0990-Homelessness Data for HHS Mainstream Programs 8-10-07

Homelessness Data for HHS Mainstream Programs

Supporting Statement for Paperwork Reduction Act Submission

Contract No. 233-02-0088

Task Order: HHSP233200600004T

February 9, 2007

Revised August 10, 2007

Prepared for

Flavio Menasce

DHHS/OS/ASPE

Office of Human Services Policy

Humphrey Building, Room 405F

200 Independence Avenue, SW

Washington, DC 20201

Prepared by

Abt Associates Inc.

4550 Montgomery Avenue
Suite 800N

Bethesda, MD 20814-3343

Table of Contents

Part A: Justification.....................................................................	............................................	2
Background.......................................................................	............................................	2
A.1 Need and Legal Basis............................................	............................................	2
A.2 Information Users..................................................	............................................	3
A.3 Improved Information Technology.......................	............................................	4
A.4 Duplication of Similar Information.......................	............................................	4
A.5 Small Businesses...................................................	............................................	5
A.6 Less Frequent Collection.......................................	............................................	5
A.7 Special Circumstances...........................................	............................................	5
A.8 Federal Register Notice/Outside Consultation......	............................................	5
A.9 Payment/Gift to Respondents................................	............................................	6
A.10 Confidentiality.......................................................	............................................	6
A.11 Sensitive Questions................................................	............................................	6
A.12 Burden Estimate (Total Hours and Wages)...........	............................................	6
A.13 Capital Costs (Maintenance of Capital Costs).......	............................................	8
A.14 Costs to Federal Government................................	............................................	8
A.15 Program or Burden Changes..................................	............................................	8
A.16 Publication and Tabulation Dates..........................	............................................	8
A.17 Expiration Date......................................................	............................................	9
A.18 Certification Statement..........................................	............................................	9
Part B: Collection of Information Employing Statistical Methods Justification................................................	............................................	9
Identification of Appropriate Respondents......................	............................................	10
Administration of the Survey...........................................	............................................	10
Maximizing the Response Rate........................................	............................................	11

Part A: Justification

Background

This request is for clearance for survey data collection sponsored by the U.S. Department of Health and Human Service (HHS) Office of the Assistant Secretary for Planning and Evaluation (ASPE). The proposed data collection will involve telephone interviews with state officials who administer the TANF and Medicaid programs. The telephone surveys will collect information about the type and quality of data related to homelessness and housing status that are collected from and recorded about TANF and Medicaid applicants.

The proposed data collection will obtain information from state TANF and Medicaid officials in all 50 states and the District of Columbia to determine whether these two largest HHS mainstream programs for low income individuals are collecting information from program applicants and/or participants regarding their housing status. Prior to conducting the telephone interviews, Abt Associates, the research contractor conducting the study, obtained copies of TANF and Medicaid applications for the 50 states and the District of Columbia and reviewed the applications to determine the extent to which items related to homelessness or housing status are included in the applications. The survey data collection for which review is requested here will involve telephone interviews to learn more about the data collected on TANF and Medicaid programs.

In the preliminary review of applications, we found that the questions included on TANF and/or Medicaid applications that related to homelessness or risk factors associated with homelessness are:

• Are you homeless?

• Are you currently living in a shelter or on the streets?

• Are you living in a domestic violence shelter?

• Do you have an eviction notice for your current residence?

• Are you living with friends or relatives?

In the survey, more detailed information will be collected about how these items are collected and maintained, definitions used for homelessness, and how the information is used by the state. For programs that do not include any of these items on the applications for TANF or Medicaid, questions will focus on reasons why such items are not collected, barriers to collecting the information, and assessments of the likelihood that such information will be added to application forms in the future.

A.1 Need and Legal Basis

The study promises to fill a gap in knowledge about whether information on homelessness is available from TANF and Medicaid program records. It will help HHS determine whether information is currently being collected by the states that could be used to benchmark the incidence of homeless persons using the Department’s mainstream benefit programs.

The Department’s efforts to expand mainstream benefit coverage to eligible homeless persons are significant and have the potential to substantially advance the Administration’s goal to end chronic homelessness and to resolve homelessness for many other households that experience it. Many states have adopted similar goals to improve access to mainstream benefits for persons who are homeless, due in part to HHS’s support of State Policy Academies and to Departmental guidance requesting states to reduce barriers to mainstream enrollment for persons who are homeless and sharing strategies on methods to do so.

The results of the study will be reported to HHS in tables that will summarize the extent of housing status or homelessness data collection for each state. Narrative will be used to interpret the data and help the Secretary consider the viability of using these existing administrative data for the Department’s performance objectives related to homelessness.

It is important to note that the information we are seeking in the telephone interviews will be used to expand on what has been learned from reviewing the application forms, including plans to begin collecting homelessness data for states not currently doing so, and definitions of homelessness used by states that currently ask applicants if they are homeless at the time of application for TANF or Medicaid. For states that include questions regarding homelessness on their TANF and/or Medicaid applications, the survey will also include questions about how these data are recorded in automated systems, data quality and consistency. All of the information we are seeking is based on informants’ current knowledge and will not pose an additional data collection burden beyond the time of the interview itself. This study will not require that any states develop new reporting procedures or collect new information about TANF or Medicaid applicants.

Data collection will occur with officials from up to 102 TANF and Medicaid administering entities, representing each state and the District of Columbia. The data collection instruments are shown in Appendix A. Two versions of the instrument have been developed for each program. One includes questions to be asked in cases where no information is collected on the application pertaining to homelessness or risk factors for homelessness, and the other will be used in cases where such items are included on the application.

The authorizing legislation for this data collection is Section 301 of the Public Health Service Act (42 U.S.C., Sec. 301).

A.2 Information Users

The information collected in this study will be used by the Office of the Assistant Secretary for Planning and Evaluation in the U.S. Department of Health and Human Services to help HHS officials determine the extent to which information is being collected by state TANF and Medicaid programs that could be used to benchmark the incidence of homeless persons using the Department’s mainstream benefit programs.

There are several other potential uses for this information. This study can help HHS develop an evaluation strategy that will answer several important research questions. First, if viable sources of information on housing status are found at the state level, Medicaid, TANF, and other HHS mainstream program data can be used to document states’ progress in reaching and enrolling persons who are homeless, which will help HHS track progress in achieving this goal at the national level. Second, these administrative data can be used to identify study populations for future research that would address more detailed, significant questions related to the impact of mainstream benefits on persons who are homeless. For instance, HHS could explore whether access to mainstream benefits helps to resolve a household’s homelessness; or conversely, HHS could examine the characteristics of people who receive benefits and remain homeless to determine what additional strategies might be effective in resolving their homelessness.

Since many states have also established goals to improve access to mainstream benefits and to address homelessness, this study will also help them determine whether their data are sufficient to track state-level progress and may raise awareness among state officials in states that do not currently collect housing status data. These states may add or change data collected on housing status as a result of this study. It is also possible that if HHS makes the information collected in this study available to all states, access to information about practices in other states might be of use to state program administrators who want to begin collecting information on homelessness or modify current practices. The long-term indirect benefits of this research are therefore likely to be substantial.

A.3 Improved Information Technology

Improved information technology has been used in this evaluation to the extent possible. To collect the initial information on TANF and Medicaid application data collection, we attempted to obtain as many of the state application forms as possible from state websites (in some cases this was not possible and we contacted state officials who provided a copy via fax or mail). The research contractor entered information about each state TANF and Medicaid application into the study database and all subsequent data collection will build on the application database. The application database includes information about contact persons in the state who provided the application, items included on the application regarding housing status, homelessness, and risk factors, and the date when the application was last updated by the state. The application database will be used to manage the survey data collection effort. It will be used to track responses and also to fill in the information already known about the content of the application, thereby reducing burden during the telephone interviews (since information already gathered during the application review will not need to be asked again of survey respondents).

Conducting the telephone interviews will also benefit from information technology. A Microsoft Access database will be used to store survey responses, and the interviewers will enter responses directly into the survey database at the time of the interview. This will allow for confirmation that responses conform to established response codes and that skip patterns are followed correctly. Given that the interviews are relatively short, the number of respondents is small, and the survey instrument includes several open-ended questions requiring probes, we will not be using Computer Assisted Telephone Interviewing for this data collection effort.

A.4 Duplication of Similar Information

The purpose of the HHS Homeless Data Collection Study is to assess whether existing data sources, the administrative data collected during the application process for TANF and Medicaid benefits, can be used to address the federal objective to understand the intersection of homelessness and mainstream benefit programs. In the absence of this research, HHS could inadvertently require state agencies to establish a separate, duplicative process to collect and report data on the extent of homelessness within their states. Thus, we are seeking to minimize duplication by determining whether data that are already being collected and maintained by the state agencies is a viable option to meet HHS’s objective. A telephone survey of TANF and Medicaid officials is necessary because centralized information on this topic is not currently available.

A.5 Small Businesses

Not applicable. This HHS-sponsored collection of information will not affect small entities, because the respondents are all state agencies that administer TANF and Medicaid programs.

A.6 Less Frequent Collection

The proposed data collection activity is to be conducted only once for each respondent.

A.7 Special Circumstances

The proposed data collection activities are consistent with the guidelines set forth in 5 CFR 1320.6 (Controlling Paperwork Burden on the PublicGeneral Information Collection Guidelines). There are no circumstances that require deviation from these guidelines.

A.8 Federal Register Notice/Outside Consultation

In accordance with the Paperwork Reduction Act of 1995, the Office of the Assistant Secretary for Planning and Evaluation in HHS published a 60-day notice in the Federal Register on May 11, 2007. The docket number was FR- Vol.72, No. 91 and the document identifier is OS-0990-0000. The Federal Register notice appeared on page 26818. The authority is Section 3506 of the Paperwork Reduction Act of 1995, 44 U.S.C, Chapter 35, as amended. The notice provided a 60-day period for public comments, and comments were due by July 11, 2007. As of that time no comments were received. A copy of the notice is shown in Appendix B.

The design of this study and the data collection procedures were developed and are being carried out with the assistance of Abt Associates Inc. Key members of the Abt team include Dr. Howard Rolston, Ms. Michelle Wood, and Ms. Brooke Spellman.

HHS-ASPE has also collaborated on the design of the study with staff from the Health Resources and Services Administration throughout all phases of the study to date. The purpose of such consultation is to ensure the technical soundness and usefulness of the data collection instruments in carrying out the aims of the evaluation.

A.9 Payment/Gift to Respondents

Not Applicable. There are no plans to provide payments to the respondents from the state agencies for providing information on their TANF or Medicaid application data collection practices.

A.10 Confidentiality

Not applicable. Since this data collection effort will ask administrators of TANF and Medicaid programs to describe the data elements included in their application forms and application data collection practices, the study team will not provide assurances of confidentiality. The information collected in these interviews will be used by ASPE to understand better the extent to which data on homelessness and housing status are now being collected by state TANF and Medicaid programs. To be useful to ASPE it will be necessary to identify responses for each state. We do not anticipate that this will present any complications since the nature of the information being collected is not sensitive and pertains only to state practices for administering TANF and Medicaid applications.

A.11 Sensitive Questions

Not Applicable. This study will be collecting general information about states’ TANF and Medicaid administrative data collection and reporting practices and will not be collecting personal information from individuals. None of the information being requested from the respondents deals with topics of a sensitive nature, such as sexual behavior and attitudes, religious beliefs, drug or alcohol use, or other matters that are commonly considered private.

A.12 Burden Estimate (Total Hours and Wages)

The collection of information about state-level administrative data collection practices is a one-time effort to be carried out in the fall of 2007.

Exhibits 12-A and 12-B show the estimated respondent burden for the data collection associated with the HHS Homeless Data Collection Study. The total estimated burden is 102 hours, based on an estimate of one hour for each of the 102 entities that will be interviewed. Since the survey instrument is the same for all state agencies, with the only variation based on whether homelessness (or risk factors for homelessness) items are currently collected on the application, we have assumed that the burden for responding to the data request will be the same for each respondent. However, since the data collection practices vary from state to state, we anticipate that some interviews may take less than the allotted time. In fact, interviews for state programs that do not collect any homelessness data are expected to require even less time, perhaps as little as 30 minutes. On the other hand, in some cases it may be necessary to talk to more than one person to obtain responses to all items on the survey instrument. Even in such cases, however, the total time needed to complete all survey items is not expected to exceed one hour per state program.

As allowed under OMB regulations, we tested the survey instrument with 8 respondents. These pretests confirmed our burden estimates.

Exhibit 12-A

Estimated Annualized Burden Hours for the Homelessness Data in HHS Mainstream Programs Research

Data Collection Form Name	Type of Respondent	Number of Respondents	No. of responses per respondent	Average Burden per response (hours)	Total Burden Hours
Provide information on state TANF application data content and application data collection practices	State official responsible for administering the TANF program	51 (each state and DC)	1 per respondent	1	51
Provide information on state Medicaid application data content and application data collection practices	State official responsible for administering the Medicaid program	51 (each state and DC)	1 per respondent	1	51
Total					102

Exhibit 12-B

Estimated Annualized Cost to Respondents for the Homelessness Data in HHS Mainstream Programs Research

Type of Respondent	Number of Respondents	Hourly Wage Rate	Total Respondent Costs
State official responsible for administering the TANF program	51	$27.80	$1,417.80
State official responsible for administering the Medicaid program	51	$27.80	$1,417.80
Total			$2,835.60

We have used the figure of $27.80 per hour based on data available from the U.S. Department of Labor Bureau of Labor Statistics from the 2000 National Compensation Survey for the average hourly wage rate for Administrators and officials, public administration (available at www.bls.gov).

A.13 Capital Costs (Maintenance of Capital Costs)

Because this is an exploratory study to understand the viability of using data that are already being collected by states for federal purposes, this data collection activity does not pose any additional recordkeeping and reporting cost to respondents.

A.14 Costs to Federal Government

The estimated cost to the federal government of the planned data collection activities for the Homelessness Data in HHS Mainstream Programs study is $63,677. This estimate represents a subtotal of the total evaluation costs of $190,930. Total evaluation costs include costs associated with research design, application review, data analysis, reporting, and briefing ASPE staff on the results of the interviews.

A.15 Program or Burden Changes

This request for clearance does not involve a change in burden due to any program changes or adjustments. It concerns a new data collection not previously submitted to OMB for review.

A.16 Publication and Tabulation Dates

The data collected for the HHS Homeless Data Collection Study will be analyzed, tabulated, and reported to HHS by the research contractor, Abt Associates Inc. Data will be used to produce information summarizing the type of housing and homelessness status information collected and recorded by each state, as well as narrative conclusions about how HHS may be able to use these data to fulfill its objectives.

The duration of survey activities will span 42 weeks. The timetable for key activities for the survey is as follows (for the purposes of this table, the point at which we receive clearance from OMB is considered the first week):

Week 1	Receive OMB clearance
Week 2	Begin data collection
Week 18	End data collection
Week 19	Begin data analysis
Week 24	End data analysis
Week 42	Publish final report

The survey data will be combined with relevant information from the application and from other sources (e.g., data on the state caseload or dollar volume of program) and presented to ASPE in a findings memorandum. These data will be analyzed to provide answers to questions on the presence and type of homeless data collected, the quality of the data, the use of the data, incentives needed to get more states to collect such data, and the accessibility of aggregate reports or data on the number of homeless persons accessing TANF and Medicaid services. The memorandum will highlight the major findings of the analysis. After discussing the findings memorandum with ASPE staff and any policy implications of the study findings, Abt Associates will prepare a final report for the study in May 2008. A preliminary outline for the final report is shown in Exhibit 16-A.

Exhibit 16-A

Preliminary Outline of Final Report

Executive Summary
Chapter 1	Introduction: Short overview of objectives and methodology of study
Chapter 2	What data on homelessness are collected in TANF and Medicaid programs?
Chapter 3	What are homeless data used for?
Chapter 4	What incentives are needed to overcome barriers (including costs) to collecting homeless status data?
Chapter 5	Are the data available from states that collect it and what do we know about homeless persons using TANF and Medicaid from available data?
Chapter 6	Other HHS Mainstream Programs: What data are collected on homeless status and what are approaches for collecting this information?
Chapter 7	Summary and Conclusion: Can HHS use available data to benchmark access to their mainstream programs by persons experiencing homelessness?
Appendices	Methodology, copy of survey instrument, and additional tables and charts

The final report will describe and interpret the findings of the study regarding access to mainstream services by homeless persons and collection of data to measure the use of mainstream services by homeless persons.

A.17 Expiration Date

Not applicable. All data collection documents created for the HHS Homeless Data Collection Study will display prominently the expiration date for OMB approval.

A.18 Certification Statement

This submission describing data collection requests no exceptions to the Certificate for Paperwork Reduction Act (5 CFR 1320.9).

Part B: Collection of Information Employing Statistical Methods Justification

Statistical methods will not be used to select respondents for this study. The universe of TANF and Medicaid programs will be included in this study and data are to be collected from all States and the District of Columbia. This section describes our plans to identify and locate appropriate respondents in each state and the District of Columbia; our plan for administering the survey; and a discussion of planned efforts to maximize the response rate to the survey.

Identification of Appropriate Respondents

The target respondents for the telephone interviews are state TANF and Medicaid directors in each state (102 respondents). However, in some cases it may be necessary to speak with more than one respondent to obtain responses to all of the questions. Prior to the interviews, we will send letters to the state directors to inform them about the study and to provide a list of topics that will be addressed in the interview. On the basis of that description we expect that some state directors may refer us to other staff that may be better able to provide the information needed.

We have made important progress in assembling information on state directors or other senior program staff and their contact information. During the application collection process in November and December 2006, Abt staff obtained contact information for directors and senior program staff on the Internet. We entered information about contact persons, email addresses, phone numbers, and mailing addresses for each state program into an application database. While we anticipate that there may be some changes in staffing by the time the survey is administered in fall 2007, the information collected in December 2006 should be reliable for the most part and will give a good starting place for contacting survey respondents. During introductory phone calls to schedule interviews, we will verify that the contact information previously obtained is still accurate, and update the database as needed.

During the application collection and review process, we also found that many states use combined applications that cover several benefits programs. In states that use these types of applications covering both Medicaid and TANF, it may be possible to collect all of the desired information from one respondent. But to confirm this, we will contact directors of both programs in each state to make sure the most appropriate and knowledgeable respondents are identified so that all needed information is collected.

Administration of the Survey

As mentioned above, prior to conducting the interviews, we plan to send two letters to state TANF and Medicaid directors: one signed by relevant ASPE officials sent 2-3 few weeks before the survey begins, and a second sent from Abt Associates 3-4 days before we begin contacting respondents. The ASPE letter will introduce the study to respondents and underscore its importance to ASPE and HHS. The second letter from Abt Associates will alert them to expect a call shortly from an Abt interviewer, and will also include a list of topics that will be discussed during the survey to help them prepare for the interview or to identify alternative respondents as necessary. The letter from Abt will provide contact information for the Abt Project Director and will encourage state officials to call with any questions or concerns they have about the interview.

Three Abt staff will be trained to conduct the interviews. The half-day training session will focus on the purpose of the study, and the desired information for each question to ensure probes and follow-up questions are asked appropriately. The training will also include a review of the application database and instructions to review the applications carefully prior to conducting the interviews so that information already available from the applications can be confirmed. During training, the interviewers will also practice introductory scripts developed for the interviews and responding to questions about the study to ensure participation. As part of the training, the interviewers will do practice “mock” interviews with other study staff before initiating their first calls. Each interviewer will be responsible for interviewing both the TANF and Medicaid director in approximately one-third of the states. After each interviewer completes his/her first two interviews, the team will meet as a group to discuss the interviews, review any problems or questions that have arisen, and address any issues or concerns with the interview instrument.

After this initial debriefing, the Project Director will monitor the interviewing process by meeting at least weekly with each interviewer to review interviews that have been completed and to assist with any questions that arise, unusual situations, or difficulties encountered by any of the states in providing the needed information. Our expert advisors and will also be available as needed to assist with interpretation of program operations or other issues.

A survey database will be developed to manage the data collection effort. Information from the application database (state contact information and data elements contained in the applications) will be included in the survey database and used to track responses. The survey database will be developed using Microsoft Access and will be used to store survey responses. It will also facilitate recoding of open-ended responses and analysis of the data. Interviewers will be able to enter responses directly into the survey database while conducting the interview. However, since many of the survey questions are open-ended questions requiring detailed responses or probes and follow-up, interviewers will also be instructed to take detailed notes as needed when conducting the interviews and in some cases it may work better to note responses on paper copies of the interview instrument, returning afterward to enter responses into the database. The survey database will include fields for open-ended responses. Entering data into the database during or immediately after the interview will ensure that responses conform to established response codes and skip patterns are followed correctly.

Maximizing the Response Rate

As with all survey data collection efforts, achieving the highest possible response rate is paramount. This is especially true for this study, which seeks comprehensive information about the application data available for TANF and Medicaid applicants across all states. Even losing a small number of states from the data collection would compromise the value of the study in a substantial way. In developing plans for conducting the interviews we have therefore given special attention to procedures that will help achieve the highest possible response rate. The key steps to promote high response rates are the pre-survey letters, flexibility in scheduling the interviews, and aggressive follow-up steps in cases where it proves difficult to obtain agreement to conduct the interview or to collect the desired information.

State officials might be reluctant to participate in the interview if they think it will be burdensome or inconvenient. The pre-survey letters should help to alleviate those concerns. The letter sent from ASPE to state directors prior to the start of data collection will be crucial to ensuring cooperation with the data collection because it will emphasize the importance of the study and of obtaining responses from all states. The letter sent from Abt will help to reinforce that message. The letter from Abt will also help to reassure state officials that the information sought is relatively straightforward for them to provide, by listing the topics to be discussed in the interview. The letter from Abt will also emphasize the study team’s flexibility in scheduling the interview. We will offer to conduct the interview at a convenient day and time for the respondent, including during the early morning or evening. Approximately 3-4 days after the letter from Abt is mailed, the interviewers will conduct introductory calls to answer any questions about the study and to schedule the interview.

Interviewers will make several attempts to contact the respondent via phone to schedule and conduct the interview. If there is no response after several attempts, we will make additional efforts to contact the person through e-mail, fax, or even a letter sent via FedEx. Our experience has shown that different people are responsive to different modes of contact. That is, some people may forget to respond to a phone call, but will respond to an e-mail. Overnight delivery of a one-page letter often gets the attention of a respondent when all else fails. If after trying all of these methods there are still states that do not respond or if any state refuses to participate in the interview, we will alert ASPE promptly and work together to develop other methods to gain cooperation. These might include identifying other respondents, contacting other officials in the state to attempt to reach the respondent, or scheduling conference calls with Abt and ASPE to discuss concerns about participating in the study.