Dying in the shadows: the challenge of providing health care for homeless people

The study of mortality among homeless women reported in this issue by Cheung and Hwang¹ (see page 1243) is a clarion call to our society and our health care community. The stunning 10-fold disparity in mortality rates between Toronto's homeless and housed women aged 18–44 is complemented by data from 7 other cities, which show that the risk of death among younger homeless women is 5–30 times higher than the risk among their housed counterparts. Previous studies by Hwang and others of homeless people in Boston and Toronto have reported overall mortality rates 3–5 times higher than those among the general public.^2,3 This smouldering public health crisis can no longer be ignored.

Homelessness is a prism that refracts the failures of society's key sectors, especially housing, welfare, education, health care and corrections. This complex social phenomenon thwarts simple definitions and resists easy solutions. The often-romanticized hobos and skid-row denizens of past lore have been joined by families with children, runaway and “throwaway” adolescents, struggling minimum-wage workers and fragile elderly people.

Only the economics seem straightforward. Housing is a scarce but highly valued commodity. Those least able to compete are doomed to fail; among them are people whose opportunity and choice are limited not only by abject poverty but also by chronic mental illness, substance abuse, physical and sexual violence, illiteracy, complex medical problems and advancing years. The complexity of need, the rich diversity and the growing numbers of homeless people have baffled researchers, bewildered policy-makers and exhausted the compassion of our society.

Homelessness magnifies poor health and exposes those huddled in crowded shelters to communicable diseases such as tuberculosis and influenza. It complicates the management of chronic illnesses such as diabetes and asthma, makes health care harder to access and presents vexing obstacles that exasperate health care providers and confound delivery systems. The burdens of mental illness and substance abuse are well documented in homeless populations. Chronic illness is also common: many homeless people have hypertension, diabetes, peripheral vascular disease, respiratory problems, and liver and renal disease. Skin diseases are extraordinarily common and can lead to costly hospital admissions because of cellulitis. Hypothermia and frostbite are dreaded hazards of life on the streets and have been shown to be risk factors for early death. Some conditions, such as diphtheria, pellagra and lice infestations resulting in endocarditis from Bartonella quintana,^4,5,6 hearken to earlier centuries. Those caring for this population must merge medicine with public health: tuberculosis and HIV/AIDS are endemic; outbreaks of communicable diseases such as influenza and infestations are ubiquitous in shelters; trauma and violence hound homeless poor people.⁷

Cheung and Hwang reviewed excess mortality in 7 cities in 4 countries: England, Canada, Denmark and the United States. A curious and troubling observation is the apparent lack of effect of health insurance on the risk of premature mortality. Shamefully, in the United States there are still over 40 million citizens without health insurance, whereas the 3 other countries have long had universal health insurance. Although universal coverage is desperately necessary, it does not appear to be sufficient to prevent premature death among homeless people. Fundamental change in the delivery of health care is essential to address the current disparities in health care for such vulnerable populations.

The medical care of homeless individuals and families poses a vexing challenge for our traditional health care delivery models. The relentless immediacy of the daily struggle for safe shelter and a warm meal relegates health needs to a distant priority. Common illnesses progress and injuries fester, leading to increased numbers of emergency department visits and acute care hospital admissions.

Treatment plans that make sense for those with homes and family support are often unworkable for homeless people: bedrest is impossible, simple dressing changes difficult, medications hard to obtain and store, and adherence to regimens requiring multiple daily dosing is daunting. For example, regular exercise and adequate control of diet, the mainstays of diabetes care, present formidable challenges to people living in shelters and subsisting on meals in soup kitchens. To further complicate the plight of homeless people with diabetes, the safe storage of medications is often difficult and the possession of syringes forbidden in many shelters.

The disparities in health outcomes for homeless people expose shortcomings in our current delivery systems. Learning from medicine's past may be important. Effective primary and preventive care for homeless people is possible if physicians are willing to venture out from traditional hospital and office settings to deliver care directly on the streets and in shelters. These “home” visits to homeless people overcome myriad obstacles and nurture the physician–patient relationship. Clinicians who use this approach have found it to be the sine qua non for continuity and quality in caring for homeless patients, but the time and mobility required are rarely funded or valued by our evolving delivery models. Community outreach, best done by physicians working or volunteering in multidisciplinary teams with other health care professionals, can be fully integrated with local emergency departments and hospitals to help reduce the frequent use of these costly services. Respite care programs, piloted in cities such as Boston, Chicago and Washington, have attempted to compensate for the lack of safe housing by providing 24-hour short-term medical care for people ready for discharge from hospitals or emergency departments but too ill and vulnerable to return to the streets. Over 30 cities in the United States and Canada are now piloting similar models to fill this gap in our continuum of care for homeless people.

Physicians willing to care for populations on the fringes of society need not be marginalized by their own profession. Academic medical centres, long a refuge for the urban poor, should heed these disparities in mortality and embrace the care of homeless and other vulnerable populations as vital to their missions and as a critical component of the teaching curriculum.

Caring for homeless people poses an uneasy ethical dilemma. As we work to prevent illness, alleviate symptoms and minimize suffering, our helplessness in influencing the fundamental determinants of health — which include housing — inevitably haunts and outrages us. As Cheung and Hwang have so dramatically shown, housing and health are intricately interwoven. Impoverished women and men without homes bear an undue and unacceptable burden of illness and are dying prematurely in our streets, in the very shadows of our towering health care institutions. The ultimate solution to homelessness will require change in many sectors. This public health crisis will not be ameliorated until housing and health care become a fundamental right for every human being.

β See related article page 1243