From Wiki Journal Club
Jump to: navigation, search
Little P, et al. "Antibiotic prescription strategies and adverse outcome for uncomplicated lower respiratory tract infections: prospective cough complication cohort (3C) study". BMJ. 2017. (357):j2148.
PubMedFull textPDF

Clinical Question

In adult patients with lower respiratory tract infections, how does the administration, timing or lack of antibiotics correlate to reconsultation for non-resolving symptoms, hospital admission, or death?

Bottom Line

For uncomplicated lower respiratory tract infections (LRTI), delayed antibiotics may decrease the rate of reconsultation for non-resolving symptoms, and either immediate or delayed antibiotics may not reduce the rates of hospital admission or death.

Major Points

Antibiotics have not demonstrated clear benefit when administered for acute, uncomplicated, ambulatory, community based lower respiratory tract infections, in clinical trials[1] or in a Cochrane review.[2] The antibiotic Prescription Strategies and Adverse Outcome for Uncomplicated Lower Respiratory Tract Infections: Prospective Cough Complication Cohort (3C) Study was a prospective, observational cohort study conducted in 522 primary care facilities in the UK. Including 28 779 patients (after eliminating those that had respiratory symptoms for other reasons such as cancer) they compared antibiotics prescribing strategies. When patients 16 years and older presented with new or worsening cough for less than 3 weeks were included. Patients were divided into one of three options, no antibiotics prescribed (n=7332), immediate antibiotics prescribed (n=17 628), or delayed antibiotics (n= 3819. Median delay was 3 days). There were several limitations in this trial. The first is that patients were recruited during the busiest time when also LRTI are most common, and time pressures to complete the performa by the physicians completed with patient care. There was no training offered to the physicians for assessing LRTI nor was there a mechanism for continuity across all participating sites and physicians. Blinding did not occur but due to the nature of the observed intervention it may not have been possible. We do not know what patient characteristics led to those patients that were or were not approached for recruitment into this trial and bias may be present. Major adverse events (reconsultation, hospital admission, or mortality) occur less than 1% of the time. A delayed antibiotic regimen should be utilized as it was significantly (P < 0.001) associated with a reduced number of reconsultations for worsening illness.


As of February 2019, no guidelines have been published that reflect the results of this trial.


  • Prospective, observational cohort study
  • N= 28,779
    • No antibiotics = 7,332 (25.4%)
    • Immediate antibiotics = 17,628 (61.3%)
    • Delayed antibiotics = 3,819 (13.3%)
  • Setting: 522 United Kingdom primary care facilities
  • Enrollment: October 2009 – April 2013
  • Mean follow-up: 30 days after initial consultation
  • Analysis:
  • Primary endpoint: Reconsultation in primary care or visit to an emergency department with progression of illness in 30 days after the index consultation, hospital admission or death


Inclusion Criteria

  • Age ≥ 16 years old
  • Acute lower respiratory tract infection with an acute infected cough as the main symptom (acute infected cough defined as a cough new or worsening for three weeks or less)

Exclusion Criteria

  • Other causes of acute cough (eg heart failure, acid reflux, fibrosing alveolitis, cancer)
  • Immunocompromised
  • Inability to consent (eg. severe mental illness)
  • Previous episodes of the same illness
  • Previous inclusion
  • admission to hospital on day of index consultation
  • confirmed pneumonia on radiography

Baseline Characteristics

  • Age ≥ 60 years
    • No Antibiotics 28.8%
    • Immediate Antibiotics 42.3%
    • Delayed Antibiotics 33.4%
  • Female
    • No Antibiotics 60.7%
    • Immediate Antibiotics 58.6%
    • Delayed Antibiotics 59.8%
  • Illness duration <7 days
    • No Antibiotics 47.7
    • Immediate Antibiotics 50.2%
    • Delayed Antibiotics 42.1%
  • Received pneumovax <10 years
    • No Antibiotics 14.1%
    • Immediate Antibiotics 20.7%
    • Delayed Antibiotics 15.7%
  • Ever Smoked
    • No Antibiotics 49.4%
    • Immediate Antibiotics 56%
    • Delayed Antibiotics 49.2%


  • No antibiotics
  • Immediate antibiotics (at time of consultation)
  • Delayed antibiotics (advised delay, median 3 days)


Primary Outcomes

Hospital admission or death after uncomplicated presentation within 30 days
No Antibiotics 0.3%
Immediate Antibiotics 0.9% (multivariable risk ratio 1.06, 95% CI 0.63 to 1.81) P = 0.84
Delayed Antibiotics 0.4% (multivariable risk ratio 0.81, 95% CI 0.41 to 1.64) P = 0.61

Secondary Outcomes

Reconsultation with non-resolving or worsening symptoms within 30 days
No Antibiotics 19.7%
Immediate Antibiotics 25.3% (multivariable risk ratio 0.98, 95% CI 0.90 to 1.07) P = 0.97
Delayed Antibiotics 14.1% (multivariable risk ratio 0.64, 95% CI0.57 to 0.72, P<0.01


  • There was no training for or mechanism to ensure quality of the diagnostic skills used by the physicians
  • Bias potentially occurred due to missing oxygen saturation data
  • Patients included in the trial were not blind to their antibiotic therapy or lack there of therapy
  • Patients were recruited during the busiest part of the year
  • Patients not approached for consent were not documented
  • Trial powered only for detecting odds ratio greater than 0.66
  • Did not describe the antimicrobial regiment utilized


  • Grant for independent research from the National Institute for Health Research (NIHR)

Further Reading

  1. Little P et al. Effects of internet-based training on antibiotic prescribing rates for acute respiratory-tract infections: a multinational, cluster, randomised, factorial, controlled trial. Lancet 2013. 382:1175-82.
  2. Smith SM et al. Antibiotics for acute bronchitis. Cochrane Database Syst Rev 2014. :CD000245.