Chicago Health Outreach

1015 West Lawrence Avenue
Chicago, Illinois 60640
Contact: Heidi Nelson
Phone: (773) 275-2060
Key words: residential program; mental health;
case management; TB; asthma; housing

Note: Chicago Health Outreachs (CHO) pilot study consisted of four major subparts on the following CHO programs: 1) the Safe Haven Residential Mental Health Program; 2) Connections Mental Health Case Management Program; 3) Primary Care Clinic TB Program; and 4) Asthma Primary Care Services. To allow for maximum clarity, each topic will be discussed separately; therefore, there are four reports for Chicago Health Outreach.

#1: SAFE HAVEN RESIDENTIAL MENTAL HEALTH PROGRAM OBJECTIVE
The purpose of this study was to examine whether or not the provision of housing in a safe environment and the support of staff encouraged people to take psychotropic medications and to participate in mental health services.

METHODS
Safe Haven is a low demand, residential program for difficultto- engage, severely mentally ill persons with a history of homelessness. The goal of the Safe Haven program is to engage participants in mental health services. The indicators for medications and participation in services were measured retrospectively through chart review for the 13 month period from the beginning of the program (June 1996) through June 1997. A total of 13 residents lived at Safe Haven during this time, for varying lengths of time. Medical records were reviewed for information about sessions with the psychiatrist and medications taken before and during patients time at Safe Haven. Attendance logs were reviewed for participation in mental health services offered at Access to Community Care and Effective Services and Support (ACCESS), including attendance at the ACCESS respite center. The Safe Haven supervisor provided demographic information; dates of stay at Safe Haven; and information about participants_ progress with medications. The associate director of the residential mental health programs, the director of the ACCESS program, and the program psychiatrist provided information about participants progress in the program and their housing status after leaving the Safe Haven. Seven women and six men comprised the study group, including eight Black, three White, and two Hispanic participants. Their average age was 40 years. During the 13 month period, the average length of stay for the 13 residents was approximately five months (155 days), and the median was approximately 2.5 months (79 days). Five of these people were taking some medication at the time they accepted housing at Safe Haven. Outreach workers met participants on the street or in shelters, and some had already attended the respite center and/or met with the psychiatrist.

RESULTS
Of the five people who came to Safe Haven already taking psychotropic medications, all of them continued taking medications during the study period. One of these people accepted intramuscular injections only and no oral medications. The average length of stay for this group was 82 days. On average, these people took medications 88 percent of the days they lived at Safe Haven. Of the eight people who were not taking medications at the time they came to Safe Haven, three decided to begin taking medications. Their average length of stay was 281 days. One of these people accepted injections only and received injections for all of his time at Safe Haven. One person took medications on 25 percent of the days and the other on 62 percent of the days. Five people took no medications while living at Safe Haven. Two of these people actually decided to begin medications but discontinued after 1 day. The average length of stay for this group was 152 days. However, the range was very large, and the median length of stay was only 14 days. The overall rate of Safe Haven participants on medication for the study period was eight per 13, or 62 percent. The average and median percent of days that the subjects were on medication were 49 percent and 62 percent, respectively. This rate cannot be considered compliance. In some cases, several medications were prescribed, and the patient accepted some and refused others. For medications taken several times a day, some doses were missed, and sometimes days or weeks were missed. Among these 13 people, only three never accepted psychiatric medical care from the programs psychiatrist. The recommended number of sessions with the psychiatrist is one per week, although residents are not required to meet with the psychiatrist and do not have weekly appointments. For the ten people who did accept psychiatric care, the average percentage of weeks they met with the psychiatrist was 19 percent, and the median percentage was 17 percent. Only two people never attended the respite program at ACCESS. On average, Safe Haven residents attended 40 percent of the time (the median was 50 percent). Daily attendance at the respite program was recorded by ACCESS staff. Participants were considered to be present as soon as they came through the door. Respite program activities include preparing and eating lunch and other social activities.

DISCUSSION
Because the total number of Safe Haven residents at the time of the study was very small, frequencies have limited usefulness. Another barrier to analyzing the results is the lack of a control group. Obviously, ethical considerations prohibit withholding services from some participants in order to draw conclusions. Also, because the program had been accepting residents for just over one year, it was difficult to measure outcomes since engagement in services can be a long-term process. The provision of safe, low-stress housing seems to have an impact on maintenance of medications as well as on the decision to take medications. Through engagement in ACCESS mental health services prior to accepting housing at Safe Haven, five people had already begun medications, and they remained medicated while at Safe Haven. Non-adherence to medications is common among homeless persons. It is unlikely that these people could have maintained such a high percentage of days on medications (88 percent) if they had been living on the street. Similarly, the access to medications, encouragement and monitoring that was provided to residents made it possible for three persons to decide to take medications. For this difficult-to-engage group, the number who decided to take or maintain medications at all is significant. Five individuals did not take medications at Safe Haven. For three of these people, the length of stay was very brief (one person stayed only 2 weeks and two people stayed only 4 weeks). Little time in a supportive environment may have been the reason that they did not make progress on this goal. One woman who was at Safe Haven for 45 weeks actually decided to accept medication but became ill upon taking the first dose and discontinued the medication. Another person agreed to take medications but discontinued after one day. These two people apparently realized their need to be medicated, which may be the first step in medication adherence. Engagement in psychiatric services was low. It is unknown why Safe Haven participants were reluctant to meet regularly, or at all, with the program psychiatrist. Perhaps meeting with Safe Haven staff in an informal manner where people live or with ACCESS staff informally at the respite center is more acceptable for these people than meeting with a medical professional one-on-one. Or perhaps, past experiences with psychiatric care have left some people afraid to engage in treatment on this level. People_s reluctance to be treated by a doctor highlights the importance of the case management and other on-site services provided at the Safe Haven. Participation in mental health programming through ACCESS was widespread and fairly high. Almost all people spent some time at the ACCESS respite center, and for those people, an average of 40 percent of the sessions were attended. Again, for this difficult-to-engage group of people, willingness to be involved in any programming is encouraging. An ultimate outcome of the Safe Haven program is long-term housing for its residents, after they leave the program. Although long-term housing was not a formal indicator for this project, it is important to discuss the results in light of this factor. Taking medications and participating in programming do not necessarily point to stability. Similarly, connecting participants to services is an overall outcome. The staff participant relationship is important because connecting participants to services is a process. A stay at Safe Haven may not result in a participant accepting all necessary services. If trust is built, however, progress can be made over time, regardless of the immediate outcome for a person. At the end of the study period, four people were still living at Safe Haven. Since the end of the study period, however, three of these people have found permanent housing in apartments, and only one remains at Safe Haven. Four additional former residents are permanently housed: three of these are renting their own apartments and one person is living with her family. Two people left the Safe Haven for long-term hospital care, and their current housing status is unknown. One person is now in transitional housing, and only two former residents are living on the streets. The housing status of former residents points to success with the overall outcome objective of long-term housing.

CONCLUSION
Safe Haven has a number of overall outcomes as well as intermediate outcomes. While persons are residents of Safe Haven they are encouraged to participate in services and to take medications. However, they are not forced to engage in any services. Building relationships with staff and others and identifying what participants feel are their needs are also actively pursued. A number of stays at Safe Haven and enrollment in other programs may be necessary before a person decides to engage in services. The safe, low-demand housing provided at Safe Haven seems to have an effect on people_s decision to be medicated and to participate in mental health services. Furthermore, the supportive environment seems to contribute positively to the long-term goal of permanent housing for former residents.

#2: CONNECTIONS MENTAL HEALTH CASE MANAGEMENT PROGRAM OBJECTIVE
The purpose of this study was to determine the effectiveness of this program by examining its relationship to clients_ maintenance of housing and benefits.

METHODS
Connections is a long-term case management program for persons with serious mental illness and a history of homelessness and uses the Assertive Community Treatment (ACT) model. The indicators--maintenance of housing and maintenance of benefits--were measured over a 6-month period for all clients in the Connections program as of January 1997. These clients included both those in the day program as well as those receiving only case management. Clients who enrolled in the program after January 1997 were not included. Housing status and benefits received were measured at Time One in January 1997 and again at Time Two in June 1997. The program director and a research assistant collected the data. Most of the information was reported directly by the program director; some was obtained through record review. Clients were considered successful in maintaining housing if they were housed in a space rented to them or provided for them, and not living in another person_s space or staying in a shelter. A person who moved to a new location during the study period was considered successful in maintaining housing. Any such movement was monitored, however, to show stability in housing and whether or not the move led to an improved level of functioning for the person. Connections staff provided representative payee services, meaning they received and disbursed benefits to those clients are entitled to them. Changes in benefits were monitored and recorded over the 6-month period. As of January 1997, there were a total of 68 Connections clients; 25 were day program participants, and 43 were nonparticipants. The same 68 clients were followed over the 6- month period. In June 1997, there were 36 day program participants, 27 nonparticipants, and 5 persons who were no longer clients. Thus, there were 49 clients whose status as a day program participant/nonparticipant did not change from Time One to Time Two, and maintenance of housing and benefits was measured for these clients. Among these 49 subjects, 23 (47 percent) were participants and 26 (53 percent) were nonparticipants. Those whose enrollment changed during the study period and those whose cases were closed will be described separately. Participants were categorized into two groups: light and heavy users of day program services. Light users attended the program less than 50 percent of the days it was open, and heavy users attended 50 percent or more of the days. At Time One, 20 of 25 or 80 percent were heavy users, and only 5 (20 percent) were light users. At Time Two, of the 36 participants, exactly half (18) were heavy users and half (18) were light users. Of the 23 day program participants, there were 19 (83 percent) men and 4 (17 percent) women. The average age was 39 years. Racial/ethnic groups included 12 (53 percent) Black, 4 (17 percent) White, 4 (17 percent) Hispanic, and 3 (13 percent) Asian. The average length of time as a client in the program was 24 months. Of the 26 nonparticipants, half were men (13) and half were women (13); their average age was 49 years. Racial/ethnic groups included 14 (54 percent) Black, 10 (38 percent) White, 1 (4 percent) Hispanic, and 1 (4 percent) Asian. The average length of time as a client in the program was 35 months. All Connections clients were housed at Time One. Clients were housed in either one of the two residential programs or in single room occupancies and apartments in the neighborhood. All but one of the day program participants received some benefits at Time One. Benefits received included Medicaid (91 percent), Supplemental Security Income (SSI) (87 percent), food stamps (57 percent), restaurant allowance (35 percent), and Medicare (4 percent). All of the nonparticipants received benefits at Time One. Benefits received included SSI (85 percent), Medicaid (69 percent), food stamps (35 percent), Medicare (23 percent), restaurant allowance (19 percent), wages from employment (15 percent), Veterans Benefits (8 percent) and pension (4 percent).

RESULTS
Among the 23 day program participants, 100 percent maintained housing over the 6-month period, although, nine (39 percent) of these persons did experience some change in housing. According to the program director, four of these housing changes resulted in a better level of functioning for the persons; three changes resulted in a worse level of functioning; and two changes did not change the level of functioning. (More details to be provided in the Discussion section.) For nonparticipants, 100 percent also maintained housing; this group, however, experienced no changes in housing. One hundred percent of the day program participants maintained the benefits they were receiving at Time One. The person who received no benefits at Time One was receiving SSI, Medicaid, and food stamps by Time Two. One person began receiving income from employment at Time Two, in addition to maintaining his other benefits. There were no changes in benefits for nonparticipants; one hundred percent maintained benefits over the study period. Fourteen subjects enrollment in the day program changed from Time One to Time Two; This group included eight men, five women, seven Blacks, five Whites, and one Hispanic. Their average length of time in the program was 33 months. Thirteen nonparticipants at Time One became enrolled in the day program over the study period. Of the 13, 11 became light users and 2 became heavy users. One person who had been a heavy user of 48 the program at Time One was no longer a day program participant at Time Two. Among the group of people whose enrollment status changed, 100 percent maintained housing. There were two changes in housing, but the move did not result in a different level of functioning. Four people experienced changes in benefits. One change was positive: a person gained income from employment. Of the remaining three, one person lost a restaurant allowance; one person lost food stamps; and one person lost all his benefits, including Medicaid and food stamps. Five people were no longer Connections clients at Time Two; these individuals included two Black women, two Black men, and one White man. The average length of time in the program was 25 months. Four of them were not day program participants at Time One, and one was a light user. One of the drop out clients went to prison; another went to a nursing home outside the catchment area for the program; a third client moved across the country; one client refused services; and one client could not be located.

DISCUSSION
All subjects maintained housing and benefits over the study period. Therefore, subgroup analyses of subject
characteristics to determine success with the indicators are not useful. Stability in housing was high in all groups. Most people maintained the same dwelling throughout the study. While some changes in location of housing did occur, few led to a lower level of functioning for the person. Level of functioning was a subjective judgment made by the program director who had regular contact with the clients during the study period. Monitoring improvement in housing was found to be problematic. While some housing changes clearly led to improvements (i.e., cleaner surroundings, better appliances), other changes such as increased security, stricter rules for the tenants, or relocation to a nursing home could be viewed differently by different people. It is encouraging to note that of the clients whose enrollment status in the day program changed, 13 became day program participants and only one dropped out of the day program. Although there was no effect on maintenance of housing, maintenance of benefits was not 100 percent for this group. It is interesting that three people lost some or all their benefits. Also of interest is that of the five people who dropped out of the Connections program completely, four were not day program participants at Time One. Therefore, only 1 day program participant left the program.

CONCLUSION
All Connections clients were successful in maintaining housing and benefits for the 6-month period. Participation in the day program does not seem to be the factor responsible for the high level of success. Case management services, which are received by all Connections clients, are more likely responsible for high levels of maintenance. In order to show the effects of the day program, different indicators must be chosen, keeping in mind the goals and the services of the day program. For example, if socialization skills are the desired outcome, indicators must be designed to measure progress in this realm.

#3: PRIMARY CARE CLINIC TB PROGRAM
OBJECTIVE
The purpose of this study was to evaluate the effectiveness of CHO_s primary care clinic tuberculosis (TB) program through the measurement of two indicators: completion of TB prophylaxis and maintenance of housing for those housed while receiving directly observed therapy (DOT).

METHODS
Prophylaxis Indicator The CHO measured the rate of TB prophylactic treatment completion for persons who had a PPD positive screening and a subsequent normal chest x-ray. Records of all persons screened by CHO, both in the clinic and in shelters, and all persons referred for chest x-rays were kept by CHO_s TB program coordinator.
In 1996, 128 people had chest x-rays at the Board of Health after being referred by CHO. The medical chart of every other person on the list was selected for review. Only those who started prophylactic treatment in 1996 were selected as subjects. This allowed the whole treatment period to be monitored before the end of the study period, June 30, 1997, since the prescribed course of treatment was 6 months in all cases. Charts missing chest x-ray information and charts for patients under 18 years of age were not reviewed. One woman died during the treatment process and was excluded. The final number of charts reviewed was 56; 27 of those actually began the TB treatment process and were chosen as subjects. Twenty-nine persons did not begin treatment. In most cases, the reason why prophylaxis was not recommended was not explicitly documented in the chart. Possible reasons for medication not having been prescribed include age; history of alcoholism or liver problems; and questionable ability to adhere to treatment. In eight cases, the patient did not return to the clinic for the chest x-ray results or did not show up for an appointment to receive treatment. Medical records contained information on the date prophylaxis was prescribed and progress with treatment. In most cases, a 1-month supply of medication was given, and the patient returned monthly for refills. Provider notes regarding completion of treatment were used to determine success. In cases of non-adherence to treatment, the patient was considered successful only if he or she eventually completed the required course of treatment. Among the 27 randomly chosen subjects who began TB prophylactic treatment, there were 19 men and 8 women. The ethnic/racial breakdown of the group was 33 percent Black, 30 percent Hispanic, 22 percent White, 7 percent Native American, and 4 percent Asian. Foreign-born persons accounted for 11 (41 percent) of the total and represented the following countries: Mexico (8), Poland (1), Pakistan (1), and Albania (1). Ten of these 11 persons had a language other than English as their primary language. The average age was 38.7 years. Half had less than a high school education, 26 percent had only a high school education, and 19 percent had some college credit or a degree. Almost three quarters (74 percent) had no medical insurance, while 22 percent had Medicaid benefits, and 4 percent had Medicare benefits. The average number of clinic visits in 1996 for this group was nine. Almost half (48 percent) had a history of mental health problems, and 33 percent had a history of substance abuse. Housing Indicator All persons receiving DOT who were assisted in obtaining housing by CHO through TB case management from June 1995 to June 1997 were subjects for this part of the study. Current case workers provided lists of these persons. Clients who were housed but still receiving TB treatment in June 1997 were not included. In several cases, clients receiving DOT did not accept housing assistance from CHO. Some people paid rent for their own housing, some lived with friends, and others preferred to remain in shelters or on the street. These people were not included as study subjects, thus the total number of study subjects was 16. Demographic and treatment information was obtained from medical charts. Information about housing status was obtained through case manager interviews. Two TB case managers, who had been at CHO since 1995, were knowledgeable about the housing status of former clients, even after case management was terminated. Among the 16 subjects, there were 14 men and 2 women. There were 12 Blacks, 3 Hispanics, and 1 White person. The average age was 43 years and the age range was 35 to 53 years. Five persons received income through employment during and after the program, and eight persons received SSI benefits during and after the program. One person was in a mental health program, one person was in a substance abuse program, and one person was in a program for persons with HIV. While receiving DOT, clients were housed as follows: 14 in single-room occupancy units, 1 in an apartment, and 1 in a residential HIV program. All 16 clients completed TB therapy.

RESULTS
Prophylaxis Indicator Of the 27 people who began prophylaxis, 12 (44 percent) successfully completed the treatment, including 7 men and 5 women. Fifteen (56 percent) did not successfully complete the treatment, including 12 men and 3 women. The average number of months of treatment completed, for those who did not adhere, was 2.4 months. Among Black patients, only 10 percent adhered to treatment while 90 percent did not. Among Whites, 83 percent adhered while 17 percent did not. Among Hispanics, 75 percent adhered while 25 percent did not. One Asian and two Native Americans did not adhere to treatment. The average age was the same (39 years) for the group which completed treatment and the group which did not complete treatment. A higher percentage of people who had less than a high school education adhered (58 percent) than did not adhere (42 percent). Of those who had no medical insurance, 40 percent adhered and 60 percent did not adhere. Of those who were Medicaid beneficiaries, 50 percent complied and 50 percent did not. The one person with Medicare benefits complied. Of the 11 foreign-born clients, 8 (73 percent) adhered to treatment and 3 (27 percent) did not adhere to treatment. The average number of clinic visits per person in 1996 was slightly higher for the adherence group (9) than for the nonadherence group (8). Only 39 percent of persons with a history of mental health problems adhered, while 61 percent did not adhere. Of persons with a history of substance abuse, only 22 percent adhered and 78 percent did not adhere. Housing Indicator Of the 16 subjects, nine (56 percent) were still housed at the end of the study period or after 12 months, whichever came first. Seven (44 percent) did not maintain housing and were homeless at the end of the study. Because completion of DOT occurred on different dates for individuals, the following categories were created. Seven clients completed TB therapy 12 or more months prior to the end of this study. Of those seven, four remained housed after 12 months; one remained housed for 6 months; one remained housed for 1 month; and one remained housed for less than 1 month after completing therapy. One person who had completed therapy 9 months before the end of the study was still housed after 9 months. Of the three people who completed therapy 6 months prior to the end of the study, two were still housed after 6 months, and one did not maintain housing after completing therapy. Three months after completing therapy, two people were still housed, while two people remained housed for 1 month and one person did not maintain housing at all.

DISCUSSION
Prophylaxis Indicator Individuals face a number of barriers to adherence to TB prophylaxis. For example, because persons with TB infection do not have active disease and may not feel ill, they may not understand the importance of completing treatment. Also, the length of the treatment process may deter some individuals from completing prophylactic treatment. This study suggests that a number of demographic and socioeconomic characteristics may contribute to adherence as well. The results suggest that women are more likely to complete treatment, while men are more likely not to complete treatment. The average number of months of treatment completed, for those who did not adhere, was 2.4 months. Black patients appear more likely not to adhere to treatment than to successfully complete prophylaxis. Whites and Hispanics are more likely to complete treatment than not to complete treatment. There were no Native Americans or Asians who successfully completed treatment, however, the number of people in these groups was too small to analyze. Age does not seem to be a factor in adherence to treatment. It is unknown why people with less than a high school education are more likely to adhere than not to adhere. It is unlikely that lack of insurance contributed to non-compliance, since the medications and clinic visits were provided at no cost to all clients. The reasons for the high rates of adherence among foreign-born persons are unknown. Perhaps these persons were given extra attention by the provider or translator to make sure they understood the importance of treatment. Perhaps these people were more aware of TB treatment because of high TB prevalence in their countries of origin. It may also be possible that if ones immigration status were an issue, a person would adhere to medical treatment so as to avoid contact with the authorities. The average number of annual clinic visits does not seem to affect adherence. Persons with a history of mental health problems seem to be at risk of non-adherence. Likewise, persons with a history of substance abuse are at risk. Because of the relatively small total number of subjects and the even smaller number in specific groups, it is difficult to determine which factors, if any, are closely related to treatment adherence/non-adherence. Overall, the rate of completion of TB prophylactic therapy was 44 percent. For CHOs population of homeless clients, this degree of success was encouraging. The rate of completion of chemotherapy for persons with active TB disease in 1996 was 83 percent. Again, the fact that persons needing prophylaxis do not have disease symptoms may result in lower rates of adherence compared to those who have active TB. As the results suggest, Black patients and persons with a history of mental illness or substance abuse may be at greater risk for not adhering to treatment. The CHOs patient population includes a high proportion of people with these characteristics. Further evaluation of CHOs patients who uccessfully/unsuccessfully complete therapy may be helpful in determining the reliability and validity of these findings as well as finding ways to create messages and programs that are likely to be successful in targeting those who are most likely to be at risk of non-adherence. Housing Indicator The sample size was small and, therefore, the analysis of factors related to the maintenance of housing had limited significance. Also, there was little variation by gender, race, or age. The six persons who did not remain housed or maintained housing for only 1 month include one woman and five men, one Hispanic and five Black clients. Of those who remained housed after 12 months or by the end of the study, there were nine men, six Blacks, two Hispanics, and one White. These groups are relatively proportional to the overall group. It is interesting that the three persons involved in other programs (mental health, substance abuse, and HIV) were all housed at the end of the study. Not surprisingly, the results suggest that income may be a factor in maintenance of housing. Of the nine persons who maintained housing after completing DOT, seven (78 percent) received SSI benefits during and after the program. Of the seven persons who did not maintain housing, only one (14 percent) received SSI benefits during and after the program. The influence of income from employment on maintenance of housing is not clear. A higher percentage of employed persons (three out of five) did not maintain housing, although again, the total number was very small. Two of the nine persons who maintained housing (22 percent) received some income from employment during and after the program. Of the seven persons who did not maintain housing, three (43 percent) earned income from employment, although they worked in low-paying and/or part-time jobs.

CONCLUSION
Homeless persons with TB who received DOT and were housed during treatment had a 100 percent adherence rate to treatment. The combination of case management, including DOT, and provision of housing are probably dual factors in the success rate. Over half (56 percent) of the former clients maintained housing for up to 1 year after DOT was terminated, an encouraging number given that they were homeless before being treated for TB. Although 44 percent did not maintain housing, these people successfully completed TB treatment, and so housing people while they undergo DOT remains an important service.

#4: ASTHMA PRIMARY CARE SERVICES
OBJECTIVE
The purpose of this study was to measure the hospitalization and emergency department (ED) utilization rates for its adult asthma patients, in part to determine the effectiveness of CHOs asthma primary care services.

METHODS
Using CHO_s management information system, study coordinators produced a list of all persons with a diagnosis of asthma who were treated in 1996, both in the clinic and at other locations. The chart of every other person on the list was reviewed. One of CHOs physicians, a pulmonologist, reviewed the charts to confirm that the patients actually were asthmatic. A total of 51 patients were identified for the study. Information on asthma-related hospitalizations and emergency department utilization in the 12-month period from August 1996 to September 1997 was obtained retrospectively by patient report. A data collection form was attached to the inside of the subjects charts. Providers were instructed to ask patients about hospitalizations and ED visits for asthma in the 12-month period and to record information including the hospital name, the date(s), and the number of days spent at the hospital. Providers were also asked to rate the accuracy of the information, based on their opinion of the patients ability to self-report. By the end of the project, only 10 data collection forms had been completed. Most of the 51 asthma patients identified for the study did not have a clinic visit during the time the data was collected. Because the number of actual subjects was so small, the data have limited usefulness, and rates and subgroup analyses are not significant. Among the 10 subjects, there were 7 women and 3 men. Three persons were White, five persons were Black, and two persons were Hispanic. Their average age was 39 years. Six persons received Medicaid benefits, and of those, two also received Medicare. Four persons were uninsured.

RESULTS
Of the 10 persons, only 1 had been hospitalized for asthma during the study period. This person reported being
hospitalized three times for an average of 2.7 days each time. Nine persons reported no asthma-related hospitalizations during the study period. Only two persons reported no ED visits during the study period. Of the eight who did visit an ED, five made only one visit to the ED and three made two visits to the ED. Providers rated the accuracy of seven of the patients selfreport information. Four of the providers felt the information was definitely accurate; three providers felt the information was probably accurate; and only one provider felt the information was probably inaccurate.

Due to the small number of subjects, it was not possible to determine the factors related to hospital and ED utilization by asthma patients. It appears from this preliminary data that while asthma patients are not often hospitalized for asthma, ED visits are common. The ten subjects charts were reviewed to monitor use of CHOs primary care clinic. During the same 12-month period, the average number of clinic visits for these subjects was 17.5 (the range was 1 to 37). It is important to note that many of the people who came frequently for clinic visits were being treated for conditions other than asthma (e.g., HIV infection, TB infection or disease, and diabetes). A larger number of subjects might have shown that those who receive regular primary care have less asthma-related crises and utilize the hospital and the ED less.

CONCLUSION
The methodology of this study failed mainly because the data collection period was not long enough to obtain data from a significant number of patients. In any case, conclusions would have to have been made cautiously due to the potential unreliability of patient self-report data. Patients may have difficulty recalling information from the past year. These indicators might be more feasible if patients were routinely ask for utilization information at each visit.