Core Performance Indicators for Homeless-Serving Programs Administered by the U.S. Department of Health and Human Services. Main Findings from Interviews with HADS Administrators and Reviews of Background Documentation on HADS

09/01/2003

Exhibit 3-1 provides a comparison of key HADS features and performance measures (as of the Summer 2002, when our interviews were conducted) in the five local sites included in this study. Below, we highlight key findings that emerge from our examination of these five HADS.

The HADS system in New York has been operational since 1986, while the other four have been designed and implemented during the past decade; all five systems are either in the process of being upgraded to use the most recent technology or were recently developed using state-of-the-art technology. As shown in Exhibit 3-1, of the five localities examined, New York Citys system is the oldest  originating in the mid-1980s. At the time of our visit to NYC, the Department of Homeless Services was pilot testing a new HADS that included basically the same data elements as the former system, but featured the latest technology in terms of hardware and software components. For example, the system is being developed using an ORACLE platform, which will enable emergency and transitional facilities located across New York City to input data directly into the system via T-1 lines. The other four HADS have all been developed and implemented within the past 10 years. The system used in Kansas City  MAACLink  originated in 1994, though during the past year its sponsoring agency (the Mid-America Assistance Coalition) has spent about $200,000 enhancing the software and other operational aspects of the system. The system utilized in Hawaii  referred to as the State Homeless Shelter Stipend Database  was initially implemented also in 1994, but is currently being substantially revised and upgraded to become a web-based application (with the new system expected to become operational by the end of 2002). The homeless agencies in Madison (WI) and Columbus (OH) have implemented the ServicePoint system, a web-based system designed by Bowman Internet Systems. The Community Shelter Board (in Columbus) implemented the system in 2000 (though 10 years of previous data was subsequently uploaded to the system), while the Bureau of Housing (in Madison) implemented the ServicePoint system in 2001. The ServicePoint system is a web-based application, with data entered at remote service sites (i.e., homeless-serving agencies in the Madison and Columbus areas) and sent electronically over the Internet for storage at secure file servers located on the premises of Bowman Internet Systems (located in Louisiana).

 

Exhibit 3-1:
Comparison of Key Features of Homeles Administraive Data Systems (HADS)
HADS Characteristics New York City, NY Madison, WI Columbus, OH Kansas City, MO Honolulu, HI
HADS Name HOMES (tracks adults), SCIMS (tracks families) BILLING (invoicing) ServicePoint ServicePoint MAACLink State Homeless Shelter Stipend Database
Year System Became Operational at Site 1986 (pilot testing new system; projected to be operational late-fall 2002) 2001 (system designed by vendor in 1997) 2000 (system designed by vendor in 1997); 10 years of earlier data uploaded to new system) 1994 1994 (being upgraded to web-based system; to be operational by end of 2002)
Administering Agency NYC Department of Homeless Services (DHS) Bureau of Housing, Special Needs Housing Community Shelter Board Mid-America Assistance Coalition (MAAC) Housing and Community Development Corp. of Hawaii (HCDCH)
Agency Maintaining System NYC Department of Homeless Services (DHS) Bowman Internet System (original system designer) Bowman Internet System (original system designer) Mid-America Assistance Coalition (MAAC) Housing and Community Development Corp. of Hawaii (HCDCH)
Partnering Agencies Who Use the System Limited to NYC Department of Human Resources (note: NY Human Resources Administration provides data on whether homeless individual is public assistance recipient  but is not a user of data.) 84 partnering agencies across WI; includes broad range of agencies, with a focus on agencies serving homeless individuals or at-risk of homelessness (includes 35 emergency shelters, 27 transitional/ supportive housing facilities, 21 DV agencies, 13 faith-based organizations, 2 tribal agencies) 28 agencies (all homeless-serving agencies)  including emergency shelters, homeless prevention programs, resource centers, housing search assistance agencies 227 partnering agencies contribute data (on-line or hardcopy) from 5 counties surrounding KC (136 agencies are connected on-line)

Partners include homeless-serving agencies, but also many other agencies serving low-income and disadvantaged individuals

 

All agencies funded by HCDCH, including emergency shelters, transitional shelters and other homeless-serving agencies
HADS Contains Data Exclusively on Homeless Individuals Yes No No No Yes
Types of Individuals for Which Data Are Maintained Families and individuals entering homeless shelters in NYC Vast majority of those entered into the system are homeless and individuals at-risk of homelessness; however, agencies may enter non-homeless individuals into the system, including low-income, individuals in employment and training programs. Any individual that comes into contact with partnering agencies, including both homeless individuals and those at-risk of homelessness

Any agency/program that the Community Shelter Board funds must use system

Only small % of those entered into system are homeless; partners may enter anyone using services at their agencies into MIS

Agencies distributing MAAC utility vouchers and agencies receiving county funds to provide homeless case management services are required to enter clients into MIS

Homeless individuals (in emergency and transitional shelter) and street outreach
Total # of Individuals Entered Into HADS to Date ~800,000900,000 ~40,000 ~54,000 ~450,000 ~100,000 (~12,000-13,000 per year for 8 years, but many duplicates across years)
# of Individuals Entered into System Each Month 7,903 families and 7,557 individuals in shelters on avg. each day (in June 2002) ~3,000-4,000 new clients entered into system each month ~750-800 new clients entered into system each month (9-10,000 in 2001) ~10,000 new clients entered into system each month (total of 112,000 in 2001) ~1000 new clients entered into system each month (12-13,000 per year)
Types of Client-Level Data Elements in System
  • Client identifiers (name, aliases, SSN, PA Case Number)
  • Current/former address
  • Client demographics (e.g., age, sex)
  • Education level
  • Household size and composition
  • Reason for homelessness
  • Special needs (e.g., substance abuse, mental health problems)
  • Health conditions (e.g., medical condition, pregnancy)
  • Vaccination data (on children)
  • Referral date
  • Facility and Room #
  • Date entered shelter
  • Date exited shelter
  • Days in facility
  • Facility information, including: # held in room, special features of room (e.g., crib), vacancy status
Extensive range of data items; state sets minimum data entry expectations (but partnering agencies may collect additional data if they choose to and develop own forms). Minimum requirements include:
  • Basic demographics  age, sex, race, marital status, veteran status
  • Current address
  • Identifiers  SSN
  • Education level
  • Household size, members, relationships
  • Current living situation, homeless status, reason for homelessness, date became homeless, whether first-time homeless, reason for leaving prior living situation
  • Employment status, hours per week, health insurance, wage, income sources and amounts
  • Medical disability
  • Service cost, duration, type of service
  • Outcomes: reason leaving, destination, length of stay
Extensive range of data items; each partnering agency develops own forms, so varies across partners). Most partners collect the following:
  • Basic demographics  age, sex, race, marital status, veteran status
  • Current address
  • Identifiers  SSN
  • Education level
  • Household size, members, relationships
  • Income/source
  • Education level
  • Reasons for homelessness
  • Last zip code
  • Service cost, duration, type of service
  • Outcomes: reason leaving, destination, LOS
  • Client and spouse identifiers (name, SSN)
  • Current address
  • Client demographics (e.g., sex, age, race/ ethnicity, veteran, handicap status)
  • Household size, members, relationships, age, SSN
  • Education level
  • Employment status
  • Whether homeless
  • Household budget (incl. income by source and actual expenditures)
  • Why help is needed
  • Client goals
  • Type of funds used
  • Services information: start date, end date, type of services received, funding source, vendor, voucher amount
  • Why left program
  • Client identifiers (name SSN)
  • Basic demographics (age, sex, ethnicity, marital status, citizenship, country of origin, Hawaii residency, veteran status)
  • Household size and composition
  • Education level
  • Employment status; reason unemployed
  • Length of homelessness and living situation at entry
  • Reason for homelessness
  • Monthly income by source
  • How referred
  • Medical and mental health history
  • Substance abuse history
  • Medical resources
  • Basic information about children (age, relationship, sex, attend school)
  • Exit information, including exit date, destination, reason for exit, income and sources, and support services received while in project
When Data Are Collected on Homeless Individuals At intake and exit from NYC shelters At intake, initial assessment, the point of service provision, and exit At intake, regular intervals, and exit At intake, point of service provision, and exit At intake and exit from shelters

Outreach staff collects data at each encounter

Who Enters Data Emergency and transitional shelter staff 84 partnering agencies enter data into web-based application

Data stored remotely on file server located at vendor site in Louisiana

28 partnering agencies enter data into web-based application

Data stored remotely on file server located at vendor site in Louisiana

136 partnering agencies enter data on-line; other 91 agencies send hardcopy forms for entry into system HCDCH collects hardcopy forms from partnering agencies and enters data into system

Under new web-based system, agencies will enter directly into system

Software Used Existing system  legacy software

New system  Visual Basic (with ORACLE platform)

Service Point (proprietary software developed by Bowman Internet Systems) Service Point (proprietary software developed by Bowman Internet Systems) MAACLink (software application developed in SQL) Current system uses DOS-based dinosaur; new system will use MS SQL type system, with Internet explorer interface
Hardware Used Existing system  Mainframe

New system  PC-based

PC-based system (partners connected through Internet) PC-based system (partners connected through Internet) PC-based system (connect via the Internet) PC-based system
Internet Access No (Shelters connect to the existing and new systems via T1 Lines) Yes (web-based application) Yes (web-based application); no software is required on computer at the remote sites where data is entered Yes (not web-based application, but connect to system via Internet) Currently not web-based; but new system will be web-based
Uses of the Data System
  • Analyze characteristics of individuals served
  • Track utilization and length of stay (days in facility, etc.)
  • Analyze readmissions
  • Monitor shelter capacity and performance
  • Assignment of individuals/families to appropriate vacant shelters/units
  • Validate invoices submitted by shelters
  • Research purposes
  • Coordinate services and streamline referrals among partnering agencies
  • Reduce duplicative client intakes and assessments
  • Partners can easily generate HUD Annual Performance Report
  • State can generate unduplicated count of homeless and analyze scope of homeless problem in WI
  • Partners/state can analyze participant characteristics, needs, services received, and some outcomes
  • MIS created standardized data across 28 partnering agencies
  • Ready availability of data for reporting and analysis purposes
  • Ability to report quickly and accurately over any facet of the data collect
  • For partners  system is easy to use; enables shelters to keep running tabs individuals entering shelters; creates ability for partners to analyze successes/failures; partners can report easily to other agencies/funders
  • MIS automatically does utility accounting  agencies can view expenditures and remaining budget
  • MIS determines if household meets eligibility guidelines for utility vouchers
  • MIS also indicates if someone is likely eligible for food stamps, Energy Assistance, TANF
  • Agencies can track services provided by other agencies  so MIS eliminates duplication of services (e.g., utility payments)
  • Helps partners report to funders and seek new funding
  • Track homeless population and characteristics
  • Analyze situation at exit (e.g., destination, reason for exit, income sources)
System Costs
  • Existing system: most costs associated data entry by shelter staff; additional $2,000/year contractor costs for system maintenance
  • New system  no cost estimates available
  • Annual costs estimated in range of $200K - $250K
  • Partners pay one-time fee ranging from $500-$3500, then annual support fee equal to 20% of initial licensing fee
  • $19K year (annual fee, plus hosting services, disaster recovery, troubleshooting/TA)
  • Consultant charges $85 per/hour (e.g., help creating new reports)
  • $200K per year (for staff and equipment)
  • Past year  also expended 200K for system improvements and upgrades
  • ~$100K (covering 2 FTEs and equipment and overhead costs)
  • No data available on estimated costs of new system, but development cost estimated at $20-25K
Implementation Challenges
  • Existing system contains most information needed, but DHS worried that given its age that MIS may crash and be difficult/expensive to recover data and repair the system.
  • Ongoing maintenance is becoming more difficult because few computer firms service the hardware or software
  • While new system is based on existing system, the new system has required substantial programming time and testing.
  • Takes longer than anticipated to get HADS up and running
  • Training is huge issue  continually training needed because of staff turnover; manual, regular training workshops, and practice data bases necessary
  • Partners often lack technological capacity and know-how
  • Issues associated with geography/scaling  partners scattered throughout the state
  • Federal reporting requirements vary at federal level  HHS and HUD need to agree on what data should be maintained
  • System turned out to be more technologically advanced than some partnering sites had a capability for  TA was needed
  • Quality controlling data somewhat of a problem (e.g., duplicate records and incorrect entry of data)
  • Standard pre-formatted reports are inadequate (though it is possible to export data to Excel or ACCESS for additional analysis)
  • Initially, some problems with domestic violence agency coming on-line, but system changed so that agency and participant information can be hidden from other users
  • Some emergency shelter directors appeared to be afraid of IT or did not want to collect information
  • Some partners were underfunded, so lacked resources to cover costs of phone lines and internet connection needed for MIS
  • Some issues emerged around sharing of data between agencies
  • Partnering agencies never used limited reporting capabilities in old system  may not have been aware that they could export data for analysis purposes
  • Some difficulties associated with getting partnering agencies to agree on standard form.
Other Comments
  • Existing system is on its last legs.
  • New system builds off of old system (containing all data items of old system), but adds some new data elements and will add new features in the future
  • The number of partners is expected to grow to about 225 by the Summer 2003. Some potential new partners include food banks, clothing pantries, and supportive service agencies
  • System designed to collect data on service needs and services received; also collects outcome data at the time of exit  but it is often difficult to collect exit information because people suddenly stop coming.
  • Each partner has freedom to develop own forms [currently only portion  10 percent  of the system variables are being utilized by agencies]
  • No problems with confidentiality  can restrict access to individuals data on any data item
  • System has 40 pre-formatted reports and users can customize own reports
  • System contains client- level data collected at exit that offers potential for pre/post comparison of outcomes, including income sources and amounts, length of stay, housing situation at exit, and reason for termination.

HADS tend to be system-wide  some cutting across a large number of partners  which avoids focusing narrowly on programs (e.g., ?silos). Several of the HADS are very large in terms of the number of partnering organizations and programs that are linked via the systems. The largest in terms of number of partnering agencies  that is, agencies providing data for entry into the HADS  is the MAACLink system in Kansas City. A total of 227 partnering agencies from a five-county area surrounding Kansas City contribute data to the MAACLink system (either on-line or by submitting hardcopy forms for entry into the system by MAAC). Partners include some homeless-serving agencies, but also other agencies serving low-income individuals or families in the Kansas City area. Nearly 60 percent of the MAACLink partnering agencies (135 agencies) are connected on-line to the HADS. The ServicePoint systems used in Madison and Columbus also have a large and diversified pool of partnering agencies. The system maintained in Madison has 84 partnering agencies from across Wisconsin contributing data. Partners include a broad range of agencies that target services on homeless individuals and others at-risk of homelessness  including emergency shelters, transitional/supportive housing agencies, local housing authorities, domestic violence service providers, faith-based organizations, and tribal agencies. Program officials (at the Bureau of Housing in Madison) anticipate that the number of partnering agencies will grow to about 225 by the Summer 2004. Some potential new partners include food banks, clothing pantries, and other agencies that provide support services needed by homeless and other low-income households. The ServicePoint system in Columbus brings together 28 partners, but partnering agencies are more narrowly focused (than in Madison or Kansas City) to include only homeless-serving agencies (such as emergency shelters, homeless prevention programs, resource centers, and housing search assistance agencies). The New York City system is focused exclusively on collecting data on homeless individuals and families served within emergency and transitional housing funded by the NYC Department of Homeless Services. The only other partner providing data for the HADS is the NYC Human Resources Administration, which provides data (merged into the HADS) to indicate whether homeless individuals/families included in the HADS are also public assistance recipients. The Housing and Community Development Corporation in Hawaii partners on the HADS with agencies it directly funds to provide street outreach, emergency shelter, and transitional shelter for homeless individuals and families.

Some HADS have accumulated substantial numbers of records on homeless and other types of disadvantaged/low-income households. These systems demonstrate that it is possible to collect and share data across a broad range of program. The NYC HADS is one of the largest (if not the largest) in the country  having accumulated an estimated 800,000 to 900,000 records on homeless individuals served by emergency and transitional housing facilities in the New York City since the inception of the system in 1986. The system includes only homeless individuals and families served in NYCs shelter system. The system creates a single case record for each individual, which displays all episodes of receipt of housing assistance through emergency or transitional facilities over the last 16 years (though there are some duplicate records because people use aliases or fail to provide accurate identifying information). On an average day in June 2002, there were 7,903 families in temporary housing and 7,557 single adults in shelters in New York City (a total of over 30,000 individuals in homeless facilities)(12)  all of which are entered into the data system. The MAACLink maintained in Kansas City also has a very large number of records in its system  an estimated 450,000 individuals (an estimated 112,000 new records were entered into the system in 2001). However, the 227 partnering agencies (many of which provide services for a wide range of low-income and disadvantaged individuals) can enter data into the system on anyone using their services  and hence, only a relatively small percentage of those entered into the system are or have been homeless.

The other HADS are much smaller in relative terms (when compared to Kansas City and NYC), but still contain significant numbers of records  about 40,000 individuals in the Madison HADS (the vast majority of which are either homeless or individuals at-risk of homelessness); about 54,000 individuals in the Columbus system (which includes mostly homeless or those at-risk of homelessness); and about 100,000 individuals in the Hawaii system (which is limited to homeless individuals in emergency or transitional shelters or contacted as a result of street outreach efforts; however, new records on individuals served each year are created, and so, there are many individuals with duplicate records from year to year).

HADS systems are not used principally for measuring program performance or outcomes  though have the capability to provide analyses of length of stay. The HADS principally serve as registry systems that facilitate tracking of program participant characteristics, services received, length of stay, and movement within emergency and transitional housing facilities. The systems can also be useful in avoiding duplication of services, reducing fraud and abuse, and facilitating payment to vendors. Exhibit 3-1 (shown earlier) provides a general overview of the key data elements collected in each of the five systems; Appendix C contains additional documentation on data elements from several of the sites that provided hardcopy forms and additional background on their data systems. All of the HADS in our survey collect client identifiers (e.g., name, Social Security Number, and address, if available), as well as a basic set of demographic characteristics. Among the core of basic demographic features being collected in most sites are gender, age, race/ethnicity, marital status, and veteran status. HADS vary in terms of other types of client characteristics data collected. Some example of other types of background information collected on the individual include educational attainment, income and income sources, employment status, living arrangement, household size, health status, substance abuse problems, mental health problems, and other special needs. Several of the systems (New York, Madison, and Columbus) collect information about reasons for homelessness. The ServicePoint system used in Madison collects what appears to be the most information concerning the individuals housing/homeless situation prior to entry into the program  including current living situation, homeless status, reasons for homelessness, date the individual became homeless, whether this episode is the first time the individual has been homeless, and reason the individual left his/her prior living situation.

The HADS in each of the five localities track some type of service data and length of stay in shelter facilities  but the types tracked and the extent to which data are analyzed varied considerably across sites. The New York City HADS collects data on the referral date, the name of the emergency or transitional facility to which the individual/family is referred, the room number, the date of entry and exit from the shelter facility, and total days housed within the facility. The New York City HADS provides the Department of Homeless Services with the data needed to validate invoices submitted by shelters for payment for specific days of shelter use for each individual/family. In addition, the system in New York City enables the Department to analyze characteristics of individuals served, track individuals into and out of shelter facilities, and monitor shelter capacity and facilitate placement of individuals/families into appropriate vacant units. Analyses by Kuhn and Culhane(13) of data from New Yorks HADS illustrate the types of outcome analyses that are possible with HADS data and some of the limitations to use of such data. For example, Kuhn and Culhane were able to analyze length of stay for users of homeless shelters for over 70,000 homeless individuals between 1988 and 1995. Using available data, the researchers identified three distinct groups of users  transitionally homeless (81 percent), episodically homeless (9 percent), and chronically homeless (10 percent).(14) For the overall population and each of these three groups, the researchers analyzed: average number of episodes of homelessness, average number of days of homelessness, average days per episode, and total and percentage of client days in shelter. The background characteristics collected at the time each individual entered the NYC shelter system enabled Kuhn and Culhane to analyze HADS data overall and for each of these three groups across the following characteristics of shelter users: age, race/ethnicity, gender, and self-reported disabilities (limited to mental illness, medical problems, and substance abuse problems). The researchers concluded that The chronically homeless, who account for 10 percent of the shelter users, tend to be older, non-white, and to have higher levels of mental health, substance abuse, and medical problems.  Despite their relatively small numbers, the chronically homeless consume half of the total shelter days. The authors note that their study is limited by its reliance upon administrative data for recording periods of homelessness and for measuring characteristics of shelter users. For example, they point out data in the HADS in New York City on mental health, medical, and substance abuse problems are self-reported (and hence, may lack reliability) and that periods of street homelessness are not captured. In addition, the number of background variables collected on each individual is limited to just a few demographic variables, and outcome measurement is limited to analysis of length of stay and whether there are multiple episodes/readmissions to the shelter system (rather than, for example, whether an individual secures and keeps permanent housing, is able to find and keep a well-paying job, is able to achieve additional education qualifications, and is able to overcome substance abuse problems).

The Shelter Stipend Database in Hawaii, which was being substantially revamped at the time of our interview in July 2002 (but was expected to be operational by late 2002/early 2003), offers excellent opportunities for outcome analyses. This is because the system employs an exit form (see Appendix C for a copy of this form and others used in Hawaii), which captures data on a number of important outcomes (some of which may permit pre/post comparisons). Specific analyses that should be possible using the exit information to be collected as part of the upgraded system in Hawaii include the following:

  • length of stay in the shelter facility (i.e., days between the date of entry into program to date of exit from the program);
  • number of individuals within the family who left and remained at the shelter at the time the household head left the shelter;
  • destination to which the individual/family was going at the time of exit, including: permanent housing (such as rental housing, public housing, a Section 8 unit, or homeownership), moved in with family, transitional housing, emergency shelter, drug treatment, unsheltered situation (e.g., street, park), hospice/care home, medical or psychiatric hospital, prison/jail, or destination unknown;
  • reason for exit, including transitioned successfully, exited voluntarily, non-renewal of lease, left before completing the program, reached maximum time allowed in program, evicted, completed program services, needs could not be met by the program, disagreement with rules/person leading to termination from facility, arrested/left for prison, left for hospital, deceased, or unknown/disappeared;
  • resources used at exit, including: public housing, Section 8, grant, loan, clients savings, financial support from friends/family, Hawaiian Homelands funding, no resources, and unknown;
  • geographic location at the time of exit, including remained in Hawaii (specific island identified), left for mainland, and unknown;
  • monthly household income by major source at the time of exit (i.e., sources include work, TANF, SSI, SSDI, retirement/pension, child support, workers compensation, unemployment benefits, Medicaid/Medicare, food stamps, financial help from family/friends, other, and unknown); and
  • support services received during the time in the project, including outreach, case management, life skills, alcohol or drug abuse services, mental health services, HIV/AIDS-related services, other health care services, education, housing placement, employment assistance, child care transportation, legal, other, and unknown.

Using the data collected in Hawaii, for example, it should be possible to make some comparisons between conditions of the household at the time of entry with conditions at the time of exit, for example, in terms of total household income and income sources, and living situation just prior to entering the shelter system with the destination to which the household was going at the time of exit.

The Kansas City system enables partnering agencies to keep track of the services being provided by other agencies  and so helps to eliminate duplication of services. The system was also designed for a very specific purpose  to enable partnering agencies to automatically process emergency utility vouchers. The system is programmed so that each partner can automatically determine if an individual meets eligibility guidelines for utility vouchers and accounts for issuance of each voucher by the partnering agency. The system also enables partnering agencies to calculate family budget, which also enables the agency to assess whether an individual is likely eligible for other types of assistance, including food stamps, energy assistance, and TANF. The Service Point systems in Madison and Columbus provide the sponsoring agencies with a wide array of reports for analysis purposes. Specific reports are available to analyze the number of homeless individuals served, participant characteristics, service needs, types of shelter facilities used, other support services provided, and length of stay.

Range of implementation challenges reported  particularly with regard to training system users to make full use of system features. Exhibit 3-1 highlights a variety of problems that agencies have encountered both in establishing their HADS and ensuring that systems are used appropriately by partnering agencies responsible for providing much of the data entered into the systems. Some of the implementation issues reported by agency officials we interviewed include the following:

  • Developing a new system requires substantial staff effort in terms of programming and pilot testing the new application (New York).
  • Training staff on how to appropriately and effectively use the system can be a huge issue, and because of staff turnover, there is a need for ongoing training. Partnering agencies often lack the technological capacity and know-how to operate systems without substantial training (Madison and Columbus).
  • Federal reporting requirements vary substantially from agency to agency (particularly between DHHS and HUD), which can complicate ways in which data elements and reports are structured within data systems (Madison).
  • Sharing of sensitive data across partnering agencies can be complicated and may require special programming so that access to such data can be limited to only certain partners and staff within agencies  for example, agencies serving individuals with domestic violence issues can be reluctant to share data that is available to other agencies (Kansas City and Hawaii).
  • Quality controlling data can be an issue when data are being collected for the same individual by different agencies (Columbus).
  • There can be problems in convincing partnering agencies to utilize standard system forms (Hawaii).
  • Standard report formats may be inadequate for the specific reporting needs or information requirements of partnering agencies (Hawaii).

View full report

Preview
Download

"report.pdf" (pdf, 1.01Mb)

Note: Documents in PDF format require the Adobe Acrobat Reader®. If you experience problems with PDF documents, please download the latest version of the Reader®