NERHMIS Meeting Minutes

June 25, 2004

Participants:

Maine: Yvonne Nichols, Kat Freeman, Cindy Namer

New Hampshire: Bob Sparks, Henry Vincent, Steve Eckberg

Rhode Island: Eric Hirsch, Don Larson

Vermont: Brian Smith, Richard Rankin, Jeanne Eicks

Connecticut: Walter Hewitt

Massachusetts: Nancy Sullivan, Julia Tripp, John McGah

HUD: Mary Ann Martel, Dick Hatin, Rosslyn Black

Agenda:

UMass TA Project (handout):

• Presentation of HMIS Technical Assistance Workshop to be held on heals of National HMIS Conference in September/October 2004
• Pre-implementation: integration of data when these agencies do not want to participate; want to be able to support the aggregation of data and roll up of report (minimal participation for agencies that don’t
• Concern about date of conference
• Important to invite those that are not SP communities
• Funding is an ongoing concern
• Maybe make more sessions group sessions
• Format: have prepared materials on each subject area; brief PowerPoint presentation followed by brown bag discussion
• Define audience for workshops; possibly break out by stakeholder groups (front line staff, researchers, DOE/DOL/Housing)
• Interagency council participation of HMIS project staff/rep on committee (have HMIS chairs on council: ME, VT, )  (do not have chairs on committee: MA, RI/CT (coalition is on))

TA Needs

• Data collection and reporting
• Rhode Island: coalition of DV providers, using HMIS by putting unique ID in name field
• Iowa: successful participation of DV providers

Participation of DV providers: 

• Massachusetts: 4 participating on/off
• Rhode Island: participating with unique ID
• Maine: 1 DV shelter will participate with batch upload
• CT: don’t have any
• NH: no DV’s participating and facing reluctant; DV providers do enter de-identified information through DOJ

National HMIS Initiatives:

• TA project updates
• Onsite visits with each implementing jurisdiction
• Phone assessments with other CoC’s
• National conference
• Invited 1 CoC rep, 1 implementing jurisdiction rep; 1 policy academy rep
• Annual Homeless Assessment Report (AHAR)
• 80 sample sites
• additional 20-30 sites that have data
• demonstrating the potential of HMIS
• will collect aggregated reports from each community
• each community will be asked to generate table shells
• using extrapolation methods to generate counts of homeless
• can still record entry and exit date for people who do not want to give information
• cannot extrapolate missing data elements
• Data standards:
• Me: largest shelter not able to participate
• Client right to not enter data (at what point is a client no longer identifiable)
• Reporting protocol
• Based on universal standards (15 fields)
• Table shells to be filled out
• Ideally years worth of services; this year 3 months
• Mostly counts of people and cross tabs by type
• Nights of shelter
• What is needed from each HMIS is a query to extract information from database
• Two big challenges in creating query
• 1) Unduplicated count at the CoC level
• Report will be broken down by program type (emergency shelter and transitional housing; followed by permanent supportive housing and street outreach)
• Still unclear on how they are differentiating between individuals and families
• Each individual family member needs to have their own client record
• 2) data quality
• not entering data on every client
• when clients don’t provide information  (especially at the low barrier shelters)
• not recording exit dates
• Issues: what about clients that are not receiving beds (i.e. supportive services only); not being required to participate
• Housing Activity Chart is going to be used for coverage calculations
• Generate one day count to see how many people in shelter on the same day and how many are using; run query on a daily basis to see if there is undercounting

Street Outreach

• Rhode Island is capturing street outreach to record
• Access Rhode Island will invite clients back to office to work with them
• Providence Center mission is to connect clients and encourage clients to go back to office and sign consent
• Mostly using unique ID
• NH is entering in smaller database
•  

Next steps:

• Make connection between RI/ME DV providers
• At what point is a client no longer identifiable without getting an ROI (using a unique identifier instead of name, SSN); when do you need an ROI (Michelle check with Jon Neiditz)
• Web development with Studio:Sckaal (committee: Michelle, Cindy, Julia, John)
• Continue development of consumer committee structure (pilot with Maine)