key: cord-0033644-ysjgk1mz authors: nan title: Homelessness and housing date: 2003-06-03 journal: J Urban Health DOI: 10.1093/jurban/jtg003 sha: ba263288331df1826fad9623057988ddd13bbbe9 doc_id: 33644 cord_uid: ysjgk1mz nan between the two groups. There were nonstatistically significant trends towards increased primary care and substance abuse treatment in the intervention group. Implications: SH maintained housing in chronically homeless person but did not decrease acute healthcare utilization. Nonrandom group assignment and the possibility that the comparison group successfully attained housing may have biased results towards the null. Further research that clarifies determinants of acute healthcare utilization in this population may help refine future interventions. Toronto Infirmary. Seaton House Canada's largest men's hostel provides health services on site. The 36 bed Infirmary is utilized as a training site for family practice residents. The Infirmary cares for clients who have been recently discharged from hospital, and have increased health care needs due to acute and chronic illness. The objective of this study is to evaluate the learning experiences of family medicine residents at the Rotary Club of Toronto Infirmary at Seaton House. Methods: A total of 13 family practice residents will be surveyed between July and August 2003. The analysis will determine if their learning has increased within each practice domain. Results: It is hypothesized that residents who had previous experiences with similar patients will have a preexisting interest in the care of the marginalized. Recommendations based on responses with suggestions for improvement will be made to strengthen the learning experiences for the next year's resident group. Implications: The learning opportunity for Family Physician Residents provided within a model of care that integrates Harm Reduction within a shelter environment will become an invaluable skill as the rate of homelessness continues to rise. Baseline information will be utilized to modify the teaching rotation and recommend further improvement strategies targeted at academic and systems programming. Background: Seaton House, Canada's largest men's shelter and St. Michael's, a large academic Inner-City hospital share a partnership caring for homeless. We recognize that many clients at Seaton House and St. Michael's do not receive coordinated care planning between the two agencies due to the many complex barriers they experience. Supported by a community partnership grant, this joint initiative attempts to improve coordination and integration of services to increase health outcomes of homeless men and stop the "revolving door" cycle of shelter, hospital, and street. The goals of the project are to improve referral and discharge planning, improve harm reduction strategies, and to provide supports for clients admitted to Emergency Department, Internal Medicine, and other "highly utilized" areas. Methods: Strategies utilized to improve integration of services over a 6-9 month period are identification of barriers and strategies by staff at both agencies; implementation of a Seaton House Liaison position at St. Michael's; development of Emergency Department and inpatient protocols that support coordination of care, as well as processes, guidelines, and information pathways that will guide referral and discharge processes of both Seaton House and St. Michael's staff. Results: Projected outcomes include regular client care summaries from shelter to hospital, discharge plans forwarded from hospital to shelter care teams, and complete program descriptions, guidelines, and summaries for shelter, hospital, and community. Implications: The success of this joint initiative can provide an "Integration Model of Care" template that can be replicated and adapted to other organizations. Golembeski C, Spielman S, Jean-Louis B, Shoemaker K, Rome M, Northridge M, Vaughan RD, and Nicholas SW Background: The primary goal of the Harlem Children's Zone Asthma Initiative (HCZAI) is to reduce asthma morbidity through improved health care utilization. The parents of 1304 children ages 0-12 have completed screening surveys; 25% (n = 316) of these children have asthma, 56% of whom have been to an emergency department in the past year because of breathing problems. Geographic information systems (GIS) are being used to describe the geographic distribution of children with asthma in the HCZ Project with high spatial accuracy. Further analysis of the built environment at both the building and unit level and its relationship to asthma prevalence is necessary. Methods: Data on housing, schools, health care facilities, and transportation systems are being integrated into the GIS to help guide environmental interventions. Furthermore, building information along with various built environment quality indicators are being analyzed in terms of their correlation to rates of asthma within the designated area. Results: 29% of children with asthma live in publicly owned buildings, a likely target for focused efforts. Both New York City Housing Authority (NYCHA) and NYC Department of Housing and Preservation (HPD) own buildings in the designated area that are linked to various tenant programs. Implications: GIS has assisted in better understanding the geographic distribution of children with asthma in the HCZ Project with high spatial accuracy. An assessment of characteristics related to structural adequacy, building history, tenant crowding, maintenance deficiencies, and building code violations at the unit level can assist in guiding interventions at the neighborhood level as well as to suggest possible policy recommendations. ICD-10. This reliable and valid instrument has been found to accurately predict youth with and without mental health disorders. The sampling methods included recruiting a group of 52 youth (mean age, 18.46; standard deviation 2.2 years) from both an innovative program (the MAC Door) and various drop-in centers throughout the city. The preliminary results of this study show that homeless street youth in Hamilton have extremely high rates of psychiatric symptoms. These rates were compared to other developed countries and to a population of housed youths from Ontario. Implications: These results have implications for ensuring that primary health care and mental health services are youth-friendly and accessible to this vulnerable population. In the future it will be necessary to investigate the presence of preexisting psychiatric and substance abuse disorders and the influence of length of time living on the streets to determine the impact of the street environment on mental health. Semogas D, Cleverley K, Thomas CF, Sanford D, North T, and Byrne C Background: In the past several years, numerous studies have reported on the adverse health status of street youth and their concomitant low rates of health service utilization. Specific comparisons on the utilization of primary health care, mental health and social services by street youth have not been as widely discussed. The purpose of this paper is to describe the utilization of primary healthcare, mental health, and social services by a sample of Hamilton street youth. Methods: Beginning in April 2002, fifty two street youth ages 16 to 25 years were recruited (from the M.A.C. Door and from community youth services) to participate in a study examining readiness to change health behaviors. The interviewer-administered questionnaires looked at a number of themes including mental health, substance abuse, and self esteem. Items related to health and social service utilization explored the number of recent visits to primary health care, mental health and social services. Primary health and mental health care included number of visits with family physicians, community nurses, psychologists/counselors, emergency rooms and counseling agencies while categories pertaining to social services included number of visits with social workers, social service agencies for youth. Results: Results indicate that street youth make far greater use of social services that health services in spite of the fact that they have significant health problems. This is particularly noteworthy for mental health services. Implications: The fragmentation of primary health care, mental health services and social service delivery has important implications for the health of homeless youth. Kertesz SG, Swain S, Posner M, Shwartz M, O'Connell J, and Ash A Background: A novel subacute care facility, the medical respite unit, offers homeless patients recuperative care while addressing psychosocial and other homeless specific needs. Such units exist nationally, but their effectiveness is unstudied and they are unrecognized by the Centers for Medicare and Medicaid Services. We used data from the largest such program in the US to test whether discharge to a medical respite was associated with delayed hospital readmission or death (R/D) for the hospitalized homeless. Methods: we examined time to first R/D for 784 homeless adults discharged from a medical/surgical hospitalization 7/98-6/01, with the follow-up to 6/02. The predictor of interest was discharge destination, in 4 categories: Respite, Own Care (shelters/streets), Other Care (e.g. nursing home), and Left Against Medical Advice (AMA). We used survival curves and proportional hazards models to quantify the effect of discharge destination on the hazard for R/D, controlling for comorbidity using Diagnostic Cost Groups, index hospitalization length of stay (LOS), number of prior hospitalizations, substance abuse & sociodemographics. Data came from crosslinking 7 administrative sources. Results: raw survival curves for the outcome of R/D did not differ by discharge destination (P = .71). Predictors of R/D included index LOS, comorbidity and alcohol abuse (all P < .01). After statistical adjustment, Respite patients had a nonsignificant reduced risk of R/D (HR 0.86, 95%CI .65, 1.13). Implications: For the hospitalized homeless, respite placement may protect against readmission, but these findings are limited by sample size and possible residual confounding. Alcohol abuse, prior hospitalizations, and comorbidity identify a homeless subgroup at increased readmission risk. Dunn JR This paper will present a conceptual framework for the role of housing in the socioeconomic production of health in the city; it is now well established that there is a strong relationship between socioeconomic status (income, education, etc.) and health status (morbidity, mortality) amongst individuals in the affluent societies. Housing costs are a significant component of the household budgets of most urban households and typically represent their single largest asset. In addition, urban households are influential upon the differential distribution of wealth and control in society, and play a significant role in shaping social identities, and conferring social status. It follows that the social and economic dimensions of urban housing would be a plausible part of the pathway between socioeconomic status and health status. This paper will present a conceptual framework which models the links between material, meaningful, and spatial dimensions of housing, on the one hand, and health status on the other, within a life-course perspective. Empirical examples will contextualize the conceptual and theoretical arguments. Kertesz SG, Horton NJ, Larson MJ, and Samet JH Background: The "chronically homeless" have been proposed for special federal hosing assistance, but this group's distinct character has received limited study. Using prospective cohort data, we tested the hypothesis that chronically homeless persons have uniquely poor health-related quality of life (HR-QOL) over time, and tested whether substance use accounted for observed differences. Methods: Subjects (N = 470) were recruited at detoxification for alcohol or drugs, and followed for 2 years. Subjects available for 3 assessments (N = 289) were classified as Chronically Homeless (CH), Residentially Stable (RS), and Transitional (TR) based on reported nights homeless. At each follow-up we measured HRQOL with the SF-36 Mental Component Score (MCS) and Physical Component Score (PCS). We tested whether housing status was associated with MCS & PCS by fitting general linear regression models for correlated data that account for multiple observations per subject, controlling for sociodemographics, time & and the interaction of housing status & time. Results: At baseline, CH subject had very low MCS scores (mean 30.7), as did RS (32.2) and TR subjects (30.6), values lower than scores found in depression. While RS MCS scores improved 13.5 points, CH showed less improvement (+5.6, P = .01 for Time-by-Housing interaction, see Figure) . Adjustment for ongoing substance use did not explain the impact of CH on MCS. A model for PCs showed moderate improvement over time for all three groups. Implications: Among persons entering publicly funded detoxification, the chronically homeless had especially poor mental HRQOL, and less improvement than others over 2 years. This uniquely vulnerable subgroup warrants targeted interventions. Tadic V, Muckle W, and Turnbull J Background: Medical record keeping for the homeless is often not available at point of care and incomplete due to episodic, crisis-driven healthcare delivery by multiple caregivers at many sites. The Ottawa Inner City Health Project is a pilot project started in July 2001 to address the complex health needs of the chronically homeless using a case management approach. This project has included the development of a web-based sharable medical record, the Electronic Health Record (E.H.R.). Methods: This information system is an internet-based application that can be accessed remotely by any appropriately configured system in Ottawa. The completed record includes demographics, client contacts, medications, medical and mental health history, substance use and client encounters. Project nurses, support workers and physicians chart for each episode of care that is provided to a client. Relevant medical information can be printed and sent with the client to hospitals and appointments. Client consent is obtained and security is addressed by utilizing Secure Socket Layer 128-bit encryption, which most financial institutions currently employ for their web banking services. Implications: Modernizing health record keeping of the chronically homeless in Ottawa with the use of this technology, has made accurate and up to date client information accessible to multiple health care providers in an efficient and timely manner. The development of the E.H.R has gained international recognition as a needed direction for health care services and patient care. Despite its unique challenges, it has been successfully introduced in the setting of homeless adults. Future directions include extending internet-based access to the hospitals in Ottawa. Psychiatric Rating Scale (BPRS). A panel of MAP staff (shelter care workers, program nurse and doctor) caring for the clients was convened and a survey was completed for each client. A psychiatric nurse practitioner also caring for program clients reviewed select surveys for internal validity. Results: Forty-one individuals (38 male, 3 female, mean age 51 years) were admitted to the MAP for an average of 215 days. Prior to the program, 20 individuals were living on the street, and 18 were in the shelter system. Subjects were rated on the BPRS as having severe symptoms of self-neglect (59%), depression (44%), anxiety (41%) and guilt (39%). 31% of subjects had diagnosed mental illness at MAP entry. 71% of clients received psychiatric assessment and treatment while in the MAP. Improvements were noted in binge drinking (83%), alcohol seeking behavior (79%) and hygiene (66%). Conclusions: The MAP provided a stable environment in which many clients decreased alcohol consumption and showed improved personal care. Harm reduction provides a novel approach to psychiatric diagnosis and treatment in chronic homeless alcoholics. Weber AE, Boivin JF, Blais L, Haley N, and Roy E Background: Prostitution among female street youth represents an important risk factor for several health problems. Little is known about the incidence and determinants of prostitution in this vulnerable population and no data have been previously reported based on a longitudinal follow-up study. The objective of this study was to determine predictors of initiation into prostitution among female street youth. Methods: Female youth aged 14 to 25 years were enrolled in the Montréal Street Youth Cohort. They completed a baseline and at least one follow-up questionnaire between January 1995 and March 2000. Girls who reported never having engaged in prostitution at baseline were followed prospectively to estimate the incidence and predictors of prostitution. Results: Of the 330 female street youth enrolled in the cohort as of September 2000, 148 reported no history of involvement in prostitution at baseline and completed at least one follow-up questionnaire. Thirty-three of these 148 girls became involved in prostitution over the course of the study (mean followup: 2.4 years) resulting in an incidence rate of 11.1/100 person-years. Multivariate Cox regression analysis revealed having been on the street at age 15 years or less (Hazard Ratio (HR): 1.8; 95% Confidence Interval (CI): 0.9-3.8), using acid or PCP (HR: 2.0; 95% CI: 0.9-4.5), using heroin (HR: 1.8; 95%CI: 0.9-3.8), the use of drugs greater than twice per week (HR: 1.9; 95%CI: 0.9-4.2) and having a regular or casual female partner (HR:3.7 ; 95%CI: 1.6-8.5) to be independent predictors of initiation into prostitution. Conclusion: The incidence of prostitution in female street youth is elevated. Early debut on the street, substance use, the frequency of substance use, and having a regular or casual homosexual partner were important predictors of initiating involvement in prostitution. Tyndall M, Sussman F,Bright V,Palepu A, Hogg R, and Montaner J Background: The use of injection drugs within large urban communities has resulted in high rates of HIV infection in many North America cities. The provision of HIV care and antiretroviral therapy (ARVs) to this population is challenging and requires novel approaches to increase uptake and optimize adherence. Methods: The MAT Program consists of a multidisciplinary team who provide supervision, monitoring and support for individuals on ARVs. In addition to the daily dispensing of ARVs, the program provides assistance with housing, finances, health concerns, and addiction treatment. Results: Between Nov 1999 and Dec 2002, 101 participants (73 males, 28 females) were enrolled in the program. The average age is 42.7 years, all are active or recent IDUs, and all are co-infected with Hepatitis C. Sixty-one (60%) remain on ARVs. Reasons for stopping therapy include toxicity and adherence problems (14), loss to follow-up (11), death (8), planned treatment interruptions (4), and imprisonment (3). At initiation, 50% had a CD4 < 200 cells/ml, 52% had a plasma Viral Load (pVL) >100,000 copies/ml. Overall, 79% achieved an undetectable pVL at some time during follow-up and 43% remain undetectable. CD4 counts are maintained above 200 cells/ml in 62%. Treatment interruptions (>1 month) occurred in 73% and at least one change in ARV regime was required by 69%. The MAT Program demonstrates that people who are drug-dependent can be successfully engaged in ARV treatment. However, ongoing drug use may jeopardize longer-term sustainability due to treatment interruptions and frequent therapeutic changes. McCoy K, Curtis R, and Weiss L Background: Failure rates for abstinence-based drug treatment programs typically range from 70 to 90 percent (Goode 1999). Most people enter treatment programs repeatedly over their drug-using careers (Hubbard et al. 1989 ). This study investigates one possible explanation for these trends; not everyone who enters drug treatment intends to "quit." Methods: Data on people's experiences with abstinence-based treatment were compiled and analyzed from qualitative interviews with subsamples from an ethnographic study of heroin users (n = 35 of 500) and from a qualitative interview study of health care access and utilization among heroin and cocaine users (n = 39 of 71), both conducted in New York City. Results: Many people sought drug treatment without the intention of stopping drug use. Rather, they were seeking food and shelter, hoping to cut down use, or "to take a break." These alternative reasons for enrollment in treatment were generally concealed from providers. When people did want to quit, however, they often felt insulted and discouraged by what they perceived as dehumanizing practices on the part of treatment professionals. Harm reduction programs that they felt valued them and honored their goals were perceived as helpful. Implications: The assumption that most or all drug users entering abstinence-based drug treatment are aiming to quit using drugs is inaccurate. What is counted as treatment failure may actually reflect the fact that programs are often used to meet other needs. Programs that value people's goals and needs may be more effective in building the self-efficacy needed to better manage and/or cease drug use. In the last decade scholarly research and publication in the United States have dramatically increased in addressing older persons in the areas of trauma and substance abuse. How substance abuse and trauma are defined and assessed in the international research literature remains quite controversial. This presentation will discuss innovative prevention, practice, and teaching in this global problem. While an overwhelming majority of older adults (age 55 and older) see physicians and other professionals often, the literature reports that those at risk for substance abuse (alcohol, prescription misuse, and other drugs) are unlikely to be identified despite frequency of contact. Additionally, symptoms of substance abuse problems exhibited by older adults are often dismissed by clinicians and family members as signs of "old age." Challenges in identifying the "hidden alcohol abuser" are important for those working with or caring for older people. According the literature, there is debate as to which screening instruments accurately detect alcohol and drug abuse in the elderly. The prevailing issue is questioning how clinicians have refined screening protocols in older adults for alcohol and prescription medication problems, quantity and frequency of use, and responses to negative emotional states and isolation. Our poster will recommend research, education, training, and clinical interventions for international social work educators who provide a wide range of services for the older adult at risk for substance abuse problems. Future education efforts would include infusing trauma, substance abuse, and older adult content into the social work curriculum, and examining instruments used with this population. Cunningham C, Sohler N, Berg KM, and Shapiro S Background: Despite a growing body of literature demonstrating that HIV-infected substance users have poorer health outcomes than other risk groups, few studies have examined the impact of specific drug use on health care. This study investigated associations between specific drug and alcohol use and health care utilization. Methods: HIV-infected individuals residing in single room occupancy hotels in the Bronx, NY were interviewed regarding demographics, HIV disease and treatment, health care perceptions, and patterns of drug and alcohol use. Results: Of the 270 participants (96% response rate), mean age was 41 years; the majority were male (58%), Black or Hispanic (93%), and recent drug users (within the past two months) (61%). Individuals reporting any drug or crack/cocaine use were less likely to have a regular doctor than those reporting no drug or no crack/cocaine use (adjusted odds ratio [AOR] = 0.50, P = 0.05; AOR = 0.39, P = 0.006 respectively). Individuals reporting any drug or crack/cocaine use were also less likely to perceive quality of health care positively (AOR = 0.50, P = 0.02; AOR = 0.41, P = 0.003 respectively). Heroin use, injection drug use, and problem alcohol use were not associated with these outcomes. When the sample was limited to recent drug users, similar patterns were found. Implications: Although substance use in general is associated with negative health outcomes, in our sample of HIV-infected SRO hotel residents, poorer utilization of health care among drug users was associated predominantly with crack/cocaine use. It is important that clinicians and researchers working with drug-using populations understand how specific drug use patterns differentially impact on health care. Nandi AK, Galea S, Ahern J, Vlahov D A substantial burden of posttraumatic stress disorder (PTSD), depression, and increased substance use has been reported in New York City following the September 11 attacks. The results of prior studies on the relation between disaster exposure and substance-related disorders have been conflicting. Using random-digit dialing to contact a representative sample of adults living in New York City, we assessed the prevalence of symptoms of alcohol and smoking dependence after September 11, 2001. In the four months after September 11, 1.6% of the 2001 respondents reported symptoms consistent with alcohol dependence and 10.4% reported symptoms of smoking dependence. Respondents with symptoms of alcohol dependence were more likely to have probable PTSD and depression than respondents without symptoms of alcohol dependence (15.1% vs. 7.1% for PTSD; 42.2% vs. 7.5% for depression). Respondents with symptoms of smoking dependence were also more likely to have probable PTSD and depression than those without (18.1% vs. 5.7% for PTSD; 23.6% vs. 6.0% for depression). This study showed the co-occurrence of symptoms of substance dependence with probable PTSD and depression after the September 11 attacks and suggests a potential association between traumatic event exposure, substance dependence, and psychopathology after a massive disaster. Roy E, Nonn E, Haley N, and Morissette C Background: There is a need to understand the social contexts around injection practices that increase risk of bloodborne infections among street youth injecting drugs. Methods: We conducted a qualitative study using in-depth interviews. Subjects were recruited through various means including outreach at street youth agencies and snowballing technique. Results: Usually youth started using drugs around age 12, before becoming street active, and initiation into drug injection occurred within 5 years of arriving on the streets. Youth's main concern regarding the risks linked to drug injection is to avoid becoming addicted. When they first start injecting, most youth do not worry about the infectious risks of injection and use any available equipment. However, concerns about infection develop rapidly among those who continue injecting. In Montréal, where sterile syringes are available, youth think that injecting with a used syringe is taking an "unnecessary risk." Most of the time, they use new syringes but this is not the case for other pieces of injection equipment which are not available. Contexts that encourage unsafe use of injection equipment include: 1) the first injection, 2) relapse into drug use, 3) intensive periods of use, 4) a high degree of intimacy among partners, and 5) injecting in the streets. Implications: In an environment where sterile syringes are available, their use has become the norm among young street IDUs. To prevent bloodborne infections, it is important to provide all sterile injection materials and to develop interventions that will reduce unsafe injection practices in specific high risk contexts. follow-up were selected from the database. Three time periods were examined: lifetime before initiation (P1); from initiation to the first questionnaire following initiation (P2); and from this questionnaire to the next one (P3). Results: By 2000/09, 81 of the 542 non-IDU at entry had initiated injection during follow-up. Sixty-six initiates provided information on P1: 95.5% had used acid/PCP, 89.4% cocaine/crack/freebase, and 28.8% heroin. Among subjects first injecting with cocaine, 88.6% (31/35) had used cocaine before; 48.3% (14/29) of heroin initiates had used heroin before (P = 0.0004). During P2 (n = 64), 59.4% most often injected cocaine, 39.1% heroin and 1.6% PCP. Participants continued injecting mostly their firstinjected drug (cocaine initiates: 88.6%; heroin initiates: 77.8%). During P3 (n = 60), 48.3% reported no injections since last questionnaire; 11 had injected only once during P2. Among subjects previously injecting mostly with cocaine (n = 33): 60.6% reported no injections (half still used cocaine), 30.3% mostly injected with cocaine, and 9.1% had switched to heroin. Among those previously injecting mostly with heroin (n = 24): 25.0% reported no injections (none still used heroin), 66.7% mostly injected with heroin and 8.3% had switched to cocaine. Implications: Certain youth may only experiment with injection. Many youth initiate heroin use by injection and heroin injectors are more likely to continue injection than cocaine injectors. Reasons for these different patterns merit further study. Farrell S, Wood B, Muckle W, and Lougheed D Background: The Ottawa Inner City Health Project provides a management of alcohol program for persons with chronic alcohol use, complex health problems and histories of homelessness. Previous research has confirmed high incidence of psychiatric symptoms in this population, but to date, no Canadian studies have examined the impact of providing a flexible model of community-based psychiatric services to clients' mental health, substance use and quality of life. Methods: Using a repeated measures case comparison design, all residents of this harm reduction program were assessed at admission, 1, 3, 6, 9 and 12 months into the program. Assessment was conducted by the Psychiatric Nurse Practitioner and consulting Psychiatrist providing services. Assessment measures included the Brief Psychiatric Rating Scale, Mini-Mental Status Examination, and Wisconsin Quality of Life Questionnaire (Client and Provider). Case comparison between clients who did and did not receive psychiatric services and those receiving different levels of service are made to examine the impact of psychiatric services on client well being. Results: Preliminary findings to date suggest that the introduction of flexible, community-based psychiatric services provides improvement in clients' symptom level and quality of life. Implications: Persons in harm reduction programs in Canada have not systematically received psychiatric services to address their concurrent mental illness. Therefore, it was important to determine the impact of this flexible model of service to assist with developing an effective intervention to improve the health of this vulnerable urban population. Implications for the delivery of psychiatric services within harm reductions will be discussed. Novelli LA, Sherman SG, Havens JR, Sapun M, Fuller CM, and Strathdee SA Objective: We hypothesized that circumstances surrounding the first injection episode ("first hit") may be associated with future needle sharing among young injection drug users (IDUs). Such findings may be important for informing preventive interventions for this vulnerable population. Methods: Participants (n = 431) were IDUs in Baltimore, MD, aged 15-30, who had first injected <5 years ago. Survey data were collected on sociodemographics, circumstances surrounding the first hit, and current drug use. Chi-square tests and logistic regression were used to determine associations between these factors and needle sharing within the last six months. Results: Participants were primarily white (70.3%), male (59.4%), and initiated drug injection at a median age of 22 years; 23.2% used a used needle for their first hit. IDUs using a used needle for their first hit were more likely than those using a sterile needle to report recent needle sharing (60.6% vs. 32.3%, P < 0.001). Factors independently associated with current needle sharing, adjusting for age, gender, and race, included: using an unclean needle at injection initiation (AOR = 3.30; 95% CI: 2.01-5.40) and having heard of a needle exchange program (NEP) prior to initiation (AOR = 0.59, 95% CI: 0.37-0.92). Implications: Injection-related risks may be established at the onset of injection initiation, support-ing the need to educate young non-IDUs about the harms associated with unsafe injection practices. The protective effect of the awareness of NEPs prior to injection initiation suggests that these programs have an important role in disseminating educational messages to drug users, in addition to providing free, sterile injecting equipment. Plitt SS, Strathdee SA, Sherman SG, Gaydos CA, Hobelmann K, and Taha TE Background: Young injection drug users (IDUs) have consistently shown higher sexual risk behaviors than older IDUs. The advent of nucleic acid amplification tests (NAATS) has facilitated the identification of STIs among these high-risk populations. This study examines the prevalence of common STIs and associated sexual risk behaviors among young IDUs. Methods: IDUs aged 18-30 years completed a questionnaire assessing sexual risk behaviors and were tested for chlamydia, gonorrhea and trichomonas infection by NAAT in urine (males) or selfadministered vaginal swabs (females). Females were also screened for bacterial vaginosis (BV) using Nugent's criteria. Risk behaviors of infected and noninfected participants were compared using contingency table analysis. Results: Of 330 participants, 65% were male, 77% were white, and median age was 24 years (Inter-Quartile Range: 22-27). In the three months prior to baseline, 33%had traded sex for money or drugs, 68% had ≥ 2 sex partners, and only 14% of those with steady sex partners and 41% of those with casual partners reported consistent condom use. STI prevalence rates were consistently higher among females than males, respectively: chlamydia: 5.2% vs. 3.3% (P = 0.39); gonorrhea: 3.5% vs. 0% (P = 0.02); and trichomonas: 8.6% vs. 1.9% (P = 0.008). BV prevalence was 67% and was highly associated with douching (OR = 3.2, 95% CI: 1.4-7.5). Implications: High levels of sexual risk were exhibited among young IDUs, however STI prevalence rates were similar to those reported from non-IDU populations. The clinical significance of the high BV prevalence warrants further investigation given the high levels of HIV risk-taking behavior in this population. Methods: We performed analyses of: (1) syringe lending by baseline HIV-infected IDU and (2) syringe borrowing by HIV-negative IDU among participants enrolled in the Vancouver Injecting Drug Users Study (VIDUS) a prospective cohort of IDU. Since serial measures for each individual were available, variables potentially associated with the outcome in each case (lending or borrowing) were evaluated using generalized estimating equations with logit link for binary outcomes. Results: Overall, 1475 IDU were enrolled into the cohort between May 1996 and May 2002. Among the 318 IDU who were HIV-infected at baseline, having been incarcerated in the six months prior to the interview remained independently associated with syringe lending during this period (Adjusted Odds Ratio: 1.33 [95% CI: 1.06-1.69]; P = 0.015). In addition, among the 1157 individuals who were HIV-negative at baseline, having been incarcerated in the six months prior to the interview remained independently associated with reporting syringe borrowing during this period (Adjusted Odds Ratio: 1.26 [95% CI: 1.12-1.44]; P < 0.001). Implications: Among IDU in the community, periods of incarceration are strongly associated with HIV transmission behavior even after adjustment for risk behavior occurring within the community. These data support recent Canadian studies indicating HIV transmission in prison, and have public health implications for the communities after inmates release from prison. Ompad DC, Galea S, Wu Y, Fuller CM, Latka M, and Vlahov D Background: Two decades after widespread introduction of recombinant vaccine, Hepatitis B (HBV) vaccination prevalence remains low among drug users (DUs). We examined correlates of previous HBV vaccination among young injection and noninjection heroin, crack, and cocaine users in Harlem and the South Bronx, and the correlates of vaccine uptake and completion. Urban injecting and noninjecting DUs were recruited between 2000 and 2002. Participants completed a demographic and risk behavior questionnaire and were tested for HBV. Logistic regression was used to determine demographic and behavioral correlates of HBV vaccination history, uptake of currently-offered vaccine and completion of the three-part HBV vaccination protocol. Participants were nominally compensated for returning vaccination documentation. Results: Among 654 DUs enrolled (mean age 28.5, 72.5% male, 54.1% Hispanic, 34.0% Black, 20.2% injecting), 26.6% self-reported previous HBV vaccination, 16.8% demonstrated serological evidence of previous vaccination, 23.1% had been previously infected and 60.1% were susceptible to HBV at baseline. Correlates of previous HBV vaccination included younger age, 60% of new hepatitis C virus (HCV) cases, with annual incidence up to 30%. Therapeutic guidelines stress the need to identify factors impeding or promoting HCV therapy among IDUs. We determined factors associated with willingness to undergo HCV therapy among young IDUs. Methods: Eligible subjects were IDUs aged 18-35 testing HCV-positive and HIV-negative who injected drugs in the past 6 months in Baltimore, New York and Seattle. Baseline surveys collected data on behaviors, attitudes, depression (CES-D score ≥16) and readiness for drug use cessation based on Prochaska and DiClemente's stages of change (e.g., "determination" reflecting acknowledgement of problem drug use and intention to quit). Factors associated with willingness for HCV therapy were identified by logistic regression. Results: Of 265 IDUs studied to date, 77% were male, 59% White, 39% uninsured; median age was 26. Most (74%) indicated willingness to initiate HCV therapy; 78% thought HCV treatment was safe. Of 75 persons who visited a provider after testing HCV-positive, 31 (41%) were offered HCV therapy, but only 20 (27%) initiated therapy. Adjusting for gender, age and perceived safety of HCV treatment, odds of HCV treatment willingness was higher among those in the contemplation stage (Ad-jOR = 4.5, 95% CI: 1.5-13.8) or determination stage for drug use cessation (AdjOR = 7.5, 95% CI: 2.2-26.1). Implications: Barriers to HCV therapy among IDUs persist. Treatment of HCV infection should incorporate treatment for drug abuse since individuals ready to quit drug use appear more willing to initiate HCV therapy. Hagan H, Latka M, Campbell J, Golub ET, Garfein RS, Thomas DA, Kapadia F, and Strathdee SA Background: Although recent studies have demonstrated feasibility and effectiveness of interferon with ribavirin as treatment for HCV infection in IDUs, treatment access has been low relative to other patient groups. We studied treatment eligibility according to the 2002 NIH Consensus Guidelines for Management of HCV. Methods: Eligible subjects included HCV antibody-positive Baltimore, New York, and Seattle IDUs aged 18-35 who injected illicit drugs during the prior 6 months. To estimate the proportion who may be ineligible for HCV treatment, we assessed depression (Centers for Epidemiologic Studies Depression (CES-D) score >16 or 23), problem drinking (Alcohol Use Disorders Identification Test (AUDIT) score >8 or 10), recent drug injection and ALT levels. Results: Of 265 subjects, 67% had CES-D scores >16, 51% had CES-D scores >23. Forty one percent scored >8 on AUDIT, 32% scored >10. Fifty-three percent had ALT values above the upper limit of normal. Mean number of days since last injection was 7.8; 93% injected during the previous month. Of those with elevated ALTs, 63% may not be candidates for HCV treatment because of depression (CES-D >23) or problem drinking (AUDIT >10); using more restrictive criteria (CES-D >16, AUDIT >8, injection in past 30 days), the proportion rises to 89%. Implications: Increasing access to HCV therapy in this population will require concurrently addressing drug use, depression and alcohol use, which are all treatable conditions. Applying recent drug injection as an exclusion criterion would eliminate access to treatment for a majority of young HCVpositive IDUs in our sample. Latka M, Hagan H, Bonner S, Kapadia F, Hough E, Golub ET, Sherman S, Garfein RS, and Strathdee SA Background: Injection drug users (IDUs) account for up to 60% of new hepatitis C virus (HCV) cases. High HCV prevalence among IDUs is a serious health problem, and represents a hazard to uninfected peers due to the high transmissibility of HCV through blood. Methods: A randomized trial is being used to test a small-group, six-session secondary prevention intervention to reduce HCV transmission risk behavior by infected IDUs to their peers. Eligible subjects are HCV+/HIV-IDUs ages 18-35 who have recently injected drugs. Intervention content is delivered in the context of training participants to mentor peers about safer injection, which we hypothesize will reduce distributive injection equipment sharing among participants themselves. Classroom-based sessions cover HCV prevention information; communication skills-building; setting safe injecting examples; role play; guided practice and feedback. This cognitive-behavioral intervention includes a single, structured community activity where participants conduct HCV-prevention outreach to their peers, thereby stimulating social pressures on index participants to avoid lending used injection equipment. Results: To date 265 subjects (mean age 26; 59% white; 77% male) have been enrolled. Session attendance is high (80% in the treatment, 77% in the attention-control arm). Recruitment continues; preliminary results are expected in 2004. This paper, a companion to analytic submissions, will detail intervention theory and content. Implications: This is the first prevention trial for HCV+ IDUs, and one of a handful to supplement cognitive-behavioral methods with a structured activity to harness social influences for reducing risk behavior. If effective, this approach holds promise for reducing the disproportionate burden of HCV among IDUs and may translate to lower HIV incidence. Background: Individuals who struggle with substance use and HIV infection in New York City have been traditionally underserved or not appropriately served by common approaches to HIV care. Harm reduction is a well-established and valuable framework in treating individuals who experience the negative consequences of substance use. Despite the fact that harm reduction came into prominence as a public health approach and a clinical approach largely in response to the AIDS epidemic, in general, harm reduction is standing on its own in syringe exchanges and some community-based programs. This paper identifies harm reduction interventions and strategies that can be integrated within HIV primary care, and offers recommendations for how they can be integrated. Methods: A variety of harm reduction approaches that can be integrated in the HIV primary care setting are identified. The methodological approach advocated in this paper draws heavily on critical reflective practices within the field of education and the practice of mental health clinicians as well as the interdisciplinary harm reduction work of McCoy, Heller, and Cunningham. Implications: Implementing HIV harm reduction practice impacts the organization of service delivery, practitioners and patients. The role of the health care provider is more collaborative and less authoritative, and the role of the patient is more participatory and less dependent than in traditional models. The implications for quality and effectiveness of care under the proposed approach need to be further mapped and evaluated. Objectives: To determine factors that may influence clients to drop out of treatment. Methods: Clients were invited to complete a survey regarding their current and past experience. Demographic data, type of substance abuse, route of referral, therapy, and social supports were documented. Results: Preliminary results are available for the first 24 clients who were surveyed at SMWMS. 10 clients "dropped-out" (DO) and 14 completed the program (C). The mean age for each group was 41. The majority had a Grade 12 education (64%), was self-referred (67%), and was homeless (85%). On average, DO clients previously accessed the center 13 times compared with 2.5 in the C group (P < 0.001). There was a trend toward a lower average income in the DO group (DO $24,500, C $42,667). Alcohol (25%) and cocaine (21%) were the most abused substances. Reasons for leaving the program included drug cravings (43%), family problems (29%), and lack of desire to continue (29%). Indicators of social support did not indicate any differences between the groups, although there was a strong trend toward fewer family supports in the DO group. Implications: Detoxification programs serving Inner City populations have unique challenges. Increased services and supports may be needed to help address issues of homelessness and isolation. Focus groups may help meet the needs of specific groups seen. Wiewel EW, Go VF, Kawichai S, Beyrer C, Vongchak T, Srirak N, Jittiwutitikarn J, Suriyanon V, Razak MH, and Celentano DD Background: Thailand's rates of HIV due to drug use are increasing, and incidence among male drug users is ten times higher for injectors than noninjectors. Nearly one-half of clients at a drug treatment center in the north are from marginalized ethnic minorities called hill tribes, who are increasingly urbanizing and having contact with majority Thais. Understanding the context of hilltribe injection drug use is critical to illuminate patterns of HIV risk. Methods: A prospective study was conducted at the Northern Drug Dependence Treatment Center, Mae Rim, between February 1999 and January 2000. Trained interviewers administered a baseline questionnaire to 1865 people admitted to the center. Males from hill tribes with at least 50 participants (Karen, Akha, Hmong, Lisu, Lahu) were included for analysis (n = 629). Sociodemographics, drug use, and sexual practices were compared across hill tribes and assessed for association with injection. Results: Between 7% and 28% of each hill tribe had ever injected drugs (P = 0.002). Younger age, speaking Thai language, having a job with higher contact with Thais, and having had prior drug detoxification were statistically significantly associated with injection in multivariate analysis, while Hmong and Lisu ethnicity were statistically significantly protective. Implications: Using occupation and language as proxies, contact with Thais and concomitant acculturation may be a risk for injection in male hilltribe drug users. Harm reduction is imperative in hilltribe settings and in areas of ethnic mixing, as is the need to address the inadvertent consequences of acculturation and urbanization in hill tribes and other migrants. Background: Since 1999, CitiWide Harm Reduction and Montefiore Medical Center have collaborated to engage homeless People Living with HIV/AIDS through harm reduction outreach at 'welfare hotels' in New York City. This model's key features include: Integrated staffing (medical providers and peers); Evening outreach hours; Useful tools (e.g. syringe exchange, self-care kits); Tailored service op-tions; Consistent services, and; Supported transportation provision. Client-level service utilization data has been collected since 1998. Program theory is grounded in the belief that this model increases access to and engagement in care for this marginalized population. Methods: Service utilization patterns among participants engaged through outreach versus those engaged through walk-ins at CitiWideHR's drop-in center are examined. Analysis measures differences in engagement in care and related services for each group. Results: Preliminary findings confirm that among those contacted through integrated harm reduction outreach, 97% accessed medical care, while only 50% of those engaged as walk-ins accessed medical care. Further, among participants receiving harm reduction services, 93.8% also accessed medical care, while only 73.3% of those not receiving harm reduction services accessed medical care. Significant relationships are found for outreach engagement, and between healthcare access, harm reduction, and housing placement variables. Implications: Program theory is accurate in assuming the integrated harm reduction-medical outreach model increases access to and engagement in healthcare for this population. Reducing barriers to care by providing such consistent, tailored service options promotes effective engagement for marginalized populations. Kunins H, White A, Terlikbayeva A, Gilbert L, Matzdorf M, and Sturkey D Background: Despite mounting evidence indicating that intimate partner violence (IPV) is a key proximal factor associated with drug abuse and poor treatment outcomes among women, most drug treatment programs do not adequately address IPV among female clients. This study elicited attitudinal, structural and systematic factors that could facilitate or impede the implementation of a structural systems intervention, including conduction of routine screening and assessment of IPV, safety planning and effective referrals. Methods: Seven focus groups with 41 staff members from 40 substance abuse treatment programs in New York City were conducted from August 2002 through December 2002. Participants represented different treatment modalities, populations and positions in programs. Five focus groups were audiotaped and transcribed. Transcripts were coded independently by two investigators using themes developed in an iterative analytic process. Results: Preliminary analysis of themes suggests that the large majority of substance abuse treatment staff acknowledged that IPV was a common problem among their female clients negatively affecting their recovery. Participants across focus groups indicated a fear of "opening the box" of IPV due to their lack of training in handling IPV, concerns about detraction from focusing on recovery, and lack of referrals for shelters serving women with drug problems. Several participants raised problems of confidentiality and safety when batterer is a client in the same drug treatment program as the woman. Implications: The study findings underscore the need to develop and test IPV-specific protocols that may be implemented by providers in different substance abuse treatment settings. Millson P, Challacombe L, Strike C, Villeneuve P, Myers T, Fischer B, Shore R, Hopkins S, Pearson M, and Raftis S Objective: To determine whether reductions in drug use and HIV risk behaviors seen six months after enrolment in low threshold methadone programs are sustained at 12 months. Method: All new enrollees entering two low threshold methadone programs in Ontario are invited to participate in a prospective cohort study, completing interviewer administered questionnaires at baseline and at six monthly follow-ups, whether continuing in the program or not. Programs accept clients' treatment goal choices, whether abstinence or continued drug use. Mean number of days using each drug in the past month at baseline and follow-up are compared using paired t-tests. McNemar's test is used to compare proportions using each drug and sharing needles at baseline and follow-up. Results: By Dec. 2002, 78 participants completed 12 months follow-up (81%): 61% male, 39% female; injected a mean of 14 years at entry; 59% injected at least once a day; 12% shared needles, 26% shared drug paraphernalia and 36% injected in a shooting gallery. Significant declines in risk behaviors were maintained at 12 months, with further significant decreases in use of heroin, other opiates, and more than one substance per day. Implications: Significant drops in HIV risk behaviors and use of heroin and other opiates seen in this cohort six months after enrolment in low threshold methadone have been sustained in those fol-per month. Contingency table analysis and logistic regression were used to assess the relationship between religiosity and addiction severity (i.e., upper tertile of CDS score). Results: To date, of 130 IDUs, median age was 41 years, 66% were male, and 76% were African American. The majority (78%) reported a Christian affiliation, 9% Muslim, 7% other religions and 7% none. Most (94%) reported being religious/spiritual but only 29% reported attending a religious institution ≥ once/month. 9% reported religion as a motivator where as 2% cited it as a motivator for entering treatment, respectively. IDUs attending religious institutions > month were less likely to have CDS in the upper tertile compared to those attending less frequently (19% vs. 39%, P = 0.03). Adjusting for age, race, and gender, frequently attending religious institutions was associated with lower CDS (OR: 0.34, 95% CI: 0.09-0.74). Conclusions: These preliminary data suggest that religious communities may be an important source of support for IDUs entering drug abuse treatment. The role of religiosity in recovery from addiction requires further study. Mino ME, Oliver-Velez D, Deren S, Andia J, and Kang SY Background: While drug treatment and related services for injecting drug users (IDUs) in New York are often insufficient, in Puerto Rico, they are much scarcer. Studies have shown that some IDUs come from Puerto Rico to NY to obtain drug treatment and medical services and that these new migrants often practice riskier HIV-related drug injection behaviors than their NY counterparts. This study examines the availability of HIV and related services for new migrant Puerto Rican IDUs in East Harlem, NY and the kinds of obstacles that they face when seeking services. Methods: Qualitative interviews were conducted with service providers at East Harlem needle exchange programs, methadone clinics, drug treatment centers and other social service agencies about the availability of services for migrant Puerto Rican IDUs. Focus group and individual qualitative interviews were also conducted with new migrant IDUs about the types of obstacles they face when seeking services. Results: Interviews with service providers show that few programs targeted migrant Puerto Rican IDUs for services and that some HIV/AIDS services were unavailable in Spanish. Interviews with migrant IDUs show that while many came to NY for drug treatment, they had trouble accessing services, often because they did not speak English or know where to find the services. Implications: Culturally appropriate interventions are needed for new migrant Puerto Rican IDUs. This presentation will discuss the plans to develop an intervention that will train peer leaders to conduct HIV risk reduction with new migrants and sensitize service agencies about the needs of this group. Barwell J, Watson W, and Stade B Background: Caused by prenatal exposure to alcohol, Fetal Alcohol Spectrum Disorder (FASD) is a serious neurodevelopmental disorder. Individuals with FASD often demonstrate poor impulse control, problems in social perception, poor capacity for abstraction, problems in memory, attention and judgment. While the research describes the typical profile of the child and adolescent with FASD, there is a paucity of literature that describes characteristics specific to adults with FASD. It is anticipated that the study results will lead to more effective interventions for adults with FASD. Purpose: The purpose of this study, currently in progress, is to describe the typical characteristics of FASD that present in adults. Methodology: Sample and Setting: Adults 19 to 46 years old, referred to the St. Michael's Hospital (SMH), FASD diagnostic clinic since November 2002 (n = 15). SMH is located in the inner city of Toronto, Canada. Data Collection: A 13 page intake and diagnostic form, and a detailed physical examination were used to collect data on prenatal alcohol history, school and work history, behavioral problems, neuropsychological profile, FASD facial features, physical health, and growth of each of the participants. Data Analysis: Content analysis of the data obtained in the intake form and the written physical examination is currently being conducted. Results: The results of the study will present typical behavioral problems, mental health problems, work history, physical characteristics, and health problems specific to adults with FASD. Implications: Implications for practice and research will be discussed. Young MA, Galea S, Stuber J, and Ahern J Experience of discrimination has been shown to affect both physical and mental health. Few studies have assessed the role discrimination may play in shaping the health of highly marginalized groups, such as minority drug users. We assessed the relation between discrimination due to drug use and mental and physical health of minority substance users in New York City. We recruited 1,008 adults (≥18 years), who were currently (last 2 months) using cocaine, crack, or heroin. Participants were asked about their experiences of discrimination due to a variety of characteristics (e.g., drug use, race, gender), and about their current state of health. The majority of participants were male (63.9%) and single (62.3%); 41.7% were Latino, 49.7% were African-American, with a mean age of 40.4 (SD = 8.2). Discrimination due to drug use, poverty, and race were the most frequently reported forms of discrimination experienced (75.3%; 32.7%; 31.1% respectively), and were associated with lower mental health status (P < .0001; P < 0.001; P = 0.04 respectively), and higher number of self-reported diagnoses (P = .007; P = 0.009; P = 0.0002 respectively) in separate multivariable models adjusted for age, income, education, gender, race, social support, social networks, and drug use history. These observations suggest that discrimination experienced by illicit drug users is negatively associated with self-reported health, even when adjusting for drug use behaviors. Policies that reduce discrimination due to drug use behavior, poverty, and race may decrease the burden of illness experienced by this vulnerable group. Fuller CM, Galea S, Blaney S, Sisco S, Dorris S, Boyer A, Canales E, Fontenez E, Love G, Lasenberg L, and Vlahov D Objective: January 1, 2001, pharmacy sale of nonprescription syringes was legalized in New York State through the Expanded Syringe Access Demonstration Program (ESAP). ESAP Evaluation results have indicated differential use by injection drug users across New York City (NYC) communities. To identify possible social structural barriers due to community opinions, a random-digit dial telephone survey was conducted among East and Central Harlem, Bedford Stuyvesant, and South Bronx residents. Methods: Logistic regression was used to determine factors significantly associated with ESAP support. Results: Of 980 community residents surveyed, 45% were male, 48% African-American, 43% Hispanic, 8% white or other race. Average age was 42, and 77% were high school graduates. Of the ESAP-supporting residents, a higher proportion were white (68.5%) compared with African-American (26.5%), Hispanic (26.3%), and other race (40%;P < 0.001). Of ESAP-supporting residents, a lower proportion favored severe drug-user penalties (20.3%) vs. those opposed (41.0%;P < 0.001); a higher proportion had knowledge of syringe exchange programs (SEPs) (47.6%) vs. those who did not (26.6%; p0.001), and lifetime drug use (39.5%) vs. never (26.8%; P < 0.001). ESAP supporters tended to have higher education, income, and live in high social cohesion-scored neighborhoods. After adjustment, ESAP supporters were less likely to be Hispanics who ever experienced racism (AOR = 0.07) compared with nonsupporters. SEP knowledge (AOR = 2.7) and opposing severe drug-user penalties (AOR = 0.4) were also associated with ESAP support. Conclusion: Hispanics who have experienced race discrimination may perceive ESAP to further stigmatize their community; however SEP knowledge translates into support among all NYC residents. Similar race-sensitive community-wide educational efforts as with SEP should be duplicated with ESAP. Miller CL, Spittal PM, Li K, Laliberté N, O'Shaughnessy MV, and Schechter MT Background: Among young adolescents in North America, females represent approximately half of all new HIV infections and unlike other populations, this trend appears to be increasing. We undertook this study to determine HIV prevalence and to characterize HIV related vulnerabilities among young female IDU. Objectives: To determine gender differences in HIV status between young (aged <30) female and male IDU and to characterize associated risk factors within an ongoing cohort of IDU. Methods: IDU were recruited starting in May 1996 through a street level study site and outreach. Participants were administered questionnaires covering sociodemographic characteristics, risk factors, health status and service utilization and underwent serologic testing for HIV and HCV at baseline and semiannually thereafter. This analysis was restricted to young injectors (age 30 or less). Results: There were 520 participants who met the age criterion, of those 232 (45%) were female and 288 (55%) were male. Young female IDU were more likely to be HIV positive (25% vs. 14%; P = 0.001); Aboriginal (36% vs. 19%; P < 0.001); engage in sex trade work (71% vs. 19%; P < 0.001); require help to inject (59% vs. 35%; P < 0.001); use crack daily (19% vs. 11%; P = 0.023) and inject heroin (59% vs. 45%; P = 0.002), cocaine (50% vs. 38%; P = 0.009), and speedballs (25 vs. 16%; P = 0.048) at least daily. Young female IDU were younger (24 years [IQR: 20-27] vs. 25 years [IQR: 22-28]; P = 0.001), however there was no difference with respect to number of years injecting (4 [IQR: 2-9] vs. 4 [IQR: 1-8]). Conclusions: Despite younger age and similar number of years injecting, young female IDU were more likely to be HIV positive. There were a number of HIV drug and sexual vulnerabilities associated with being female suggesting that for young IDU, a gendered response to the epidemic is required. Miller CL, Li K, Braitstein P, Spittal PM, Frankish JC, Shoveller J,Wood E, Laliberté N, Montaner JSG, and Schechter MT Objectives: To determine the prevalence and incidence of HIV and Hepatitis C co-infection among young (aged 29 and under) injection drug users (IDUs) and to compare sociodemographic and risk characteristics between co-infected, mono-infected or negative youth. Methods: Data were collected through the Vancouver Injection Drug Users Study (VIDUS). To date, over 1400 IDU have been enrolled and followed, 479 were aged 29 years and younger. Semiannually, participants have completed an interviewer-administered questionnaire and have undergone serologic testing for HIV and HCV. Univariate and multivariate logistic regression analyses were undertaken to investigate predictors of baseline co-infection. Cox regression models with time dependant covariates were used to identify predictors of seroconversion for a secondary infection. A Cochran-Armitage Trend Test was used to determine risk associations across three categories; no infection, mono-infection and co-infection. Results: Of the 479 young injectors, 78 (16%) were co-infected with HIV and HCV at baseline and a further 45 (15%) with follow-up data became co-infected during the study period. Baseline positivity was independently associated with being female, Aboriginal, older age, greater number of years injecting, and living in the IDU epicenter. Factors independently associated with secondary infection seroconversion were borrowing needles, greater than once daily cocaine injection and accessing methadone maintenance therapy in the previous six months was protective. There were clear trends across the three categories for increasing proportions of females, Aboriginals, older age, greater number of years injecting, living in the IDU epicenter, and daily cocaine use. Interpretation: There were a shocking number of youth living with co-infection, particularly female and Aboriginal youth. The median number of years injecting for youth seroconverting to a secondary infection was 3 years suggesting that for high-risk youth the window of opportunity to prevent infection is exceedingly small. Background: Many studies recognize the relationship between sexual abuse, drug involvement and HIV/AIDS risk behavior, but few highlight or explore the ethnic differences which emerge. An understanding of these differences will contribute to the development of prevention and treatment programs likely to be effective among low-income women. This will also lead to longer lasting alternatives to arrest and incarceration. Methods: This study sample includes 821 women interviewed about self-reported drug use in a New York City (Manhattan Only; 1996-1997) booking facility as part of the Drug Use Forecasting (DUF) Program. A DUF-Manhattan supplement obtained information about sexual abuse and HIV/ AIDS risk behavior. Urine specimens were collected from DUF participants and tested using EMIT analysis for 10 drugs and for pregnancy. Results: Eighty-three percent were urine positive for at least one illicit drug. Twenty-five percent of women in the study sample reported at least one lifetime experience of sexual abuse. Logistic regression revealed that white women who reported sexual abuse were more likely to report involvement in sex work and injection drug use, suggesting that white women have a unique pattern of risk behavior resulting from sexual abuse. While many studies find that African Americans and Latinas are at the greatest risk for drug involvement and HIV/AIDS, the current study found that white women who reported adult sexual abuse were more likely to be involved in the high risk behaviors investigated. Understanding these differences can inform successful prevention, interventions and treatment programs for this ever growing population of women. Welsh S, Altenberg J, Balian R, Lunansky L, Magee W, and Welsh S Background: Services for substance use are typically provided by different agencies than services for mental health. These agencies usually demand that consumer's deal with the other issue before they receive services, e.g. consumers are told they must deal with their drug use before receiving mental health services. This leads to many drug users/consumers falling through the cracks of the service system and receiving inadequate health care services. Our research focuses on the health care service experiences and needs of urban dwellers with concurrent mental health and illicit drug use issues and examines the effect of integrating the philosophy of harm reduction into mental health services. Methods: This community-academic partnership and participatory action research (PAR) project utilized four focus groups of service users and two focus groups of service providers. Each group was asked a range of questions about their experiences with mental health and drug services. Focus group transcripts were coded using a constant comparative analysis and were analyzed using NVivo. Results: The following themes emerged from the focus groups with services users: the need for services that ensure safety, the desire that staff in service agencies possess a range of lived experience and knowledge as well as an attitude of respect, and the need for flexibility in services. Service users reported that most current services, in particular hospital-based services, have not met their needs. Service providers mention that a harm reduction approach is frequently undermined in a system that is governed by a medical model that requires "abstinence" before dealing with mental health problems. Implications: Our recommendations include integrating harm reduction philosophy into mental health services. A shared care model where psychiatric and community medical services are provided together and that also assists in providing direct education to service users and providers is ideal. This type of model has the potential to have a higher rate of success than current services. Friedman R, Keem M, Tempalski B, and Friedman SR Background: Street-level outreach tries to prevent HIV transmission and other complications that may arise from injection drug users (IDUs). Programs rely on both private and public funding. Methods: Qualitative interviews concerning services for IDUs were conducted as part of research on local HIV epidemics in 96 large metropolitan areas (MSAs) in the United States. Fifty-eight in-depth telephone interviews with researchers, public heath officials and service providers in 14 of the MSAs were examined for discussion of outreach. Results: Harm reduction is the prevailing philosophy advocated for outreach services by those interviewed. While most MSAs offer some harm reduction outreach for IDUs, most respondents considered services inadequate. They explained the challenges of obtaining funding for these services, particularly during a time of general budget cuts. In several MSAs, funding has never been accessible because harm reduction is viewed as condoning drug use. In metropolitan areas facing less opposition to harm reduction, funding problems included reallocation of money to programs targeting other populations, decreased funds for small community-based programs, and grants awarded to interventions more easily "proven" successful. Harm reduction outreach has survived using partnership with other programs, providing services that require minimal funds, and building strong relationships with the community. Implications: In the short run, these strategies can maintain harm reduction outreach services on small budget. Further research should study the ongoing impact of budget cuts on outreach viability and coverage and what strategies succeed in maintaining outreach services. Children who breathe comparable levels of ETS have higher relative levels of cotinine than adults. As the home is the major site of the ETS exposure, home smoking restrictions may greatly diminish exposure to children. Purpose: As Korean Americans (KAs) have one of the highest rates of smoking, a study of household smoking restrictions was conducted. Methods: A community convenience sample of KAs recruited from Korean-specific community centers, religions institutions and cultural events (n = 272) were surveyed with assistance as needed. Results: 62% of KA men reported having smoked <100 cigarettes; 53% are current smokers. 7% of KA women reported smoking <100 cigarettes in their life; 3% are current smokers. 57% of KA women reported smoking is completely prohibited in the home, 22% reported some restriction, and 21% reported no restrictions. 52% of KA men reported smoking is completely prohibited; 33% reported some restriction, and 15% reported no restrictions. Females <50 years, who have health insurance were more likely to have some smoking restriction (OR = 3.4, CI: 1.2-10.0 and OR = 0.3, 1.9). Males who had a check up < 2 years and a college education of more were more likely to live in smoke-free homes (OR = 4.0, CI: 1.3-12.0 and OR = .2, CI: .05-.6)]. Conclusions: The high number of homes with some smoking restrictions indicates that culturally sensitive targeted tobacco control interventions to promote smoke-free homes may further decrease ETS exposure to children and may provide adults' incentives to quit. Kwon SC, Kim S, and Roh H Background: Previous research in California has documented that tobacco advertising is more pervasive in Asian American (AA) neighborhoods than in other ethnic neighborhoods; no data exist for NYC. Purpose: This study proposes to document the amount of point-of-purchase (POP) tobacco advertisements in AA communities in NYC and explore the relationship between neighborhood characteristics and the level of POP tobacco ads found. Methods: 161 tobacco retailers in six predominately AA neighborhoods were surveyed by members of a Korean teen tobacco education program. Neighborhood socioeconomic factors were compared. Results: 78% of the surveyed stores had some exterior tobacco advertising and/or promotional items clearly visible from the street; 87% of the stores ha d some interior tobacco advertisement. More tobacco displays were found that reported in national POP tobacco surveys. Of the interior ads, 21% were placed near candy displays, and 23% were located at 3 feet or below (eye-level with a child). 40% of the poorest surveyed neighborhoods had 1 or more Marlboro ads, the most heavily advertised brand, compared to 23% of the richest neighborhoods (P < .01). 43% of the wealthier neighborhoods displayed cigarette products near candy compared to 54% of poorer areas (P < .001). Other neighborhood effects such as number of foreign-born, English proficiency, and education attainment of residents were explored. Implications: Virtually all surveyed stores in these predominately AA neighborhoods displayed some type of tobacco advertisement with sores in the poorest neighborhoods disproportionately represented. Study findings are being used to implement community education and intervention to address these issues. Pindera C Background: Drug use is a critical risk behavior for women. Despite high rates of diverse and complex health needs, little is known regarding injection drug users' (IDUs) perceptions of or context for their health needs. This study's primary emphasis is to give "voice" to the understandings and experiences of the everyday lives of women IDUs in Winnipeg. Methods: A purposive sample of 8 women IDUs was selected for this study. A qualitative research design approach adapted from Carolyn Wang's Photovoice Methodology was used. This assisted with the identification of how, from the woman's perspective, her context and life experiences functioned to create meanings to explain her risk-related behaviors. Unstructured interview, participant's photographs and life narratives were the data collection methods used. Results: Women's drug use was connected to the personal and social circumstances in which they lived. Their drug use was linked to the process of managing losses and to live with integrity under threatening life conditions. The women relied heavily on drugs as a means to escape the painful feelings resulting from childhood and on-going traumas. They took responsibility for their decision to use drugs; but also held a common perception that social and physical environments limited their ability to manage their drug use and find alternative coping strategies. Conclusions: This study demonstrates that research sensitive to women's feelings about their social environment is required to provide effective and meaningful services for women who use drugs. Knowledge gained from this study can be used to assist in the development of appropriate health and social services strategies and interventions. Implications: Addictions treatment and interventions that enhance social supports in marginalized populations should be further explored since implementation of such programs may reduce the risk of hospital discharge AMA. Lankenau SE and Clatts MC Background: Injection drug users (IDUs), an enduring hidden population given the illegality and stigma associated with streets drugs and syringes, are an important population to reach from both epidemiological and harm reduction standpoints. Ketamine, a "club drug" popular among youth as an intranasal anesthetic, has been identified as a drug increasingly common among a new hidden population of IDUs. Methods: We used participant observation and a semistructured interview guide to recruit a sample of 40 young Ketamine injectors (18-25 years old) from two sites in the East Village (the "Cube" at Astor Place and Tompkins Square Park) located less than one mile apart. The interview guide focused on Ketamine injection and recent drug using events. Results: Despite the close proximity of the two recruitment locations, an analysis of the narrative accounts and descriptive statistics are revealed important differences between the youth interviewed at each site. In particular, two distinct subpopulations were uncovered: The Tompkins Square Park recruits (n = 13) were typically female, mobile, older, homeless IDUs who had fewer material resources and initiated ketamine injection outside of New York. In contrast, the Cube recruits (n = 27) were more commonly youth of color, younger, less mobile, novice IDUs who were more resourced and initiated ketamine injection in New York. Implications: These findings point to the emerging trend of ketamine injection among young injectors. Additionally, the results demonstrate that ethnography is an effective research methodology for uncovering hidden populations and describing risk behaviors among new and varied populations of IDUs. Podymow T, Turnbull J, and Yetisir E Background: Chronic homeless alcoholics suffer increased health problems and have high use of emergency services. There is a low likelihood of rehabilitating chronic public inebriates. Harm reduction is a policy to decrease the adverse consequences from substance use without requiring abstinence. The Ottawa Inner City Health Project (ICHP) is a shelter-based project created to deliver health care and harm reduction to homeless adults. This includes a managed alcohol program (MAP). Methods: Chronic public inebriates were referred by a community panel. MAP is shelter-based, used as stable housing, and alcohol is dispensed on an hourly basis. Hospital charts were retrospectively reviewed for all emergency visits and admissions for three years prior and two years of program enrollment and were statistically compared. Trends in bloodwork were analyzed. A questionnaire was administered to subjects and staff regarding patterns of alcohol use, health and activities of daily living prior to and while in the program. An economic evaluation is ongoing. Results: Seventeen adults (15 males, 2 females), average age 50.7 years were enrolled in the MAP for a mean of 16 months. Mean duration of chronic alcoholism 35.2 years; ten subjects reported regular use of nonbeverage alcohol. ER visits decreased from a mean total of 13.5 visits/month to 8 visits/month (P = .004) and bloodwork values did not change significantly while in the program. All subjects reported alcohol consumption was less than while on the streets, and the majority perceived improved hygiene and health. Client care staff noted improved hygiene, nutrition, compliance to medication and attendance to medical appointments in the majority. Conclusions: A managed alcohol program as a harm reduction measure in chronic homeless alcoholics can stabilize alcohol intake, improve quality of life and significantly decrease ER visits. Strathdee SA, Bishai D, Huettner S, Cornelius L, and Latkin C Objective: This study estimates the value that clients place on drug rehabilitation services at the time of intake and how this value varies with the probability of success and with the availability of social services. Method: During intake we asked 114 drug rehabilitation clients to state a preference among three hypothetical methadone programs that varied across 3 domains: weekly fee paid by the client out of pocket ($5 to $100), presence/absence of social services, and time spent heroin free (3 to 24 months). Subsequently each subject was asked if they would enroll in their preferred choice. Each subject was asked to complete 18 orthogonal comparisons. We compute the median expected willingness to pay as the weighted average fee with weights being the probability of enrollment from both raw and adjusted multivariate logistic models. Results: The median willingness to pay for a program that offered on average 10 months of heroin free time was $9.50 per week, rising to $15.33 per week for programs that offered 24 months of heroin free time. The availability of social services increased median willingness to pay by $4.81 per week. The fee was the most important predictor of the probability of enrollment with a price elasticity of -0.57 (SE 0.62). Conclusion: Clients median willingness to pay for drug rehabilitation falls short of the average program costs of $60 per week, which reinforces the need for continued subsidization. Clients will pay more for higher rates of treatment success and for the presence of social services. Background: Providing effective substance abuse treatment remains one of the major challenges of urban health. Women enter substance abuse treatment with a complex set of social and medical problems including increased rates of co-morbid mental health problems and traumatic exposures that have been associated with poor substance abuse treatment outcomes. However there are few resources available to front-line substance abuse providers to assess these problems. To address this need we conducted a study of the predictive validity of a brief screening instrument we designed for use by substance abuse treatment counselors. Methods: 354 screens were administered to women entering substance abuse treatment and 80 received an independent DSM-IV diagnostic assessment. Analyses were conducted to assess the predictive validity of the screening questions for psychiatric diagnoses including major depression, bipolar disorder and PTSD. These included four one-sided, 2.5%-sized, stratified tests of the null hypotheses for the diagnoses of interest, and estimates of the 95% confidence intervals for the sensitivity and specificity of the screen for each condition. Results: Participants reported significantly higher rates than the general population for mental health symptoms (88%) and exposure to violence (89% intimate partner violence, 71% sexual assault, 61% childhood abuse) with significant race/ethnic differences. Analysis of the screen validation will be presented. Implications: The use of a brief, easy to administer, screening instrument may be an important tool for front-line substance abuse treatment staff to provide the best care for women in substance abuse treatment. Hagan H, Thiede H, and Des Jarlais D Background: HCV infection in IDUs is an important urban health problem, with high prevalence of infectious carriers in most IDU-populations. In this study, time to hepatitis C virus (HCV) seroconversion was estimated in initially-seronegative injection drug users (IDUs), to assess the window of opportunity for prevention of HCV infection, and examine factors that affect timing of infection. Methods: 484 HCV antibody-negative IDUs were followed a median of 2.1 years to observe HCV seroconversion. Time to seroconversion was examined in relation to subject characteristics, using the Kaplan-Meier method and Cox's proportional hazards regression. A weighted average time to HCV seroconversion was calculated among new injectors (injecting <2 years) using seroprevalence and seroincidence data. Results: There were 134 HCV seroconversions (11.6/100 person years at risk; the 25th percentile of time to seroconversion was 26.2 months). Daily injection (adjusted hazards ratio (AHR) = 1.4), injection with a syringe used by another IDU (AHR = 1.8) and sharing a cooker or cotton (AHR = 1.7) were significantly associated with time to HCV seroconversion. Using the estimate of the mean time to seroconversion in new injectors (4.8 years), and the median duration of injection in new injectors who were HCV-positive at enrollment (1.2 years), the weighted average time to seroconversion after beginning to inject was estimated to be 3.4 years. Conclusion: The period of susceptibility to HCV infection in the majority of IDUs appears to be sufficiently long to justify the allocation of substantial resources toward interventions that may reduce drug injection frequency and injection-related risk behavior in these individuals. Feldman B, Fowler H, and Chin N Background: Health disparities between racial populations in the United States are known to transcend financial access to medical care. It is widely accepted that these health disparities are in part facilitated by attitudes and beliefs among racial minority populations, yet there have been few studies characterizing racial minorities' attitudes and beliefs toward healthcare. Methods: Minimally structured patient-centered interviews of thirty African American, Hispanic, and American Indian adults living in urban shelters or visiting urban soup kitchens. Interviews centered on attitudes and beliefs toward healthcare among these populations. Results: Recurring perceptions were that providers are prejudiced and stereotype patients, are poor listeners, spend too little time with patients, and inappropriately use patients as teaching material for health care providers in training. Participants also expressed a wide range of beliefs about generic medications, organ donation, HIV/AIDS, and medical research. Participants described the characteristics of their ideal doctor and suggested ways in which current health care providers could improve their interactions with patients. Implications: The attitudes and beliefs described by participants serve as significant barriers towards accessing medicine and contribute to healthcare disparities in the United States. Strategies to decrease health disparities should address both patient-provider interactions and community interactions that serve as the sources of negative perceptions among these populations. Continued research is needed to further characterize the views of marginalized populations that contribute to health disparities. Background: This is a qualitative, grounded theory study about the ways in which residence in a low-income, predominantly white, urban neighborhood affects the health and well-being of women. This research used formal qualitative research methods to explore how the urban environment influences parenting stress and shapes behaviors related to child health and development. Methods: Data collection included in-depth interviews that incorporated standardized measures along with open-ended questions, journals, photographs, and neighborhood checklists, as well as analysis of Census and city data. Research focused on 10 mothers living in Pigtown, a low-income, predominantly white neighborhood in Baltimore, Maryland. Additional data were collected during individual and group interviews with community service providers, business owners and employees, and longtime residents. Results: Findings clustered around three main themes: (1) that stressors are generally routine but also include confrontation with frequent and unpredictable crises and that this combination of routine stressors and crises are part of several overlapping domains, (2) that environment plays an important role in women's lives and that this environment includes not only the immediate neighborhood of residence but also the larger city and sociopolitical community, and (3) that women have few social supports or mechanisms for coping with stressors and crises. Implications: Policy and programmatic recommendations focus on efforts to promote neighborhood stabilization and pride, improvements to core city services, improvements in community health, provision of affordable entertainment for families, and income support and asset development. Background: "Increasing Outreach and Decreasing Disparities in African Americans" (IODA) is a three-year demonstration project funded by the American Cancer Society (ACS) to reduce disparities due to colorectal cancer among African Americans in two urban settings: Bridgeport, CT and East Baltimore, MD. Methods: The ACS hired and trained two community outreach specialists (COS) to implement a seven-step outreach process. After interviewing over 100 leaders identified through a key informant process, COS distributed a written summary and presented highlights at well-attended community fo-rums. Community planning groups are implementing action plans appropriate to identified local needs and assets. Results: Community planning groups in the two sites have developed and implemented interventions that include community education, advocacy among policymakers and creation of affordable, accessible cancer screening services. Preliminary evaluation data indicates increased colorectal cancer screening in participating communities. Final data collection and analysis will be completed in 2004. Implications: The IODA project has been effective in opening doors to two African American urban communities experiencing a disproportionate cancer burden. A trusting relationship has been built through outreach with community residents, and work is underway on locally appropriate cancer control initiatives. This process can be replicated in any community. Simich L, Beiser M, Mawani F, and Wu F Background: This presentation will describe perceptions of health and social support among recent immigrants to Canadian cities from Hong Kong and Mainland China using data from the multisite, qualitative research project, Multicultural Meanings of Social Support among Immigrants and Refugees. The presentation will focus on the role of culture and social context on shaping perceptions of social support and its impact on mental health among Chinese immigrants in Toronto. Methods: Phase 1 consisted of in-depth interviews with 60 policy makers and service providers, Phase 2 of in-depth interviews with 120 Chinese immigrants and Somali refugees, and Phase 3 of focus groups with all stakeholders to generate policy and program recommendations. Results: Chinese immigrants described challenges and frustrations, particularly in finding suitable employment, finding information, navigating the urban social environment and recreating supportive networks in Canada. They expressed impacts on mental health including stress, depression, low self esteem, and need for emotional support. Distinct difference emerged between Hong and Mainland Chinese immigrants (as well as between immigrants and refugees) with respect to expectations of supports, coping strategies, and concepts of health. Implications: Immigrants judge social support by its effectiveness in overcoming specific obstacles. Absence of social support is keenly felt in declining mental health and reduced potential for social integration. Analyzing the role of culture and social context in shaping perceptions of social support adds significant dimensions to our understanding of social support and social capital in immigrant communities and suggests ways to enhance supportive health and social services. Lai W, Uddin S, Brown JB, and Nisker J Background: Advocacy is one role that physicians are expected to fulfill. We sought to elucidate the practical meaning of advocacy in medicine and apply it to medical education, so that future physicians include advocacy in their daily practice. Methods: This qualitative study used both in-depth interviews and focus groups with persons qua patients, community service providers, physicians, and medical students. To elucidate community needs, focus groups were recruited from local community agencies. The medical community was also consulted. A semistructured interview guide was used to guide the discussion about the responsibility of the physician as advocate, motivations and barriers to advocacy, skills and knowledge required, and how to teach advocacy. Transcripts were coded independently by two investigators and analyzed iteratively for common themes. Results: Emerging themes include three areas. First, communities expressed a need and an expectation that physicians be a voice for them when necessary. The second domain addressed the modes of advocacy, from individual patient care to public/political roles. It includes both motivations and barriers. Third, an integrative model of advocacy training was described. It includes multiple levels of advocacy education and promotion, starting at selection for medical school, continuing with patient-centered teaching, and reinforcement in clinical teaching and role-modeling. It is perpetuated in continuing medical education and faculty development. Implications: Knowledge, skills, and positive attitude towards advocating for marginalized patients can be promoted by education, experience, and exposure at all levels of medical training. A multimodule curriculum on advocacy would prepare physicians to fulfill their advocacy roles. Creatore MI, Glazier RH, Agha MM, and Moineddin R Background: Income distribution and relative income differences may affect health independent of absolute income levels. There is inconsistent evidence, however, of such effects on mental health. We investigated whether income disparity between recent immigrants and other residents was associated with mental health hospitalizations. Methods: Income disparity was defined as the percent difference between the mean individual income of recent immigrants and that of all other residents of each census tract in south-central Toronto, Canada. Sex-specific mental health hospitalization rates were generated using 1996 Canadian Census denominators for four groups representing different levels of relative and absolute income differences. Results: Recent immigrants' income was 37% (range: 15-55%) less than other residents. Mental health hospitalization was highest for women (4.0 per 1000) in the least advantaged group (low incomehigh disparity), and for males equally in the least advantaged group and in the high income-high disparity group (both 3.1 per 1000), though neither finding was statistically significant. Implications: Preliminary results suggest that males experience higher mental health hospitalization rates in areas with high income disparity regardless of the overall neighborhood income. In females, a combination of material disadvantage and relative disadvantage resulted in the highest rates. Due to the rarity of mental health hospitalizations in the study area, further analysis using a larger geography is necessary to confirm these results statistically. Significant income disparity exists for recent immigrants in Toronto and this may be related to mental health particularly in males and low income females. Background: This presentation will explore the relationship between women living with HIV/AIDS and their experiences with gender-based violence with the goal of (1) identifying the characteristics of HIV-related violence and (2) promoting the appropriate legal intervention services of a local urban service provider. In doing so, this study engages a holistic and broadly defined notion of health. Methods: A qualitative research methodology was employed consisting of eight in-depth audiotaped interviews with women participants totaling fifteen hours of tape. An interview guide was created and tested. The researcher also followed the narrative strategies of each participant. Six African-American and two Puerto Rican women ranging in age from 32 to 49 were interviewed. The participants were clients of a Bronx service provider. Results: Important patterns of abuse were identified that provide a window into understanding the complexities of women's HIV-related violence. Seven areas of HIV-related violence were identified: (1) Housing (2) Substance Use (3) Sex and Pregnancy (4) Disclosing HIV Status (5) Access to Benefits (6) Orders of Protection (7) Custody Implications: The study will suggest that there is a significant need for legal intervention in addressing HIV-related violence. Several steps could be taken. Training of both lawyers and case managers in identifying HIV-related violence and pursuing appropriate questions is a critical first step. Equally important is the need to encourage the creation of a range of referral services. Furthermore, coordinated efforts between program areas would greatly benefit HIV+ clients as would interagency efforts addressing the relationship between HIV/AIDS and gender-based violence. This could help close the gap between those providers that focus either on HIV or violence. Background: Daily mortality is typically higher on hot days, and certain groups experience disproportionate risk. Air-conditioning has been recommended to mitigate heat-related health effects. We examined whether air conditioning (AC) prevalence in US cities is associated with mortality on hot days and explains any disparities by age and race. Methods: Poisson regressions were fit to daily mortality (1986) (1987) (1988) (1989) (1990) (1991) (1992) (1993) in Chicago, Denver, Detroit, Minneapolis, New Haven, Pittsburgh, and Seattle. Predictors included natural splines of time, barometric pressure and day of week; and linear terms for particles and mean daily apparent temperature, lag 0. Separate models were fit to death counts stratified by age (>65 years, <65 years) and race (Black or White) to derive the percent change in mortality at 29o C, relative to 15oC. City-wide AC prevalence was also evaluated for influence on this effect, and further stratified by race and age. Combined effect estimates across all cities were calculated using inverse variance weighted averages. Results: Deaths among blacks were more strongly associated with hot temperatures than deaths among whites; age did not modify the effect. White households had over twice the AC prevalence than black households in most cities. AC prevalence differed little by age. Cities with higher AC prevalence experienced 40% lower heat-related mortality, and AC prevalence explained part of the differences in heat effects by race. Implications: Improved access to air conditioning, in addition to efforts to reduce social isolation and address urban heat island effects, may reduce disparities in health responses to hot weather. Akiyama C As the population of Asian/Pacific Islander Americans (APIA) continues to rise in the U.S, so do their healthcare needs. Unfortunately, not all APIA are as uniformly educated, acculturated, and financially stable, as the myth of the "model minority" would have us suggest. Although adults from many nationality groups between APIA have adapted well to life in the U.S, serious problems have emerged among our youth. In particular, gang violence in the APIA community has increased dramatically nationwide. In Los Angeles County California alone, there are currently 155 Asian gangs (membership over 6,000). In neighboring Orange County, gang involvement has reached an all time high with over 65 documented gangs (membership over 2,000). Even more disturbing is the increase of Asian females involved in gangs. In Orange County, where the Asian gang population makes up 12%, there are 140 Asian female gang members, up 60% from last year. The author interviewed 400 gang members out in the streets, jails, and juvenile halls, using a target questionnaire; concomitantly went a step further disguised as a gang member. This study identified several distinct differences between the ideology and the "state of mind" between Asian male/female gang members. Moreover, the author identified six contributing factors, which lead to involvement in gangs (i.e., lack of adult supervision, family breakdown, victimization due to racism, culture shock, need for survival, and monetary profit). The purpose of this study is to present timely data on APIA youth gangs; offer strategies for prevention/intervention to help control this rising problem. Crafton C and Ibrahim Q Background: There exist differences in traditional attitudes towards medicine among the immigrant communities. The differences, though less dependent on religious views, are largely due to differences in economic and cultural traditions. Secondary migration of immigrant and refugee communities into urban areas with resettlement into extended clan and family groups makes these communities somewhat 'hidden' populations. It is essential for these populations to be related to with a needs assessment based approach that focuses not only on community deficiencies, but also cultivates the resources and talents within the community. This pilot project was funded through the Minnesota Women's Foundation and written by Leadership, Development and Empowerment Group (LEAD). LEAD realizes that generating community system building efforts, both from within and from outside of the community and strengthening the existing community organizations are of paramount importance. Women are traditionally and have been the health care providers, as well as the brokers and protectors of their families. There are a large number of women who are first-time heads of their households, who cannot read or write, and do not know about navigation of the health care system. They are also in need of health care services and use them more frequently than men, particularly during their childbearing years. The goal of the study is to increase the access to health care services, particularly breast and pelvic cancer screening, develop relationships with targeted communities, and health care providers systems, and educate the providers about culturally competent health care for these communities. Metropolitan State University School of Nursing with support from the Metropolitan State University Community-Based Learning Center partnered with LEAD on all aspects of this project. Methods: The seven focus groups with East African older women, over thirty-five years of age, met at their perspective community organization's centers. The participants were gathered by the facilitator/ interpreters from these communities including the Eritrean Community Center of Minnesota, Confederated Somali Community of Minnesota, and the Oromo Community Center. The Sudanese community was contacted several times through community leaders but declined to participate in the project. The women participants were asked questions about health perceptions, knowledge, beliefs, barriers to care, and their interaction with preventive care and the health care system. Results: The focus group findings inevitably reflect the experiences and views from the women themselves. However, these findings provide some insight into East African immigrant women's perceptions, barriers, and experiences of the health care system in Minneapolis-St. Paul. The majority of women stated that they seek health care only when they are sick. Women were not aware of needing screening or have knowledge about illness for screening. Many said they waited until symptoms persisted for several weeks or until they were very ill before seeking care. Many women expressed that good health is a strong value that is embedded in the cultural and religious traditions of their ethnic communities. A few women expressed distrust of the American health care system because they believed that as refugees they were not given the best care or that they were subjected to experimental treatments or unnecessary surgery. Older women speaking in a group of women in their own community center, gave these women, some of them community elders, an opportunity to speak in a safe setting. Implications: This information will be incorporated into development of health promotion, and culturally competent interventions. Interventions will be focused on the provider as well as community for education on the importance of preventive health care practices and screening, and to encourage women to discuss practices that they use to maintain or improve their health that are consistent with their culture. Idioms, sayings and unintended outcomes will offer insights into women's health beliefs. Barry J and Breen N Medical services are not evenly distributed by geographic location; consequently, not all women receive cancer preventive services. We evaluate the usefulness of three different residential indictors for identifying areas in Atlanta, San Francisco and Detroit where women are not receiving preventive cancer services. Using cancer registry data (SEER) matched to US Census data, we found that residence in a Census tract that met the federal government's definition of a medically underserved area (MUA) increased the probability of a late-stage diagnosis of breast cancer. Additionally, residence in an extremely poor tract (poverty rates 40% or more) or in an underclass tract (rates of high poverty and social distress) increased the probability of a late-stage diagnosis for both breast and cervical cancer. However, not all of these tracts have been designated as Medically Underserved Areas (MUAs). This study suggests there is a need to reform and broaden the criteria traditionally used by the federal government for designating MUAs. Existing federal programs need to be integrated so that geographic areas in need of cancer screening services can also become neighborhoods where follow-up care is available. How race/ ethnicity plays an independent role in determining diagnostic outcomes depended on the city of residence. More research is needed to examine differences in the availability and delivery of health care and possible discriminatory practices against minority women. Bayoumi AM and Hoch JS Background: Advocates for disadvantaged populations have expressed concerns about how costutility analyses can incorporate concerns about fairness and equity. We examined how a decision maker's choice would vary depending on definitions of justice. Methods: We imagined a decision maker with a fixed budget to be spent on one of three interventions: highly active antiretroviral therapy (HAART) for HIV, sildenafil for erectile dysfunction, or augmentation therapy for alpha-1-antitrypsin deficiency. The cost-effectiveness of these interventions was $23,000, $11,200 and $207,800 per quality adjusted life year (QALY), respectively, based on published studies. We examined eight decision rules; four neglected efficiency considerations, but allocated resources according to health and demographic outcomes, two used only measures of efficiency, and two used "weighted QALYs" to incorporate equity concerns. Results: Without considering efficiency, sildenafil was preferred when the decision maker's objective was to maximize QALYs, augmentation therapy was preferred when individuals with the worst quality of life were given first priority, and HAART was preferred when individuals with the lowest expected survival or those with the youngest mean age were given priority. Considering efficiency, sildenafil was preferred when outcomes were measured as QALYs, while HAART was preferred when outcomes were measured as life years. Results were generally unchanged when QALYs were weighted by patients' quality of life, but tended to favor HAART when QALYs were weighted by patients' baseline quality-adjusted survival. York City. Among cases with a known motive/perpetrator, forty percent were intimate partner femicides. While the rate of nonintimate partner femicide has decreased between 1990 and 1999, the rate of intimate partner femicide has remained stable. Conclusions: Intimate partner femicide exhibits a unique epidemiology, as compared with homicide more generally, and represents a point of prevention of premature mortality among women in New York City. This study demonstrates that young, foreign-born, and ethnic minority women are overrepresented among intimate partner femicide victims. In order to reduce deaths among these population subgroups, funding of intimate partner violence against women prevention and intervention activities must be increased. The New York City experience reveals that surveillance data can have such an impact on funding and program design decision-making. Semogas D, Cleverley K, Rice C, Roelofsen D, Jensen C, Hjartarson K, Thomas H, and Szatmari P Background: The M.A.C. Door (Making a Change) program is an academic and community collaboration that applies an incentive based approach to assist homeless street youth to successfully leave the street environment. However, what constitutes quality of life within the context of living on the street is not known. The purpose of this paper is to: 1) describe quality of life as perceived by youth as compared to housed youth and 2) examine variables associated with variations in quality of life (e.g., time on the streets, level of education, age, gender, and motivation to change as well as perceptions of physical and mental health) Methods: Approximately 50 street youth are currently being surveyed using The Quality of Life Profile: Adolescent Version Questionnaire. The 54 item instrument covers nine areas of life under three major sections: Being (physical, psychological, spiritual); Belonging (physical, social, community) and becoming (practical, leisure, growth). Each of the 54 items is scored according to its importance to the adolescent and level of satisfaction with current status. Additionally, the nine areas of life are rated according to level of control of areas of life and available opportunities for change. Results: Pilot data indicate that the tool is acceptable to youth and provides variation in scores. The poster will describe data on 50 homeless youth compared to controls. Implications: Understanding how street youth perceive their quality of life and knowing whether certain factors affect their quality of life will assist with interventions to help those ready to exit the street. Background: Stigma regarding mental disorders is quite prevalent. Recently, the Internet has been used for depression screening and we used the Internet to study mental disorder stigma beliefs among depressed Asian Americans (AA) and whites. Methods: We placed the CES-D depression scale on the Intelihealth website. All scoring above the depression cutoff score, were asked about stigma relating to friends, family, and employer. Using AN-OVA and ANCOVA, we compared AA to whites and we also stratified by AA category to compare AA stigma beliefs with regard to urban status and region of the US. Results: Sample size was 1,839 AA and 66,820 whites. AA had higher mean stigma scores than whites in all three areas. Stratifying for AA, there was a trend (P = .055) for urban AA to have higher mean stigma scores relating to family than non-urban AA. No differences were noted for stigma relating to friends or employer. Percentages for those indicating either "agree" or "strongly agree" for each stigma category were the following: family (urban = 44.2%, non-urban = 39.1%), friends (urban = 54.3%, non-urban = 51.7%), and employer (urban = 72.0%, non-urban = 67.4%). Also, no US regional differences were noted for all three stigma areas. Implications: Urban AAs have greater stigma beliefs than non-urban AA. This may be because those in urban areas are less acculturated and have a higher AA ethnic density than those in non-urban areas. Culturally sensitive mental disorder screening by health care providers may help detect and improve the health of those urban AAs who may initially be hesitant to discuss their mental disorder. Lewis S, Cano I, and Watkins BX Background: Health disparities are associated with a wide array of factors such as race, gender, environment, socioeconomic status, and culture. However, the underlying mechanisms of these factors and how they work together is not fully understood. Longitudinal data on the lifespan of elders is needed to better understand the multitude of factors that effect elderly health and well-being. This exploratory study examines the social and cultural dimensions of health in two ethnic Mew York City communities. Methods: Qualitative and quantitative data on Central Harlem, Manhattan and Boro Park, Brooklyn between 1950 and 2000, were collected, including detailed personal information from a sample of Black and Jewish elders, environmental neighborhood data, and historical data regarding access to social and medical services. Vital statistics, census data and other documentary evidence were used to create demographic profiles of each community over the course of five decades. Results: For Harlemites, the years between 1950 and 2000 are marked by community disintegration, increases in crime, physical and social decay, and ill health, while the same years are marked by community integration, decreases in crime, physical and social improvements, and good health in Boro Park residents. Preliminary data shows that while both communities are similar in age distribution, educational attainment, and income distribution, elderly Harlemites death rates are almost twice the death rates for the elderly in Boro Park. Implications: Understanding how cultural differences inform beliefs and shape social interactions is critical for developing public health interventions that can reduce health disparities across racial and ethnic groups. Background: Lack of health insurance, insufficient coverage, and underutilization of US health system are some of the main problems faced by immigrant populations in the US. The migratory experience is characterized by stressful circumstances (e.g., lack of English proficiency, undocumented status, etc.) that dually impact on immigrants' physical and mental health. Because of access barriers, many immigrants do not receive mental health treatment despite their needs. Specific Aims: This paper aims to develop hypothesis in two areas: (1) immigrants' access barriers to mental health services; (2) identification of immigrants' perceptions of their mental health needs expressed via indirect ways, such as somatic susceptibility (e.g., abdominal pain caused by nervousness) Methods: Data will be obtained from in-depth interviews (from an original sample of 50 immigrants in NYC) and field notes from participant observation conducted in Argentine social enclaves. Results and Implications: This paper will develop the notion of "silent demand" to refer to Argentine immigrants' awareness of their mental health needs that are not translated into an active request for, and access to, mental health services. Immigrants' uninsured status, lack of financial means, and their unfamiliarity with the US health system frequently discourage them from seeking mental treatments. However, immigrants' everyday difficult circumstances also make them wish for professional support that could help them cope with everyday stressors. These results could lead to the design of health interventions addressing immigrants' access barriers to metal health services, including the consideration of their own perception of symptoms and their unsatisfied demands. Hussain TM The economies of many migrant-sending countries would have collapsed, and those of the migrant receiving countries would not have attained their impressive economic performances were it not for the presence of women migrant workers in urban areas. There is an increasing trend towards the feminization of migration in Asia's mega-cities. The increase in the participation of women in the regional labor migration (from 15% in 1976 to 27% by 1987) revealed an upward trend in the 1990s, and is evidenced by the feminization of migrant work in Asia. For instance, overseas Filipino workers increased from 61% in 1998 to 68% in 1999. The increasing number of women migrant workers engaged in the informal labor sector, both as domestic help and as sex workers, raises health issues. While the vulnerability of migrants to HIV and other infections is recognized, their lack of capacity to negotiate safe sex and their continuous denial of rights to make free choices regarding their bodies exacerbate their risk of contracting HIV. Access to health care is not considered a right of migrant workers in receiving countries. For undocumented migrant workers, their fear of arrest effectively restricts the use of medical facilities, especially that of state-run hospitals. Lack of poor access to health care by migrant workers precipitates their vulnerability to HIV. In essence, the right to health is the right to life, and of which no migrant worker or migrant family should be deprived. The UN convention on the Protection of the Rights of All Migrant Workers and Members of Their Families provides a comprehensive guideline to the treatment to which migrants are entitled as guest workers in a host country. The paper will attempt to look deeply at the current situation by using various international sources and information, and to provide a forum for discussion and recommendations. All cases were geocoded according to address of residence. Data on population density, risk areas using household income levels from neighborhood block groups were obtained, respectively, in the 2000 census report and from the specific study carried out by the municipality. Time and space clustering were analyzed using the scan method of Kulldorf. Results: High rates of hospital admissions (112.8) and readmission (25.0) per 100,000 children aged 0-5 were observed. Age and gender did not differ by hospitalization and rehospitalization rates. A remarkable seasonal distribution was observed for admission but not for readmission rates. Space clusters coincided with high risk areas for low socioeconomic profile either for admissions or readmissions; as well as a high correlation between the occurrences of both events with low socioeconomic profile. Implications: We found high rates of asthma admissions and readmissions especially for those living in critical areas of the city represented by low socioeconomic populations at high risk of exposure related to symptoms triggering and exacerbations for asthma. These findings might be related to low health care access and utilization. Grant T, Soriano Y, Nelson I, Williams E, Remirez D, Burg J, Marantz P, and Nordin C Background: Inner city immigrant and mixed ethnic communities are known to be a high risk for cardiovascular disease and diabetes. Such communities may also be underserved for preventive medical care. We hypothesized that (1) Community based screening, using partnerships between physicians and community groups, can detect new cases of diabetes and persons at risk for diabetes and cardiovascular disease; (2) Hemoglobin A1c (HbA1c) can be used as screening test in this setting; and (3) The result can be used to identify subgroups at particular risk. Methods: Screening took place in neighborhoods throughout the Bronx at five types of locations: shelters, group homes, street fairs, and outpatient clinics. Data were analyzed for number of cases of new diabetes (HbA1c >7%), persons at risk for diabetes (Hba1c 6%-6.99%), effect of ethnicity of participant, and type and location of screening. Results: HbA1c and lipid profile were obtained on 539 persons at 23 screening, which formed the cohort for this analysis. Mean HbA1c for the cohort was 6.00%. 32% of cohort had HbA1c >6%, and 11.4% had >7%. Excluding known diabetics (n = 59), 24% had HbA1c >6% and 3.4% had >7%. There was significant effect of location within the Bronx on the value of the HbA1c for the cohort as a whole (P = 0.032). The mean HbA1c for persons living in the south Bronx (which has a higher poverty income ratio) was significantly greater then those living in the North Bronx (6.08±1.38% vs. 5.74± 1.01%, P = 0.013). When we analyzed the value of HbA1c as function of type of screening, we found a significant group effect (P < 0.001). Post hoc analysis showed that persons screened in shelters had significantly lower HbA1c than those screened in all other locations (5.46±0.89%) (n = 105)vs 6.13± 1.39% (n = 434, P < 0.001). Persons who identified themselves as Hispanic had slightly higher HbA1c than Caucasians and Black, but the difference was not significant. Using the data, we found the highest yield for persons at risk for diabetes was from nonshelter screenings in the South Bronx: in this cohort (n = 181), 32.9% (n = 59) had HbA1c >6% and 5% (n = 9) had a new diagnosis of diabetes. Conclusion: Community based screening with HbA1c can be used as a method for identifying high percentages of patients at risk for diabetes or with undiagnosed diabetes in an inner city, immigrant, mixed ethic population. Targeting specific populations by neighborhood and type of screening may increase yield for diabetes and cardiovascular risk. Background: There is a high prevalence of hypertension with very low rate of awareness and inadequate control of blood pressure among Korean-Americans. Hypertension predisposes to all of the major cardiovascular disease outcomes including cardiac failure, stroke, coronary artery disease and peripheral artery disease. Without early intervention in the high normal or stage 1 hypertension, the costly cycle of managing hypertension and its complication will be continued. Therefore, there is a need to establish a systematic approach for community-based strategies to guide prevention, treatment, and control of hypertension with emphasis on differences in culture, heritage, and local influence in Korean-Americans. Methods: The purpose of this two-group, repeated measures quasi-experimental pilot study was: to examine the effectiveness of a culturally relevant, theory-based lifestyle modification intervention in hypertension prevention and control for community dwelling older Korean-Americans from baseline and at 3 moths after initiation of intervention. The sample was drawn from 5 community organizations serving Korean-Americans. Initially mini-health screening was done on 156 persons with aged 50 or over. Thirty-seven persons consented among 60 who met the inclusion criteria and evaluated the effects of intervention on knowledge, health beliefs and attitudes, diet pattern and physical activity level, and BP, BMI, and cholesterol and compare between intervention group and comparison group. Result: T test and Multivariate Repeated Measure ANOVA for comparison of variables from Baseline (TB) to Evaluation Time (TE) between 2 groups were done. The results showed that there were significant differences in knowledge, diet patter and physical activity level at P < 01, and more reduction of blood pressures in intervention groups. However, there were no significant changes in health beliefs and attitudes over time. Implications: Even though the study being a pilot with very small sample size, the finding is significant in that the culturally relevant intervention plays an important role in modifying diet patterns and physical activity, which results in reducing BP levels. Knowledge seems to be important mediator in this process. Additional research is needed to strengthen implications for practice and further the design of culturally relevant interventions for hypertensive management in vulnerable populations. Kwon SC, Siegel K, and Senie, R Objective: The purpose of this exploratory descriptive study was to elicit the illness representation of breast cancer from a community sample of Korean American (KA) women living in New York City. Methods: Weinman's Illness Perception Questionnaire, based on Leventhal's model of illness representation, was modified and adopted to reflect the study population, and then pilot tested to ensure accuracy and comprehension of the instrument. The survey was administered with assistance as needed to a convenience sample of 150 Korean American women over the age of 18 who attended community health fairs, cultural events, religious institutions, and senior centers. Results: Respondents under the age of 50 years were more likely to believe that treatment is effective in curing the cancer (OR = 2.4, 95% CI 1.1-5.6). Individuals who engaged in breast screening were less likely to believe that there is little one can do to get well from a cancer diagnosis (OR = .1, 95% CI .0-.7) and less likely to believe that one gets breast cancer due to chance (mammography OR = .09, 95% CI .01-.9; self breast exam OR = .06, 95% CI .01-.5). Conclusion: Preliminary analysis suggests that individuals with a negative view of cancer treatment and diagnosis are less likely to engage in breast cancer screening behavior. Creating interventions and health education also address how breast cancer is conceptualized and to specifically target held misconceptions on the illness may increase breast cancer screening practices among Korean American women. Fahs M, Shelly M, Qu C, and Burton D Background: Little is known about health risks in Chinese Americans, the fastest growing immigrant group in the US. The discrepancy between adult male smoking rates in Mainland China (67%) and among Chinese Americans (33%) raises questions about the relationship between acculturation and health risk behaviors. This paper presents baseline data, from an NCI funded longitudinal study, on the impact of acculturation on health risk behaviors among Chinese Americans. Methods: In-person interviews using a comprehensive household-based survey are conducted with 2500 adults ages 18-74, constituting the largest probability sample of Chinese in the United States. Interviews are conduced in English, Mandarin, Cantonese and other dialects. Two NYC communities, with a 100% increase in their Chinese populations over the past decade, constitute the sample frame: Sunset Park, Brooklyn, a relatively recent immigrant community; and Flush Park, Queens, a more established immigrant community. Results: Chinese male smoking prevalence is 50% higher than general population smoking prevalence in both communities (34% in Sunset Park, and 31% in Flushing Park). Preliminary data (n = 712) indicate that in 15.3% of households, English is the primary language spoken at home. Other acculturation measures include: years in US, use of Western versus Easter medicine; and use of language-specific media. The paper presents results of descriptive and multivariate analyses of the relationship of acculturation to health indicators including smoking, physical activity, obesity, alcohol consumption, general health status (SF12), and health risk knowledge, controlling for socioeconomic and demographic characteristics. Implications: This paper provides the first population based analysis of the relationship between acculturation and health risk disparities among Chinese Americans. Findings will inform public heath practice targeting effective health initiatives among urban immigrant populations. Background: A project exploring health literacy and health behaviors is interviewing 200 Latinas age 40+ of varying nationalities in New York City. Methods: Participants complete a survey administered in Spanish and the TOGHLA-S (Test of Functional Health Literacy in Adults, in Spanish). Results: Of the 169 interviewed thus far, 51% had elementary education or less, 58% no health insurance, 38% no regular source of health care, and 7% were US-born. TOFHLA-S scores indicated a population with low health literacy: 27% had inadequate health literacy in Spanish, 20% marginal, and 53% adequate. Women with an elementary education or less an older women were significantly (P < .001) more likely to have inadequate health literacy. Only one of the 15 women age 65+ had adequate health literacy. Characteristics not associated with health literacy included years in the US, nationality, insurance status, having a regular source of health care, and marital status. Women with inadequate health literacy were significantly (P < .01) less likely to have heard of a Pap smear or to ever have a pelvic exam. No association was found between health literacy levels and the number of visits to a health care provider in the last year. Implications: Almost half of the Latinas we studied will have difficulty interpreting medical materials, even if made available in Spanish. Therefore, providers who care for Latinas age 40 and up, regardless of their nationality or insurance status, need to be aware that relying on written materials, even in Spanish, may not be an effective way to deliver health messages. This vulnerability is compounded if the woman is an adolescent. The concern for this population was illuminated by the starvation death of a newborn living in a shelter with his young mother. St. Michael's Hospital (Toronto) together with the community group, Young Parents of No Fixed Address, has developed a comprehensive program to reach this population. Purpose: To develop strategies to make obstetrical care more accessible and attractive to street youth thereby improving the health outcomes for mothers and their infants. Methods: The St. Michael's designated team, including Social Workers, an Acute Care Nurse Practitioner, Family Physicians and Obstetricians work along with Public Health Nurses to facilitate obstetrical care. The team's collaboration with City Hostel Services has helped establish a priority shelter system for this population. The Rotary Centre, a haven located in the emergency department, provides short term respite until care is arranged. Results: Process and outcome evaluations will report on numbers seen, birth outcomes and efficiency of service access. Action research using qualitative methodology will explore feedback from the women themselves. Implications: Realistically a safe place to birth may be all that can be achieved. However, with the collaborative efforts of the hospital and community, it is hoped that the health of both the mother and newborn may be maximized and that they can go on to become a successful family. Bonner S, Matte T, Rubin M, Fagan JK, Ahern J, and Evans D Background: Under use of inhaled corticosteroids and overuse of inhaled or nebulized bronchodilators has characterized the medical management of asthma among children in socially disadvantaged communities. The frequency of oral bronchodilator use has not been documented. Methods: We interviewed 149 parents/guardians of children aged 2-5 years old enrolled at subsidized preschools in East and Central Harlem, NYC. The children were identified through a brief respiratory questionnaire as having asthma or probable asthma. We classified 76 (51%) as having current persistent asthma based on frequency of respiratory symptoms in the last 14 days and/or short-acting inhaled bronchodilator use in the last 4 weeks. Results: Only 17 (22%) of children with current persistent asthma used controller medications regularly-of whom only 2 used inhaled corticosteroids. 44 (58%) used oral bronchodilators in the last 4 weeks, often (68%) in conjunction with short-acting inhaled or nebulized bronchodilators. Use of oral bronchodilator was associated with more severe symptoms. Implications: This study is the first to document the prevalence of oral bronchodilator use for treatment of children with asthma. We found a reliance on oral bronchodilators in a pediatric population where the impact of asthma is great and appropriate asthma controllers (i.e., inhaled corticosteroids) are underutilized. Oral bronchodilators, with slower onset and more systemic side effects than short-acting inhaled bronchodilators, are not the recommended treatment for quick relief. The side effects of reliance on oral bronchodilators and their impact on asthma prevention and control deserve further study. Houseman C Violence is a major urban health issue. Being a victim of or even being a witness to violence is considered to be a major risk factor in developing mental health problems. On a larger scale, the prevalence of violence in cities saps human and social capital and results in tumbling real estate values and decreased quality of life. Interventions to prevent violence are generally developed from a single disciplinary perspective, e.g. psychological or sociological. Such a perspective is limited in tackling such a complex problem rooted in the many areas currently being explored by different fields. This paper will propose a multilevel, multidisciplinary model for researching the topic of violence. Examples of research methods and results of studies at each level will be summarized in order to produce a more integrated model to guide research regarding the phenomenon of violence. Information from biology, psychology, sociology, and other disciplines will be used to develop a true urban health approach. Contributions to the problem from hereditary factors, intrapersonal phenomena and interpersonal experiences will be discussed. Included will be areas of research related to families of origin, present family and organizational interactions. The role of community disorganization will also be outlined including the contribution of neighborhoods from both a citizenry and urban planning perspective. The overarching economic and cultural factors as they relate to understanding violence will also be discussed. Suggestions regarding the use of coalitions in researching, planning and evaluating violence interventions will be presented. Ways of involving the public and professionals in the prevention of violence will be discussed. Markham Piper T, Galea S, Ahern J, Tardiff KJ, and Vlahov D Homicide rates in New York City (NYC) have decreased dramatically in the past decade. We were interested in examining if changes in homicide rates are associated with urban neighborhood income and income inequality. We used homicide data from 1990-1996 from the Office of the Chief Medical Examiner of New York City (OCME). We calculated median income and the Gini coefficient for 59 city neighborhoods from the 1990 US Census. Homicide rates were stratified by fourths of income and income disparity and compared over time. Overall, the highest homicide rates were in the poorest neighborhoods and in neighborhoods characterized by the greatest income inequality. However, over the 7-year period, the greatest decreases in homicide rates occurred in the poorest neighborhoods and in those with the greatest income inequality. In neighborhoods with the greatest income disparity, homi-cide rates decreased from 76.98 to 29.21 deaths per 100,000, a 62% decrease. In neighborhoods with the least income disparity, there was a reduction from 17.2 to 9.2 deaths per 100,000, a 47% decrease. In 1996, homicide rates remained 5.7 times higher in the poorest neighborhoods than in the wealthiest neighborhoods and 3.2 times higher in neighborhoods with high inequality compared to neighborhoods with low inequality. These findings suggest that there was differential reduction in homicide in urban neighborhoods with different incomes and income distribution; variable enforcement efforts or sociodemographic changes may partly these observations. Additional research is needed to explore mechanistic relations between urban neighborhood context and homicide. Lyman JM, McGwin G, Malone DE, Taylor AJ, Brissie RM, Davis G, and Rue LW Background: The purpose of this study was to determine the epidemiology of homicide among children younger than six years in Jefferson County, Alabama. In 1990, nearly 90% of Jefferson County's population lived inside an urbanized area. Birmingham, the largest city in Alabama, is located in Jefferson County, the most populous county in Alabama. Methods: Cases included deaths attributed to homicide or that were of undetermined intent among children younger than age six who were born and died between January 1, 1988 and December, 31, 1998. Victim and offender characteristics were obtained from Jefferson County Coroner Medical Examiner Office records. Environmental factors and circumstances surrounding the death were also noted. Results: The 53 victims were primarily female (55%), Black (69%), younger than two years of age (85%), had single mothers (38%), and a history of abuse (53%). Offenders were more likely to be male (64%), Black (73%), and the victim's parent (53%). Homicides primarily resulted from an angry impulse (61%), with hands the most common weapon (61%). Implications: The majority of homicides resulted from an angry impulse. Two possible intervention methods that might prevent child homicide in this type of at-risk family are crisis child care and home visitation by trained nurses. Home visitations during pregnancy and the first two years of life reduce the frequency of child abuse and neglect among firstborn children of unmarried adolescents of low socioeconomic status. This may be critically important considering 85% of the cases in this study involved children younger than age two. Holtsclaw E, Avila S, Valencia Y, and Williamson M Background: Families in the Trauma Department suggest that there are few outlets to express emotions during a time of crisis. Research further indicates detrimental emotional and physiological effects of suppressing feelings. As a result, staff established a journaling program to address these concerns, improve the quality of services, and facilitate coping during times of acute stress. Cook County Trauma Department is a Level 1 urban trauma center. Patients typically are men from underserved neighborhoods. There is no program provide the acute emotional support of patients and their families. Method: Family members consenting to participate in the journaling program were interviewed about coping mechanisms. The journal provided for expressive open-ended writing and recording thoughts and feelings. Follow-up interviews were conducted about the experiences journaling and impacts on the coping and grieving process. Data from the interviews was synthesized and themes were extracted. Results: Six family members of trauma patients participated, three African American and three Latino women. The women find it difficult to talk about their problems and many used unhealthy means to cope, such as increased smoking, eating, or sleeping. During the program all of the women actively used journaling in individual ways. During the follow-up interview participants stated that the process aided in successful coping. Implications: The Trauma Department aspired to gain insight into the coping skills of family members of urban trauma patients. Expressive writing has been documented to assist in coping with life stressors, however; little data is available for applying journaling in an urban trauma setting. The results from the project suggest that journaling improves coping for family members during acute life stressors. There are direct implications for the design of support systems in an urban trauma setting. Further research is needed to better understand the applicability of journaling in an urban trauma center. Luke's-Roosevelt Hospital Center in New York City has developed a collaborative project to deliver culturally sensitive IPV services to HIV positive patients. The project seeks to improve CCC staff capacity to identify and respond to IPV. Collaborators include a community-based organization (CBO) for crime victims, a theater-based training program, and a program evaluation center. Methods: An eight-hour training, which utilized didactic and interactive learning models, was developed for primary care and multidisciplinary staff of the CCC. Existing training manuals were adapted to create empowerment-based didactic portions. Interactive learning was accomplished through theaterbased training. Results: Integrating different orientations of a CBO with a hospital-based HIV center was challenging. The hospital "culture of medicine" values empiricism, while the CBO "culture of social work" emphasizes experiential learning. We learned it is critical to address provider resistance by "speaking their language", using the "evidence" to encourage adoption of skills being introduced, and enlisting "scientists" to present data that support the need for a new standard of care. We also learned that the theater-based method was a more effective learning tool for staff of particular disciplines. Implications: Creating a culturally sensitive IPV program requires not only sensitivity to the cultures of our patients, but also the ability to examine the biases we bring from the various professional cultures to which we claim membership. Castelino-Pinto C This exploratory project sought to study the posteviction well-being and safety of women who "stay put" in their site of battering and evict their batterers. The findings are based on qualitative analysis of two-hour long interviews with 36 battered women in New York City, who responded to fliers posted in all the police precincts, victim service offices, and emergency rooms in New York City hospitals. Some participants were recruited from special pendant alarm programs (coordinated community interventions). Respondents included women who fled, as well as women who were still living with their batterers. Their narratives revealed that those who "stayed put" and evicted their batterers were motivated and fortified by their longstanding socioenvironmental connections to their housing and neighborhoods. Posteviction, some women solicited the vigilance of doormen even in the neighboring buildings, some estranged neighbors became the much needed eyes at the back, some building managements levied and enforced entry bans against the batterers, while some women in special programs were not extended that same level of support from the police/district attorney. The presentation will address the transformation of site-of-abuse to home. It will also highlight the need for tailored neighborhoodlevel interventions directed at inculcating an intense, "not in my backyard" kind of intolerance and resistance to battering-the kind faced by local governance when proposals of waste disposal facilities and homeless shelters are presented to neighborhood boards and block associations. Methods: We collected suicide data from 1990-1998 from the Office of the Chief Medical Examiner (OCME) of New York City and assessed trends and determinants of firearm-related suicide deaths. Results: Between 1990 and 1998, there were a total of 5171 suicides in New York City; 1074 of these deaths (20.8%) were firearm-related suicides. Firearms were the third leading cause of suicides, after hanging (1316 deaths or 25.5%) and long falls (1218 or 23.6%). Over time, total suicides decreased from 650 in 1990 to 491 in 1998 (a 24% decrease); total firearm-related suicides also decreased from 131 in 1990 to 102 in 1998 (a 22% decrease). The proportion of firearm-related suicides remained unchanged, however, (between 20.2 and 20.8%) in the period studied. In multivariable modeling, characteristics of suicide decedents that were associated with a greater likelihood of firearm suicide among all suicides were: male (OR = 3.48), black (OR = 1.29 compared to white referent) and Latino (OR = registered society dedicated to reducing injury and death of youth resulting from MVCs. The mission of the society is to create a vibrant network of students and adults across the province working together to prevent car crashes. Purpose To reduce injuries and death in the youth in British Columbia due to motor vehicle collisions. Methods: Joining the C.A.R.S BC network, four grade 11 students have formed a team to begin a 2-Year MVC prevention strategy in their high school. Focus will start with 4 main road safety issues: speeding, aggressive driving, distracting loud music and cell phone use. With help of an adult supervisor and support from their school, the teen team will plan events, design and produce posters and brochures, book guest speakers and organize fund-raising activities. These students will also participate in the annual C.A.R.S. BC provincial youth conference further developing networks and generating enthusiasm about Road Safety initiatives. Summary: Implementation of an injury prevention program in high schools which involves teens of eligible driving age will help reduce injuries and death in British Columbia caused by motor vehicle collisions. Background: Most batterers' programs are based on the Duluth Model (Pence &Paymar, 1993), which is built on the premise of a universal culture that teaches men to dominate (Everingham & Schenk, 1995) . The model has been criticized for omitting the importance of substance abuse and its contribution to violent behaviors (Young, 1999) . The model also fails to adapt batterers' programs to the cultural environment of the intended population. Miami-based PAiirS, a federally funded program to devise prevention strategies for domestic violence, established the first culturally designed voluntary men's group as a primary and secondary intervention. Methods: The participants are inner city African American men. The 12-week curriculum evolves around interpersonal skills. A satisfaction survey is used to assess intervention effectiveness. Results: Eight out of the eighteen males registered for the group complete all sessions (44.4% retention rate). Participants reported to apply skills learned in various aspects of life. They also expressed that the programs' cultural design and the cofacilitators' nonjudgmental attitude increased their level of trust. Implications: The voluntary support groups indicate a good initial response and a nontraditional intervention that could potentially decrease the level of domestic violence. The voluntary aspect of the program yields more participation and better learning predisposition. The absence of negative connotations increases openness and trust among participants. By self analyzing behaviors, men are more likely to make more consistent changes. Steele LS, Glazier RH, Lin E, Austin PC, Moineddin R, and Mustard CA Background: In 1996, the government of Ontario instituted a 21.6% reduction in welfare benefits to most recipients. We sought to examine the relationship between the benefit reduction and rates of mental health service use in the inner city of Toronto, 1992-1998. Methods: This study was an ecologic time-series analysis using generalized estimating equations to explore neighborhood quarterly rates of mental health service use by deciles of welfare dependency. Data sources included a municipal public assistance database (1996) , census data (1996) and a provincial health insurance claims database (1992) (1993) (1994) (1995) (1996) (1997) (1998) . The setting included 1220 census enumeration areas comprising the inner city of Toronto, Canada. We measured 1) visits to psychiatrists per 1000 claimants 2) individuals with a visit to a psychiatrist per 1000 claimants 3) mental health visits to general practitioners per 1000 claimants and 4) individuals with a visit to a general practitioner per 1000 claimants. Results: There were no significant relative SES differences in rates of mental health service use before the policy change compared with rates of mental health service use after the policy change. Rates of psychiatric service use were higher in neighborhoods with the lowest levels of welfare dependency {Rate Ratio: women = 3.21 (95% C.I.: 2.64 to 3.89); men = 2.81 (2.24 to 3.54)}. number and proportion with a physician visit in the year prior to admission, proportion with a physician visit 14 days prior to admission and 14 days one year before admission. Results: Age-sex adjusted ACS admission rates were considerably higher in the lowest income quintile than in the highest (rate ratio = 1.7, 95% CI 1.6, 1.8), yet income differences were not found in annual physician visits (17.3), percent with a visit (96.2) or usual source of care (84.9%). Relatively fewer in the lowest income group had a visit prior to ACS compared with marker admission (Q1:Q5 ratio 0.86, 95% CI 0.64-1.08) and compared with one year before admission (Q1:Q5 ratio 0.78, 95% CI 0.61-0.93). Implications: We found a large excess in ACS admissions with social disadvantage despite good access to care. Lack of timely care prior to admission may be an important cause of excess ACS admission among disadvantaged groups. Kiesel R, Markham Piper T, and Galea S Access to health care has been shown to affect the health of urban populations. This paper provides a systematic review of the issue of physician distribution in urban areas in the United States. We conducted a thorough search of the peer-reviewed and grey literature regarding the issue of intra-urban physician distribution. Although the evidence for physician distribution is limited, most published work suggests that physician maldistribution exists in urban areas and that the reasons for this maldistribution are complex. Physicians cluster close to academic training centers, preferentially practice in higher income areas, and tend to work in areas where persons of similar racial/ethnic backgrounds work and live. Many interventions have been implemented in order to address physician distribution in urban areas. Policy changes such as financial incentives for physicians locating to underserved urban areas, increase in inner-city medical school funding, and the use of foreign medical graduates have all generated mixed results. There are several methodological and logistical issues that make the study of physician distribution in urban areas challenging. The variations in the definition of "urban" in different national and international contexts make cross-study comparisons difficult. Determining the optimal number of physicians for particular urban areas is also difficult due to intra-urban differences in morbidity and the mobility of both physicians and urban residents. The projected increase in urban populations in the coming decades suggests that questions pertaining to access to equitable care for urban residents are timely and important. Additional research is needed to understand the factors and solutions related to physician maldistribution to improve the health of urban populations. Hirschman JA, Fuller JF, Barnickol L, Srinivasan G, Singh J, and Justicz J Background: High-risk infants and their families frequently encounter barriers to important medical and welfare services and may often need legal intervention to access these services. Hospital-based legal aid programs could be effective in improving access to critical services for these infants. Yet, few such programs exist, and none have been rigorously evaluated. This presentation presents the results of a randomized controlled trial of a post-NICU case management and legal services (CM/LS) intervention at two large, inner city hospitals. Methods: To date, 122 infants have been enrolled and randomly assigned to receive either CM/LS or routine care. Infants/families in the CM/LS group are screened by a case manager for their eligibility to receive health and welfare services. Referral to a lawyer (on-site in the clinic) is made if the family is unable to obtain a health/welfare service for which they were eligible. All infants are followed for 1year corrected age. Instruments are administered to measure: compliance with scheduled care, rehospitalization and ED visits, growth and developmental outcomes, utilization of services, maternal quality of life and parental stress. Results: Data analyses address the questions: (1) does providing CM/LS improve the high-risk infant's receipt of services after NICU discharge, and (2) does it improve outcomes for the high-risk infants and mother after NICU discharge? Implications: Providing on-site legal aid may be one effective strategy in improving access to critical services for high-risk infants, but further research is needed to identify underlying cause(s) and to develop effective strategies to reduce the barriers in access to health and welfare services. provider suspicion, 47% on scores less than 2SD on the Bayley or OWPVT, and 17% by parents themselves or outside sources. 6.7% of 15-month-olds and 40% of 24-month-olds had been referred to EI. Significant predictors of EI referral included male gender and a baseline measure of family stress. Implications: The remarkably high rates of EI referral resulting when standardized testing is combined with usual sources of identification suggests that in poor urban areas developmental delay may often go undetected, and underscores the importance of developing broad, low-cost public health approaches to prevent or reduce delay. 12149 Predictors of Satisfaction with Primary Care and Improving Homeless Men's Access to Primary Care: a Randomized Controlled trial Svoboda T, Bloch G, Knowles H, Hwang S, Glazier R, Kelly M, Chow K, Jones K, and Tynan AM Background: Many homeless men at Seaton House, Canada's largest shelter for homeless men in inner city Toronto, remain on the street due to lack of coordination between health and shelter care resulting in a cycle between shelter, hospital, and street. Barriers to care include difficulties in finding a physician, setting up appointments, and reduced physician propensity to provide care to the homeless. A randomized controlled trial supported by the City of Toronto Supportive Communities Partnership Initiative (SCPI) will be conducted to provide some understanding of the use of primary care resources in the homeless population. Methods: A database of primary care physicians within close proximity of Seaton House will be created and a referral list of physicians/sites willing to accept new clients, developed. Seaton House clients who fit study criteria will be referred to a randomly selected primary care site for an appointment or to usual hostel care. Both groups will be followed up with structured interviews at 4 weeks and 6 months. Results: Groups will be compared to determine the propensity to use primary care vs. emergency department sites before and after the intervention. Data will be analyzed to determine if a primary care referral program impacts on client use of and satisfaction with primary care services. Implications: The study will provide an understanding of the use of primary care resources in the homeless population and help identify client and physician specific barriers to the access of these services. This information can further impact policy development and programming for this vulnerable group. Islam N Background: Taxi drivers, who comprise a workforce of over 100,000 in NYC, largely uninsured and experience low access to health care. A partnership between a university health project and the NYTWA, a taxi workers rights group, was created to address health care access issues for NYC taxi drivers. A survey developed in conjunction with NYTWA and administered at LaGuardia Airport found 80% of drivers were uninsured, 25% had never had a routine checkup, and 20% had experienced a barrier to obtaining health care in the last year; a decision was mad by the partnership to initiate a health campaign to increase access to care for drivers. Methods: A 3-pronged approach was initiated by the partnership to increase health access: (1). Organize the first-ever health fair for taxi drivers; (2). Enroll eligible drivers into government and community-sponsored health insurance programs; and (3). Use survey data to advocate for health benefits for uninsured drivers. Results: The health fair held in July 2000 at the JFK Airport holding lot served over 300 drivers and offered over 15 different preventive screenings, and a resource list of low cost clinics and hospitals. The partnership has begun mobilizing drivers to enroll into available health insurance programs, and survey data has been presented to the NYC Taxi and Limousine Commission in support of health benefits for uninsured drivers. Implications: Future plans include biannual health fairs, further advocacy for an industry-owned health fun; and provision of health education materials to drivers. By using a direct serve and advocacy tactic this successful partnership has increased health care access to a largely underserved community. The Bush administration has proposed adding 1,200 new and expanded community health centers over the next five years, doubling the number of patients treated at them. About half the patients treated at the community health centers have no insurance coverage, while many other have limited or inadequate coverage. The purpose of this paper is to describe how Geographic Information Systems (GIS) can be used to maximize eligibility for federal Medically Underserved Area (MUA) designation. Methods: GIS was used to maximize areas that are eligible for federal MUA designation. The process included: obtaining a "clean" list of primary care providers by specialty and census tract, identifying the practice sites and calculating the number of full-time equivalent practicing physicians, identifying the poverty rate, percent of population over 65, and infant mortality rates, computing Primary Care Physician-to-Population Rate, and redrawing Rational Service Areas. Results: Approximately 24 Rational Service areas in Los Angeles County, ranging in population from 75,000 to 150,000 each, were identified using GIS that have an Index of Medical Underservice score of 62.0 or less, making them eligible for federal MUA designation and receive federal funds to establish or expand community health centers. Use of GIS increased the number of persons residing in areas eligible for federal MUA designation from 1 million to over 2.5 million. Implications: GIS provides a process for maximizing geographic areas populations that are eligible for federal MUA designation and opportunity to access federal resources to increase access to care for medically undeserved populations. Background: The configuration of hospitals, and specifically, their types and locations, strongly influence overall access, cost, and quality of health care in US cities. In large districts with few physicians in private practice, hospitals have become vital providers of ambulatory care, particularly to African-Americans. Methods: Using published data and local informants on 51 large and mid-size US cities, we have identified acute general and other special hospitals open in 1936, 1950, 1960, 1970, 1980, 1990, 2000, and 2003 . Some 1,200 hospitals have been tracked. We have identified closings, relocations, mergers, conversions to nonacute uses, and other reconfigurations. To predict hospital reconfigurations and assess their implications, we have identified: financial characteristics of each hospital at the start of each period (margin and reserves), operating characteristics (beds, occupancy rate, efficiency), demographic characteristics of residents of surrounding census tracts, competing hospitals, and other characteristics. Results: The African-American share of the surrounding population was one of the most salient predictors of closing. Teaching hospitals were less likely to close. Contrary to some expectations, efficiency never predicted survival; more efficient hospitals tended to close. Implications: Cumulatively, decades of urban hospital reconfiguration have markedly reduced the number of hospitals in African-American districts. Surviving urban hospitals are disproportionately large teaching hospitals. The observed reconfigurations appear to have substantially increased travel times to inpatient, outpatient, and emergency services. Costs appear to have increased considerably. Potential quality of care has probably improved. We identify opportunities and methods for reconfiguring hospitals to improve access and cut cost. Background: This study examines the performance of hospitals based on location (geographical region, rural, urban). In this study, recent data has been used to better understand the hospitals performance after the introduction of Prospective Payment System (PPS). The data set used by the study is much comprehensive in its coverage and information on a number of relevant variables. Methods: Piecewise regression model to capture the effect of changes within range of the hospital size. We have included a number of new economic and financial variables in the analysis and examined the effects of conversion of hospitals from not-for-profit to for-profit on hospital performance. Results and Implications: Our empirical findings suggest that the size of hospitals, occupancy rate of hospital beds, ownership status, degree of competition faced in the market, teaching status, and measure of financial indebtedness of hospitals are significant determinants of hospital performance holding location constant. The empirical model also suggests that the relationship between hospital efficiency measure and its various determinants is actually nonlinear in nature and therefore, it is important to adopt appropriate nonlinear econometric models for empirical estimation of the performance function. Finally, our findings show that large urban hospitals face significant factors that hinder its performance in comparison to rural hospitals such as the lack of (DSH) payments and the cease of the benefits of the economy of scale. The findings should be taken within the study limitations, however, it results show that further support to the urban hospitals by the state and federal legislatures is essentials for its survival and the continuation of its mission to serve the indigent population. Muller M, Ornstein K, and Boal J Homebound status is a final common pathway for a variety of medical conditions, including dementia, advanced cardiopulmonary disease, cancer, neuromuscular disorders, and depression. Men and women who are homebound are frequently unable to access traditional healthcare services. Urban settings can exacerbate this already enormous problem by introducing variables such as inadequate housing, fragmented families, cultural and language barriers between patients and physicians, and unsafe environments. Finally, strong community and social support is often lacking in many urban areas and this growing population can suffer from isolation, depression, high degrees of symptom burden, and caregiver burnout. The Mount Sinai Visiting Doctors program cares for 600 patients annually and delivers primary medical care to homebound persons in Manhattan through the use of physician and nurse practitioner home visits. Services include primary medical care, urgent visits, social work services (counseling, entitlements, caregiver support), and Spanish language translation. Reasons for referral include dementia (34%), refusal to go to a doctor or inability to get out of a walk-up apartment (20%), endstage neuromuscular or cardiovascular disease (15%), and terminal illness (5%). Patient demographics include: average age 79 years, 78% female, 30% Latino, 30% African American, 65% have Medicaid, 34% live in public housing, and 28% report Spanish as their primary language. The annual mortality is 25% and, in stark contrast to rates of at-home death reported around the country (24%), 60% of our patients die at home. We have designed and implemented a clinically successful and financially viable model of care that is replicable in other urban settings. Queens (OR = 2.25 compared to Manhattan referent) and Brooklyn (OR = 2.11 compared to Manhattan referent), using cannabis (OR = 2.10), and using alcohol (OR = 1.24). Conclusions: These findings suggest that during the 1990s, firearms played a role in about one fifth of suicides in New York City, with a greater use among minorities and those living in the outer boroughs. Additional research is necessary to examine reasons for the low proportion of suicides caused by firearms in New York City compared to other parts of the US A 10-Year Review of Penetrating Traumatic Injuries (PTI) at an Urban Level Trauma Center Lawless B and Ahmed N The incidence of PTIs in urban areas is increasing and becoming a more prevalent problem in the inner city. There is no Canadian data to validate this hypothesis. The purpose of this study is to review and characterize the cases of PTI presenting to an inner city Level-1 trauma center over a 10-year period. Trends in incidence, types of PTI, and associated demographic information were recorded.Methods: Patients with PTI who presented to our trauma center from January 01, 1992 until August 31, 2002 were identified using the trauma registry database. Charts for those identified were reviewed and the data collected. Data from the city police department was obtained to review the incidence and locations of inner city crimes involving PTIs. A clinical epidemiologist was consulted for statistical methods.Results: PTI comprised 8% of all traumas in 1992 compared to 13.8% in 2001. PTI increased overall by 5.7% from 1992 to 2001.Implications: There was a 6-fold increase in PTI from 1992 to 2002. Dynamics of inner city life contribute to this increase in violent behavior. Underlying social reasons must be examined and changes in community and health policies implemented. Boparai D, Cannon WG, Gotto G, Lee P, Cannon-Rodriguez JA, Verchere C, and Carr NJ Background: Coyotes (Canis latrans) inhabit most of North America and are one of the world's most adaptable predatory animals. Coyote's natural prey includes rats, mice, birds, rabbits, house cats or domestic dogs. In urban settings without their natural predators e.g. bears, cougars or wolves; the coyote becomes a real threat to humans-especially children. In British Columbia, numbers of "urban coyotes" has increased, and young children are at risk from attack. Our purpose is to raise public awareness about the danger to children from wild coyotes living in urban settings.Results: In September 2000 an eighteen-month-old boy playing near a soccer field was attacked and bitten by a wild coyote. 7 stitches were required to close the wound above his eye. In December 2000 a coyote stalked and charged a 3-year-old boy near an elementary school. In July 2001, 6 yearold girl playing near a tennis court was attacked by a daughter from the animal's grip and she received treatment for her puncture wounds and bruises. The same month, a coyote viciously attacked a 14month old girl playing in her own front yard. The child required stitches for facial wounds. Adults were nearby during all attacks.Summary: Wild coyote's fear of man fades as the number of "urban coyotes" grows; attacks on children are increasing. Coyotes socialized by contact with humans (especially if fed) become bolder and a threat to our children's safety. Boparai D, Cannon WG, Cannon-Rodriguez JA, Jacobsen K, Chu A, Fung A, Hallgren NK, Bristol S, and Brown EE Introduction: In British Columbia during 2000, nearly 90 youths (from 13 to 21 years of age) died as a result of motor vehicle collisions (MVCs) and over 10,000 were injured. Motor vehicle collisions are the Number 1 Killer and Number 1 health issue for young people in British Columbia. Youth represent 12% of the population and those over 16 years of age are eligible to drive. New inexperienced drivers are twice as likely to cause motor vehicle collisions as are more experienced drivers. The 3 leading causes of MVCs in our province are 1) Speeding 2) Impaired driving and 3) Risk-taking while driving. Youth Counterattack and Roadsense Society of British Columbia (C.A.R.S BC) is a provincially Implications: No increases in neighborhood rates of mental health service use were demonstrated after a large reduction to welfare benefits. Individual level data would be helpful to determine if the benefit reduction was related to service use or psychiatric morbidity at the individual level. Crawford M and Mostashari F Background: The Institute of Medicine estimates that 18,000 American adults 25-64 die each year because they lack health insurance coverage. Not having ready access to needed medical care and the resulting discontinuity makes being uninsured potentially the sixth leading cause of death among people under age 65. This analysis characterizes the uninsured adult population in New York City (NYC) and investigates the extent of the health care access challenges they face.Methods: We examined data from the New York City Community Health Survey (NYCCHS), a random sample telephone survey of 9,674 noninstitutionalized adults conducted from May-July 2002.Results: Overall, nearly 700,000 (14%) New Yorkers between 18 and 64 reported having no health insurance. Particularly high proportions of individuals were uninsured in foreign-born populations (28%), the Asian-American community (24%), low-income households (22%), and among the unemployed (24%). Most of the uninsured (62%), however, were employed. Neighborhoods in NYC ranged from a low of having a 4% uninsured rate (Gramercy/Upper East Side, Manhattan) to a high of 24% (Greenpoint, Brooklyn). A high proportion of the uninsured had an unmet healthcare need in the past year (31%) compared to others (7%). In terms of preventive services, the uninsured received less care; for example, fewer had their blood pressure checked (68% vs. 91%), or were taking medications if diagnosed with high blood pressure (27% vs. 59%).Implications: Identification of groups most at risk to having large numbers of uninsured people can inform and focus effective interventions. Results: Overall, over 90% of the children were insured. The insured were more likely to have a regular source of care than the uninsured (94% vs. 80%; p = .00). Over 90% of the insured and the uninsured go to their regular source of care for preventive care and for sick care. Of those who did not have a regular source of care, 60% of the insured and 66% of the uninsured went to an ER or community health center for care. In the year before the survey, the insured were more likely to have had a preventive visit (77% vs. 64%; p = .02); dental care (72% vs. 60%; p = .01), prescription (61% vs. 39%; p = .00), and vision care (39% vs. 30%, p = .05). The insured and the uninsured had similar use of illness care (28% vs. 26%) and mental health care (9% vs. 12%). The uninsured reported more unmet need for care than the insured. Implications: In this sample, presence or absence of insurance predicted access and utilization. It is important that gains made due to expansion of public insurance programs in the late 1990s not be rolled back during more difficult economic times. Methods: Data from the Ontario physician billing database was used for the analysis. Visits during the SARS period were compared against the visits during the same period the year before. Additionally, two months before the SARS period was compared against the same two months in the previous year.Results: Two months previous to the SARS period, 9,540 patients made 9,789 visits which was a 1.6% increase from the previous year. During the SARS period, 6,957 patients made 7,158 visits which was a 19.7% drop from the previous year. The rankings of the top five diagnoses (anxiety, hypertension, HIV, drugs addiction and common cold) changed slightly during the SARS period compared with the year before but were no different from previous months. There was a significant increase in the number of mental health disorders encounters during the SARS period (20.3% versus 17.2%). No major changes occurred in the number of annual physicals performed.Implications: Utilization of health care services decreased during the SARS outbreak but there were only minor changes in the overall case-mix of the urban population. Patients did not appear to change their reasons for seeking medical care during the outbreak. Results: During the study period 1178 H-UH patients were seen in the ED. 933 were directly discharged to the community and 245 to the RC. (1) 76% of RTC users completed follow-up compared with 31% who were discharged to the community (p < 0.0001). There were significant improvements in follow-up to FP, SW/Psychiatry, ED, Detox/Shelter. There was no significant difference in follow-up for IV therapy or dressing changes. (2)The satisfaction survey was completed by 126 clients (51%), and 79% reported excellent care overall.Implications: The RTC has improved compliance with follow-up care for H-UH. RTC users have a high level of satisfaction with the facility. Huberman H, Tineo W, Rosenberg T, Sharif I, and Mendelsohn A Background: To address the problem of early childhood developmental delay, a randomized trial in NYC is evaluating a low intensity preventive intervention. Based on a public health / primary care partnership, this intervention combines parent-based (ASQ) screening, monthly Building Blocks parenting newsletters and toys (BB), and Reach Out and Read (ROR) clinic-based distribution of children's books.Methods: Families of 4-7 month old children attending a South Bronx health center serving a poor (median income $20,000) Black and Hispanic community are offered enrollment and randomized to a control, an ROR-only, or an ASQ/BB+ROR group. Outcome measures include the Bayley MDI, and the OWPVT language test. Preliminary data on EI referral rates are presented on a sample of 56 children with 15 month outcomes and 68 children with 24 month outcomes completed (irrespective of study group).Results: Overall, 25% of the children had been referred to EI. Of these, 37% were based on health