4308.0: Tuesday, November 14, 2000 - 8:45 PM

Abstract #8353

Challenges in evaluating demonstration housing programs for multiply-diagnosed homeless people with HIV

Angela A. Aidala, PhD, Joyce Moon Howard, DrPH, Jo L. Sotheran, PhD, Jo C. Phelan, PhD, Fleur H. Lee, MPH, and Maria C. Caban, MA. SPNS Evaluation Technical Assistance Center, Joseph L. Mailman School of Public Health, Columbia University, 600 W. 168 St, New York, NY 10032, (212) 305-7023, aaa1@columbia.edu

INTRODUCTION: Very little housing has been created specifically for multiply-diagnosed homeless or unstably-housed people who also have HIV. This presentation will outline major challenges in attempting to identify and document needs and outcomes in measurable ways in order to improve access to scarce housing funding. METHODS: The experiences of the Mailman School of Public Health (Columbia University) Evaluation and Technical Assistance Center for the Special Projects of National Significance in developing both individual project-specific evaluation plans and, and a cross-site evaluation plans are used as illustrations. RESULTS: As a basic service and human need, housing presents both analytic and practical challenges for evaluation based on survey research methods. Major analytic challenges include: isolating intervention effects on multiply-diagnosed populations in low-threshold services, locating appropriate comparison groups, unmeasured differences between target populations and available samples, small but diverse samples, rapid population turnover, lack of pre-defined sampling frames, translating programmatic goals into measurable outcomes of individually-determined packages of services, and developing meaningful measures for these populations. Practical challenges include locating and following up a mobile and hard-to-reach population, developing new evaluation capability among service providers, and assuring data quality assurance while minimizing the burden of research on providers. CONCLUSIONS: Challenges stem from both the nature of the population studied and features of the housing-service programs. No single evaluation model is appropriate, but triangulation of sources and careful attention to basic sample and measurement can produce meaningful results.

Learning Objectives:

  1. 1. Recognize three common challenges in evaluating relationships between housing and health outcomes.
  2. 2. Describe two major difficulties in measurement of psychosocial outcomes for formerly homeless people.
  3. 3. Discuss the importance of identifying appropriate comparison groups for research design.

Keywords: Evaluation, Housing

Presenting author's disclosure statement:
Organization/institution whose products or services will be discussed: None
I do not have any significant financial interest/arrangement or affiliation with any organization/institution whose products or services are being discussed in this session.

The 128th Annual Meeting of APHA