Impact of an infection consultation service for bacteraemia on clinical management and use of resources

Since 1993, the infection consultation service for bacteraemia has seen 310patients in the Medical and Surgical Directorates at Ninewells Hospital andKings Cross Hospital. A random sample of 100 was audited. Case-notes wereincomplete for five patients, leaving 95 fully-audited patients. Clinicaloutcome measures were death from infection, and readmission within 2 weeksof discharge. Initial treatment was inconsistent with antibiotic policy in46 patients (48%). Antibiotic treatment was changed in 37 (80%) of thesepatients: increased in intensity in 19 (41%) and decreased in 18 (39%).Changes were also made in 30 (61%) of the 49 patients whose initialtreatment was consistent with sepsis policy-increased in seven (14%) anddecreased in 23 (47%). Median daily antibiotic costs were lowered inpatients whose initial treatment was consistent with sepsis policy (pounds10.10 vs. pounds 7.28, p = 0.0274). However, in the other patients, savingswere balanced by increases (p = 0.7696). Consultation required oneconsultant session per week (3.5 h) and the audit required an additional 16consultant sessions. Seven patients died, but only one death was directlyrelated to infection. Six patients were readmitted to hospital within 2weeks, in three due to recurrence of infection. Changes to treatment wererecommended in the majority of patients, regardless of whether initialtreatment complied with the sepsis policy. The service primarilyredistributed resources rather than reducing costs. A fully audited servicerequires considerable consultant time, but we believe such time is wellspent.