Delaying Antibiotics for Respiratory Infections
What effect does delayed prescribing of antibiotics have on the clinical course of respiratory infections and the likelihood of complications?
Delayed prescribing appears to be a reasonable and safe option in patients with cough and in those who do not appear very ill, but it cannot be recommended for children with sore throat unless streptococcal pharyngitis has been ruled out. Delayed prescribing of antibiotics results in a small increase in symptoms in some groups of patients at day 3. However, this risk must be balanced against the benefits of fewer antibiotic side effects, less antibiotic resistance, and lower cost.
Delayed prescribing is one strategy for reducing antibiotic use for common respiratory tract infections (RTIs). Most of these infections are viral and do not benefit from the use of antibiotics. Delayed prescribing means giving patients a prescription but suggesting that they not fill it unless they begin to feel worse or develop specific symptoms, or requiring that patients call the practice or pick up a prescription at a later date if their symptoms persist or worsen. A previous systematic review1 showed that delayed antibiotic prescribing for RTIs reduces by one half the number of patients who take an antibiotic. In this review, Spurling and colleagues identified seven good-quality randomized controlled trials that assigned patients to immediate or delayed antibiotics and followed them prospectively to determine their clinical outcomes.
The studies included patients with sore throat, common cold, otitis media, and cough. The results were mixed. In three studies of unselected children with sore throat (many of whom likely had streptococcal infection), patients in the delayed antibiotic group were more likely to have a fever on day 3. Findings were similar for pain and malaise: some studies found a benefit with immediate antibiotics, and some did not. Studies showing a benefit tended to enroll sicker patients.
Delayed antibiotic prescribing often is used in adults with cough; only one study considered this group, and it found no difference in outcomes. There were no consistent differences in reconsultation rates or complications between patients who received delayed treatment and those who received treatment immediately.