Delaying antibiotic prescribing made little difference to most symptoms of respiratory infection. It reduced antibiotic use and did not affect patient satisfaction compared with immediate prescribing of antibiotics.
Increasing antibiotic resistance is a global health concern. Many people don’t realise that viruses cause most respiratory infections and that antibiotics won’t help. The strategy allows some time for symptoms to improve naturally.
This review of the latest evidence on delayed prescribing for self-limiting respiratory infections is in line with current guidance. On the whole delaying antibiotics made little difference to symptoms compared with immediate use although certain symptoms, like malaise and fever in sore throat, might last a bit longer.
The 11 studies differed widely by patient populations, delay strategies, antibiotics given and settings. This makes it difficult to draw firm conclusions on where delayed prescribing is most appropriate.
Nevertheless delaying antibiotics seems a worthwhile strategy to reduce antibiotic use.
Why was this study needed?
Antibiotic use is rising in England, which is driving an increase in resistant bacteria. Between 2011 and 2015, antibiotic use increased by 6.5%. It’s estimated that one in five people expect to receive antibiotics when they go to the doctor.
Strategies to reduce antibiotic use are urgently needed to prevent a situation where serious infections can no longer be treated, and even standard surgical procedures become hazardous. Otherwise, it’s estimated that antibiotic resistance may account for 10 million deaths worldwide each year by 2050.
Delayed antibiotic use involves prescribing antibiotics, but advising people not to use them unless symptoms worsen ('patient-led delay'), or asking them to collect a prescription at a later date if they are getting no better. This strategy could educate the public while giving a sense of security to both patients and practitioners.
This Cochrane update adds one trial to the last 2013 review.
What did this study do?
The review included 11 randomised controlled trials involving 3,555 people with respiratory tract infections. Most studies compared delayed with immediate antibiotics, though four compared advice on a delayed prescription with no prescription.
Five studies included children only, two adults only, and four mixed populations. Six were conducted in primary care, three in paediatric clinics and two in the emergency department. Studies came from the UK, New Zealand, US, Spain and Jordan.
Delayed prescribing involved advice to use antibiotics only if symptoms worsen (or to return for a prescription) after more than 48 hours. Studies varied considerably by patient population, treatment protocol and follow-up time. These differences precluded meta-analysis for most outcomes.
Trials were of moderate quality for the main outcomes. The most likely sources of bias were that study personnel were aware which group participants were allocated to, either at time of randomisation or data collection.
What did it find?
- Delayed prescription reduced antibiotic use. Thirty-five people per 100 used antibiotics in the delayed group compared with 93 per 100 in the immediate group (odds ratio [OR] 0.04, 95% confidence interval [CI] 0.03 to 0.05; seven studies, 1,963 people). Antibiotic use was higher in the delayed group for studies where the comparison was with a no-antibiotic group (four studies, 1,241 people).
- Five studies (1,573 people) looked at sore throat. Three showed no difference in pain between delayed and immediate antibiotics, while two found increased pain or pain duration with delayed use. Two studies found that delaying antibiotics increased fever on day three (by a mean 0.53 degrees C, 95% CI 0.31 to 0.74). Another study also found malaise was more common in the delayed group on day three. There was no difference in symptoms in two studies comparing delayed with no antibiotics.
- Three studies (830 people) looked at middle ear infection. Two compared delayed with immediate antibiotics with conflicting results. One study found no difference in pain and fever on days four to six; the other again found that more children in the delayed group reported malaise on day three. One study found no difference in pain or fever between the delayed and no antibiotics.
- Three studies (1,402 people) found no difference in cough (bronchitis) between delayed, immediate and no-antibiotic groups. One person in the no-antibiotic group developed pneumonia. For the common cold, two studies (534 people) showed no difference in symptoms between delayed and immediate groups. One study (405 people) showed an advantage of delayed over no antibiotics for reducing pain, fever and cough duration.
- There were similar levels of patient satisfaction between the delayed and immediate groups (OR 0.65, 95% CI 0.39 to 1.10; six studies, 1,633 people). Satisfaction was, however, better in the delayed group compared with the no-antibiotics group (OR 1.49, 95% CI 1.8 to 2.06; four studies, 1,234 people).
What does current guidance say on this issue?
NICE guidance recommends advising people with self‑limiting conditions such as common cold and sore throat about self-management and the adverse effects of overusing antibiotics.
NICE recommends using a delayed strategy as an alternative to immediate prescribing if there is uncertainty whether a condition is self-limiting or the person is likely to deteriorate. This still encourages self-management but allows a person to access antibiotics without another appointment if their symptoms get worse.
What are the implications?
Most respiratory infections are self-limiting, and complications are unlikely. A delayed prescribing strategy seems an effective approach to tackle overuse of antibiotics. It could also reassure patients who may have concerns about managing without antibiotics.
The variety of studies makes it difficult to establish in which settings (e.g. general practice or hospitals) and for which patient groups delayed prescribing could be used safely and most effectively.
It is possible that the delayed prescription strategy reinforces a belief that antibiotics might be needed and this theory was not tested here.
Continuing to educate people that antibiotics aren’t likely to make a difference to their symptoms, with advice to return if symptoms get worse, still seems an essential part of care.
Citation and Funding
Spurling GKP, Del Mar CB, Dooley L, et al. Delayed antibiotic prescriptions for respiratory infections. Cochrane Database Syst Rev. 2017;(9):CD004417.
The Cochrane Acute Respiratory Infections Group is supported by The National Health and Research Council (NHMRC) of Australia, Cochrane Australia, Bond University, University of Queensland, General Practice Education and Training, Australia and previously by NIHR infrastructure funding.
NHS Choices. Respiratory tract infections. London: Department of Health; updated 2015.
NICE. Antibiotic stewardship. QS121. London: National Institute for Health and Care Excellence; 2016.
NICE. Respiratory (self-limiting): prescribing antibiotics. CG69. London: National Institute for Health and Care Excellence; 2008.
Public Health England. Guidance: Health matters: antimicrobial resistance. London: Public Health England; 2015.
Public Health England. Management and treatment of common infections. Antibiotic guidance for primary care: for consultation and local adaptation. London: Public Health England; 2017.
Produced by the University of Southampton and Bazian on behalf of NIHR through the NIHR Dissemination Centre