Thursday, November 10, 2016

HMIS and Bed Usage Rates

Revised: January 2019
 

By Sarah Owens and Michael Livingston


Image: Kitwe Co, UK
In October, CANDO made a recommendation to the Mid Willamette Homeless Initiative Task Force (MWHITF) to advocate for a systematic approach to improving homeless services delivery, starting with wider use of ServicePoint, a Homeless Management Information System (HMIS) application in use in Oregon, and the development of a coordinated assessjment and entry system (CES).

Those who don't live in the world of homeless services delivery, or haven't made a study of it, might be unfamiliar with these tools.  The purpose of this post, and the next one, is to give readers some idea what they can do.

HMIS is concerned primarily with housing.  HUD requires every continuum of care (CoC) to inventory the housing available in their community, and issue an annual Housing Inventory Count report.

When an Oregon provider enters bed-usage data in ServicePoint, HUD counts those beds as "HMIS-covered."  A high bed coverage rate means the data will be meaningful.  CoCs that have high coverage rates are better able to compete for HUD funding.  Here's what the HMIS-coverage situation looks like in Region 7, based on the last bed inventory count (shaded = not covered):

Region 7 2016 Housing Programs

Program
*
Fam Units
Fam Beds
Child Only Beds
Adult Only Beds
Total Beds
S
Salem IHN

4
14


14
S
Center for H&S
DV
3
13

2
15
S
UGM Simonka

5
18

78
96
S
UGM Men’s




114 (+94 overfl)
208
S
Sable House
DV



10
10
S
Polk County




3
3
S
NWHS SOS



15

15
S
Total Beds





361


Program
*
Fam Units
Fam Beds
Child Only Beds
Adult Only Beds
Total Beds
TH
St Francis

14
46


46
TH
St Joseph

17
51


51
TH
Salvation Army




84
84
TH
Shangri-La

2
4

14
18
TH
Grace House




9 (W)
9
TH
UGM Men’s




56 (M)
56
TH
Titus Hse
AD



6
6
TH
Shelly’s Hse
CH



17 (W)
17
TH
Rstration Hse
CH



48 (M)
48
TH
HOB
V



5 (M)
5
TH
NWHS



8

8
TH
Total Beds





351


Program
*
Fam Units
Fam Beds
Adult Only Beds
Total Beds
PSH
SHA-CHDV

3
15
6
21
PSH
SHA-VASH

6
20
48
68
PSH
NWHS-HOAP



9
9
PSH
Shangri-La 0

4
8
12
20
PSH
Shangri-La 2

3
5
4
9
PSH
Shangri-La B



5
5
RRH
CAA-SSVF

2
8
3
11
RRH
CAA-OHCS

3
10
7
17
RRH
CAA-ARCHES

10
23
20
43
RRH
CAA-ARCHES

2
4
3
7

Total Beds




210

S-Shelter  TH-Transitional Housing   PSH-Permanent Supportive Housing   RRH-Rapid Re-Housing   DV-domestic violence   AD-alcohol/drug addiction  CH-criminal history  V-veterans  SHA-Salem Housing Authority   
CH&S-Center for Hope and Safety  CAA-MWVCommunity Action Agency     HH-Household   NWHS-Northwest Human Services      HOB-Home of the Brave (closed summer 2016)  CHDV-Chronically Homeless Disabled Vet SSVF-Supportive Svces for Vet Families

Not included, Father Taafee Homes, Woodmansee Community, River of Life House, Safe Families, Polk CDC, Oxford Houses

Based on the above data, Region 7's bed coverage rates look like this:

Housing Type
Total Beds in HIC
Total DV Beds in HIC
Total HMIS Beds
HMIS Coverage Rate
Shelter
361
25
14
4%
TH
351
0
124
35%
PSH
132
0
34
26%
RRH
78
0
78
100%

These rates are so low (RRH excepted) as to limit the usefulness of the data as systems measures.  Region 7's rates also drag down ROCC's rates (see below) (though Region 7 isn't the only one doing this), and limits the usefulness of ROCC's data as systems measures.  For that reason,  they also harm ROCC's ability to compete on a national level for CoC Program funding.

ROCC's 2016 Consolidated Application

ROCC's 2016 Consolidated Application
For the past several years, HUD has steadily raised the bar as CoCs improve their data collection and fine-tune their tools and methods.  If a CoC's bed coverage rate is below a certain percentage, the CoC "loses points" and below-par data won't be accepted for HUD's statistical purposes.   Here to the left is what ROCC told HUD this year about its low bed coverage rates (BOS CoC is another name for ROCC).

Translated, ROCC is acknowledging that it has not been able to persuade non-grantee providers (i.e., providers who don't receive HUD funding) to use ServicePoint.  We would question whether ROCC is "diligently working" to obtain those providers support.  Barriers include 1) license fees ($350/yr), 2) staff time (a few hours/wk), 3) philosophical objections, 4) security objections, 5) culture of non-cooperation among providers.   

To get some idea what coverage rates could tell us, at right is a draft of the most recent report on ROCC's Bed Coverage Rate (BCR) and Bed Utilization Rate (BUR).  The BUR, if based on accurate inputs, indicates how efficiently  programs are operating, and whether they're meeting the needs of the homeless community.

However, the BUR here is not based on sufficient  inputs.  The BCRs for ESFAM, ESIND, THFAM and THIND are too low because too few providers participated in HMIS.   

Other ROCC measures (Average HH Size, Length of Stay), are similarly unreliable.

(When HUD has finished reviewing the data in this report, asterisks will appear by "ESFAM, ESIND, THFAM and THIND" to indicate those coverage rates are unacceptable.) 

Next:  HMIS and Coordinated Entry

No comments:

Post a Comment