Infectious Diseases PHARMACOEPIDEMIOLOGY OF URINARY TRACT INFECTIONS IN IOWA MEDICAID PATIENTS IN URBAN LONG-TERM-CARE FACILITIES Darrel C Bjornson, John P Rovers, Julie A Burian, and Nancy L Hall OBJECTIVE: To describe the therapeutic management of Medicaid patients with urinary tract infections (UTIs) in urban long-term-care facilities (LTCFs) and to link individual therapies to patient outcomes. DESIGN: Retrospective review of medical records in LTCFs of patients who had documented UTIs. METHODS: Patient data were collected from 17 LTCFs in the Des Moines, lA, metropolitan area during a I-year period starting January I, 1995. Patients with UTIs were selected from the LTCF infection control logs. Data collected on patients included demographics, concomitant diseases, type of UTI (i.e., symptomatic, asymptomatic, catheter-related), process measures for management, UTI treatment, patient outcomes, and follow-up. Patient outcome data were defined as either cure or no cure. A UTI cure was defined as a negative urine culture while taking antibiotic therapy and/or complete resolution of signs and symptoms, as well as no further treatment given within 2 weeks after the end of treatment. RESULTS: Data were collected on 310 patients who had at least one UTI over the I-year study period. Patients were primarily elderly (mean age 82.2 ± 12.3 y), white (95.1 %), and female (83.9%). Concomitant diseases were common and about one-fourth (23.0%) of the patients were catheterized. There were 536 UTI events (the unit of analysis) documented over the I-year period, with about one- half (45.9%) being UTIs with symptoms consistent with uncomplicated lower UTI. Nearly two-thirds (62.3%) of the patients were cured, based on the study definition; there was no association between cure and type of antimicrobial therapy (p = 0.99). Over one-third (35.2%) of the UTIs were treated with a quinolone antibiotic. Others were treated with trimethoprim/sulfamethoxazole (24.4%), nitrofurantoin (13.9%), a cephalosporin (10.4%), or ampicillin/amoxicillin (9.8%). Sixty-day follow-up showed no association between type of therapy and hospital readmission, physician follow-up visits, or subsequent UTIs. CONCLUSIONS: There were no differences in cure rates when comparing LTCF UTI patients receiving various regimens. With Darrel C Bjornson PhD. Associate Professor. College of Pharmacy and Health Sci- ences. Drake University, Des Moines, IA John PRovers PhannD, Assistant Professor, College of Pharmacy and Health Sci- ences. Drake University Julie A Burian, PhannD Student, College of Pharmacy and Health Sciences, Drake University Nancy L Hall PhannD. Pharmacy Resident. St. John's Hospital, Springfield. It. Reprints: Darrel C Bjornson PhD, 116 Fitch Hall. 2507 University Ave.. Drake Uni- versity, Des Moines. IA 50311, FAX 515/271-4171 This study was supported by the Iowa Medicaid Drug Utilization Commission under the auspices of the Department of Human Services, State of Iowa. outcomes being the same, the clinician should closely consider costs of drug therapy in selecting a treatment preference. KEY WORDS: long-term-care facilities, urinary tract infections, Medicaid, pharmacoepidemiology. Ann Pharmacother 1997;31 :837-41. ALTHOUGH DRUG UTILIZATION REVIEW studies can describe how a drug is used in patients, our previous work has shown that it is not an effective method to determine how a given disease state is being treated.' To determine how physi- cians are treating a specific disease, it becomes necessary to designate the disease state, rather than the drug, as the variable of interest. The investigator then determines the various drugs used to treat the disease, rather than the vari- ous diseases treated by a given drug. Few studies have used this pharmacoepidemiologic ap- proach to study the management of urinary tract infections (UTIs), Those that did studied only hospitalized patients>' Selection of an appropriate antimicrobial agent for UTIs has become complex because of the increasing number of compounds available, each with its characteristic spectrum of activity and adverse effect profile. Some authors suggest there is no evidence to support any superiority of bacterici- dal drugs over bacteriostatic drugs in UTIS.5 Studies that have specifically addressed the treatment of UTIs in the elderly are rare, and many reports do not dis- tinguish among the very old, ambulatory, chronically ill, or institutionalized elderly."! The diagnosis, pathogenesis, epidemiology, and treatment of U'TIs in the elderly are complicated by advanced age, organ dysfunction, and ex- posure to antibiotic-resistant organisms. Changes in dis- ease features or outcome are typically caused by the debili- ty from the chronic diseases that accompany advanced age, not age itself. The prevalence of bacteriuria in noncatheter- ized elderly residing in long-term-care facilities (LTCFs) is 15-60%.' Nearly one-half of residents of LTCFs are funded through the Medicaid system. to Data on the management of U'TIs with subsequent outcomes in the LTCF population are The Annals of Pharmacotherapy • 1997 July/August, Volume 31 • 837