In HUD’s 2007 report to Congress, Affordable Housing Needs 2005, the American Housing Survey (AHS) proxy used to estimate the worst case housing needs  of disabled non-elderly very-low-income renters without children was incomplete because it did not incorporate a new 2005 AHS question about disability income.  Moreover, the AHS proxy results were not compared to independent sources of better data on numbers of very-low-income renters with disabilities and increased to agree with these control totals, as had repeatedly been done in previous HUD Worst Case reports.  

This study uses data about households with severe rent burdens from the 2005 American Community Survey (ACS) to overcome these two weaknesses and develop more accurate estimates of worst case needs among households containing non-elderly adult renters with disabilities.  The ACS identifies disabilities through direct questions about six disabling conditions, and thus has better data on persons with disabilities than any AHS proxy could provide. Yet the ACS does not have all the data elements needed to measure worst case needs as well as the AHS does, so the estimates developed here are based on AHS relationships between severe rent burdens and worst case needs. Then, because two other national surveys have better questions about disabling conditions than does the ACS,  the estimates of worst case needs made from the ACS were adjusted to be consistent with control totals from those two other surveys.