Patient information leaflets to reduce antibiotic use and reconsultation rates in general practice: a systematic review

Journal Club Summary

8 July 2015

Facilitated by Dr Amanda McCullough

1.  Background

  • Antibiotic resistance is a major global health threat.
  • Information leaflets about infection and antibiotics are a simple intervention for clinicians to use during consultations.
  • Existing evidence for the effect of information leaflets on antibiotic use and reconsultation has not been synthesised.
  • This paper was of interest as Dr McCullough and Prof. Glasziou are part of a team currently conducting a related Cochrane review (O’Sullivan et al. 2014).

2.  Paper presented

  • Type: Systematic review of randomised controlled trials and non-randomised studies. Results were presented narratively.
  • Participants: Patients with a range of common infections presenting at general practice (GP) consultations in developed countries
  • Intervention: Information leaflet given by GPs or other GP staff during a consultation
  • Comparator: No leaflet
  • Outcome: Antibiotic prescription rates and antibiotic use, reconsultation during same illness episode and intention to reconsult

We appraised this paper using the FAITH method:

Find: We were satisfied that authors searched two databases using appropriate search terms and did not exclude based on language. We thought they may have also considered searching CINAHL for studies of nursing and allied health professions.

Appraise: The authors used the Cochrane Risk of Bias tool to appraise included studies. The group were satisfied with this approach. However, we discussed the update to this tool which states that the blinding domains (personnel and outcome assessors) should be completed for each outcome. The authors did not appear to do this.

Include: The authors included all 8 studies in the analysis irrespective of quality. A single non-randomised study was presented. We were satisfied with this approach.

Total: The authors did not meta-analyse any of the included studies because they believed there was too much methodological heterogeneity. Some members of the group would have been interested to see a meta-analysis of the respiratory disease studies (given their similarity), but accepted that the research question was to explore the effect on all common infections.

H: The authors noted there was methodological heterogeneity – statistical heterogeneity not explored. See comment above.

3.  Summary of results

Of 2512 articles screened, 23 full texts were reviewed and 8 studies (5 for respiratory tract infections) of 3407 patients were included. The authors rated all studies as having a high risk of bias for blinding of participants and personnel.  Reduced antibiotic prescribing was recorded in one of two studies that measured it. Reduced antibiotic use was reported in two out three studies that measured it. Lower risk of reconsultation was reported in one out of four studies that measured it and intention to reconsult was reduced in one out of two studies that measured it.

4.  Discussion/Journal Club commentary

The authors concluded that the “use of written information leaflets in general practice consultations are effective in reducing antibiotic prescribing, and actual antibiotic use by patients and their intention to reconsult for future episodes of illness. It is unclear whether information leaflets also actually affect reconsultation rates.”

The group generally agreed that this was a useful paper and liked the arrangement of the data in Table 3. We discussed a number of areas:

  1. In this review, two reviewers screened 200 titles and abstracts (with no disagreements) and a single author screened the rest with double checking by a second. We noted that individual reviewers miss around 3% of articles; thus, this approach may have led to a number of missed articles (Doust et al 2005).
  2. The authors stated they would exclude multifaceted interventions but included a study by Francis et al. 2009. It included both information and training in communication skills (and could be classed as a shared decision-making intervention). However, they did acknowledge this and stated that it may have affected the results of this study.
  3. The authors assessed the risk of bias as high for blinding of participants and personnel for each study. GRADE processes suggest that the risk of bias for blinding should be assessed for each outcome (rather than study) so that the objectivity or subjectivity of the specific outcome measure can be taken into account when deciding on whether bias was likely.

Overall, we believed this was a well conducted review. Clinicians in the journal club had a lively debate about whether they would use information for their patients with common infections. We concluded that it was one tool that you could have in your toolkit to help minimise antibiotic use and the risk of antibiotic resistance.

5.  Reference(s)

De Bont et al. Patient information leaflets to reduce antibiotic use and reconsultation rates in general practice: a systematic review. BMJ Open 2015; 5: doi:10.1136/bmjopen-2015-007612

Doust et al. Identifying studies for systematic reviews of diagnostic tests was difficult due to the poor sensitivity and precision of methodologic filters and the lack of information in the abstract. Journal of Clinical Epidemiology 2005; 58 (5): 444-449

Francis et al. Effect of using an interactive booklet about childhood respiratory tract infections in primary care consultations on reconsulting and antibiotic prescribing; a cluster randomised controlled trial. BMJ 2009; 339:b2885

O’Sullivan et al. Written information for patients on the use of antibiotics in acute upper respiratory infections in primary care. Cochrane Database of Systematic Reviews 2014; Issue 10: CD011360.