Modified Reporting of Positive Urine Cultures Collected From Indwelling Catheters, a Randomized Controlled Trial

The proposed study is a randomized trial of two methods of laboratory reporting in which physicians are the main research participants. At the time of positive urine culture results, the patient will be randomized by computer generated random number placed into serially numbered sealed, opaque envelopes into two equal groups. One group will receive modified reporting, with a report that states "This POSITIVE urine culture collected from an indwelling catheter may represent asymptomatic bacteriuria or urinary tract infection. If urinary tract infection is suspected clinically, please call the microbiology laboratory for identification and susceptibility results." The other group will receive conventional reporting of identification and susceptibility. Physicians will then have the option of calling the laboratory to receive the results or not. Complete results will be released by telephone and laboratory information system to physicians who call to request them. Physicians will be informed about the study prior to initiation, and debriefed about the study after the results have been collected.

Inclusion criteria will include consecutive positive urine cultures collected from indwelling catheters among patients admitted to acute care, that are greater than or equal to 18 years of age. Inpatients must be admitted to Health Sciences Center or St. Clare's Mercy hospitals only, in order to facilitate access to inpatient records.

Exclusion criteria will include pregnancy, antibiotic treatment at the time of collection, patients in the Intensive Care Unit and patients with blood neutrophils <1.0 within 7 days, which will help protect immunocompromised individuals.

Investigators will review patient charts at 24 and 72 hours and 7 days after collection. After randomization and reporting, a physician investigator will assess inpatients for the true diagnosis of asymptomatic bacteriuria (AB) or urinary tract infection (UTI). Health records will be accessed including demographics, treatment decisions and outcomes (untreated UTI or pyelonephritis). Frequency of physician calls requesting complete reporting will be recorded.

The research hypothesis is that restricted reporting will reduce the rate of inappropriate treatment prescribed by physicians. Among inpatients, the expected rate of inappropriate treatment in the control group will be 50 percent, and 20 percent in the intervention group. Accepting a risk of type 1 error of five percent, and a risk of type 2 error of twenty percent, the study will recruit 72 patients. In order to account for missing data, recruitment will be increased to 100 patients. The statistical test to be used is a comparison of proportions between two groups (T test, two sided analysis). An intention to treat analysis including all patients randomized will be performed.