Journal of Clinical Epidemiology 54 (2001) 710–718 0895-4356/01/$ – see front matter © 2001 Elsevier Science Inc. All rights reserved. PII: S0895-4356(00)00352-8 Urinary tract infection among women aged 40 to 65: Behavioral and sexual risk factors Betsy Foxman a, *, Patricia Somsel b , Patricia Tallman a , Brenda Gillespie c , Raul Raz d , Raul Colodner d , Deepika Kandula a , Jack D. Sobel e a Department of Epidemiology, University of Michigan School of Public Health, 109 Observatory St., Ann Arbor, MI 48109, USA b Regional Medical Laboratories, Battle Creek, MI, USA c Center for Statistical Consultation and Research and Department of Biostatistics, University of Michigan, Ann Arbor, MI, USA d Infectious Disease Unit and Microbiology Laboratory, Ha’Emek Medical center, Afula, Israel, and Faculty of Medicine, Technion, Haifa, Israel e Division of Infectious Diseases, Department of Internal Medicine, Wayne State University School of Medicine, Detroit, MI, USA Received 11 January 2000; received in revised form 25 September 2000; accepted 14 October 2000 Abstract We conducted a case–control study to explore the role of health behavior and sexual and medical history on urinary tract infection (UTI) risk among otherwise healthy women aged 40–65. Cases and controls were recruited from nine practices and clinics in Michigan and a single clinic in Israel. In both countries, several factors were reported significantly more frequently among UTI cases than controls: a previous UTI within 12 months, incontinence symptoms, a recent episode of 30-plus minutes of cold hands, feet, back or buttocks, and recent antibiotic use. Cases were less likely than controls to report recent estrogen use, but the results were only statistically significant in Michigan. Sexual activity during the previous 2 weeks and having ceased menses were modestly, but not statistically significantly, protec- tive at both study sites. Risk factors for UTI among women 40–65 differ from those for younger women and these differences cannot be attributed solely to changes in menopausal status. © 2001 Elsevier Science Inc. All rights reserved. Keywords: Sex behavior; Incontinence; Antibiotic use; Cold exposure; Cranberry juice 1. Introduction Urinary tract infection (UTI) is a common condition among women in their middle years; an estimated 10% of women 40 to 65 report a physician-diagnosed UTI that was treated with prescription medication during the previous year [1]. Despite the high incidence, most UTI research has focused on younger women (40 years of age). Therefore, it is unknown whether the most common UTI risk factors among younger women (frequent vaginal intercourse, sper- micide, diaphragm and condom use, UTI history and recent antibiotic use) are associated with UTI among middle-aged women. There are hints in the literature that some risk factors may be specific to UTI among women aged 40 to 65. Sev- eral chronic conditions, including diabetes, stroke, demen- tia, incontinence, hysterectomy and atrophic vaginitis in- crease with age, and were found more frequently among women with first UTI in this age group than matched con- trols in a large record review study [2]. Women with diabe- tes, stroke and hysterectomy are more likely to have voiding defects and increased risk of bacteriuria and subsequent symptomatic disease [3]. The prevalence of hysterectomy almost doubles between ages 35–39 and 40–44: from 8.8% to 16% [4]. Risk of urinary incontinence [5] and bacteriuria [6] both rise with age; whether incontinence increases risk of UTI or UTI leads to incontinence is unknown. Women 40 to 65 are peri-, pre-, or postmenopausal, and likely to be initiating estrogen replacement therapy. Estro- gen replacement therapy minimizes physiologic changes as- sociated with menopause that lead to genitourinary symp- toms and potentially UTI, but whether estrogen replacement therapy is an appropriate therapy for preventing recurring UTI among postmenopausal women is uncertain [7]. We conducted a case–control study of women aged 40 to 65 to explore the relationship of various hypothesized be- havioral, sexual and medical history variables and risk of UTI. Our age range was chosen to bracket the menopausal years so we could also investigate the effects of menopause on UTI risk. Because of collaborative relationships, partici- pants were recruited from Michigan, USA and Israel. A * Corresponding author. Tel.: 734-764-5487; fax: 734-764-3192. E-mail address: bfoxman@umich.edu