Data Standards Comments Summarized with Responses

Comments, Responses, and Revisions

Section 2. Program Descriptor Data Elements

Organization: National Alliance to End Homelessness

Comment: The program descriptor data element should provide more information about programs. Programs funded under the Rapid Re-housing demonstration and other rapid rehousing programs should have a separate code from transitional housing programs.

Homelessness prevention should be a separate category

Response: Program Type code does have a code distinguishing HPRP programs. Rapid re-housing and homeless prevention are considered the same program source and are therefore combined.

Modification to Standards: None.

Section 2. Program Descriptor Data Elements

Organization: No More Shelters

Comment: I would also add the fee structure of all homeless agencies in the Housing Inventory. This is extremely important. And I would add payments during the household’s duration in the exit form.

Response: This is a valuable option but should not be a minimum standard.

Modification to Standards: None.

Section 2.2 Organization Name

Organization: The Partnership Center, Cincinnati, OH

Comment: Having organizations use their “legal name” in e-snaps is sometimes challenging. If you are trying to line up HMIS, APR’s, E-snaps, etc. – using the legal name of the entity would be the only way for consistency. If HMIS adds another naming convention e.g. agency initials or common name – how will that line up with e-snaps? Change instructions to read the organization name is equal to the entities legal name.

Response: Covered under consistency clause, but burdensome and confusing to users to require legal name.

Modification to Standards: The "name that user will recognize" clause was removed, because it may be inconsistent with the recommendations to achieve consistency with e-snaps and other HUD activities. Use of legal names is encouraged. An HMIS software is encouraged to allow aliases to ensure that user will recognize the organization.

Section 2.4 Program Name

Organization: The Partnership Center, Cincinnati, OH

Comment: Common names for programs and “grant names” for program are often not the same. Can a CoC in creating a naming convention change the “grant name” within e-snaps for the purpose of lining up the names? Change instructions to read the CoC may institute naming conventions to ensure that HMIS and e-snaps are consistent. Changes made to program names as they were recorded in e-snaps will require a correction in the e-snaps records by the CoC and to the funded agency in writing.

Response: Covered under consistency clause, but burdensome and confusing to users to require legal name.

Modification to Standards: The "name that user will recognize" clause was removed, because it may be inconsistent with the recommendations to achieve consistency with e-snaps and other HUD activities. An HMIS software is encouraged to allow aliases to ensure that user will recognize the organization.

Section 2.5 Direct Service Code

Organization: Northrop Grumman

Comment: It notes that if the program does not enroll clients directly the direct service code should be "No", the next sentence then states (Food pantry, out reach, drop-in, etc) should code "Yes" Is this accurate or should they be "No"? If they are "yes", what programs would be "no"?

Response: The answer to the question is already listed in the instructions.

Modification to Standards: None.

Section 2.6 Site Information

Organization: Northrop Grumman

Comment: Program Site Configuration Type - Should there be an option that is NA? For example, street outreach, what would they use for a code here?

Response: Outreach could use "multiple sites" option

Modification to Standards: Add "outreach" to parenthetical under option 3.

Section 2.6 Site Information

Organization: Cambridge (MA) Department of Human Services

Comment: A single program serving individuals may lease a mix of SROs and/or 1BR apartments and/or 2BR apartments shared by two clients and/or 3BR apartments shared by three clients. These apartments are not separate sites or facilities: they are simply different sized apartments rented for clients of the same program. There needs to be a category for "mix of housing types"

Response: Standards recognize that there are hybrids. The response categories refer to the type where "most individuals and/or families reside."

Modification to Standards: None.

Section 2.9 Bed and Unit Inventory Information

Organization: Alameda County, CA Housing and Community Development

Comment: Section 2.9D Bed Inventory, requires that programs that serve a mixed population without a fixed number of beds per household type should divide the beds based on average utilization. This average utilization calculation is an arbitrary distinction that can fluctuate on a weekly or monthly basis. For example, in programs that serve single women AND single-parent women and their children, accommodations are made in beds and units every time a vacancy occurs. It is very common to sway from a large majority of single women to a large majority of families with few to no single women within one program. This dynamic becomes extraordinarily problematic with the additional "Special Issues" guidance for Section 2.9 that states on page 33 that "the number of beds participating in HMIS must not exceed the total number of corresponding beds recorded in the Bed and Unit Inventory Information record at any point in time." The programs' ability to operate "above capacity" should be allowed and recorded in the HMIS at various points in time and should trigger an alerting message for the user to confirm over-capacity, rather than triggering an error message. In addition, HUD should review and consider all residential programs' unit occupancy and not simply bed occupancy. Family programs in particular may have 100% of the units occupied but with different family configurations that don't maximize their bed capacity. Programs do intentionally try to optimize their bed capacity, but in past years the COCs have been penalized when a program's unit occupancy is 100% but their bed capacity is lower than the 75% HMIS participation threshold or the continuum wide occupancy rates.

In the case of family housing, HMIS should be used to track the number of units of housing, but not the number of beds, since a unit of family housing can typically add or remove beds depending upon the size of the family and the ages of family members. In the case of family housing, HMIS should be used to track the number of units of housing, but not the number of beds, since a unit of family housing can typically add or remove beds depending upon the size of the family and the ages of family members. In fact, scattered site programs funded to providing leasing assistance and services to 10 individuals may, at times, be able to serve 11 or 12 individuals by virtue of their ability to raise supplemental funds, or in cases where participants are able to contribute a portion of their housing costs. This ability to operate "above capacity" should be allowed by the HMIS, rather than triggering an error.

Response: Inventory data should be updated as capacity increases. HUD will continue to track inventory at the bed level.

Modification to Standards: None.

Section 2.10 Target Population A (Optional)

Organization: Northrop Grumman

Comment: If you are requiring the housing chart to be created from the HMIS, wouldn't this field be required? Is this data field not a part of the housing chart?

Response: Target Population A has been removed from the housing inventory, and is therefore an optional data element.

Modification to Standards: None.

Section 2.10 Target Population A (Optional)

Organization: State of Montana

Comment: I was wondering about the Target Population A list of data elements. One of our

programs serves Teenage Pregnancy participants --- children with children. Which category would that fall in on page 34 (Table 2.10) --- would that be "Households with Children"?

Response: Standards are clear about how to categorize households with children.

Modification to Standards: None.

Section 3.1 Name

Organization: Cambridge (MA) Department of Human Services

Comment: Given the target population of street outreach programs and large shelters serving a transient population, it is unrealistic to expect that programs will be uniformly able to capture their clients' "legal names," as is prescribed. An effective HMIS needs to be able to recognize clients who identify themselves with a nickname or other alias, and to subsequently match them with previously stored identities that include an alternate version of the name, or differently-spelled-but-similar name.

Response: Standards already build in flexibility with the phrase "seek to obtain legal names."

Modification to Standards: None.

Section 3.1 Name

Organization: Northrop Grumman

Comment: There should be an option for none on middle name and suffix.

Response: It is okay to leave middle name blank. Explicit mention of this is not necessary.

Modification to Standards: None.

Section 3.2 Social Security Number

Organization: Cambridge (MA) Department of Human Services

Comment: A full or partial SSN may or may not be reliable. If it is reliable, it can be used in the de-duplication process (particularly if by "partial" SSN, we mean the last four digits).

However, if the SSN is not reliable, it should not be used in the de-duplication process.

We would recommend the following re-labeling of the SSN Type values: · "SSN

Type #1" = Full and Reliable SSN Reported, •"SSN Type #2" = Partial and Reliable SSN

Reported, •"SSN Type #3" = Unreliable Full or Partial SSN Reported The proposed revised labels would give software a tool for filtering SSNs that are not reliable enough to use as identifiers in the de-duplication process.

Response: As a general practice, data that are not reliable should not be recorded, and the collection of unreliable data should not be considered a legitimate alternative. SSN should not be used alone for de-duplication.

Modification to Standards: None.

Section 3.3 Date of Birth

Organization: Cambridge (MA) Department of Human Services

Comment: We appreciate your decision to allow for incomplete birthdates and to add an indicator about the quality of birth date information. We would recommend the following re-labeling: •"Date of Birth Type #1" = Full and Reliable DOB Reported, •"Date of Birth

Type #2" = Approximate, Unreliable, or Partial DOB Reported The proposed revised labels would recognize that a Full DOB is not inherently reliable, and, unless it is reliable, should not be used as an identifier in the de-duplication process.

Response: The current language of "approximate" suffices. The collection of unreliable data should not be considered a legitimate alternative.

Modification to Standards: None.

Section 3.4 Race

Organizations: Cambridge (MA) Department of Human Services, hopeFound, and Pine Street Inn (Boston, MA)

Comment: In today's world, an increasing number of people are of "mixed race" and may not be able to or wish to indicate the combination of races involved in their lineage. The HMIS needs a "mixed" race or "other" race to accommodate the multiplicity of people that programs serve. This category would also assist in categorizing street outreach clients and transient clients in large shelters who may not be able to provide the specific information about their racial lineage.

Response: Response values follow OMB requirements

Modification to Standards: None.

Section 3.9 Residence Prior to Program Entry

Organization: Cambridge (MA) Department of Human Services

Comment: Please change the descriptors for choices 10 and 11. When staff/clients hear "Apartment or house that you own" they interpret that as "apartment ..... or house that you own" rather than "apartment that you own... or house that you own". The word "apartment" connotes "rental arrangement" to most people. Most people think of an "apartment that they own" as a "condominium". AHAR data is probably skewed by this misconstruction of terms. Better ways to describe these options are: 10 = Client-rented room, apartment, or house, 11 = Client-owned apartment, condominium, or house

Response: Modification incorporated into revision.

Modification to Standards: Re-worded response options to: 10 = Rental by client: room, apartment or house 11 = Owned by client: apartment, condo or house. To be consistent with the destination field, we incorporated questions about whether or not clients were receiving a housing subsidy in their previous residence.

Section 3.9 Residence Prior to Program Entry

Organization: Cambridge (MA) Department of Human Services

Comment: Please add "Domestic violence situation" to your list of options. This would more accurately describe a situation than a prior living situation that is defined by who is on the lease (e.g., if a woman was living with an abusive partner in an apartment, we shouldn't need to figure out which name(s) were on the lease or mortgage for the purpose of the universal data element).

Response: The response value requested would not be mutually exclusive with other options.

Commenter is advised that users do not have to look up the actual name on the lease.

Modification to Standards: None.

Section 3.9 Residence Prior to Program Entry

Organization: Cambridge (MA) Department of Human Services

Comment: Please clarify how you wish the following prior living situations to be categorized (i.e., as "Other" or as a different category): •Residential program for ex-offenders who have left jail/prison •Residential program for persons in recovery from substance abuse ... after they have left a treatment facility •Residential program for persons with mental illness (distinct from "facility"-based care) If you would like these prior living situations to be grouped with the more facility-based situations that may have preceded them, we would recommend adding the words "or post-release residential program" to the following: •Jail, prison, or juvenile detention facility or post-release residential program •Substance abuse treatment facility or detox center or post-release residential program •Psychiatric hospital or other psychiatric facility or post-release residential program. Presumably, any of these residential programs incorporate some kind of "subsidy".

Response: Communities can decide these marginal cases locally.

Modification to Standards: Added statement that communities can add other categories, as long as they can map to standard categories.

Modification to Standards: None.

Section 3.9 Residence Prior to Program Entry

Comment: Please clarify how you wish the following prior living situations to be categorized (i.e., as "Other" or as a different category): •Nursing or rest home, •School dormitory, •Military •"Traveling"

Response: Communities can add other elements that can map to "other" variable.

Modification to Standards: Add statement that communities can add other categories, as long as they can map to prescribed response categories.

Section 3.9 Residence Prior to Program Entry

Organization: No More Shelters

Comment: I would also urge you to ask the last two housing locations of the client - the last one is often not telling - and without doing a complete history - adding just one more with duration is much more effective and many agencies already do this.

Response: Too burdensome.

Modification to Standards: None.

Section 3.10 Zip Code of Last Permanent Address

Organization: Bowman Systems

Comment: Section 3.10 of the Revised Draft Standards does constitute a change from previous standard with the additional response category “2=Partial Zip Code Reported.”

Response: This response value is removed.

Modification to Standards: Remove response value, change name of first value to "Full or Partial Zip Recorded".

Section 3.10 Zip Code of Last Permanent Address

Organization: Cambridge (MA) Department of Human Services

Comment: Clients do not always know/remember the zip code of their most recent permanent address. In the case of immigrants from another country, that permanent address may not have had a five-digit zip code. Please consider allowing users to record the name of a city, state, or country when the five-digit zip code is not known or is inapplicable.

Presumably there could be a "Zip Code Type = 3 = place name"

Response: Place names do not belong in a field intended for zip codes. The benefits of adding extra fields and quality codes do not outweigh the loss of simplicity.

Modification to Standards: None.

Section 3.10 Zip Code of Last Permanent Address

Organization: hopeFound

Comment: People in crisis often don’t know/remember the zip-code of where they lived prior to becoming homeless, especially when that was some time ago. The option to enter the name of a town/city or even county could counteract too many “don’t know” entries.

Response: Place names do not belong in a field intended for zip codes. The benefits of adding extra fields and quality codes do not outweigh the added complexity.

Modification to Standards: None.

Section 3.11 Homeless Status

Organization: Cambridge (MA) Department of Human Services

Comment: We appreciate your well-reasoned distinctions. Please clarify whether "couch surfing" belongs in the "imminently at risk" category or the "precariously housed" category. "Couch surfing" is typically used to construe a day-at-a-time situation that is much more temporary than "living with friends" or "doubled up".

Response: Data element is revised.

Modification to Standards: Homeless status question is revised to clarify definitions of categories.

Section 3.11 Homeless Status

Organization: No More Shelters

Comment: Integration of a homeless/housing status question is long overdue, however, the exit form for such at-risk people (in addition to the housing at entry) may still not conform to the experiences of non-homeless people.

Response: The data standards distinguish between questions geared toward programs targeting homeless persons, and HPRP, which may target non-homeless persons.

Modification to Standards: None.

Section 3.11 Homeless Status

Organization: The Partnership Center, Cincinnati, OH

Comment: You have created with these instructions a new “category of homeless persons” i.e. imminently at-risk of becoming literally homeless. THIS WILL CAUSE HOMELESS

DEFINITION CONFUSION and should be removed. HMIS data standards should not be

setting policy – but capturing information. This is a policy change in the way we “define homelessness” that combined with recent definitional issues lobbying will increase

confusion. Remove 1- Literally homeless and 2- “Imminently at-risk of becoming

literally homeless” to HOMELESS. Alternative issue: If you do not choose to remove

“at risk of becoming literally homeless” then you must move “fleeing domestic violence”

to be listed as literally homeless.

Response: We agree that the previous formulation may cause confusion.

Modification to Standards: The name of the data element and the “imminently at risk” response value is revised to ensure that they are neutral on homelessness definitions and that statuses can map to various past and present definitions. “Fleeing Domestic Violence” is moved to the “literally homeless” category.

Section 3.11 Homeless Status

Organization: Pine Street Inn

Comment: It seems that this means that an “at risk of homelessness” response would be added to the existing Homeless- Yes or No question. If this is correct, then we believe that this needs to be added to Out Reach, Emergency and Permanent Housing Programs as well as special programs that service at risk people.

Response: There is no existing element in the data standards. Commenter may be referring to a locally developed element.

Modification to Standards: None.

Section 3.12 Program Entry Date

Organization: Cambridge (MA) Department of Human Services

Comment: Generally speaking, your guidance is helpful. In particular, we appreciate the flexibility of allowing non-residential service programs to define a program-specific threshold of client non-participation for triggering discharge. We have the following concerns: •For non-residential programs, we would recommend that a one-time service should have a "length of stay" of one day, rather than the zero days that would result if entry date = exit date. Software which links assessments and services to an enrollment period may be unable to make that linkage if discharge date and assessment date are the same.

Response: This is a software programming issue.

Modification to Standards: None.

Section 3.12 Program Entry Date

Organization: Cambridge (MA) Department of Human Services

Comment: For large, day-at-a-time shelters serving a transient population, it is impractical to

record a full set of entry data each time there is a break in the continuity of services. A

client who returns to such a shelter after an absence of a day or even a week should be

counted as receiving services within the same program enrollment. Clients in a

metropolitan area may routinely relocate between multiple shelters (and unsheltered

settings). We would recommend that intermittent stays in a shelter should all be treated

as part of the same enrollment, unless there is a significant discontinuity that

accompanies a known change in circumstances (e.g., incarceration or treatment), or

unless the discontinuity exceeds an agreed-upon threshold.

Response: Programs that are not using entry and exit dates to track length of stay may determine the period of non-attendance that constitutes a program exit, after which returns to the program require new program entry data.

Modification to Standards: Added language clarifying that, programs not using entry and exit dates to track length of stay may determine the period of non-attendance that constitutes a program exit, after which returns to the program require new program entry data.

Section 3.15 Household Identification Number

Organization: The Partnership Center, Cincinnati, OH

Comment: To what end are you asking for the creation of a new household number? Children are very often born in shelter and housing. Children often are left with relatives at the intake into shelter of parents, until they determine it is a safe space and have their

children join them. Family members do come and go. They tend to balance each other

out. They remain the same family. The implication of an additional household id is that

this is a different family… which is simply not true. How are we to look at length of stay

for this family? How are we to look at housing and financial outcomes? If we treat

them as a new family with the addition of a new member, we need new intake data to

apply to this new family – and therefore an exit for the ‘old’ family. This would seriously skew outcomes. Example: a family arrives in transitional housing and a child is born. The family’s record must now contain an exit from TH with an exit destination of TH. The mother was clearly not working at the birth of the child – so the income level declines. So both housing and income outcomes for that record are not positive. Now they are reentered into the system – as a new family, starting all over again. Recommendation: As long as the primary client / head of household remains stable and the family vs. individual status remains the same, the household is regarded as the same. In the event that the primary client/head of household changes, or family vs. individual status changes, the original household should be exited and a new intake / household created.

Response: This formulation was in error and fixed

Modification to Standards: Revised text reverts to the original formulation stating that when a person joins a household that person should receive the same household identification number as the other family members.

Section 3.15 Household Identification Number

Organization: Rhode Island Coalition for the Homeless

Comment: In the new Data Standards sec 3.15 (pp. 60-61) it states, a household is a single

individual... and asks for a household identification number. It also states that there are

no changes from the precious notice. In the old notice 2.14 (pp.45910-45911) it states,

a household is a group of persons... If one wants to know how many individuals there are

as opposed to how many families, why give an individual a household identifier and why

classify an individual as a household at all. How would you propose to separate individuals

from households? Do the same rules apply as before, i.e. a household only contains adults

with someone under 18. and does this apply to both the AHAR and the APR?

Response: No change was made to the actual data element, thus “no change” was indicated. A "household" is distinct from a "family". Households have one or many persons. A single individual is a household consisting of one person.

Modification to Standards: None.

Section 4.1 Income and Sources

Organizations: The Wilder Foundation, Bowman Systems, Cambridge (MA) Department of Human Services, hopeFound, The Partnership Center, Pine Street Inn, State of Montana

Comment: There was a near universal agreement among persons submitting comments that adding response categories for each possible income source was too burdensome. To reduce the burden, the standards should be changed to allow communities to a. collect income data at the household level, rather than for each client and b. ask respondents to affirm the sources of income they are receiving, and have the HMIS software automatically record a No response for all other income sources.

Response: Based on the level of concern from respondents, we are inclined to agree that the benefit of distinguishing “No” responses from “Don’t Know”/”Refused” responses does not justify the additional burden communities. TheYes/No question, along with the “Don’t Know” and “Refused” response categories for the general question on whether a client has received any income in the past 30 days should be kept. But it is not necessary to require that the question be answered for each specific income source. Instructions indicate that only data regarding the income sources received needs to be recorded. To further reduce burden, recording income amounts for each specific source has been made optional, except for the earned income amount, which is required for program reporting.

Modifications to Standards: The “Don’t Know” and “Refused” response values have been removed in the “Receiving Income from Source” question for each potential source of income. The “Amount from Source” question has been made optional, except when clients indicate they are receiving earned income.

Section 4.1 Income and Sources

Organization: Bowman Systems

Comment: Section 4.1 and 4.2 of the Revised Draft Standards includes methodology for income which limits the income applicability to 30 days at a time. Income captured with start and end dates would be more useful and more representative of a person’s income at any point in time rather than limiting to a single point in time.

Response: Our presumption is to not change the current standards unless there are widespread problems with it. Also, we think that most people would find it more onerous to have to start a new record every time there was a change in income from any source.

Modification to Standards: None.

Section 4.1 Income and Sources

Organizations: Cambridge (MA) Department of Human Services, hopeFound

Comment: Food stamps, fuel assistance, and the Earned Income Tax Credit should all be listed as Sources of Income.

Response: HUD does not consider these forms of assistance as income for the purposes of determining eligibility for its subsidized housing programs. For the sake on consistency we are not including them as income sources here.

Modification to Standards: None.

Section 4.1 Income and Sources, 4.2 Non-Cash Benefits

Organization: The Partnership Center

Comment: Income received from any source in the past 30 days should be changed to ‘Income received or awarded from any source in the past 30 days’ because it is possible that a client has been awarded income but has not actually received the check. This change will more accurately reflect a client’s financial situation. The same is true for non-cash benefits.

Response: This change may open the door to speculation about what kind of income and benefits a client may receive. Additionally, this would represent a change in how HUD defines income and benefits, which would make year-to-year comparisons more difficult and there is not sufficient concern about the current standards to warrant a change.

Modification to Standards: None.

Section 4.2 Non-Cash Benefits

Organizations: The Wilder Foundation, Bowman Systems, Cambridge (MA) Department of Human Services, hopeFound, The Partnership Center, Pine Street Inn, State of Montana

Comment: Similar to the comments for Section 4.1, respondents felt that adding “Don’t Know” and “Refused” response categories for each possible income sources was too burdensome.

Response: We agree and recommend removing the “Don’t Know” and “Refused” response categories for all but the general question of whether the client has received any non-cash benefits from any source in the past 30 days. However, we will note in the instructions that when data is being collected through client interviews, that clients should be prompted to answer Yes/No to whether they receive each of the listed non-cash benefits.

Modifications to Standards: Remove “Don’t Know” and “Refused” response categories for all but the general ‘Is client receiving any non-cash benefits’ question.

Section 4.2 Non-Cash Benefits

Organization: The Partnership Center

Comment: ‘Section 8, Public Housing, or other rental assistance’ as a non-cash benefit is redundant since this information is already captured by the proposed Housing Subsidy Type sub-question in the Destination field.

Response: ‘Section 9, Public Housing, or other rental assistance’ is already being collected in the current standards so we see little reason to remove it, and it may be even more relevant now for homeless prevention programs, that may be working with persons in subsidized housing.

Modification to Standards: None.

Section 4.2 Non-Cash Benefits

Organization: Cambridge (MA) Department of Human Services

Comment: The HMIS should document any source of health coverage, and not just government-funded sources. Program staff need to know how all of their clients are covered, and not just how those with Medicaid, Medicare, or SCHIP are covered.

Response: While there are valid reasons for wanting to collect more information about health coverage, we do not feel that they merit changes to the data elements. In the Special Issues section, we state that ‘programs may choose to record additional information about non-cash sources of income, including: information related to benefit eligibility…’

Modification to Standards: None.

Section 4.3 Physical Disability

Organization: Pine Street Inn

Comment: What level of services qualifies as ‘currently receiving services or treatment’ e.g., going once to a clinic or doctor, on medications, etc.?

Response: We feel that program staff should use their own judgment in determining what qualifies as ‘currently receiving services or treatment’.

Modification to Standards: None.

Section 4.3 Physical Disability

Organization: hopeFound

Comment: The language used for the ‘currently receiving services or treatment… prior to exiting the program’ does not apply to the entry assessment. In the exit assessment, language like ‘…or received services/treatment while enrolled in the program” would be less confusing.

Response: We feel that the current guidance is sufficient and that communities will understand that the treatment question may need to be phrased differently at program entry, annual updates, and program exit.

Modification to Standards: None.

Section 4.4 Developmental Disability

Organization: Cambridge (MA) Department of Human Services

Comment: Please clarify: would a learning disability (e.g., dyslexia, ADHD) be categorized as a developmental disability, if it impaired the ability of the client to function independently. If not, would it be captured as a chronic health condition?

Response: We feel that the standards already provide sufficient guidance to programs on defining developmental disabilities and chronic health conditions. We cannot provide specific guidance on all possible cases.

Modification to Standards: None.

Section 4.9 Domestic Violence

Organization: Tri-Valley Haven (CA)

Comment: I am seriously concerned about section 4.9 of the data standards which appears to encourage providers of homeless services to collect domestic violence histories about their clients and enter it into HMIS. This seriously compromises client safety and confidentiality and violates the spirit of VAWA 2005. If, as the rationale states, the intent is to learn how many recipients of homeless services have experienced domestic violence, this information should be collected and submitted as aggregate data to protect survivors of domestic violence in the event of a breach of the system whether intentional or accidental.

Response: We do not feel that this data element represents a security risk. In the likelihood that a stalker were to access the HMIS data of a non-DV program it seems like they could discern the identity of their victim more easily from the existing HMIS data (name, SSN, dob). Plus, this information is already collected in the current standards.

Modification to Standards: None.

Section 4.9 Domestic Violence

Organization: The Partnership Center

Comment: Our local data shows that male perpetrators often reply yes to DV experience – which is not the question you are asking. Change field label to: Domestic Violence Victim Experience.

Response: Agreed, although Domestic Violence Victim/Survivor seems more descriptive.

Modifications to Standards: Change label of Section 4.9 to Domestic Violence Victim/Survivor.

Section 4.10 Destination

Organization: No More Shelters

Comment: Public Housing and Section 8 should be treated as distinct housing destinations.

Response: We feel that this may make things overly complicated as not all program staff and clients know the distinction between Public Housing and Section 8, and that the added benefit for national reporting purposes is not enough to merit changes to the standards.

Modification to Standards: None.

Section 4.10 Destination

Organization: Northrop Grumman

Comment: What values should be selected for Tenure and Housing Subsidy Type if the Destination Type is “Deceased”? Should there be a “Not Applicable” or “Deceased” option for those, too?

Response: There are numerous Destination Types, where “Tenure” and “Housing Subsidy Types” are not applicable or could cause logical conflicts with the actual destination, which could lead to reporting problems. The most reasonable course is to integrate the three questions into one question, incorporating tenure and subsidy information into the response values in the few cases where these data are applicable. Integrating into a single question also reduces data entry burden.

Modification to Standards: Modified response categories to roll-up the destination type, tenure, and housing subsidy elements into one data element.

Section 4.10 Destination

Organization: Cambridge (MA) Department of Human Services

Comment: There is a need for a greater distinction between "permanent" and "temporary" destinations, beyond whether there is any prescribed length of stay, even with your caveat that "these terms are not meant to predict length of stay." Although a living arrangement may not be explicitly time-limited, staff may reasonably assess that it is unstable, and likely to fall apart within a short time frame. An example furnished by my colleague was of a client who moves into a friend's apartment without letting the landlord know about it. Although there is no explicit time limit, this situation is a ticking bomb, in that the landlord could order the former client out of the apartment, or claim that the tenant had violated the lease by adding a long-term "guest". Unless we are misreading the definition, this outcome would nonetheless be considered “Permanent.”

Response: These are all good points and that is why we say in the standards that ‘these terms are not intended to predict length of stay’. However, for reporting purposes it is important to have an objective standard and that is why we draw the distinction at whether or not the destination has a specific limit on length of stay.

Modification to Standards: None.

Section 4.11 Date of Contact

Organization: hopeFound, Pine Street Inn

Comment: Given the lack of technology available for most Outreach Programs, where data could be entered in real time with a system generated time stamp, staff will most likely have to transfer their encounters from a paper log into the HMIS application, entering the time for each encounter manually. Databases allow for multiple entries for the same record set with the same date, so multiple encounters can still be counted without the time stamp. What couldn’t be produced is a sequence of the encounters but we believe that this would be a marginally relevant detail not warranting the staff time needed to enter such data.

Response: Date of contact is necessary so that street outreach programs can report the number of times each client is contacted. The hour and minute of each contact is collected to distinguish between duplicate records and cases where the same client was contacted multiple times in the same day.

Modification to Standards: None.

Section 4.12 Date of Engagement

Organization: Pine Street Inn

Comment: The addition of “Date of Contact” and “Date of Engagement” for Out Reach Programs is potentially confusing. The Date of Engagement really is the Program Start Date. The Date of Contact could be captured in the bed-list-type contact list. Maybe Site Admins would have the permission to switch Date of Engagement for Program Start Date but both names would, in fact, function the same way in the database.

Response: The instructions allow for program staff to define the date of engagement as concurrent with the program entry date.

Modification to Standards: None.

Section 4.13 Financial Assistance Provided

Organization: Alameda County (CA) HCD

Comment: Clarification is needed on the requirement for a new HPRP record at least once every three months. It is clear that service and financial assistance data must be collected at least once every three months during program enrollment. It is not clear from the guidance if a service is entered in time period one and extends past three months whether the new record would require that the previous record show an end date of the service corresponding to the start date in the new record, or whether the end date of the service would be the same in all records. For example, if a client received legal services for a four month period, would the agency be required to enter an end date for the service at the end of the first three months and then a new start and end date reflecting the fourth month, or would the end date in both cases be the final end date for the service? If each record has a distinct start and end, will it be possible to easily determine what the actual length of the assistance was?

Response: The end date is the end of the three month period where financial assistance is being tracked, not the program exit date. The total length of assistance would be the time elapsed between the initial financial start date and the final financial end date.

Modification to Standards: None.

Section 4.13 Financial Assistance Provided

Organizations: Cambridge (MA) Department of Human Services, Alameda County (CA) HCD

Comment: We appreciate HUD's cautious and flexible approach to requiring data in conjunction with the HPRP initiative. It is our intent to separately track payment of rental arrearages vs. provision of prospective rental assistance, and likewise to separately track payment of utility arrearages vs. provision of prospective assistance with utility payments. To the extent that HUD standards shape the design of HMIS software, we would recommend that HUD guidance explicitly mention that possibility. It is also our intent to track the use of other sources of client financial assistance which HPRP funds leverage. To the extent that HUD standards shape the design of HMIS software, we would recommend that HUD guidance explicitly mention that possibility, as well, given HUD's ongoing interest in encouraging programs to leverage a diversity of resources.

Response: Programs can choose to track arrearages separately from prospective rental assistance as long as they can aggregate everything up to Rental Assistance for reporting purposes. We do not see the need to track these types of assistance separately at the national level, but do not want to discourage programs from doing it if they feel it is important.

Modification to Standards: None.

Section 4.15 Additional Program-Specific Data Elements: Client Outcome Measures

Organization: Pine Street Inn

Comment: We looked at the Arizona Self Sufficiency Matrix. Our opinion is that this type of matrix could be helpful if the person evaluates his/her situation with a case manager. If the case-manager is required to complete this and does so without the ‘client’, then our opinion is that it is not an objective tool…

Response: Agreed.

Modifications to Standards: When the client outcome measures are re-introduced in the full OMB package, we will remove ‘self-administered form’, and ‘case manager records’ as eligible data sources, requiring that this information be collected through client interviews.

Section 4.15 Additional Program-Specific Data Elements: Client Outcome Measures

Organization: Alameda County (CA) HCD

Comment: Scales such as these have been analyzed in other fields and demonstrated to be unreliable by researchers because of inter-rater reliability issues. The movement in the field toward a housing first approach that recognizes that homelessness is a housing problem and is ended by obtaining and sustaining housing is potentially set back by the inclusion within HMIS of so many pieces of subjective information that are unrelated to housing outcomes. I strongly urge HUD to reconsider the inclusion of these domains within HMIS.

Response: The domains from the Arizona Self-Sufficiency Matrix were chosen because it has performed well in field tests of inter-rater reliability. These domains were added primarily based on input from communities at the Annual Performance Report focus groups requesting that they be given more freedom to report on their successes in domains that are critical to their clients. Programs will also continue to be evaluated based on their success at helping clients find and retain permanent housing.

Modifications to Standards: The Client Outcome Measures will be removed from this version of the Standards because HUD feels that communities need a full 60-day public comment period before requiring programs to capture this information.

Section 4.15 Additional Program-Specific Data Elements: Client Outcome Measures

Organization: National Alliance to End Homelessness

Comment: The inclusion of so many indicators not related to housing stability may encourage programs to focus their homeless assistance on activities not related to housing stability. HUD should communicate, in this notice and in other guidance, that the function of homeless assistance programs is to first and foremost address housing stability, and that other types of assistance should be provided and measured when there is a compelling connection to housing stability.

Response: The Client Outcome Measures were added largely based on feedback programs gave in focus groups across the country that they wanted more flexibility to report on outcomes that were critical to the success of their programs. These outcome measures are not intended to replace any of HUD’s current performance measurement metrics for homeless assistance programs.

Modifications to Standards: The Client Outcome Measures will be removed from this version of the Standards because HUD feels that communities need a full 60-day public comment period before requiring programs to capture this information.

Section 4.15F Children’s Education Domain

Organization: Alameda County (CA) HCD

Comment: On Page 94, Section 4.15F Children’s Education Domain, the Definitions and Instructions Section states this data element is required for “programs that report on income as a performance measure in the Annual Performance Report”. The word “income” is an error. This data element should pertain to those programs that report on “children’s education” as a performance measure in the Annual Performance Report.

Modification to Standards: Change “income” to “children’s education” in the Definitions and Instructions section, when the field is re-introduced with a 60-day comment period.

Section 4.16A Employment

Organization: Cambridge (MA) Department of Human Services

Comment: Rather than collecting the number of hours a client worked in the past week, we believe that it is more useful to categorize whether the position is full time or part time… We would recommend that the “seasonal” answer be re-classified as “seasonal or day labor”… In our HMIS, we ask whether the employment offers a) both sick time and health insurance b) sick time but no health insurance c) health insurance but no sick time d) neither sick time nor health insurance… To better reflect the possibilities of clients who are unemployed, under-employed, or in need of supplemental employment, we would recommend re-framing your last question on employment as “Is the client looking for a job, a second or third job, a better job, or additional hours”.

Response: We don’t see a compelling reason to change the number of hours per week do a drop-down response. Day labor fits more logically into temporary employment. In the Special Issues section we already encourage programs to collect additional information about benefits received through employment. We agree that the last question needs to be re-worded since if does not make sense to ask a person who is not currently employed if they are seeking more hours.

Modifications to Standards: Change final question to ‘Is the client looking for work and/or increased hours?’

Section 4.16G Reason for Leaving

Organization: Bowman Systems

Comment: Section 4.16G of the Revised Draft Standards states that Reason for Leaving is now an optional data element. What is the rationale for making this optional instead of required?

Response: This data element was made optional because it is no longer required to complete the APR.

Modification to Standards: None.

Section 4.16H Services Provided

Organization: Bowman Systems

Comment: Section 4.16H of the Revised Draft Standards states that Services Provided is now an optional data element. What is the rationale for making this optional instead of required?

Response: This data element was made optional because it is no longer required to complete the APR.

Modification to Standards: None.

Section 4.16H Services Provided

Organization: Northrop Grumman

Comment: Is the data element "Services Provided" the same as the existing "Supportive Services Types"?

Response: It is the same as the existing “Services Received” data element.

Modification to Standards: None.

Level of Effort/ Burden

Organization: Bowman Systems

Comment: These changes will be costly for vendors to implement in terms of time and resources

Response: HUD believes that the results will outweigh the burden.

Modification to Standards: None.

Level of Effort/ Burden

Organization: Alameda County (CA) HCD

Comment: The estimate of 210 reporting hours per grantee has no associated time frame. Is this annually or is it for the entire three year program reporting period for the HPRP program? Does it include the time needed by subgrantees? If annual and including subgrantee time I think it is unlikely that all subgrantees and the grantee will be able to meet the reporting requirements with a combined total of 17.5 hours per month for these tasks. In the past, HUD's time estimates for required data collecting and reporting have generally been significantly understated on other projects/programs.

Response: HUD believes the estimated reporting burden is accurate.

Modification to Standards: None.

Level of Effort/ Burden

Organization: Committee on Temporary Shelter

Comment: Unfortunately, HUD’s proposed expansion of HMIS data elements goes well beyond the initial Congressional directive to collect unduplicated counts of homeless persons and analyze patterns of use. Still worse, these changes will impose undue administrative and financial burdens on human service agencies at a time they can least afford them.

Finally, even the best HMIS will do nothing to stem the tide of homelessness in our country. The proposed HMIS revisions add a significant staffing and resource burden to shelter and service organizations without improving program quality or opportunities for the homeless. Further, the proposed revisions will not provide a more accurate count of the homeless than is currently available through the universal level of data already collected by HUD-funded agencies.

Response: The Department believes the comprehensive data collection and reporting is critical to addressing the homeless problem.

Modification to Standards: None.

Level of Effort/ Burden

Organization: Committee on Temporary Shelter

Comment: The initial HMIS requirement was developed, in part, to determine the extent of homelessness, develop an unduplicated count, and provide an overall view of homelessness. HUD’s proposed revisions, however, go far beyond what Congress initially requested. The new requirements will make it necessary for nonprofits to invest in costly software and computer systems or force them to duplicate program level databases that they’ve used for years. Still worse for rural states, organizations who receive small sub-grants from HUD Continuum funds may not be able to afford to participate. They’ll lose critical funding at the worst of all times and we’ll lose the universal level data that they provide on the current HMIS systems. The unexpected byproduct of this burdensome requirement is that HUD may wind up with less, not more, data to quantify the extent of homelessness in America. Another consequence of HUD’s proposal is that it will greatly increase the amount of time needed to collect and enter the program level data- time that could be better spent on vocational counseling or housing search assistance.

Response: The new Standards are unlikely to create more than minimal additional burdens for programs already participating, or need for new data systems beyond the existing level.

Modification to Standards: None.

Level of Effort/ Burden

Organization: Elim Transitional Housing, Minneapolis, MN

Comment: While we appreciate and need the Recovery and Reinvestment Funds, it will be difficult enough for communities to implement the HPRP in the time frame provided. Adding additional data elements, along with potentially doubling the number of participants into local HMIS systems will create a bureaucratic nightmare.

Response: The number of data elements added for HPRP is minimal and necessary to track both outputs and outcomes of the significant investment.

Modification to Standards: None.

Level of Effort/ Burden

Organization: Elim Transitional Housing, Minneapolis, MN

Comment: We request that no further data elements are added to HMIS and that Congress immediately holds oversight hearings across the country with direct service providers who are currently trying to deal with the HMIS bureaucracy and develop a data collection system that is outcome based not a case management tool. We request an at least an additional 60 days of public comment period be provided on any data element changes in HMIS.

Response: The revisions to the data standards were approved for emergency review because of the new data elements required for reporting for HPRP programs. There will be a full 60-day public comment period for the standards later this year.

Modification to Standards: None.

Level of Effort/ Burden

Organization: Elim Transitional Housing, Minneapolis, MN

Comment: The Federal Government needs one simple, flexible definition of homelessness prior to expending any more tax dollars on a data collection system. Homeless means: without a home. The U.S. Department of Education’s definition of homelessness with some minor modifications provides that flexibility. The most recent list of Federal definitions on homelessness is available from the Acting Executive Director of the U.S. Interagency Council on Homelessness. (I attempted to attach it; but the document was too large and not accepted when I emailed this document). Minnesota, as well as other States, also has different definitions of homelessness. The recently passed Senate Amendment of the McKinney Vento Reauthorization will add additional definitions of homelessness to the current list of Federal definitions. It is almost impossible to collect good data and to coordinate Federal, State, and Local resources with so many different definitions of homelessness.

Response: This is a legislative issue and outside of the scope of the data standards.

Modification to Standards: None.

Appropriateness

Organization: Committee on Temporary Shelter

Comment: Much of the program level data proposed by HUD requires service providers to ask highly sensitive and invasive questions of individuals who are seeking shelter or services. These questions are not only patently offensive, but would never be asked of any other American requesting other types of federal assistance (e.g., student loans). Some examples of the invasive questions HUD will require nonprofits to ask of applicants are: “Are you suffering from mental illness? Do you have a developmental delay? Have you been abused? “Have you ever been diagnosed with AIDS or have you tested positive for HIV?”

Response: The Standards state that HPRP does not require these questions. These questions are required for other programs that provide services in these areas.

Modification to Standards: None.

Appropriateness

Organization: Elim Transitional Housing, Minneapolis, MN

Comment: Will all recipients that receive Recovery and Reinvestment Funds be required to provide this type of personal data - Including but not limited to: Banks, Financial Institutions and Homeowners? The additional data collection elements are self reported on personal issues. Why does HUD need to know this information from people who are at risk of homelessness or experiencing homelessness and need legal, utility, and/or rental assistance to obtain or maintain housing? They need financial assistance! This is a Homeless prevention and assistance program, not a treatment program. We believe the request for these additional data elements may violates HIPPA laws. The data is collected by Para-professionals in the field, (who are not typically Licensed Social Workers) and who have limited training on data privacy laws.

Response: The Standards clearly state that HPRP does not require these questions. These questions are required for other programs that provide services in these areas.

Modification to Standards: None.

Suggested additional data elements

Organization: hopeFound

Comment: We believe that the combination of data collected in the elements “Residence prior to Program Entry” and “Zip Code of Last Permanent Residence” does not produce one important piece of information, especially for clients whose answer to the question

“Residence prior to Program Entry” is “Emergency Shelter” or “Place not meant for

habitation.” From these two data elements alone, we don’t know whether clients spent

some time in institutions such as jail/prisons, hospitals (including psychiatric,) etc. before

they were discharged from these institutions to wherever they stayed the night before

entering the program. A question like “Have you lived in an institution after becoming

homeless” yes/no, and if yes, when was the last discharge date and from which type of

institution (choice from list) would paint a clear picture of homelessness caused or

extended by discharges to inappropriate destinations (aka “homeless dumping”.)

Response: The elements mentioned are already being used for analysis. CoCs are welcome to add local requirements to provide additional data for planning purposes.

Modification to Standards: None.

Suggested additional data elements

Organization: No More Shelters

Comment: Where is the "length of homelessness questions"?

Response: HUD expects that as time goes on length of homelessness will be captured as client entry exits and service encounters in HMIS.

Modification to Standards: None.

Separate standards for outreach programs

Organization: No More Shelters

Comment: Outreach services entered into HMIS should not be conceptualized as entering into a formal program since most outreach is an individual event - not a continuous time period with delivery of services (typically unless the client is formally tracked regularly).

Outreach should be treated as individual contracts (not entry and discharge dates).

Outreach should also have fewer data standards to minimize burdens. Using an outreach

duration between entry and discharge assumes that the person remains with the same

housing or homeless situation throughout - and that is non-verifiable. The entire

outreach HMIS standards need to be reconceptualized and minimized in keeping with low-

demand outreach - as opposed to case mgmt.

Response: New data elements on outreach contacts and engagements address the comenter's

concern.

Modification to Standards: None.

DV Exemption

Organization: Northrop Grumman

Comment: Are DV programs still exempt from using the HMIS or are they now required if they are receiving HPRP funds... Is there a distinction?

Response: HPRP requires the use of a comparable database other than HMIS.

Modification to Standards: Added a section on exemptions for victim services providers.

DV Exemption

Organization: Tri-Valley Haven

Comment: I would also like to see it clarified in the standards that agencies whose primary mission is the provision of victim services are prohibited by VAWA 2005 from participation in HMIS. California law (civil Code 1798.79.8) further prohibits our participation, and forbids funders from requiring providing personally identifying client information into HMIS or other databases.

Response: Victim service providers must not enter data in HMIS. HPRP requires the use of a

comparable database

Modification to Standards: Added a section on exemptions for victim services providers.

Don't Know/Refused Response Categories

Organization: Toledo Lucas County Homeless Board

Comment: The addition of Don't Know and Refused Response Categories are generally warranted and understood, but will that give some data entry personnel an easy "alternative" to doing the needed inquiry of getting a useable and measurable response from the client?

Response: The Standards clearly states that this should only be used when clients don't know the answer.

Modification to Standards: None.

HPRP QPR: Data Elements, Response Categories and Justification

Organization: Bowman System

Comment: The “HPRP QPR: Elements, Response Categories and Justifications” needs to have a definitions section. For example, “households” needs to be clearly defined.

Response: Definitions will be provided either as a separate technical assistance document or in the complete revised data standards that will be published later this year.

Modification to Standards: In the final notice, we will include definitions of key QPR terms such as household, grantee, and sub-grantee.

HPRP QPR: Data Elements, Response Categories and Justification

Organization: Bowman Systems

Comment: The “HPRP QPR: Elements, Response Categories and Justifications” section 6,7,and 9 need to have a crosswalk of activity or assistance terms to that of AIRS taxonomy terms which the majority of HMIS vendors use as the primary source of service activity definition. This crosswalk will help to insure accurate and consistent reporting across all jurisdictions and vendor systems.

Response: This is a good suggestion and appropriate as a technical assistance product.

Modification to Standards: None.

Reporting Requirements

Organization: Bowman Systems

Comment: It would be helpful and instructive for all data elements contained in the Revised Draft Standards to have an additional column of information defining which reports would make use of each data element. For example, does the Program Entry Date get used in the APR, QRP, AHAR, or any other reports?

Response: This is a good suggestion and appropriate as a technical assistance product.

Modification to Standards: None.

Reporting Requirements

Organization: Northrop Grumman

Comment: Our biggest question is how do all these changes impact the current AHAR, APR processing? Does everything need to be changed retroactively to 10/01/2008 for the AHAR about to be done? Are changes to be implemented after the next AHAR scheduled date? What about programs that are just getting ready to do their APRs or need to do them in July or August and they have different sets of data they are looking at?

Response: Clarifications on timeline will be provided as a Technical Assistance product.

Modification to Standards: None.

General Comments

Organization: Bowman Systems

Comment: In the future it would increase the effectiveness of vendors’ technical response and deployment of data collection standards if all applicable and related reports were delivered at the same time of the Revised Draft Standards. Without this information, it makes it difficult to insure the best methodology for data collection design and workflow is implemented and deployed for HMIS end users.

Response: Due to the requirements of ARRA, this Standard went through emergency clearance.

Modification to Standards: None.

General Comments

Organization: Pine Street Inn

Comment: Like that “Services Received” is no longer required on the APR. Comment – We hope that this will still be required for other reporting reasons and made clear before agencies jump the gun and stop recording this data.

Response: Services received can be required locally as needed.

Modification to Standards: None.