigations Assessments Counseling Case management |
There are no actual cost data in this area. Various means of estimating directly provided services are described below. | No - costs would be derived/estimated. |
| Emergency Outlays | ||
| Food | DCS Automated Financial Systems | Yes |
| Shelter | DCS Automated Financial Systems | Yes |
| Transportation | DCS Automated Financial Systems | Yes |
| Indirect Public Agency Rate | DCS Finance Administration | No - an average rate would be applied to all cases. |
External Service Costs - those funded by the agency and by external sources |
||
| Mental health Psychiatric exams Counseling | There are multiple sources depending on the funding stream including: DCS Automated Financial Systems, contract information system, and case records. | Case-specific data would only be available if services are paid for by the state automated financial systems. All other costs would be derived/estimated. |
| Substance Abuse Treatment | There are multiple sources depending on the funding stream including: DCS Automated Financial Systems, contract information system, and case records. | Case-specific data would only be available if services are paid for by the state automated financial systems. All other costs would be derived/estimated. |
| Child day care | There are multiple sources depending on the funding stream including: DCS Automated Financial Systems, contract information system, and case records. | Case-specific data would only be available if services are paid for by the state automated financial systems. All other costs would be derived/estimated. |
| Parent Training | There are multiple sources depending on the funding stream including: DCS Automated Financial Systems, contract information system, and case records. | Case-specific data would only be available if services are paid for by the state automated financial systems. All other costs would be derived/estimated. |
| Homemaker Services | There are multiple sources depending on the funding stream including: DCS Automated Financial Systems, contract information system, and case records. | Case-specific data would only be available if services are paid for by the state automated financial systems. All other costs would be derived/estimated. |
| Respite Care | There are multiple sources depending on the funding stream including: DCS Automated Financial Systems, contract information system, and case records. | Case-specific data would only be available if services are paid for by the state automated financial systems. All other costs would be derived/estimated. |
1. Automated Financial Systems
In 1998, Tennessee was operating several different automated financial accounting systems that were not linked but all used the child’s social security number as the case identifying number. (The state continues to use social security numbers as the case identifier with its new consolidated system.) Financial data systems are also not linked with administrative data systems, although information has been “matched” in the past. Counties send all finance information in hard copy to Nashville for data entry. All claim systems are in FoxPro, which is similar to Dbase.
The Claim 13 System tracks all “flex fund” payments for prevention or reunification services as well as concrete items, such as furniture and utility payments provided by community providers to keep a family intact. The Community Service Agency maintains this system.(16) Payments to the state's family preservation program, Home Ties, were maintained on this system.(17) For families not eligible for Medicaid, this system would also support substance abuse services for parents.
At the time of the initial review, the state office was willing to generate extracts of all case-specific expenditures listed on each of the claiming systems described above. The child's social security number is used as the case identifier. If the child has entered foster care, the child's social security number will be located in the CORS administrative system, described above. For other cases, the numbers must be retrieved from the case record or, as a fallback, the number could be requested from DCF personnel who would conduct a match against the state's AFDC system called ACCESS.(18)
Pros and cons: The automated financial system is the best single source for client specific cost data on a range of child welfare funded services and will reportedly account for the vast majority of contract service costs. The only burden to study staff will be the collection of all social security numbers.
In addition to its flex fund system that is claimed on a case-specific basis, DCS funds a large number of community based support services through contracts funded by Social Services Block Grant and state dollars. A list of contracted services is available in both Shelby County and the state finance office. These services are not claimed on a case-specific basis and include mental health services, counseling and day care. DCS does not collect identifying information about clients for these services.
In order to collect cost information on DCF funded services that are not claimed on a case-specific basis, study staff would need to collect service utilization information from case records (see below) and retrieve service unit cost data from the state's finance office. Utilization information would include dates of service, length and intensity of intervention.
As mentioned above, study staff reviewed all available public agency case records for study families in Shelby County to look for supplemental information on administrative data and to gather information on the utilization of contracted services not claimed on a child specific basis. During the entire period that administrative data were collected for the Family Preservation and Reunification Programs study in Tennessee, Shelby County was experiencing severe staffing shortages especially in its Child Protection Unit. From the full case record review, case records were found to have very limited and generally inconclusive information about both casework contact and external service receipt. As in New Jersey, case records contained very inconclusive information about service receipt (whether or not a family accepted the service referral), and only rarely contained information on service utilization (length and/or intensity of service receipt). It was not clear whether the limited information was due to reduced time spent by caseworkers with families or due to a lack of time to record all casework activity.
On average, each case record took nearly two hours to review for case opening and closing dates, allegations, results of investigations, services received and social security numbers. Nearly half of the time was spent reading the case notes looking for information on service receipt.
Pros and cons: While often inconclusive, case records still offer the most thorough sources of service receipt. While labor intensive, all references to service referrals could be verified with the service providing agency to ensure service receipt, payment source and retrieve actual cost data. The drawbacks to this approach include the burden of case record review and the need to verify the information gathered from several service providers across the county. It is unclear how willing the provider agencies will be to research and supply the requested cost information retrospectively.
An alternative approach to contacting all service providers would be to use agency records to construct an average or estimated service cost, using average service duration, and applying these costs to each service referral found in the case record. However, such estimates would not be client specific.
DCS has never tried to collect service activity or workload information on directly provided services. The State Finance Director strongly discouraged study staff from using the random moment IV-E survey to calculate worker time because it uses an extremely small sample that is not representative of Shelby County where caseloads are higher and often remain open longer than in other parts of the state. Instead, he recommended the use of an annual, aggregate average cost of direct service labor per child welfare case.
Due to the limited amount of information on casework contact contained in the case records, direct service costs would have to be estimated by using one of the following options:
Through the use of focus groups with various types of workers, establish the average time devoted to different types of clients (for example, substance abusing), at different stages of DYFS involvement (screening, investigation, ongoing, placement). Establish both average number of contacts and length of time of each contact (home visits, office visits, court appearance, paperwork). Using average salary levels, and total time (as measured by number of casework activities) devoted to a case, derive value of labor.
Pros and cons: Supervisors and workers appear willing to participate in these focus groups and this approach is much less labor intensive for study staff than reviewing individual case records. However, this option produces only average estimates and requires multiple assumptions of time requirements and client characteristics.
b. Determine Value of Length of Stay:
Alternatively, one can refine the aggregate average cost per child by applying the actual length of time a case remained open in the child welfare agency. Using the length of time a case was opened, and monthly DCS personnel costs related to individual DCS units (e.g. Investigators, Ongoing CPS, Placement), create average monthly labor costs (see Appendix A for a fuller description).
Pros and cons: Presents only limited burden to state staff, but only produces rough, average estimates contingent on length of stay.
c. Compute an aggregate average cost per child.
This would be done by dividing the total child protection budget (or for those children placed in foster care, the foster care budget) by the total number of children served each year.
Pros and cons: The major drawback to this approach is that it will produce the same direct service costs for both experimental and control cases. Therefore, this option is least desirable.
Medicaid information is maintained within the Department of Health. The system is called TennCare. Cost information on services funded by Medicaid will not be available for the cost analysis because TennCare pays agencies a capped rate and information is not maintained on a child specific basis. The DCS TennCare Coordinator estimated that 95% of children in custody are TennCare eligible and therefore, DCS would not be able to retrieve their medical and some behavioral health costs. However, the Claim 15 System within DCS tracks payments for case-specific medical services including doctors, therapists, prescriptions and surgery for children not eligible for TennCare or for services which TennCare will not cover.
Between February and July 1998, members of the NEFS study team conducted a series of telephone interviews and a site visit to the Department of Human Services (DHS) in Philadelphia, PA to explore the availability of cost data. Philadelphia operates a single automated payment system for nearly all contracted services. The system is directly tied to its case management information system, the Family and Child Tracking System (FACTS).
Over the past ten years, Philadelphia has made its child welfare system increasingly reliant on externally provided services, most of which are paid for by outside sources. There is consequently a wide array of services for which little cost information is available within DHS. The only services listed in FACTS are those funded by DHS on a per diem basis. These services include: Services to Children in their Own Home (SCOH),(19) foster care, independent living, adoption and limited day care services. All other contract service costs, including family preservation, and costs for services delivered and funded by external providers must be identified through other sources.
FACTS also has a new visitation component that acts as a tickler system for ongoing protective and foster care caseworker visits. Workers are required to enter actual visitation information in the system including information on the type and number of visits made to each family. Systems staff believe they can gather this information as part of the case-specific data.
This section describes the sources of cost data available in Philadelphia and various methods for cost data collection. The primary source of data (the per diem payment system) is considered first, and possible options for filling remaining gaps are then explored with brief discussions of pros and cons of each approach. Exhibit C-1 provides a crosswalk for each cost data element and the data source(s) that should be considered.
The sources of cost data explored in this section include the city's automated payment system for services listed in FACTS, direct service contracts, case records, the Material Aid and Support program (a flexible funds program), and Medicaid.
Philadelphia operates an electronic billing system directly tied to its FACTS administrative system that accounts for roughly 90 percent of the city's DHS expenditures for contracted child welfare services. Services funded by this system include:
| Service Category(20) | Cost Information Source | Availability of Case-specific Data |
|---|---|---|
| Child Welfare Service Costs | ||
| Family Preservation Treatment and Administration | City's Family Preservation Coordinator | Yes |
Foster Care |
||
| Maintenance and Administration | FACTS | Yes |
| Clothing | Voucher System - would require case record review | No - costs would be derived/estimated. |
| Miscellaneous | Voucher System - would require case record review | No - costs would be derived/estimated. |
DCS Directly Provided Services |
||
| Investigations Assessments | There are no actual cost data in this area. Various means of estimating directly provided services are described below. | No - costs would be derived/estimated. |
| Counseling Case management | The Automated Visitation System. Other means of estimating directly provided services are described below. | No - costs would be derived/estimated. |
Emergency Outlays |
||
| Food | The Material Aid and Support Budget (MASB) and case record review. | Case-specific data are available if the item is funded by MASB. Other expenditures in this area would be derived. |
| Shelter | FACTS or the Material Aid and Support Budget (MASB). | Yes |
| Transportation | Most transportation services are provided by community providers and funded through the per diem/unit cost. DHS also has a small transportation fund, but expenditures are not tied to cases. | No |
| Indirect Public Agency Rate | DHS Finance Administration | No - an average rate would be applied to all cases. |
External Service Costs - those funded by the agency and by external sources |
||
| Mental health Psychiatric exams Counseling | Medicaid and case record review | More research is needed, see explanation below. |
| Substance Abuse Treatment | Medicaid and case record review | More research is needed, see explanation below. |
Child day care |
Day care funded through the per diem system | Yes |
| Day care funded through vouchers will require case record review | No | |
| Day Care funded through TANF will require CIS review | No | |
| Parent Training | Parent training would either be offered by the provider agency and be part of the per diem or would be offered by a community referral and will require case record review. | No |
| Homemaker Services | Homemaker services would either be offered by the provider agency and be part of the per diem or would be offered by a community referral and will require case record review. | No |
| Respite Care | Respire services would be lumped into the provider agency per diem. | No |
Pros and cons: The FACTS payment system is the best single source for client specific cost data on a range of child welfare funded services. There are no drawbacks to this data source.
Services such as family preservation are paid for by direct service contracts (known as “250's”). The Administrator of Program and Policies estimates that direct service contracts account for approximately 10 percent of all DHS contract expenditures as the vast majority of agency expenditures on contracted services are listed on FACTS. While paid for by DHS, these contracts are funded by a unit cost not by a per diem and DHS finance does not claim these expenditures by case. Each provider agency has a contract to serve a set number of families a year and length of stay does not affect payment.
In order to retrieve cost information on services funded by direct service contracts, it will be necessary to retrieve service utilization data from one of the following sources: DHS case records, provider agency records (described below), worker interviews or from families directly.
Pros and cons: Due to the burden of a case record review, described below, combined with the limited number of services funded by this mechanism, this data source has significant drawbacks.
Case records contain service utilization information on both services funded by direct service contracts, as well as services provided and funded by community based providers. Commonly used services in these categories include parent education, respite care, mental health and educational assessments, sexual abuse counseling and substance abuse services.
While we did not review a sample of case records during the visit, we were told that workers are not required to list all service referrals, and service information is spread throughout the record in notes, reports and memorandum.
If the study was to review records, most service information is located in the SCOH and family preservation provider reports. SCOH programs use quarterly reports and family preservation programs prepare mid-point and discharge summaries. Case records may also contain: copies of canceled DHS checks, ?Authorization for Payment’ memos for services not available through contract, and case notes describing services already received or anticipated.
The Director of Family Preservation Services has offered to collect the case records for our review. She will need a full list of case numbers and names in order to publish a list of records to be collected across the Department. The Director believes that gathering all records will be very time consuming.
Pros and cons: Due to the limited number of services listed on the FACTS system, case records would be the only source of service information on a majority of support services including parent education, respite services and mental health care. However, the Director of Family Preservation discourages us from conducting case record reviews for several reasons; the two primary reasons being:
Family preservation and SCOH provider reports supply a variety of information including names, and often addresses of service providers with whom the client worked during the family preservation or SCOH intervention. Family preservation reports also include provider telephone numbers and dates of service receipt. While the reports do not supply any cost information and will require follow-up contact with each agency, they supply the most detailed account of actual service history found by the study team at any of the study sites.
The Director of family preservation services has offered to give the study team copies of all family preservation reports for all study families. SCOH reports and records would be gathered from the individual provider agencies.
Pros and cons: The family preservation director believes these reports supply the most thorough account of service information outside of FACTS and will be much easier and faster to retrieve and review than full DHS records. The main drawback to this source is that the reports will only cover the period that the case was in treatment. Also, there is some question about how consistently thorough the reports are across providers and workers.(21)
While many concrete needs (food, shelter, and clothing) are paid for out of the provider agency’s per diem, in 1995, a Material Aid and Support program was set up with a budget of approximately $100,000 annually to provide additional flexible funding for SCOH and FP families. Accounts are maintained on a manual system but the City's Family Preservation Director is willing to tabulate all expenditures made on behalf of study families. The majority of program funds purchase furniture such as cribs, refrigerators and other concrete household items.
Pros and cons: This data source is client specific and will be relatively easy to access. It is highly recommended for use.
Like all other study sites, Philadelphia does not require caseworkers to track the actual amount of time spent on individual cases.(22) As described above, one approach is to assemble focus groups to determine service units and service unit costs and gather case-specific casework information from a review of the case records (see Appendix A for a description of this option as well as the pros and cons of using this approach). If it is determined that case records do not supply the necessary level of detail on casework activities, an alternative approach is to use focus groups to classify individual families problems or reason for referral and determine number, frequency and length of time spent on individual casework activities by family classification (see Appendix A).
Alternatively, the Administrator of Policy and Programs has offered to construct two separate aggregate costs related to directly provided services for the study. The first is a cost-per-case for investigations that includes all intake, investigation and assessment services. The second is an annual cost-per-case for case management activities, including foster care. This approach will help to establish estimates of time spent on certain subgroups of cases.
Pros and cons: As with other study sites, these will be very rough estimates based on a Division’s budget divided by its actual caseload. The major drawback to this approach is that it will produce the same direct service costs for both experimental and control cases (for a fuller description of issues and recommendations, see Appendix A).
Begun in 1996, FACTS has maintained a separate visitation system designed to be a tickler system to ensure compliance with mandated visitation time frames. Once the investigation is complete and the case is accepted for service, workers in each of the city's Family Centers are required to enter when a visit actually occurred and code the type of visit conducted.(23) Systems staff are willing to provide us with case-specific information since the program was initiated.
Pros and cons: This information will not account for all time spent on families (i.e. paperwork, telephone calls, meetings, etc) but will provide some actual level of effort in casework contact. The major drawback to this data source is that system staff report that the quality control for the information is uneven across offices; Section Administrators monitor the systems and the data entered. There is minimal central oversight for consistency of this system.
Services funded through Medicaid would include the majority of health, mental health and substance abuse services. Philadelphia does not operate a single Medicaid Program. Different state and city offices operate separate fee-for-service, managed care and behavioral health programs. For example, physical health services are administered by the state of Pennsylvania. Mental health services, which include the majority of substance abuse services are administered by the city, as part of the Department of Mental Health and Retardation. Accessing data would require getting authorization and case identifiers from multiple agencies.
Pros and cons: Due to the complexity of the multiple data sources and the consequent complexity of getting authorization for data collection, the pursuit of Medicaid data for this study is questionable.
Between December 1996 and January 2000, members of the study team conducted a site visit and follow-up telephone interviews with Kentucky state officials of the Cabinet for Family and Children, Department of Social Services (DSS), to identify existing data sources for the NEFS administrative and cost analysis. Kentucky maintains statewide central administrative and financial payment systems. During the study period, Kentucky transitioned from collecting child welfare data on older social services management systems to a new SACWIS system. Kentucky’s new SACWIS called TWIST, The Worker Information SysTem, was implemented statewide in October 1997. Random assignment of study cases for this evaluation ran from May 7, 1996 to February 13, 1998.
Kentucky has also recently designed a new fiscal payment system called MARS, the Management Accounting and Reporting System. This system was implemented in January 2000 and interfaces with TWIST to make foster care board and adoption subsidy payments. Prior to this implementation, the state had used an invoice payment system, the State Central Accounting System (STARS). For both the MARS and STARS accounting systems, workers submit paper invoices or vouchers to pay for other contracted and concrete services funded by the child welfare agency such as counseling, parent training, homemaker, psychological evaluations/assessment, independent living, shelter, substance abuse treatment for children, and capital outlays for food, transportation, and clothing.
This section describes the various sources of cost data the study team found in Kentucky, describing both the old and the new systems and the methods required for cost data collection. The sources of cost data in Kentucky include the state’s central accounting systems, STARS and the new MARS, the Payment Ledger, and invoices for contracted direct and support services. The recording of client information and the provision of the child welfare services by DSS workers are contained in the DSS Family Activity Tracking System; Child Abuse and Neglect Registry (CAN); and TWIST. Like other study sites, Kentucky does not require workers to record the time they dedicate to individual cases. Exhibit D-1 provides a crosswalk for each data element and the data source(s) that should be considered.
Service Category(24) |
Cost Information Source | Availability of Case-specific Data |
|---|---|---|
Child Welfare Service Costs |
||
| Family Preservation Treatment and Administration | State Family Preservation Coordinator (?) | Yes |
Foster Care |
||
| Maintenance and Administration | State Central Accounting System (STARS), Invoices, Management Reporting and Accounting System (MARS) | Yes |
| Clothing | STARS, Invoices, MARS | Yes |
| Miscellaneous | STARS, Invoices, MARS | Yes |
DCS Directly Provided Services |
||
| Investigations Assessments Counseling Case management | There are no actual cost data in this area. Various means of estimating directly provided services are described below. | No - costs would be derived/estimated. |
| Emergency Outlays | ||
| Food | Payment Ledger, STARS, Invoices | Yes |
| Shelter | Payment Ledger, STARS, Invoices, MARS | Yes |
| Transportation | Payment Ledger, STARS, Invoices | Yes |
| Indirect Public Agency Rate | DSS Finance Administrator | No - an average rate would be applied to all cases. |
External Service Costs - those funded by the agency and by external sources |
||
| Mental health Psychiatric exams Counseling | Payment Ledger, STARS, Invoices, MARS, Medicaid, Mental Health Agency System | Yes |
| Substance Abuse Treatment | Children -- Payment Ledger, STARS, Invoices
Adults – Mental Health Agency and Medicaid |
Yes
Yes--if access to Medicaid System |
| Child day care | Day Care System | Yes |
| Parent Training | Payment Ledger, STARS, Invoices, MARS | Yes |
| Homemaker Services | Payment Ledger, STARS, Invoices, MARS | Yes |
| Respite Care | Payment Ledger, STARS, Invoices, MARS | Yes |
Throughout most of the study period in Kentucky, payments for foster care maintenance and other contracted services were made through the submittal of invoices and vouchers to the State Central Accounting System, STARS. The Office of Program Support, Division of Fiscal Services was responsible for maintaining the data relating to the Cabinet for Human Services (CHS) programs in the STARS system. Financial transactions are maintained in a manner to reflect separate accountability for each program administered by CHS. Automated files are kept on all payments for services submitted on the invoice/voucher system through STARS and MARS (below), however, the payment files from STARS have been archived.(25)
Pros and cons: In order to obtain the cost data files from this system, a request must be submitted to the Cabinet for Family and Children Services Financial Director. As stated before, the STARS files have already been placed in archives, which makes it much more difficult to access. STARS files, however, contain the majority of the records from the study period. Before the implementation of MARS, payments for foster care maintenance and other contract provided services were paid through invoices and vouchers manually submitted to the state’s central payment system (STARS). The child’s case number may or may not be recorded in the older payment system in use during the first two years of the study period. For services other than foster care placements, service dates were not always captured on the invoices. The payment systems only record the payment date, not the actual service date.
Since January 2000, foster care maintenance and adoption subsidy payment invoices have been generated by TWIST and transmitted to the automated payment system, Management Accounting and Reporting System, (MARS) for the checks to be issued. All other contract services paid for by the child welfare agency are still manually submitted on invoices and are now paid through MARS.
Pros and cons: MARS files are more accessible but would also require a special request to the Financial Director and contain very little of the information we need for the cost study. The advantage to MARS is the accuracy of payment to service provision as well as being an automated file.
TWIST automatically produces foster care board and adoption subsidy invoices that are sent to MARS for the payment, however, this interface between the systems has only been operating since January 2000. The placement date and discharge date by child would be obtained from TWIST. The cost or payment for the placement would be obtained from MARS. The MARS data would cover only a very small portion of the foster care claims for the study cases. Payments for all other contracted services, other than foster care maintenance and adoption subsidies, are still made with manual invoices and vouchers submitted for payment through MARS.
Another source of cost data is the Payment Ledger that contains all or most of the same contracted and concrete services provided on behalf of children and paid for by the child welfare agency as recorded on STARS and MARS. The difference is that services are recorded on the ledger manually and the ledger is not part of an automated payment process. Payments to family preservation contract agencies are made through invoices submitted to the financial accounting systems and are recorded on the payment ledger by child.
Pros and cons: A much more accessible and obtainable source is the Payment Ledger, however obtaining the data from the ledger requires a very labor-intensive process since it is not automated. Services provided by contract providers paid by the child welfare agency on behalf of a child are also tracked manually in a payment ledger but the child’s client ID is not always recorded on this source. Payments for these services are made through the statewide payment systems and could be manually gathered possibly by using a social security number. Service dates are not captured on the payment ledger.
Day Care services are on a separate system Local Office and Contract Information System, LOCIS, handled administratively by another agency. This system is not interfaced with TWIST; however, they are able to identify child welfare children receiving day care. To obtain the data files we would again make a request to the Cabinet for Family and Children Services Financial Director.
Pros and cons: Obtaining the day care service payment file is the only barrier because it is an automated payment file. Most of the files have been archived. The child welfare agency would have to request the information from the day care agency in order to allow programmers to create a file with day care service payments made for our study clients during the study period.
Pros and Cons: All Payment Systems: The State Central Accounting System, the Management Accounting and Reporting System, the Payment Ledger, and the paper invoices are very complete sources of client specific (child and family) cost information for contracted and concrete services paid for by the child welfare agency. All of these financial sources are available for cost data collection, but not without difficulties and labor intensive scrutiny of files and records. Cost data would have to be aggregated from the multiple systems looking for the appropriate service periods on the old DSS systems and TWIST, manually compiling the services from the invoices and voucher that were submitted for payment, checking the payment ledgers for verification and concrete services cost, and checking STARS and MARS to find the cost of those services. Service periods would have to be associated to a payment date. The services found would include only those services paid for by the child welfare agency.
As is true at the other sites, cost information on services provided to clients by referral to another agency that pays for that service, will be more difficult to gather. For example, although child welfare clients can be identified on the Day Care system and on the Medicaid system, it is not clear whether these agencies would grant priority to this project or allow access to the files.
Obtaining service cost information from these many different systems is difficult because: 1. The fiscal payment systems do not always track by client identification number, some use social security numbers or another agency’s number, 2. Cases have different IDs on the old and the new child welfare systems, 3. Some of the payment systems are under the management of other agencies, not the child welfare agency, and 4. Fiscal billing cycles and payment dates often do not track service dates.
2. Administrative Systems—Direct Services
One option to collect client specific direct service information would be to use the automated case management files that record many of the direct service activities delivered by line workers. Prior to TWIST implementation in 1997, the Department of Social Services recorded services to clients and families on the Family Based Services, Family Activity Tracking form DSS 887. This form opened the case for service and track clients by grouping them by the type of services they were receiving: adoption, child protection in-home services, child protection out of home, adult protection, day care, spouse abuse, juvenile, general family, general adult, or alternate care. The investigation of abuse and neglect was recorded on a separate system, the Child Abuse and Neglect Registry. Since October 1997, child welfare workers have been using TWIST to record client specific information by the major categories of direct service such as child abuse and neglect reports and investigations, cases opening for ongoing family services beyond investigation, foster care and adoptive placements, reunification and independent living.
Pros and cons: In Kentucky the implementation of the new administrative systems did not include a transfer or conversion of all historical records from the old systems to the new. Details of services provided would have to be pieced together from a case record review or the matching of files case by case in multiple old and new systems. Additionally, in the transition, cases were closed on the old system and then re-opened on the new system. If the worker did not close the case on the old system, the system automatically closed the case with a prescribed system’s date. This process either created an artificial break in service or, for those cases that should have been closed for service, an artificially extended length of service. Although this causes a deficiency in calculating an accurate average length of service, these systems can be used to obtain the aggregate number of reports of abuse and neglect, investigations, the number of children in foster care, adoption, and the number being served in their own homes. A compilation of data from both the old DSS system and TWIST would provide fairly accurate client counts by the major service categories. However, this could represent only part of the equation to estimate the costs of direct service work. To obtain the cost of each of these service categories, the time a worker spends on each activity and a good estimation of the cost per staff per activity is needed.
There does not appear to be any way to retrieve case level service activity or service costs (other than for family preservation service and foster care placement) for services provided directly by DSS workers. As described above, one approach is to assemble focus groups to determine service units and service unit costs and then gather case-specific casework information from a review of the case records (see pages 18 and 19 for a description of this option as well as the pros and cons of using this approach). If it is determined that case records do not supply the necessary level of detail on casework activities, an alternative approach is to use focus groups to classify individual families by problems or reason for referral. Then, determine frequency and length of time spent on individual casework activities by family classification. (See page 22 above.)
Other options are less viable. An old Department of Social Services cost allocation service activity reporting system was not used or maintained for several years prior to the implementation of TWIST. A replacement system for reporting worker time spent on a case has been designed in TWIST, however, it is not fully functional and workers have not been consistently entering casework activity. Staff suggest that the only current alternative source for direct service costs is to divide the child welfare budget by the number of cases or children served to get an aggregated average cost. Data on the total number of cases receiving direct services, intake, investigation, and ongoing services, are accurate on both the old DSS systems and TWIST. However, average costs of service over the entire caseload are unlikely to be applicable to the particular case involved in this study since cases referred to family preservation are likely to differ in systematic ways from cases as a whole.
In Kentucky, child welfare clients can be identified on the Public Assistance/Child Support/Medicaid files. The systems are interfaced through nightly batch runs.
Pros and cons: Since the child welfare systems and the Medicaid system are interfaced through nightly batch runs, it would be possible to retrieve the services funded by Medicaid. Child Welfare clients are identified on the Medicaid file by client identification numbers. There are two difficulties in obtaining the data. First, the necessary files are stored and kept by a separate agency, Health Services, not the child welfare agency. Getting clearance and priority to have historical files run for the study period plus the difficulty in tracking these services over time with multiple client ID’s would make the programs more complex and less reliable. Secondly, the problems discussed earlier with the transition from the old DSS systems to the new TWIST also causes problems with capturing Medicaid services received by study clients during the study period. The program to access the Medicaid records would require an ID conversion methodology that supports tracking of clients across systems and eliminates the false service periods created when clients were closed out on the old system and re-opened onto the new system during TWIST implementation. Overall, the most daunting task would be going through the bureaucracy to obtain approval to access the files.
Case Managemant Activity/Case Type |
Initial Assessment/Intake |
|---|---|
Current Effort |
Expended Effort |
Observed Time: 69 minutes The completion of this document requires the worker to first compile all the relevant information gathered during the assessment process and then transpose it, in either handwritten or typed form, onto the assessment format. Given the extensive documentation process and the length of the form, the 69 minutes does not seem sufficient to adequately complete the initial assessment. This conclusion is supported by Round 4 Quality Assurance Review findings which indicated that only 45.5 percent of the sampled cases documented initial assessments that were done timely and adequately. |
Recommended Time: 120 minutes |
Observed Time: 121 minutes |
Recommended Time: 270 minutes
assist in the child's initial adjustment to the foster home |
(5) The list of cost data elements presented in this chart is shorter than the original data collection framework due to the structure of local payment systems operating at study sites. For example, family preservation contracts do not distinguish between therapy and administrative costs and study sites are unable to breakout child welfare caseworker time by type of activity.
(6) As described later in this report, in order to retrieve data from the Medicaid system the study team would need to retrieve all Medicaid ID numbers.
(7) Each child specific entry in the semi-annual report begins with a line that summarizes all child specific expenditures to date. In order to retrieve individual components of that total, prior ledgers must be reviewed.
(8)CPAS is actually a financial program which every month matches services listed in the agency's administrative database- SIS with vender specific rates contained in a separate system, The Provider Caretaker Information System (CPIS).
(9) The study team retrieved all historic data for one county, Bergen, as a test case. Bergen's ledger was smaller than most because it has one of the state's smallest child welfare caseloads. Still, on average, each six-month ledger took one person one hour to review and copy the relevant pages.
(10) In order to verify the thoroughness of the records, the study team also compared printouts of child ledger expenditures with services listed in the case records.
(11) From meetings with the state, we had expected to find one of three types of forms for external services: service referral forms made out by the worker; service approval forms signed by supervisory staff; and/or service summary reports completed by the provider agency that describe what happened during the intervention. However, only a small fraction of services mentioned in the contact sheets had any additional documentation.
(12) Using a random sample of 425 cases selected from field offices across the state, case workers were asked to record the amount of time required for various activities conducted on behalf of families at various risk levels and at various stages of the child welfare continuum (pre-custody, ongoing/in-home services, custody cases). Once baseline data was gathered, a panel of local and national experts led by the Child Welfare League of America, worked to establish case practice standards and duration of individual casework activities.
(13) The list of cost data elements presented in this chart is shorter than the original data collection framework due to the structure of local payment systems operating at study sites. For example, family preservation contracts do not distinguish between therapy and administrative costs and study sites are unable to breakout child welfare caseworker time by type of activity.
(14) The one exception is state operated group homes, as described in the text.
(15) Foster care maintenance payments are made twice a month to foster parents and once a month to adoptive homes.
(16) The Community Service Agency is a quasi-governmental agency created by legislation in 1996. The agency is under contract with DCS to provide both case management and financial oversight services to families at risk, or in the custody, of DCS.
(17) Home Ties providers are paid a set amount per case (in Shelby County, almost $2,600) once that case has crossed a seven day threshold of service. DCS pays the full amount regardless of the actual length of service and what services were provided.
(18) During the review of case records, NEFS study team members gathered the social security numbers on nearly all study participants.
(19) While less intensive than family preservation services, SCOH services are homebased counseling and parent education delivered by several private providers across the city.
(20) The list of cost data elements presented in this chart is shorter than the original data collection framework due to the structure of local payment systems operating at study sites. For example, family preservation contracts do not distinguish between therapy and administrative costs and study sites are unable to breakout child welfare caseworker time by type of activity.
(21) If this approach is selected, a decision must be made about the period of service review. Family preservation cases frequently receive SCOH as a transitional service after FP is complete. A decision must be made about whether the review of services should be limited to the treatment period to make data collection comparable or, whether all available service data should be gathered.
(22) While actual numbers of home visits are maintained in FACTS, it’s value is limited for the cost analysis because it does not account for all of the actual casework time involved with a specific case. More on this in the following section.
(23) Between March and October 1997, workers only recorded the minimum number of visits required to meet the state requirements (generally quarterly). Since November 1997 however, workers have been entering all visits made to families.
(24) The list of cost data elements presented in this chart is shorter than the original data collection framework due to the structure of local payment systems operating at study sites. For example, family preservation contracts do not distinguish between therapy and administrative costs and study sites are unable to breakout child welfare caseworker time by type of activity.
(25) STARS, MARS, and the Payment Ledger track the cost of DSS funded substance abuse treatment services provided to children. Adult treatment is provided through the Mental Health agency and tracked through a different system.
Contents | Top
of Page
Main Page of Report | Contents
of Report
Home
Pages:
Evaluation of Family Preservation and
Reunification Programs
Human Services Policy (HSP)
Assistant Secretary for Planning and Evaluation
(ASPE)
U.S. Department of Health and Human Services
(HHS)
