Care of Injection Drug Users With Soft Tissue Infections in San Francisco, California

Context  Illicit injection drug use results in serious soft tissue infections that are the number one nonpsychiatric reason for admission to San Francisco General Hospital (SFGH), San Francisco, Calif.

Objective  To establish a specialized clinic to provide accessible, high-quality, and cost-effective medical care to patients with soft tissue infections.

Design, Setting, Intervention, and Outcome Measures  The Integrated Soft Tissue Infection Services (ISIS) Clinic was established to provide coordinated surgical intervention, substance abuse counseling, and social services for patients with soft tissue infections treated in a public hospital. Demographic information, treatment outcome, and hospital utilization data were analyzed.

Results  In the clinic's first year of operation, there were 3365 patient visits and 2255 surgical procedures. A large number of patients reported recent injection of illicit drugs (61%), were homeless (30%), and either had hepatitis C, hepatitis B, or human immunodeficiency virus infection (62%). Patients using heroin were enrolled in either a detoxification or maintenance program (42%). Few patients were designated as treatment failures (2%) or were lost to follow-up (14%). The ISIS Clinic dramatically reduced emergency department visits (−33.9%), surgical service admissions (−47.3%), inpatient acute care bed days (−33.7%), and operating room procedures (−71%), saving approximately $8 765 200 in the first year of operation.

Conclusions  This clinical intervention was notably cost-effective while preserving a high quality of medical services. Owing to limited data, we can only assume that other communities are similarly confronted with this public health problem. The ISIS Clinic could serve as a model intervention and thus have significant impact on the treatment of this prevalent but often overlooked challenge.

DURING THE past decade, the use of heroin and other illicit injection drugs has become an irrepressible feature of the urban landscape in the United States. By 1999, an estimated 3.1 million people reported using heroin at some time in their lives, with approximately 154 000 active users and some 104 000 heroin initiates per year.¹ The injection of heroin and other illicit drugs is associated with the transmission of blood-borne diseases, drug overdose, and death. Injection drug users are also at risk for the development of serious soft tissue infections. Although infrequently life threatening, cellulitis and abscesses commonly require treatment.

The precise number of soft tissue infections from injection drug use in the United States is not known, but anecdotal evidence suggests that it represents a significant public health problem.^2-4 This lack of detailed information likely reflects the low priority typically assigned to the care of injection drug users by medical institutions. Individuals suffering from drug addiction are often simultaneously battling complex psychosocial and medical problems, including poverty, homelessness, psychiatric diseases, and viral infections.⁵ The seemingly intractable nature of these problems, combined with apathy from the public and medical professionals alike, has relegated care for injection drug users to the realm of afterthought and benign neglect. Thus, injection drug users have limited access to medical care and frequently perceive their treatment within traditional health care settings as harsh and dehumanizing.⁶ Furthermore, the paucity of governmental, institutional, and societal interest in the health of injection drug users constitutes a profound disincentive for those individuals and organizations interested in the study of drug addiction, its complications, and treatment. Despite or perhaps because of these obstacles to health care, drug users utilize more than their share of health care expenditures.⁷

San Francisco, Calif, is home to an estimated 17 000 heroin addicts and 10 000 to 20 000 recreational users.⁴ Consequently, soft tissue infections are commonly treated throughout the city's health clinics and its public hospital, San Francisco General Hospital (SFGH). Since 1992, coincident with the increased availability of inexpensive high-purity heroin, the incidence of cellulitis and soft tissue infections has progressively increased.³ By 1999, these diagnoses represented the number one nonpsychiatric reason for admission to the hospital, annually accounting for about 3000 emergency department (ED) visits, 2400 hospital admissions, and 1000 cases in the operating room.⁸ Costs to the SFGH for ED and inpatient care alone were conservatively estimated to be in excess of $20 million per year.⁹ In addition, there was the less readily quantified but nonetheless significant expenditure of time by the physicians, nurses, and ancillary staff engaged in the care of these patients. Despite these efforts, both patients and providers voiced considerable dissatisfaction with the system of care. Herein we report the results of a proactive effort to improve the care of patients in San Francisco with soft tissue infections due to injection drug use. We postulated that an integrated approach to the problem, designed in concert by physicians, nurses, substance abuse counselors, and health department administrators, would enable the delivery of accessible, high-quality, yet cost-effective medical care to patients with soft tissue infections in a user-friendly, nonjudgmental environment that patients would voluntarily utilize.

The fundamental goal of the Integrated Soft Tissue Infection Services (ISIS) Clinic was to provide compassionate, efficient, and cost-effective care to patients with soft tissue infections. In addition to rendering timely medical care, the clinic was designed to provide access to other relevant services, including substance abuse counseling and treatment, selected social services, and wound care. The clinic is open every day on a walk-in, first-come, first-served basis and is staffed by academic faculty surgeons, 3 licensed nurses, a substance abuse counselor, a social worker, and an administrative support person. A pain management specialist, pharmacist, and infectious disease physician were available for consultation as needed.

Patients requiring surgical intervention were treated on-site under local anesthesia or moderate sedation in compliance with SFGH guidelines. All clinical decisions were made by the attending surgeon and included the option to admit patients. The nursing staff was trained to interview patients, assist with surgical procedures, administer sedation, assist with patient recovery following procedures, and provide patient education. All patients with wounds were provided follow-up appointments with an adjacent clinic where dressing changes were performed.

The nursing staff interviewed patients at intake, prospectively collecting a comprehensive panel of data that included age, race, sex, type and pattern of drug use, human immunodeficiency viral and hepatitis viral seropositivity, residential status, source of the patient's referral to the clinic, and source of medical insurance. These data were entered into a computerized database (Microsoft Access; Microsoft Corporation, Redmond, Wash). The surgeons entered physical findings, diagnoses, treatment plans, and details of all operative procedures.

Data comparing ED visits, hospital admissions, operating room volume, and the costs associated with the care of patients with soft tissue infections before and after institution of the ISIS Clinic were obtained from the SFGH information database. Information from a computerized lifetime clinical record was used to assess patient follow-up and clinical outcome. A clinical success was defined as either a healing wound without evidence of persistent infection or resolving cellulitis. Persistent or progressive infections were considered clinical failures. Patients were designated as lost to follow-up if there were no additional visits recorded in the lifetime clinical record or paper medical records following their ISIS Clinic visit. Clinical outcome was determined by reviewing the lifetime clinical records and medical records for 50 randomly selected patients. In addition, clinical outcome was stratified according to whether patients reported active drug use. Patient follow-up rate as a function of drug use was further analyzed by reviewing the records for 75 randomly selected patients with abscesses 8 cm or larger in diameter. To evaluate the patients' perspectives of the clinic, 75 randomly selected patients completed a brief anonymous questionnaire about services received in the clinic using a graded scale. The questionnaire focused on timely access to health care providers, quality of the patient interactions, pain control, and social services.

The Committee on Human Research at the University of California, San Francisco, approved all procedures for this study.

Data are presented as mean ± SD and were compared using t tests and χ² analysis. P ≤.05 was considered significant.

The ISIS Clinic was developed with an emphasis on maintaining high-quality patient care while converting from a predominantly inpatient model of care to an outpatient design. In the clinic's first year of operation, there were 3365 patient visits comprising 2861 new patients and 504 follow-up visits. Demographic data are presented in Table 1. Within our patient population, several factors were associated with injection drug use. The incidence of human immunodeficiency viral infection, acquired immunodeficiency syndrome, and hepatitis virus seropositivity were significantly greater in those patients reporting injection drug use than in those patients denying drug use (Figure 1). This difference was especially pronounced when comparing the rates of both hepatitis B and hepatitis C viral infection where injection drug use was associated with 4-fold and 6-fold increased incidence compared with nonusers, respectively (P <.001). Drug use was also associated with a 40% increased incidence of homelessness compared with nonusers (50% vs 35%; P <.001).

The ISIS Clinic successfully provided methadone treatment to patients using heroin. Only a few active drug users (9%) refused substance abuse counseling. A total of 310 patients were enrolled in a 21-day methadone detoxification program with a mean participation of 17.2 days. In addition, 33 patients were referred to a methadone maintenance program with a more than 90% retention rate. Of the 815 patients admitting to heroin or polysubstance abuse, 343 patients (42%) were enrolled in either a detoxification or maintenance program (Table 1).

The volume of patients treated in the clinic steadily increased during the first 7 months of operation before stabilizing at 275 to 300 visits per month (Figure 2). Only 6% of patients evaluated each month were admitted to the hospital, a rate that remained constant throughout the year. Although approximately one third of the patients (32%) were referred from the SFGH ED, an equal number of patients were walk-ins (self-referrals, 30%; Figure 3). The remaining patients were directed from city-operated clinics (16%), the SFGH inpatient wards (14%), private hospitals or clinics (4%), and the San Francisco County Jail (4%). Of note, multiple visits to the clinic by the same patient for unique infections were uncommon during the year in review (Figure 4). Two thirds of patients (68%) were treated only once, whereas 15% and 11% recorded 2 and 3 visits, respectively, accounting for nearly 95% of the total clinic population. Thus, only a small fraction of patients utilized the clinic with an inordinate frequency.

The ISIS Clinic provided comprehensive treatment of soft tissue infections with good clinical outcomes. The scope of clinical care is presented in Table 2. There were 7 cases of necrotizing fasciitis identified, and each patient was admitted to the hospital and treated in the operating room under general anesthesia. One of the patients with necrotizing fasciitis subsequently died due to adult respiratory distress syndrome within 48 hours of admission and accounted for the only death in the clinic. A total of 190 patients (6%) were admitted to the hospital, 68 (2%) were treatment failures, and 471 (14%) were lost to follow-up. Further analysis of the medical records for 75 randomly selected patients revealed no difference in the lost to follow-up rate for patients as a function of injection drug use (users vs nonusers, 9/58 vs 2/9; P = .84; drug use unspecified for 8 patients) with a mean follow-up of 162 ± 127 days.

The ISIS Clinic effectively treated soft tissue infections and decreased utilization of hospital resources previously used to treat these patients. The clinic significantly decreased the number of ED visits, hospital admissions, and inpatient days (Table 3). Although the total number of patient encounters in the hospital database (ED, inpatients, and ISIS Clinic) increased from 4709 encounters before opening the clinic to 6437 encounters after opening the clinic (+36.7%), the annual number of ED visits decreased by 34%. The number of admissions and inpatient bed use for soft tissue infections also decreased after the ISIS Clinic opened. Admissions to the surgical service decreased from an annual mean of 1326 to 699 (−47.3%). Inpatient bed use decreased by one third (−33.7%), constituting a decrease of 2150 acute care beds. The ISIS Clinic significantly reduced use of the operating room. Historically, an average of 977 soft tissue infections were treated in the operating room per year. During the clinic's first year, this number decreased by more than 70% (Table 3).

To determine the financial impact of the ISIS Clinic, charges for ED visits and inpatient bed and operating room use were calculated based on estimates provided by the SFGH administration. Thus, considering decreased ED visits ($792 200), acute care bed days ($6 450 000), and operating room use ($2 073 000) minus the cost of operating the ISIS Clinic (−$550 000), this proactive clinical intervention yielded an estimated net savings of $8 765 200 (Table 3).

When surveyed, 86% of patients expressed a positive view of the ISIS Clinic, reporting an excellent or very good overall impression of the services (Table 4). Most patients (92%) were treated in a timely fashion and evaluated by the surgeon within 60 minutes of arrival. Most patients were satisfied with pain management during and after their surgical procedure. In contrast, assistance with basic living needs was frequently inadequate; 23% of patients indicated that they rarely or never received adequate help in this regard. Yet, respondents noted that they always or usually had access to substance abuse counseling (82%) and were either definitely or very likely to recommend the clinic to a relative or friend (92%).

Heroin use in the United States has dramatically increased during the past decade. By the late 1990s, the number of soft tissue infections related to injection drug use treated at SFGH had reached epidemic proportions. In addition to consuming considerable resources, the customary approach to caring for these challenging patients was taxing health care providers. In response, the ISIS Clinic was created with the goal of providing compassionate, quality, yet cost-effective medical care to these patients. Also, we postulated that providing such services as outpatient care would reduce ED visits, hospital admissions, operating room cases, and costs to the hospital.

The clinic significantly reduced use of hospital resources while providing quality medical care for patients with soft tissue infections. Inpatient days and ED visits were each reduced approximately 34%. Operating room use was reduced by more than 70%. The clinic provided a ready alternative for patients who historically used the ED for care of their infections. Corroborative evidence that the ISIS Clinic had a significant impact on ED visits came from the annual Drug Abuse Warning Network report, which revealed a 12% decrease in drug-related ED episodes in San Francisco.¹⁰ During the clinic's first 3 months of activity most patients were referred from the ED. However, as the clinic entered its second quarter of operation the volume of walk-in patients (self-referrals) had noticeably increased. At year's end, the number of self-referrals equaled the number of referrals from the ED. The number of patients referring themselves to the clinic, combined with the very favorable ratings of the clinic by the satisfaction survey, strongly indicated the successful creation of a service preferred by our patients.

Everyone involved in the ISIS Clinic shared concerns regarding the potential risks to the patients of converting from an inpatient-based model of care to an outpatient-based model. Thus, the collection and analysis of prospective, detailed patient information, including outcome data, was deemed critically important. Overall, 84% of patients evaluated in the ISIS Clinic were successfully treated, with a mean follow-up of more than 5 months. Only 2% of patients were deemed treatment failures, with a 14% lost to follow-up rate. The high therapeutic success and low treatment failure rates aptly underscore the clinic's delivery of high-quality, effective care for soft tissue infections on an outpatient basis. The predominant concern regarding conversion to an outpatient-based model of care delivery was that many of our patients would be lost to follow-up and suffer from incompletely treated infections. In essence, this concern is rooted in the assumption that patients who use injection drugs are irresponsible and incapable of exercising good judgment when it comes to their own medical care. This fundamental, albeit paternalistic, attitude toward substance-abusing patients imbued our previous practice of admitting most patients with serious soft tissue infections to the hospital. Once patients were admitted, we could assume responsibility for them and ensure that they received antibiotics and timely care. To render care on an outpatient basis eliminated the clinical safety net that hospital admission represented. Consequently, monitoring patient follow-up was a critical component of the ISIS Clinic's initial and continuing evaluation.

Few patients (14%) did not return for their scheduled wound clinic appointment or interface with the county-sponsored health care system within 6 months of being treated in the clinic. This lost to follow-up rate is comparable to that observed for other clinics throughout the San Francisco Community Health Network. Interestingly, patients who admitted to substance abuse were no more irresponsible for follow-up treatment than other patients. In fact, injection drug users were equally as compliant with their medical care plan as nonusers. These data not only contradict the assumption that injection drug users are inherently irresponsible when it comes to their health care, but also suggest that patients battling drug addiction, like all patients, will readily utilize services provided in a user-friendly, nonjudgmental environment.

Nowhere in the country are the consequences of managed care and its attendant effects on health care finances more evident than in the San Francisco Bay Area. Thus, the need to redesign health care systems with an eye toward increased efficiency and cost-effectiveness played a prominent role in the design and implementation of the ISIS Clinic. After 1 year the clinic is estimated to have yielded net total savings of almost $9 million, representing a 45% reduction in the costs of treating soft tissue infections. Because we are unaware of data that document the incidence and magnitude of injection drug use–related soft tissue infections throughout the United States, we can only assume that other communities are similarly confronted with this medical and public health problem. Therefore, we believe that the ISIS Clinic could serve as a model intervention and thus have a significant impact on the treatment of this prevalent but often overlooked challenge.

Box Section Ref ID

Accepted for publication August 17, 2002.

We acknowledge the invaluable contributions of the following individuals whose efforts greatly contributed to establishing the ISIS Clinic: Mitchell Katz, MD, Melissa Welch, MD, MPH, Jacqui Caesar, RN, Gayling Gee, RN, Chen Lee, MD, Joshua Bamberger, MD, Hyjasmine Blanco, RN, Gene Marie O'Connell, and Anson Moon.

Corresponding author and reprints: Hobart W. Harris, MD, MPH, San Francisco General Hospital, 1001 Potrero Ave, Ward 3A, San Francisco, CA 94110 (e-mail: hharris@sfghsurg.ucsf.edu).