Reading Helping America's Homeless brings to mind the old story about the sausage factory: The factory has performed adequately for several years, but when increased production taxes the machinery, sausages often come out malformed. The owner wonders whether the machine is flawed but doesn't want to buy a new one. Eventually he comes to believe that the problem stems from the sausages themselves and hires workers to reshape each sausage. In the long run, of course, it would have been more cost-effective and efficient to replace the machine.

Although homelessness is a condition of circumstance rather than of character, we tend to attribute it to perceived shared characteristics among homeless people. But Burt and her colleagues use structured, thoughtful research to show that the opposite is true. In fact, the only common denominator in homelessness is poverty.

The 1996 National Survey of Homelessness Assistance Providers and Clients (1) serves as the basis for Helping America's Homeless. That study was wide ranging and meticulous, enabling both a bird's-eye and an up-close view of homelessness. Burt and her colleagues describe the methodology in detail, with candid explanations of possible inaccuracies.

The authors find that services for homeless persons have increased across the United States. However, they also clearly demonstrate that we have not solved—or even diminished—the homelessness problem. Although we have the cure, we have directed our resources toward treating the symptoms. The authors point out that because cross-sectional studies often fail to capture people who are temporarily homeless, they falsely inflate the proportion of chronically homeless people within the homeless population. The authors suggest that policy design should differentiate crisis situations from chronic homelessness. Sometimes a "quick fix," such as emergency financial assistance, really is the answer for a particular individual or family. This approach is more rational than providing services only after the individual or family becomes homeless. For many homeless people, the answer is simply affordable housing. For the chronically homeless, the evidence argues for a "continuum of care" built around affordable housing. Finally, prevention is essential to eradicate the risk of homelessness.

What does all this mean for community mental health professionals? According to the authors of Helping America's Homeless, having an alcohol, drug, or mental health problem greatly increases the risk of homelessness. Homeless people with dual diagnoses seem to fare the worst. Systems of care for the homeless therefore must include treatment services. Childhood abuse, neglect, and placement in foster care also increase the risk of homelessness. Thus at-risk children should have access to early mental health services as well as life-skills education. Perhaps most important, community mental health professionals can use their expertise to inform policy decisions about housing and mental health services for at-risk populations.

Helping America's Homeless is an excellent book for policy makers, resource developers, and public program administrators. Social service professionals who lack a strong research background may find little use for the book's detailed descriptions of methodology. However, each chapter ends with a clear summary of the research and its implications.

The book concludes on a fairly pessimistic note. In asserting the necessity of increasing affordable housing and individual earning power, the authors say "we have not done so, and probably will not." Whether this assessment is realistic remains to be seen. Paradoxically, however, the more this book is circulated and its cynicism understood, the less likely it is that the pessimistic predictions of the authors will come true.

Dr. Bradford is resident physician in the department of psychiatry of University of North Carolina Hospitals in Chapel Hill.