Health and Service Utilization Patterns among Homeless Men in Transition: Exploring the Need for On-Site, Shelter-Based Nursing Care
Brush, Barbara L. Rnc, PhD, Faan, Powers, Emily M. Rnc, Msn, Scholarly Inquiry for Nursing Practice
Between September 1999 and May 2000, as part of a larger faculty/student teaching project, data describing the demographic, health problem and service utilization patterns of 183 newly sheltered homeless men (mean age = 42) seeking on-site nursing clinic services were collected and analyzed. Upon arrival to the transitional shelter, 46% of the study participants were medically uninsured or received state subsidized health benefits (49%). Almost all (99%) were in recovery from substance addiction. Despite numerous self-reported health problems, 44% had no primary care provider and 35% were seen only sporadically in local hospital clinics or in emergency departments. During the study period, nurse practitioner students and faculty managed most of the residents' episodic illnesses on-site, while assisting them to secure insurance coverage and medical appointments as a prelude to independent living. This was an attempt to curtail residents' previous pattern of waiting until medical problems became serious before seeking treatment. These findings mirrored those in studies of similar populations equating on-site delivery of health care to cost reduction. Unfortunately, many nurse-managed clinics, like our study site, fail to generate data supporting nurses' clinical efficiency and cost effectiveness in caring for homeless individuals. As such, nurses' work remains invisible and underappreciated. This article argues that future studies examining nurses' work and worth in caring for homeless individuals are necessary in determining future health care service planning with this vulnerable population.
With estimates of the number of homeless persons in the United States ranging anywhere from 0.5 to 3 million (Hwang, Orav, O'Connell, Lebow, & Brennan, 1997), there is a pressing need to compile accurate national health statistics about the magnitude of homelessness and its impact on the health care system. Part of the problem in creating a comprehensive database of the homeless, however, relates to variations in the definition of homelessness (Burt, 1996), difficulty measuring the characteristics of homeless individuals among urban, rural, and suburban communities (First, Rife & Toomey, 1994; Phelan & Link, 1999), and the invisibility of many homeless persons in society (Piliavin, Westerfelt, Wong, & Afflerbach, 1994). In addition, health and social issues range widely among homeless individuals of varying ages (Bissonette & Hijjazi, 1994; Cohen, 1999; Crane, 1998; Ensign, 1998); gender (Burg, 1994), and racial/ethnic backgrounds (Rosenheck & Seibyl, 1998). Growing numbers of a more heterogeneous population of homeless individuals demand unique approaches to successfully manage their complex health care and social needs.
HOMELESSNESS, HEALTH, AND HEALTH CARE
While earlier studies of homelessness and health care showed evidence of poorer health among homeless compared with non-homeless individuals (Brickner, Scharer, Conanan, Savarese, & Scanlon, 1991; Winkleby, 1990), newer research focuses on the predictors of health status and health care utilization among homeless individuals as evidence for present and future health care service planning (Glied, Hoven, Moore, & Garrett, 1998-1999; O'Toole, Gibbon, Hanusa, & Fine, 1999; Rosenheck & Seibyl, 1998; Spanowicz, Millsap, McNamee, & Bartek, 1998). For example, Piliavin and colleagues. (1994) identified the health attributes of 336 homeless adults in Minneapolis, Minnesota. The inclusion criterion for homelessness was broadly defined on a continuum from 1-day temporary residence in a shelter, automobile, shanty, or public building to 1-week stays with friends or relatives. Demographic variables collected for study participants included age, gender, race, education, military service, and marital status. In addition, the presence or absence of two health problems deemed to be associated with length of homelessness was documented: mental health status and alcoholism.
Participants in the Piliavin and associates. (1994) study were asked to rate their health status on a 4-point Likert scale, with "4" signifying "very good" health and "3," "2," and "1" corresponding to good, fair, and poor health, respectively. Individuals' use of health care services was ascertained by a yes/no response to the question: "Are you being treated by a doctor or at a health clinic for any problem right now?"
While more than one third of the study participants (36.3%) noted significant health problems, only 53% of that group received medical care for their conditions (Piliavin et al., 1994). Unfortunately, the study did not ascertain potential barriers to medical care, such as participants' health insurance status, access to free or sliding scale services, or ability (i.e., transportation) to keep appointments. Data regarding individuals' past and present medical history and other social and economic variables were also unavailable.
Gelberg, Gallagher, Andersen, and Koegel (1997) attributed poor utilization of health services among 1,563 homeless individuals in Los Angeles, California, to the problem of "competing priori ties" (p. 217). That is, the homeless person's basic need for food, shelter, and safety took precedence over issues of health and illness. Among the 80% male, 56% African American subject population between the ages of 18 and 41, 37% rated themselves to be in fair or poor health, 47% reported a regular source of care, and 31% had gone without needed medical care in the previous 12 months (Gelberg et al., 1997). Gelberg and colleagues. (1997) argued that locating walk-in clinics where homeless individuals congregated for meals, clothing, shelter, and bathroom facilities not only helped increase health care utilization, but took into account the competing priorities associated with homelessness.
Sachs-Ericsson, Wise, Debrody, and Paniucki (1999) found similar problems and patterns among 292 homeless men and women at a free clinic in a small southeastern community and identified numerous barriers preventing homeless individuals from seeking health care, including lack of affordable, accessible service and disrespectful treatment from health care providers. They argued that locating health facilities in close proximity to homeless shelters and staffing them with sensitive providers improved service utilization and treatment compliance.
In contrast, Wojtusik and White (1998) reported that on-site, rather than proximal, health care access increased compliance with treatment regimens and decreased homeless adults' use of emergency rooms in San Francisco. Segal, Gomory, and Silverman (1998) also supported on-site health care in their analysis of homeless and marginally housed users of mental health self-help agencies. Onsite services for the homeless increased access and utilization of health services, enabled detection and treatment of illness at earlier stages, promoted compliance with prescribed interventions, and reduced overall associated costs (Segal, Gomory, & Silverman, 1998).
Indeed, the availability of on-site services targeting common illnesses among the homeless has proven particularly successful. Griffin and Hoff (1999) studied the effect of tuberculin screening in five on-site homeless shelter clinics in Kansas City, Missouri. Voluntary tuberculin skin testing was provided to 856 homeless men and women with on-site radiographic testing available to all who tested positive. Of those tested, 654 (76.4%) returned to the clinic for follow-up readings. Eighty-nine (13.6%) had a positive induration (greater than 10mm), 42 initiated treatment, and 8 completed the treatment course. Although the researchers reported poor overall chemotherapeutic compliance among the study participants, they noted that the on-site clinic clients with positive skin tests received radiographic evaluation more readily than a comparable group cared for at a local Health Department. The comparable group was comprised of homeless individuals living on the streets or in shelters without on-site facilities, thus necessitating their use of the public health department. Griffin and Hoff (1999) argued that on-site medical services were therefore more successful than other more traditional sources of proximal care delivery for homeless individuals with tuberculosis.
Macnee, Hemphill and Letran (1996) reported the outcomes of nurse-managed, shelter-based clinics specifically designed to screen hypertension, diabetes, foot problems and tuberculosis among their homeless populations. Over the 9-month operating period of the screening clinics, 22% (47) of the 214 clients screened were found to have one of the four conditions. Although patient participation varied from clinic to clinic, screening was found to be a cost-effective means to address disease prevention and health promotion among the shelter clientele.
Nurse-managed clinics, in particular, have proven effective in increasing access to health care among the homeless (Macnee, Hemphill, & Letran, 1996; Patti, McDonagh, & Porter O'Grady, 1990; Skelly, Getty, Kemsley, Hunter, & Shipman, 1990). Many of these clinics, staffed by registered nurses, nurse practitioners (NPs), and midwives, provide primary health care to individuals with otherwise nonexistent, limited, or fragmented health care options (Davis, 1993; Holthaus, 1993; Pearson, 1988).
CARE AT A PRICE
Over the past decade, however, many researchers have examined the high cost of providing health care to the homeless and argued for new approaches for delivering care to this population. Rosenheck and Seibyl (1998) found that the Veteran's Administration incurred a $3,196 higher per-person cost when caring for homeless compared to domiciled veterans, primarily because when homeless veterans did seek health care, they were sicker, needed more crisis care, and required long lengths of hospitalization.
Glied, Hoven, Moore, and Garrett (1998-1999) and Little and Watson (1996) found that even when homeless men and women received Medicaid services, their use of non-hospital medical care increased only slightly; medically insured homeless adults continued to utilize emergency rather than primary care services for basic health care needs. Little and Watson (1996) reported that the 135 homeless persons in their study attended the emergency room (ER) 233 times during a 6-month period. Ninety-one percent were males with an average age of 40 and 81.5% of their presentations to the ER were for minor problems. Glied and colleagues (1998-1999) also found that treatment delay and utilization of emergency services by Medicaid insured homeless individuals often led to costly hospital admissions.
In another study, the hospitalization costs associated with homelessness in New York City were twice that for all types of patients using general hospitals between 1992 and 1993 (Salit, Kuhn, Hartz, Vu, & Mosso, 1998). Eighty percent of the admissions among the homeless were related to or exacerbated by substance abuse and mental illness, requiring longer and more expensive hospital stays. Even when substance abuse and mental illness were controlled statistically, homeless patients stayed 4.1 days, or 36% longer, than other patients and cost between $2,424 and $4,094 more per admission (Salit et al., 1998).
Although many of the problems faced by homeless individuals are similar in scope and magnitude, the cost of providing health care to homeless individuals and families has widespread implications for local communities. Harris, Mowbray and Solarz (1994) described the physical illnesses, mental health issues, and substance abuse problems among users of Detroit's four largest homeless shelters. Many of their findings, such as higher rates of tuberculosis compared to the general public, untreated mental illness, and chronic alcoholism have been reported by others caring for the nation's homeless (First, Rife, & Toomey, 1994; James, 1992). Regardless of the community location and types of problems, however, care to the homeless is costly.
Historically, nurse-managed clinics have been key contributors in the provision of accessible, affordable, quality health care to both rural and urban poor and homeless (Glass, 1989; Walker, 1994). While studies often focus on the process and outcomes of nursing care in clinics serving traditionally underserved populations, however (Carter, Green, Green, & Dufour, 1994; Davis, 1993), few have examined the costs associated with such care (McGrath, 1990; Nichols, 1992)
Setting: The ECHO Project
As part of the larger Expanded Care for Healthy Outcomes (ECHO) project, demographic, health problem, and service utilization patterns of 183 homeless men living in a transitional housing shelter in the Boston area were collected between September, 1999 and May, 2000. The shelter, in operation for over a decade, is one of five clinics affiliated with a large, multi-service, not-for-profit organization serving many of Boston's homeless men and women. Despite its longevity, the clinic did not regularly collect data regarding patients' demographic profiles or clinical problems and presentations. Thus, although the ECHO project was designed as a faculty/student learning model, we collected data as a means to describe the sample of patients seen in the shelter during a nine-month period. The purpose of collecting data, based on initial intake history and physical examinations and logged clinical visits during the study period, therefore, was to describe residents' demographic and other correlates of health service utilization, general health problems, reasons for clinic utilization, and referral and follow-up patterns.
The ECHO project was a two-year collaborative effort between the Boston College School of Nursing and the Pine Street Inn Nurses' Clinic. As an academic/clinical alliance, the ECHO project prepared 15 family nurse practitioner (FNP) students, under faculty supervision, to incorporate spiritual assessment into their clinical encounters with homeless guests at the Anchor Inn nurses' clinic. An affiliate shelter of the larger Pine Street Inn Nurses Clinic system, the Anchor Inn is a transitional program for homeless men who are newly sober and committed to working on the root causes of their homelessness (Brush & McGee, 1999). Transitional shelters like the Anchor Inn reflect a multifaceted intervention that addresses both homelessness and addiction (Braught et al., 1995; Conrad et al., 1998; Davis, 1997). Residents are offered food, clothing, shelter, health care, job training, educational opportunities, and case management services at no cost. Program participants must be detoxified from drugs and/or alcohol prior to entering the facility, however, and must remain abstinent throughout their stay.
Sample and Procedures
The study sample was 183 male residents of the Anchor Inn, a 200-bed transitional shelter in Quincy, Massachusetts affiliated with the Pine Street Inn, residing at the Inn between September 1999 and May 2000. As part of the Inn's policy regarding new residents, all individuals had a baseline history and physical examination at the on-site nurses' clinic within 1 month of entering the program. The nurses' clinic is located next to the residents' cafeteria and is normally staffed by 2 registered nurses 5 days per week, from 11 AM to 7 PM During the study period, an additional 15 students and 4 faculty members rotated to staff the clinic each evening from 4-7 PM, with one student and one faculty member consistently present. All residents were eligible to use the clinic for their health care needs during clinic hours, although only those who consented to participate in the study and were seen by FNP students and faculty were included in data collection and analysis.
The FNP students administered all new resident history and physical exams, screened residents for psychosocial, spiritual, and primary health care needs and made appropriate referrals for follow-up. An on-site minister, part of the ECHO project, cared for many of the residents' identified spiritual needs (Brush & McGee, 1999), while clinic counselors tended to individuals' mental health needs. The FNP students were responsible for evaluating and providing episodic physical care, where appropriate, under faculty supervision. Referrals to physicians were made where prescription medication was indicated and when ongoing follow-up was deemed appropriate. A daily clinical log was maintained to document each patient visit to the clinic and included residents' reasons for accessing clinic care, treatment rendered, referrals given, and patient follow-up with scheduled appointments. Information gathered in the clinical log was also recorded in the patient record.
Data Collection and Analysis
Data collected from admitting history and physical examinations and logged clinical visits were entered and analyzed via the SPSS computer statistical program. Descriptive statistics were calculated for demographic information such as age, race, education and health insurance status; individuals' length of stay at the shelter, addictive substance of choice, and number of clinic visits was recorded. The latter were broken down into discrete categories of "acute and treated visits," "acute and referred visits," "chronic evaluation visits," "chronic referred visits," "medication review visits," "history and physical visits," and lastly, "follow-up of referred visits" as a means of categorizing reasons for health care utilization by residents. All acute and chronic diagnoses were then subcategorized by the affected body system to establish patterns of illness.
Of the 183 men who entered residence between September 1999, and May, 2000, 47% identified themselves as Caucasian, 36% as African American, 13% as Hispanic, and 4% as Asian, American Indian, or multiracial. Eighty-seven percent reported English as their primary language and 13% were Spanish speaking. Ages ranged from 20 to 69 with an average age of 42 (median age = 41). Thirty percent had less than a high school education, 20% were high school graduates, and 7% had earned a GED. Fourteen percent reported completion of some or all years of college. Because they entered a shelter that specifically promotes individuals' transition from active substance use to sobriety, it was not surprising that 99% of the study participants were in recovery from substance addiction. Thirty-nine percent identified alcohol as their primary "drug of choice," 9% and 12% named cocaine and heroin, respectively, and 37% reported polysubstance abuse. At 3 months, 81% of the participants were still in the program (19% relapse rate), compared to 60% at 6 months (39% relapse rate) and 41% (59% relapse rate) at 9 months.
Upon shelter admission, 46% of the participants had no health insurance, 49% were Medicaid recipients, 2% received Medicare, and 3% had private medical insurance. Forty-four percent could not identify a source of health care and 35% reported sporadic use of local hospitals, free-clinic services or emergency rooms. Only 21% identified a regular primary care provider. Forty-nine percent reported one to three hospital admissions in the prior 12-month period, usually related to traumatic injury.
Regardless of health care coverage or follow-up, over half of the residents reported dental disease (56.3%) and 47% and 44% reported emphysema or asthma, gastrointestinal reflux disease, or dyspepsia. Twenty-eight percent had untreated dermatologic conditions and 31 % reported hypertension or other cardiac disease. Given the mean age of the population, only 14% suffered from arthritis, 3% from Type 2 diabetes, and 9% from prostate disorders. Only 1.4% reported a known positive HIV status and 5.6% reported a diagnosis of Hepatitis C. No HIV testing was done at the clinic site.
The clinical log documented 176 visits over the course of the data collection period. Of these, two study participants alone made 33 visits. The data from these two individuals were not entered into the system and were analyzed individually. Thus, only 143 visits were entered for analysis. Of these, seventy-nine (40%) were labeled as "acute and treated visits," thirty-six (18%) as "acute and referred," and twenty (10%) as "chronic evaluation visits." Four percent (6) of the study participants were seen for follow-up after referral and 1% were "chronic referred" or "medication review" visits.
Upper respiratory infection constituted the majority (65%) of "acute and treated" visits to the clinic, followed, in order, by gastrointestinal complaints (24%), musculoskeletal injuries (5%), and dermatologic conditions (4%). Respiratory tract conditions (e.g., to rule out pneumonia) also accounted for the bulk (91.4%) of acute referrals, with referrals for gastrointestinal complaints a distant second (8.6%). Among chronic and evaluated episodes, the most common complaints also concerned the respiratory system. The high rate of chronic respiratory conditions presenting to the clinic (84.2%) correlated with individuals' past medical history of asthma and chronic bronchitis. Chest pain, hypertension, and shortness of breath were categorized as cardiovascular system complaints and represented 10.5% of chronic visits. Endocrine system conditions (diabetes and thyroid) ranked third, at 5.3% of chronic visits.
The two individuals excluded from analysis had high volume visit patterns far exceeding those of the other Inn residents. The first individual, a 69-year-old recovering alcoholic and 3-pack-per-day smoker with emphysema, was evaluated 22 times by nurse practitioners during the data collection period. Of these visits, four were "acute and treated," four were "acute and referred," 13 were for "chronic evaluation" and one was for "follow up of a referral." The second individual excluded from data analysis was seen in the clinic 11 times. Similar to the first individual, he was older than the average Inn resident (age 54) and had significant cardiac and respiratory illness likely exacerbated by chronic alcoholism and longstanding cigarette smoking. Among the 11 visits made to the clinic, one was "acute and treated," two were "acute and referred," five were "chronic and referred," and three were "chronic and evaluated." Despite the fact that the two aforementioned participants totaled 33 visits to the on-site nurses' clinic, neither individual was hospitalized during the study period.
Many of the health problems for which individuals sought care or treatment at the nursing clinic were managed on-site by the clinic nurse practitioner faculty and students and were short-term events. This is an important finding given data demonstrating that many homeless individuals fail to access health systems until acute problems reach chronic and often intractable stages (Brickner et al., 1991). At that point, they usually seek care in emergency rooms rather than in primary care facilities (Gelberg, Linn, & Usatine, 1990). This finding also supports research that advocates the use of on-site services as a means to help the homeless increase access to care (Jackson & McSwane, 1992). By screening them for psychosocial, spiritual, and physical illnesses or problems and by seeing individuals in more acute stages of illness, the NPs were able to link residents to appropriate providers for routine follow-up and health maintenance instead of episodic care only. Individuals requiring complex health treatment and/or physician management were referred to outside providers by written referral so that clinic nurses could coordinate care and follow-up when appointments were needed, missed, or needed to be rescheduled. Residents were encouraged to maintain appointments with proximal health care providers as well as in anticipation of discharge from the shelter to independent living situations. Follow-up and encouragement to keep scheduled appointments was facilitated in the system by the fact that all study participants lived on-site, were sober, and were able to access clinical services when necessary. Thus, issues of competing priorities and active substance abuse or mental illness preventing access and follow-through for health services was not problematic in our sample.
Screening and early intervention among younger homeless adults such as those in our study population may also help prevent many costly health problems known to increase morbidity and mortality in older homeless persons. Indeed, many researchers in the field of elder homelessness define older homeless individuals as older than 50 and report significant untreated or underdiagnosed physical and psychological health problems among that cohort (Cohen, 1999; Crane, 1994). Older homeless adults who are substance addicted are particularly vulnerable to increased morbidity and mortality (Atkinson, Turner, & Tolson, 1998).
The two individuals in our analysis who fell outside the normal range of visit patterns to the clinic, for example, were older individuals with chronic illness and long-term substance abuse histories. Had on-site services not been available, it is likely that these individuals would have used local emergency services for care at later stages of intervention or would have been hospitalized, incurring far greater costs to the health system than were incurred at the shelter clinic. Neither individual had previously identified a primary care provider before entering the program and had received care only sporadically through emergency and public health departments. By seeing them regularly and promptly on-site and coordinating ongoing care with their newly identified primary care providers, both individuals remained relatively healthy during the study period.
While our findings support other claims that early recognition of disease through case finding and appropriate use of the health care system are essential ingredients in preventing or managing illness among homeless individuals, they also support the relative invisibility of nurses in the care of vulnerable homeless individuals. The Anchor Inn clinic, like many nurse-managed clinics serving homeless populations, operates as a nonprofit facility with free of charge service and lack of direct billing. Costs of providing care to individuals, therefore, are not easily compared or tracked across facilities. The lack of a formal data collection system at the site, even after 10 years of continuous operation, equates with 10 years of lost information about the work and worth of the nurses in this setting. Clinical settings employing nurse-managed models must gather statistics regarding patient demographics, health problems, and service utilization patterns to clearly demonstrate the magnitude of care needs and the quantity and quality of nursing intervention among this growing population.
Given the economically driven focus of our health care system today, nurses must measure their effectiveness in reducing the high costs associated with a population notorious for delayed care and irregular follow-up. To make this determination, cost analyses must be included in further studies of on-site nursing services to homeless men, women and children across racial/ethnic backgrounds. With the number and diversity of homeless individuals expected to rise in the future, data about the value and efficiency of on-site services that are sensitive, emphasize health promotion and illness prevention, promote compliance with treatment regimens, and support healthy lifestyles are critical. Nurses, who are often at the front lines caring for this vulnerable population, must seek direct reimbursement for their services as a means to track the process and outcome of their interventions. Only then can we accurately determine the role of advanced practice nursing in reducing costs and increasing quality services to this growing population. Only then will the outcomes of nursing intervention be visible and valued.
Atkinson, R. M., Turner, J. A., & Tolson, R. L. (1998). Treatment of older adult problem drinkers: Lessons learned from "the class of '45," Journal of Mental Health and Aging, 4, 197-214.
Bissonnette, A., & Hijjazi, K. M. (1994). Elder homelessness: A community perspective. Nursing Clinics of North America, 29, 409-416.
Braught, N., Reichart, C, Geissler, L., Bormann, C, Kwiatkowski, D., & Kirby, M. (1995). Effective services for homeless substance abusers. Journal of Addictive Diseases, 14, 87-109.
Brickner, P., Scharer, L., Conanan, B., Savarese, B., & Scanlon, B. (Eds.). (1991). Unde the safety net: The health and social welfare of the homeless in the United States. New York: Norton.
Brush, B. L. & McGee, E. M. (1999). The Expanded Care for Healthy Outcomes (ECHO) Project: Addressing the spiritual care needs of homeless men in recovery. Clinical Excellence for Nurse Practitioners, 3(2), 116-122.
Burg, M. A. (1994). Health problems of sheltered homeless women and their dependent children. Health and Social Work, 19, 125-131.
Burt, M. R. (1996). Homelessness: Definitions and counts. In J. Baumohl (Ed.), Homeless in America (pp. 15-23). Phoenix, AZ: Oryx Press.
Carter, K. F., Green, R. D., Green, L., & Dufour, L. T. (1994). Health needs of homeless clients: Accessing nurisng care at a free clinic. Journal of Community Health Nursing, 11, 139-147.
Cohen, C. I. (1999). Aging and homelessness. The Gerontological Society of America, 39,5-14.
Conrad, K., Hultman, C., Pope, A., Lyons, J., Baxter, W., Daghestani, A., Lisiecki, J., Elbaum, P., McCarthy, M., & Manheim, L. (1998). Case managed residential care for homeless addicted veterans: Results of a true experiment. Medical Care, 36, 40-53.
Crane, M. (1994). Elderly homeless people: Elusive subjects and slippery concepts. Aging and Society, 74,631-40.
Crane, M. (1998). The association between mental illness and homelessness among older people: An exploratory study. Aging and Mental Health, 2, 171-180.
Davis, E. (1993). Establishing a nurse-managed health center: Assuring excellence. Nurse Practitioner Forum, 4, 151-157.
Davis, H. (1997). Health care services for alcohol and drug dependent adult male homeless populations: A nursing framework for social model programs. Journal of Addictions Nursing, 9, 9-16.
Ensign, J. (1998). Health issues of homeless youth. Journal of Social Distress and the Homeless, 7, 1, 159-174.
First, R. J., Rife, J. C, & Toomey, B. G. (1994). Homelessness in rural areas: Causes, patterns, and trends. Social Work, 39(1), 97-108.
Gelberg, L. Gallagher, T. C., Andersen, R. M., & Koegel, P. (1997). Competing priorities as a barrier to medical care among homeless adults in Los Angeles. American Journal of Public Health, 87, 217-220.
Gelberg, L., Linn, L.S.,&Usatine,R. P. (1990). Health, homelessness and poverty: A study of clinic users. Archives of Internal Medicine, 150, 2325-2330.
Glass, L. K. (1989). The historic origins of nursing centers. In National League for Nursing, Nursing Centers: Meeting the Demand for Quality Health Care (pp. 117-132). New York, NY: Author.
Glied, S., Hoven, C., Moore, R. E., & Garrett, A. B. (1998-1999). Medicaid and service use among homeless adults. Inquiry, 35, 380-388.
Griffin, R & Hoff, G. (1999). Tuberculosis screening in Kansas City homeless shelters. Missouri Medicine, 96, 496-499.
Harris, S., Mowbray, C, & Solarz, A. (1994). Physical health, mental health and substance abuse problems of shelter users. Health & Social Work, 19, 37-45.
Holthaus, R. (1993). Nurse-managed health care: An ongoing tradition. Nurse Practitioner Forum, 4, 128-132.
Hwang, S. W., Orav, E. J., O'Connell, J. J., Lebow, J. M., &Brennan,T. A. (1997). Causes of death in homeless adults in Boston. Annals of Internal Medicine, 126, 625-628.
Jackson, M. P. & McSwane, D. Z. (1992). Homelessness as a determinant of health. Public Health Nursing, 9, 185-192.
James, F. J. (1992). New methods for measuring homelessness and the population at risk: Exploratory research in Colorado. Social Work Research and Abstracts, 28, 9-14.
Little, G. F. & Watson, D. P. (1996). The homeless in the emergency department: A patient profile. The Journal of Accident and Emergency Medicine, 13, 415-17.
Macnee, C., Hemphill, J., & Letran, J. (1996). Screening clinics for the homeless: evaluating outcomes. Journal of Community Health Nursing, 13, 167-177.
McGrath, S. (1990). The cost-effectiveness of nurse practitioners. The Nurse Practitioner, 15, 40-42.
Nichols, L. (1992). Estimating costs of under using advanced practice nurses. Nursing Economics, 10, 343-51.
O'Toole, T. P., Gibbon, J. L., Hanusa, B. H., & Fine, M. J. (1999). Utilization of health care services among subgroups of urban homeless and housed poor. The Journal of Health Politics and Policy Law, 24,91-114.
Patti, J., McDonagh, K., & Porter-O'Grady, T. (June 1990). Streetside support. Health Progress, 60-62.
Pearson, L. (1988). Providing health care to the homeless: Another important role for NPs. Nurse Practitioner, 13, 38-48.
Phelan, J. C. & Link, B. G. (1999). Who are "the homeless"? Reconsidering the stability and composition of the homeless population. The American Journal of Public Health, 89, 1334-1338.
Piliavin, I., Westerfelt, A., Wong, Y. I. & Afflerbach, A. (1994). Health status and healthcare utilization among the homeless. Social Service Review, 68(2), 236-253.
Rosenheck, R. & Seibyl, C. L. (1998). Homelessness: Health service use and related costs. Medical Care, 36, 1256-1264.
Sachs-Ericsson, N., Wise, E., Debrody, C. P., & Paniucki, H. B. (1999). Health problems and service utilization in the homeless. Journal of Health Care for the Poor and Underserved, 10, 443-452.
Salit, S. A., Kuhn, E. M., Hartz, A. J., Vu, J. M. & Mosso, A. L. (1998). Hospitalization costs associated with homelessness in New York City. The New England Journal of Medicine, 338, 1734-1740.
Segal, S. P., Gomory, T., & Silverman, C. J. (1998). Health status of homeless and marginally housed users of mental health self-help agencies. Health and Social Work, 23,45-52.
Skelly, A., Getty, C., Kemsley, M., Hunter, J., & Shipman, J. (1990). Health perceptions of the homeless: A survey of Buffalo's city mission. Journal of the New York State Nurses Association, 21, 20-24.
Spanowicz, M. J., Millsap, G., McNamee, M. J., & Bartek, J. K. (1998). Health problems of sheltered homeless men using a mobile health van: A 4-year study. Clinical Excellence for Nurse Practitioners, 2, 279-285.
Walker, P. H. (1994). Dollars and sense in health reform: Interdisciplinary practice and community nursing centers. Nursing Administration, 19, 1-11.
Winkleby, M. A. (1990). Comparison of risk factors for ill health in a sample of homeless and nonhomeless poor. Public Health Reports, 105, 404-410.
Wojtusik, L. & White, M. C. (1998). Health status, needs, and health care barriers among the homeless. The Journal of Health Care for the Poor and Underserved, 9, 140-152.
Barbara L. Brush, RNC, PhD, FAAN
Boston College School of Nursing
Chestnut Hill, MA
Emily M. Powers, RNC, MSN
Medical Care Affiliates/Health Promotion Affiliates
Offprints. Requests for offprints should be directed to Barbara L. Brush, RNC, PhD, FAAN, Boston College School of Nursing, Gushing Hall, 140 Commonwealth Avenue, Chestnut Hill, MA 02167.…