Impact of guidelines for the diagnosis of urinary tract infection on trimethoprim susceptibility of Escherichia coli

In an attempt to improve immediate patient management and develop more rational use of the laboratory, we promoted the use of biochemical reagent strips for screening urine and introduced guidelines for result interpretation in both community and hospital settings. Clinical criteria were used to allocate patients into high or low UTI risk groups which, together with the result of strip tests for the presence of urinary leucocyte esterase and nitrite, guided clinicians on the need for empirical therapy and use of the laboratory for urine culture (see Table). Trimethoprim was advised for first-line empirical therapy for UTI uncomplicated by the sepsis syndrome or pregnancy in both the community and hospital settings. The mean (S.D.) quarterly usage by GPs in the 12 months before introduction was 13 338 urine samples compared with 11 449 in the following 12 months, a drop of 14.2% (95% confidence interval: 10.6‐17.7% P � 0.0001). Hospital clinician usage dropped from a mean 12 269 per quarter before