Frequency Report: There were five (.29%) records (n=1754) that indicated a date of a periodic review in 1999 and 76 records indicated a date of 2000.

State uses the latest date of review or permanency review.

Program code includes a parameter to exclude dates that are not in the current removal episode. If a date of review occurs prior to the removal date, then the date of review is set 1/1/1900, and this is then mapped to blank.

There is a field for the worker to enter an "A" for an approximated date of birth.

The Database Manual states that if a date of birth has to be approximated, then "use January 1 of the approximate year as a date of birth." This is incorrect for AFCARS reporting. The State must use the 15^th of the estimated month and year.

State's age of emancipation is 18. State may be including youth over age 19 in the population. Staff will evaluate and exclude those over age 19.

Post-site visit evaluation: The State has corrected the date used for an estimated date of birth, and now excludes individuals over the age of 19.

The information system has the capability to collect information on multiple races.

The State's code "?" stands for "unknown," which means the information is not known. A "U" stands for "unable to determine," which also has the same definition as the AFCARS value. The definition is included in the user manual, but it currently shows the definition for the "?." The State staff will update the manual to reflect training and practice.

Workers cannot enter a race code and a "?" or a "U" at the same time.

Frequency Report: n=1754; Yes = 47; No = 110; Not yet Determined = 1565; Missing = 32

The State does not have a policy that indicates children who are in foster care must have a medical/psychological evaluation done within a certain time frame.

The agency is making a modification to the system by adding a "health history screen," which is where the information for elements #10 - 15 will be recorded. The plan is to have this screen implemented by the fall of 2002. Additionally, this information will be reviewed and updated, if necessary at the time of the periodic review.

The State staff believes the frequency for this element is under-reported. This is probably due to workers not updating this information. Staff believe that this may be due to its location in the system, which is on the "face sheet" screen. This is not a screen that workers are likely to update.

In the data dictionary there are codes "A" and "O." These are old codes and are no longer used. These codes did not appear on the entry screen for the workers to select. State will update the data dictionary.

Frequency Report: n=1754; Does not Apply = 1717; Applies = 5; Missing = 32

At the time the State implements the "health history screen," review the AFCARS resource document for mapping disability information to AFCARS codes to ensure that all possible conditions are being captured and accurately reported to AFCARS. (The document is located on the Children's Bureau's AFCARS web page.)

Frequency Report: n=1754; Does not Apply = 1720; Applies = 2; Missing = 32

See notes in foster care element #11.

Frequency Report: n=1754; Does not Apply = 1715; Applies = 7; Missing = 32

See notes in foster care element #11.

Frequency Report: n=1754; Does not Apply = 1692; Applies = 30; Missing = 32

See notes in foster care element #11.

Frequency Report: n=1754; Does not Apply = 1714; Applies = 8; Missing = 32

See notes in foster care element #11.

Selection list used by workers includes "L" for "specific learning disability" but it is not included in the program code. This should be mapped to "other diagnosed condition."

Post-site visit evaluation: The State has corrected the program code to include "learning disability" and it is mapped to this element.

Frequency Report: n=1754; Yes = 3; No = 1750; Unable to Determine = 0; Missing = 1

The response to this question in AFCARS is based on the worker completing the field on the screen that indicates the age the child was adopted. The responses available to the worker are: "N, not adopted," "?, unable to determine," or the age of the child at the time he/she was previously adopted. If the program code identifies an age, then this element is mapped to "yes." If an "N" is identified, then this element is set to "no." If the worker enters a "?," then the response to element #16 is blank. However, due to screen design and the mapping for element #17, this element will default to "no." See notes in element #17.

Post-site visit evaluation: The program code for element #16 and #17 have been corrected to reflect an entry of "unable to determine" for both elements, if applicable.

Frequency Report: 0 =0; 1 = 0; 2 = 2; 3 = 1; 4 = 0; 5 = 0; Not reported = 1751.

In order to respond to the AFCARS question as to whether a child has been previously adopted, the State's system is designed to collect the age of the child. When worker accesses this screen the field "age adopted" is set to "N" for "no". If an age is entered, then element #16 is mapped to "yes" and the age is mapped to the appropriate age category for element #17. The worker can enter an "?" for unable to determine, which is mapped to blanks in elements #16 and 17.

A response of "N" results in blank information in this element, as indicated by the frequency report.

A blank in this field will be mapped to blanks in elements #16 and 17.

Due to the screen default to "no" for this element, the State's data is reflecting a false "no." The data for "no" cannot be analyzed with a level of confidence that the response truly means that the child had not been previously adopted.

This default should be removed and the field should be blank when the screen is accessed. The program code should continue to map missing information to blanks.

The program code does not include the code "U." This value needs to be added and mapped to the AFCARS value "5, unable to determine."

Post-site visit evaluation: The default to "N" has been removed for element #16. The program code includes a "U" for "unable to determine" that is mapped to "3, unable to determine" in element #16 and a "five" for #17.

The program code looks for dates that have reasons of removal associated with them, and distinguishes them from dates of placement. It then selects the earliest date.

State's definition of a removal episode matches the AFCARS definition.

See note regarding juvenile justice population in element # 21. This change affects the accuracy of the data in element #18.

The program code looks for dates that have reasons of removal associated with them and distinguishes them from dates of placement. It then selects the latest date prior to the end of the report period.

Based on AFCARS policy and the organizational structure of the Department of Social and Rehabilitative Services, youth that are placed in the care, placement and supervision of the agency are to be reported to AFCARS. If the youth's first placement is a "detention facility," the date of removal should be the date the child is placed in a community-based placement (i.e., a group home or foster home). See the Child Welfare Policy Manual at 1.3 (Question: Are children in juvenile justice facilities included in AFCARS reporting? Answer: Children who are removed from their home and initially placed in a juvenile justice facility are not to be included in AFCARS reporting....)

Placement and removal information is on the same screen.

The program code looks for placement change reasons and excludes "exit reasons," then the program code extracts the date based on there being a "move" reason present.

State enters the date a child enters a "trial home visit" and when a child runs away.

Pre-AFCARS submission reports generated would find data entry errors for discharge reason.

Frequency Report: The number of placement settings ranges from 1 to 27. There are 95 records that were blank. The majority of records had between 1 and 4 placements.

The State is counting all placement settings, except for trial home visit, runaway and pre-adopt changes if the child is being adopted by its foster parent(s).

The program code identifies unique provider identification numbers.

The State should not include respite placements.

The State also includes other non-hospital short-term emergency placements/emergency group homes for kids in crisis. The State asked how they should treat these cases for AFCARS and what is meant by "short-term." ACF will follow up on these questions and provide the State with a clarification. Response: The Children's Bureau has updated the AFCARS section in the Child Welfare Policy Manual. The following question and answer was added:
Question: How should the State count brief periods spent away from the child's foster care provider in foster care element 24, " Number of previous placement settings during the removal episode"?
Answer: In general, the State is required to count a placement that lasts more than 24 hours while the child is in foster care under the placement, care or supervision responsibility of the State agency. See 45 CFR 1355.40 and CWPM 1.2B.7 and 1.3. This includes moves that may be made on an emergency or unplanned basis, such as shelter care placements, treatment facility placements, and certain placements for juvenile justice purposes. See CWPM 1.3. However, there are certain temporary living conditions that are not placements, but rather represent a temporary absence from the child's ongoing foster care placement. As such, the State must exclude the following temporary absences from the calculation of the number of previous placement settings for foster care element 24.

• Visitation with a sibling, relative, or other caretaker (i.e., preplacement visits with a subsequent foster care provider or preadoptive parents)
• Hospitalization for medical treatment, acute psychiatric episodes or diagnosis
• Respite care
• Day or summer camps
• Trial home visits
• Runaway episodes

If the type of placement described by the State does not fit into one of the bullets above, then the State must increment the placement count. This policy is to be implemented by States effective October 1, 2002.

Frequency Report: n=1754; Voluntary = 34; Court Ordered = 1720; Not Yet Determined = 0; Missing = 0

The State does not have "voluntary" removals. The agency would have to get a court order to place a child. The only voluntary placements are those for special circumstances, such as treatment in a hospital and no one is available to care for the child.

Workers are able to only enter up nine "circumstances associated with removal." The system should support the recording of all possible combinations of circumstances associated with removal.

"Other Entry" on the "entry" list in the user manual on page 33 needs to be removed.

Database Manual has two codes with the same definition, "AC, child alcohol abuse" and "SC, child drug abuse." This is a typo, the Manual needs to be updated.

"AC, alcohol abuse child" (as a reason for removal) is listed in the program code, but not on the mapping form. The mapping form and Database Manual need to be updated.

Frequencies: Pre-adopt home = 112; Foster Family Home, relative = 178; Foster Family Home, non-relative = 883. Total AFCARS values 1 - 3 = 1173. Group Home = 251; Institution = 62; Supervised Independent Living = 48; Runaway = 27; Trial Home visit = 193. Total for AFCARS values 4 - 8 = 581.

Workers enter if child is on a "trial home visit" and if the child is on runaway status.

The user manual needs to be updated to reflect that if a child is removed from a caretaker that is a relative, and not the parents, and the child is returned to this caretaker for a trial home visit, then code would be "RE, relative."

Temporary visits to a relative (from whom the child was removed) are mapped in the program code as relative; these need to be coded to "8, trial home visit."

Post-site visit evaluation: Program code corrected to map temporary visits to a relative from who the child was removed to "trial home visit."

Frequency Report: n = 1754; Reunify = 1025; Live with other relative = 9; Adoption = 306; Long term foster care = 176; Emancipation = 93; Guardianship = 0; Not yet established = 145; Missing = 0

Codes "PR", "IL" and "LG" are not listed in the database manual, but are in the program code. These are old codes and are no longer used, but are in the program code in order to capture older records that have them as the case plan goal.

Default is "not yet established."

Vermont has "Care and Protection" (CP) option, which is to provide immediate care and protection to a child in the first 60 days of a removal episode. This code is not currently included in the program code. It should be mapped to "not yet determined" in the code. It does get extracted as "not yet established" because this is the default. Staff explained that most of the cases would have a goal of "reunification" set after the 60 days.

State should remove the default of "not yet established" or map to missing data after 60 days from a child's removal from home.

Map "care and protection" to "not yet established" and after 60 days set to blank.

Post-site visit evaluation: The State corrected the program code to map "care and protection" to AFCARS "not yet established." The default to "not yet established" was removed and missing information is mapped to blanks.

The frequencies indicate that there are no single male or female caretakers. It appears the code is mapping these to "unmarried" due to the default setting. The frequencies in the 2001A file are okay. There was a program code change made between 2001A and 2001B files. The program code was corrected during the site visit.

Default is set to "unmarried couple." The program code has the final default as "unable to determine" but the code would never get that far.

Program code has an option of "?" (unknown), which is mapped to blank.

"Civil unions" are not included in the program code, but has been added to data entry screen. Civil unions should be mapped to "unmarried couple."

This element is not gender-dependent in Vermont.

See the frequency report regarding dates back to 1900.

Frequencies: Not Applicable = 581; Married Couple = 89; Unmarried Couple = 30; Single Female = 180; Single Male = 38; Not reported = 836. Total for 1 - 4 = 337.

The number of records reported with a marital status for foster parent family structure is considerably less then the number of records reported in element #41 for foster home settings. The number reported for "not applicable" matches the number reported in element #41 for the values 4 - 8. The State staff believe the number of records with blank information are provisional foster homes that are not yet licensed (children can be kept here for 30 days). The demographic information on foster parents is located on the license screen. Therefore, information on provisional foster parents is not on the license screen until the home is licensed. The program code extracts from the license screen.

For "trial home visit" and "runaway," Vermont does not pick up previous foster parent information.

Post-site visit evaluation: State has corrected program code to collect demographics on all foster parents. The program code now maps "civil unions" to "unmarried couple."

The user manual on page 33 lists the code "OT, other termination." Staff explained that this is an old code and is no longer used. They will update the manual.

"Runaways" remain in the State's care and placement responsibility until age 18. At that time the State considers them discharged.

Frequency Report: n = 1754; Does not apply = 1236; Applies = 518

This is being collected from the screen that shows the payment distribution for substitute care. The frequency for this element seems to be to low compared to the number reported as applies in element #59.

State believes they probably interpreted this as Medicaid contributing to their cost of substitute care. Almost all children in care receive Medicaid. For this element there does not have to be a "payment," just be eligible. State needs to re-examine the collection of this information. The State has this information and will need to modify the extraction code.

Post-site visit evaluation: The State has corrected the program code to capture Medicaid eligibility.

This element is to reflect if there is a child that has no State or other Federal sources of income. "Applies" means there is a source of income, general funds (State funds) or other Federal funds.

The State's program code reports this as "applies" if all of elements #59-64 are coded "does not apply."

The state explained that children who are at home or who are runaways will not be receiving income.

Once the Medicaid figures are adjusted for element #63, these numbers will drop in element #65.

Post-site visit evaluation: Program code corrected by removing mapping of this element to "applies" if elements #59-64 are coded "does not apply." The program code looks for a specific payment.

The State is reporting partial payments that are added together for the month if there were multiple types of care.

The State is to report zeroes for children that enter foster care and have not been in care a full month prior to the end of the report period. If a child is discharged mid-month, then the full month's payment prior to discharge month should be reported.

Post-site visit evaluation: The program code has been corrected to report a full monthly payment for only one setting and does not include partial payments.

Derived from the field "special needs." If there is a code for one of the specials needs in element #10, then this element is coded as "yes."

The accuracy of the data for this element is affected by the reporting of element #10.

The worker is able to enter all the codes that apply or can leave the field blank. The program code maps blanks to "no" for element #9. Missing information should be mapped to blank in element #9. This provides a false "no" in the data.

The user manual does not include "age" in the list of the options to be reported for this element.

There is a system-generated hierarchy for determining the primary basis for special needs. Based on the following order, the system determines which one to map as the primary basis: Race, Age, Sibling, Disability and Other.

If "D" (disability) is selected, the program code extracts the disability information from the face sheet screen. (Disability information is actually reported to AFCARS even if it is not the primary reason for special needs.)

Selection list used by workers includes "L" for "specific learning disability" but it is not included in the program code. This should be mapped to "other diagnosed condition."

See notes in element #11.

The State code "O, Other," and "unknown" are mapped to null.

If a child is abandoned, the worker codes it as a "?" and it is mapped to blank in the program code. A "?" should be mapped to "unable to determine" instead. State needs to re-code and distinguish "unknown" from "unable to determine." "U" in this case is "unmarried" instead of "unknown." State will re-examine and modify code abbreviations to be consistent throughout the system.

On page 48 of the Database Manual, there is an option "unknown." Since this is information on the person adopting the child, it would seem that this information is known and "unknown" should not be an option.

The State code "other" may be the code workers are using for "unmarried couples."

State needs to add "civil unions" to the program list of options and map to "unmarried couple."

The user manual needs to be updated to include instruction regarding civil unions and unmarried couples. State is going to modify the entry screen by changing "mother/father" to "Parent 1" and "Parent 2" and will need to collect gender.

Frequency Report: n = 70 Within State = 62; Another State = 8

The State staff explained that those reported as "another State" are children placed in Vermont by a private agency in another State. When the staff verified the records in the system, it was found that only three met this criterion. The staff suspect that the other five are ICPC cases. State will follow up on this and clarify the instructions and follow up with training.

"State/Public," "Tribal," or blank are the only options the worker can enter into this field.

"I, independent person" is not included as an option for the worker. The staff explained they would never have an individual directly placing a child for adoption with the agency. The agency would refer the person to a private agency.

Post-site visit evaluation: During the post-site visit evaluation it was determined that the State is correctly collecting and reporting this information.

If Medicaid is the only subsidy, it is not being mapped to this element. State needs to include Medicaid only subsidies and the response for element #36 would be zero and for #37 it would be "no."

Post-site visit evaluation: The program code now maps "Medicaid only" subsidies to this element as a "yes."