NHSDC Comments on Proposed Data Standards

May 31, 2013 in Latest News by admin

At our recent Spring Conference in Seattle, NHSDC renewed its commitment to being a forum for exploring and building consensus among community based management information systems administrators.  Of particular relevance at this conference was the release of draft regulation from the US Department of Housing and Urban Development Special Needs Assistance Program Office on the new definition of chronic homelessness and the Data Standards for Homeless Management Information Systems (HMIS).  NHSDC has, and continues to be one of the few places that system owners and operators can gather to speak directly with each other.  We make a special effort to include a variety of stakeholders, from software providers, to service recipients; from federal officials to local elected representatives.
A broad cross-section of those groups and others were present during two excellent sessions at the NHSDC Spring 2013 Conference.  Participants working closely with the NHSDC Board of Directors have developed the below draft of language to be formally submitted to HUD on behalf of NHSDC.  While over 120 of your peers participated in the development of this language, we want to ensure the broadest possible review to support buy-in and relevance of the content.  The language below is from the session focused on the proposed HMIS Data Standards.  Where possible NHSDC conference participants have identified alternatives to the proposed methodology, participants also identified practices that we support to bolster the case for finalizing language.  Please review the language below and if in support of it let us know by commenting on this post; if you feel this is dramatically going in the wrong direction or disagree with a statement please let us know that as well.  We will attempt to integrate any trends of substantive changes that resonate broadly.
We apologize that this message is going out at such a late date. These comments are due to HUD by Monday, 3 June. As such, we hope that these comments are merely the beginning of a dialogue about the regulations.
On behalf of the Board of Directors thank you for your participation.
David Canavan
NHSDC Board Chair
NHSDC Proposed Data Standards Feedback
The National Human Services Data Consortium (NHSDC) has been a long time supporter of the US Department of Housing and Urban Development’s (HUD) implementation of Homeless Management Information Systems (HMIS) across the United States.  Long before the 2004 implementation of the formal rules by HUD, NHSDC founding members worked with the Department to try to understand the landscape and the impact this new tool would have on communities.  More than a decade has passed since those plans were first developed; HMIS and other community-based data systems have proved a valuable tool to communities with many seeking to expand coverage and increase the quality of data collected.
With the publication of the 2004 Data and Technical Standards, the Department set the expectations of software providers, communities, and the individuals and families using HUD funded housing.  Since that publication, HUD has updated various aspects of the Data and Technical Standards.  NHSDC has worked with communities to ensure that these shifts come with enough notice and enough training to be effective.  Indeed, NHSDC has become the primary forum for the delivery of high quality training and technical assistance provided by communities, expert practitioners, and federal partners.
With each iteration of HUD guidance, the focus, as published in the Federal Register, allowed HMIS partners to understand the parameters of the government’s interest in the data collected.  The Data Standards defined the questions that would be asked and the response categories; the Technical Standards defined the manner and technology with which the HMIS would be operated.  A critical element to the initial publication in the July 2004 Federal Register was the inclusion of HUD’s responses to comments from the Notice and Comment process.  These responses reassured many future partners in HMIS that the values expressed by HUD and memorialized in the Federal Register would govern the use of the data.
Some of those essential voices may be absent from the comments on the 2013 Data Standards.  Utilizing a forum other than Regulations.Gov to publish potential policy changes may prevent organizations that are focused on the formal site established for the promulgation of regulation from participating in the final decision making.  Additionally, partners who find significant value in reading and processing comments by different constituencies’ professional advocates will miss out on their wisdom as the OneCPD website does not permit the advanced collaboration that Regulations.Gov does.
Important to note, this set of Data Standards breaks new ground; no previous Standard integrated the requirements of other Federal Departments.  While HUD has historically used a collaborative process, engaging technical assistants, HMIS users, and HMIS administrators to review changes, NHSDC has no knowledge that other federal departments use a transparent collaborative process.  As new partners are added to the HMIS effort, the decisions made in this first step will set precedent moving forward.  If HUD grantees are expected to enforce and support the participation by new local providers, it is important to recognize that the context in which they do so may have changed. This changing context includes questions around the following.
  • Funding- Can HUD Continuum of Care (CoC) money be used to pay for Health & Human Services (HHS) or Veteran Affairs (VA) reports?
  • Privacy- HUD articulated values around sharing identified data with other Federal Partners in the 2004 Standards; how has this value changed in light of the VA’s Supportive Services for Veteran Families (SSVF) program or inclusion of minors through HHS Runaway and Homeless Youth Network?
  • Contracts- Will HUD articulate a single set of expectations for collaborating with VA, HHS, and other Federal Department? Or should administrators apply their local policies and procedures to every new program?
  • Staffing- In the event that other Federal Partners add significant burden to the HMIS project, will HUD add funds to HMIS budgets for the addition of new staff?
These are a few of the question our constituents were discussing at the April 2013 NHSDC conference in Seattle, Washington.  The adoption of new Data Standards without underlying guidance on infrastructure leaves HMIS administrators in a position of supporting the HMIS without all the information required.  While some of these questions may be answered by HUD in other forums, asking grantees and HMIS stakeholders to evaluate changed Data Standards in their absence does not allow for a full consideration of the possible concerns.
It was at this recent April 2013 Spring NHSDC Conference, with participation by over 100 CoCs, long time volunteer David Eberbach and new board member Danielle Wildkress worked to review and summarize the perspective of our diverse HMIS stakeholders.  We have summarized those comments here for ease of reading and to maintain a clear understanding of the overall document.  For ease of collation HUD requested the information be inserted into a formatted spreadsheet for submission; we have also populated that document for review.
  • For questions related to length of homelessness, NHSDC members suggest that HUD provide more nuanced answer options. We do not believe that “less than one year” is specific enough. Perhaps, HUD can require programs to estimate in terms of months and years. Additionally, members also pointed out that length of homelessness will be hard to document across the CoC where the CoC has a closed HMIS.
  • NHSDC members spent a significant amount of time debating the merits and potential impacts of asking about sexual orientation. First, HUD needs to provide clarity about whether this question is asked for clients under 18 years of age. Second, while some members think that asking about sexual orientation can help us to understand if LGBTQ clients are over/under-represented, many NHSDC members are concerned that some organizations will use sexual orientation status to reject clients. Members expressed deep concerns about how this very sensitive information could have negative consequences for clients in need of housing and services. NHSDC suggests that HUD make this data element optional. One NHSDC member also posited that asking about sexual orientation may not be legal in some states.
  • Similar to the question about sexual orientation, NHSDC members have deep concerns about asking clients about pregnancy status. NHSDC agrees that this is important data; however, it may be better for this data element to be optional. Asking pregnancy status of everyone may not be appropriate, particularly for certain programs (like those that only serve men). More importantly, some women may be afraid to disclose pregnancy status when receiving services from faith-based or other organizations who may have unfavorable feelings about unwed mothers. Moreover, this data element combined with household data over a series of months/years, could unintentionally reveal which women had miscarriages or abortions.  Additionally, NHSDC members point out that HMIS vendors will need to create a streamlined means of updating this data. Potentially, there will be many data quality issues if pregnancy is not flagged for update on a regular basis.
  • Determining project funding source prior to entry and for destination will be difficult to determine, and research time will not be time well spent. The people collecting this data often do not even know how a client is funded internally. Figuring out pre and post funding simply does seem to offer sufficient value to warrant to incredible data collecting burden.
  • NHSDC would like to see definitions for services to guide responses to the “services provided” question. Alternatively, HUD should use the AIRS taxonomy.
  • Similarly, NHSDC would like a definition of “referrals provided.” Members posited that “linkages made” is a more appropriate data point worth tracking, since a referral does not necessarily insinuate that a client was successfully connected with a service. Finally, ostensibly, someone’s HIV/AIDS status could be revealed if that client is referred to a provider that is listed as a HOPWA program.
  • Area Median Income (AMI) is captured twice in the proposed standards. This data is captured in 4.19 and 4.41, but the question is asked in slightly different ways. The question should only be asked once. Having two approaches to the same data is duplicative and potentially confusing.
  • For question 2.13, SHP and S+C should simply be listed as “CoC Program.” If HUD wants to track rental assistance vs. master leasing projects, then that should be a separate question that is not limited to McKinney funding.
  • For veteran status, it is unclear why this question is asked each time a client enters a program. Veteran status is unlikely to change very frequently. Ultimately, different providers will inevitably enter conflicting data without a clear way to reconcile.
As a last overarching comment, NHSDC members want to emphasize that when HUD articulates a policy reason for collecting a particular data element, implementation of that data collection is much easier at the local level. Providers, leaders, and other stakeholders need to understand why this information is important. Otherwise, many key constituents are left to wonder and are, in some circumstances, unwilling to comply.
Thank you for your consideration. We hope to see a more formal comment period on Regulations.gov. In the meantime, we appreciate your support and collaboration in the effort to end homelessness nationwide.