Nursing home practitioner survey of diagnostic criteria for urinary tract infections.

Objectives: To identify clinical and laboratory criteria used by nursing home practitioners for diagnosis and treatment of urinary tract infections (UTIs) in nursing home residents. To determine practitioner knowledge of the most commonly used consensus criteria (i.e., McGeer criteria) for UTIs.

Design: Self-administered survey.

Setting: Three New Haven-area nursing homes.

Participants: Physicians (n=25), physician assistants (PAs, n=3), directors/assistant directors of nursing (n=8), charge nurses (n=37), and infection control practitioners (n=3).

Measurements: Open- and closed-ended questions.

Results: Nineteen physicians, three PAs, and 41 nurses completed 63 of 76 (83%) surveys. The five most commonly reported triggers for suspecting UTI in noncatheterized residents were change in mental status (57/63, 90%), fever (48/63, 76%), change in voiding pattern (44/63, 70%), dysuria (41/63, 65%), and change in character of urine (37/63, 59%). Asked to identify their first diagnostic step in the evaluation of UTIs, 48% (30/63) said urinary dipstick analysis, and 40% (25/63) said urinalysis and urine culture. Fourteen of 22 (64%) physicians and PAs versus 40 of 40 (100%) nurses were aware of the McGeer criteria for noncatheterized patients (P<.001); 12 of 22 (55%) physicians and PAs versus 38 of 39 (97%) nurses used them in clinical practice (P<.001).

Conclusion: Although surveillance and treatment consensus criteria have been developed, there are no universally accepted diagnostic criteria. This survey demonstrated a distinction between surveillance criteria and criteria practitioners used in clinical practice. Prospective data are needed to develop evidence-based clinical and laboratory criteria of UTIs in nursing home residents that can be used to identify prospectively tested treatment and prevention strategies.