Title: Lack of Positive Association Between Falls and Bacteriuria Plus Pyuria in Older Nursing Home Residents
Abstract: To the Editor: Falls are common in older nursing home residents. The 2004 National Nursing Home Survey estimated that 36% of residents had had at least one reported fall in the prior 6 months.1 Although multiple factors account for these falls, they often prompt empiric antibiotic treatment for urinary tract infection (UTI).2 Because antimicrobial prescriptions are frequent in nursing homes, and inappropriate use fosters the development of antibiotic resistance, it is important to examine the relationship between falls and UTI.3 Previous studies have reported a positive association between falls and UTI in older adults.4 The rate of UTI in older adults who fall has been reported to be as high as 18%, making it one of the most commonly associated diagnoses,5 but the diagnostic criteria for UTI have not been explicitly defined in these studies.5, 6 The rate of falls in residents with suspected UTI and subsequently confirmed with bacteriuria (>100,000 colony forming units/mL) plus pyuria (>10 white blood cells per high-power field on urinalysis) has not been systematically examined. The association between falls abd bacteriuria plus pyuria was investigated in a previously described cohort of noncatheterized nursing home residents with clinically suspected UTI.7 In this prospective cohort of 551 participants, there were 397 episodes of clinically suspected UTI in 228 residents. Nursing home staff caring for participants at the time of clinically suspected UTI identified falls. The longitudinal association between falls and bacteriuria plus pyuria was examined. Of 397 clinically suspected UTI episodes, 45 falls occurred in 39 participants (34 participants fell once, 4 fell twice, and 1 fell three times). Twelve participants had at least one urinary tract–specific signs or symptoms (costovertebral tenderness, suprapubic pain, hematuria, new or increased urinary incontinence, urgency, and frequency), 13 had changes in urinary characteristics (color or odor), eight had change in mental status, and four had dysuria. Table 1 shows a cross-classification of episodes of falls and bacteriuria plus pyuria. Of the 45 fall episodes, nine (20.0%) were cross-classified with bacteriuria plus pyuria. Of the 352 episodes without a fall, 137 (38.9%) were cross-classified with bacteriuria plus pyuria. A Rao-Scott chi-square statistic, adjusting for nesting of episodes within participants and participants within nursing homes, showed a statistically significant negative association between episodes of falls and bacteriuria plus pyuria (χ2 = 6.69, degrees of freedom = 1, P = .01), but in a multivariable regression model, using a generalized estimating equation approach to account for serial correlation between recurrent episodes of bacteriuria plus pyuria and controlling for important covariates (dysuria, fever, and change in mental status), the negative association between falls and bacteriuria plus pyuria lost its statistical significance (results not shown). Of the 45 falls, 22 had urinary dipstick testing performed. Seven were negative to leukocyte esterase and nitrate; none had bacteriuria plus pyuria. In this study, contrary to previous reports,4, 5 falls were not associated with bacteriuria plus pyuria. Eighty percent of the 45 fall episodes did not have bacteriuria plus pyuria, suggesting that UTI was unlikely to be associated with the fall. In addition, seven of the 22 participants who fell and were evaluated with a urine dipstick (32%) had a negative test for leukocyte esterase and nitrate, subsequently with no bacteriuria plus pyuria. Urinary dipstick testing for leukocyte esterase and nitrate has been shown to have a negative predictive value of 100%,8 consistent with the current findings. These results suggest that the majority of individuals in the cohort suspected of having UTI because of a fall would not have benefited from empiric antibiotics. Although participants in this study were from New Haven-area nursing homes only, they are representative of nursing home residents in the United States.7 Antibiotic resistance in nursing home residents is increasing and is often attributed to overuse of antibiotics.9 Nursing home residents are particularly susceptible to overuse of antibiotics because of nonspecific symptoms associated with infection such as altered mental status and falls. The results of the current study do not support claims of a positive association between falls and UTI, so empirical treatment with antibiotics for falls is not warranted and may contribute to the overuse of antimicrobials in nursing homes with negative consequences, including isolation of increasingly drug-resistant bacterial pathogens, adverse drug reactions, and secondary infections due to overgrowth of organisms such as Clostridium difficile. Conflict of Interest: The editor in chief has reviewed the conflict of interest checklist provided by the authors and has determined that the authors have no financial or any other kind of personal conflicts with this paper. K23 AG028691 (MJM), Claude D. Pepper Older Americans Independence Center P30 AG021342; National Institute on Aging, National Institutes of Health (MJM, VT, PHVN), T32 AI007517–12 (TR). Author Contributions: Theresa Rowe and Manisha Juthani-Mehta: study concept and design, analysis and interpretation of data, preparation of manuscript. Virginia Towle: acquisition of subjects and data, preparation of manuscript. Peter H. Van Ness: analysis and interpretation of data, preparation of manuscript. Sponsor's Role: None of the funding agencies had any role in the design, methods, subject recruitment, data collections, analysis, or preparation of the manuscript.