Background
Method
Criteria for considering studies for this review (Table 1)
Types of studies
Study characteristics | Include | Exclude |
---|---|---|
Participants | Members of general public or patients of all age groups with upper respiratory tract infections (such as sinusitis, pharyngitis, sore throat, otitis media, common cold, and acute cough) who seek treatment in any general practice setting. | Patients with lower respiratory tract infections (LRTIs) and those with chronic lung conditions (such as chronic obstructive pulmonary disease (COPD)) will be excluded. |
Interventions | Any intervention that is directed to patients, parents of patients (when the patients are children), public, or healthy individuals to reduce unnecessary use of antibiotics for URTIs in the primary care setting. | Interventions that are directed to healthcare providers or clinical staff will be excluded. The interventions that target patients indirectly (the primary and main effect of the intervention are directed to healthcare providers and patients benefit secondarily from that effect) will be excluded. Patient decision aids |
Comparisons | These comparisons will be included: • Interventions directed at patients/public versus no intervention. • Interventions directed at patients/public versus standard or usual care. • One form of intervention directed at patients/public versus another. | Other comparisons will be excluded. |
Outcomes | Primary outcomes: • Prescription or use of antibiotics for URTIs in the primary care setting. Secondary outcomes: • Public/patients’ satisfaction with the treatment/consultation. • Public/patients’ beliefs that antibiotics are effective for URTIs. • Re-consultation for the same illness in the next 2 weeks. | Studies that do not report the primary outcome will be excluded. |
Study designs | Randomized controlled trials (RCTs) Quasi-RCTs (a trial in which randomization is attempted but subject to potential manipulation, such as allocating participants by day of the week, date or birth, or sequence of entry into trial). CBA (controlled before and after) studies are included if: • There are at least two intervention sites and two control sites; • The timing of the periods for study for the control and intervention groups is comparable (that is, the pre- and post- intervention periods of measurement for the control and intervention groups should be the same); • The intervention and control groups are comparable on key characteristics. ITS (interrupted time series) studies will be included if: • The intervention occurred at a clearly defined point in time, and this was specified by the researchers; • There were at least three data points before and three data points after the intervention was introduced. | CBA and ITS will not be included if they do not meet the mentioned criteria. Other kinds of studies (e.g. observational, reviews) will be excluded. |
Language | English studies will be included. | Studies of other languages will be excluded. |
Types of participants
Types of interventions
Types of comparisons
Types of outcome measures
Type of language
Search methods for identification of studies
-
MEDLINE (OVID): 1946 to 2016 November 11
-
EMBASE (OVID): 1974 to 2016 November 11
-
Cochrane Central Register of Controlled Trials (CENTRAL) including the Cochrane Effective Practice and Organisation of Care (EPOC) and Cochrane Consumer Network (CCNet): inception to 2016 October 23
-
CINAHL: 1981 to 2016 October 24
-
Web of Science: 1900 to 2016 October 28
Data collection and analysis
Selection of studies
Data extraction and management
Risk of bias assessments
Measures of intervention effect
Dealing with common methodological issues
Unit of analysis issues in cluster-allocated studies
Inappropriate analysis of ITS studies
Dealing with missing data
Data synthesis
Assessment of heterogeneity
Assessment of reporting biases
Sensitivity analysis
Results
Description of studies
Results of the search
Included studies
Study | Methods | Participants | Interventions | Primary outcome |
---|---|---|---|---|
Delayed prescription | ||||
Arroll et al. [34] | RCT, 2 arms. Settings: New Zealand, one family practice (15 physicians) | Patients presenting with the common cold. Number of participants: 129 | Delayed antibiotic prescription | Antibiotic use |
Little et al. [35] | RCT, 3 arms. Settings: UK, general practices | Patients with sore throat. Number of participants: 716 | Delayed prescription of antibiotics, no antibiotics | Antibiotic use |
Little et al. [36] | RCT, 2 arms. Settings: UK, general practices | Children with acute otalgia. Number of participants: 315 | Delayed prescription of antibiotics | Use and collection of antibiotic prescriptions |
Little et al. [37] | RCT, 5 arms. Settings: UK, primary care setting | Patients with a respiratory tract infection. Number of participants: 556 | Delayed patient-led, post-dated prescription, delayed-collection, delayed-re-contact, no prescription | Symptom severity |
Pshetizky et al. [38] | RCT, 2 arms. Settings: Israel, primary care setting | Parents of children with acute otitis media. Number of participants: 81 | Delayed prescription of antibiotics | Antibiotic use |
Poza Abad et al. [39] | RCT, 4 arms. Settings: Spain, primary care setting | Patients with uncomplicated respiratory infections. Number of participants: 398 | Delayed patient-led, delayed collection, no antibiotic | Duration and severity of symptoms |
Worrall et al. [40] | RCT, 2 arms. Settings: Canada, primary care setting | Adult patients with acute URTIs. Number of participants: 149 | Post-dated delayed antibiotic prescription | Filling the prescription by the patients |
Patient/public information and education | ||||
Alexandrino et al. [41] | RCT, 2 arms. Settings: Portugal, 10 private daycare centres | Caregivers. Number of participants: 177 | Health education session (HES) | Impact of HES on the indicators of individual health and health care utilization |
Francis et al. [42] | Cluster RCT, 61 clusters, 2 arms. Settings: UK, general practices | Children with a respiratory tract infection. Number of participants: 558 | Interactive booklet | The proportion of children who attended a face-to-face consultation about the same illness during the 2-week follow-up period |
Lambert et al. [43] | ITS. Settings: UK, a single geographical population in the North East of England | People of the community | Mass media education | Prescribing rates for all microbial agents |
Lee et al. [44] | RCT, 2 arms. Settings: Singapore, primary care setting | Patients presenting RTI symptoms. Number of participants: 914 | Education through pamphlets and counselling scripts | Antibiotic prescription |
Little et al. [45] | RCT, 2 arms. Settings: UK, primary care setting | Adult patients. Number of participants: 2923 | Interactive website | General practitioner consultation |
Taylor et al. [46] | RCT, 2 arms. Settings: USA, primary care setting | Healthy children. Number of participants: 499 | Parental education through pamphlets and videos | Number of diagnoses of otitis media and sinusitis per study child, number of visits per child for which antibiotics (oral or intramuscular) were prescribed |
Interventions
Outcomes
Funding sources
Risk of bias in included studies
RCT/cluster RCT
ITS
Effects of interventions
Primary outcome
Secondary outcomes
Study | Patients’ satisfaction1 | Antibiotic Beliefs2 | Re-consultation3 |
---|---|---|---|
Delayed prescription | |||
Arroll et al. [34] (delayed vs. immediate) | Intervention 96% Control 94% OR = 1.47 (CI 0.32 to 6.85) | Intervention 76% Control 76% OR = 1.02 (CI 0.45 to 2.28) | Intervention 73% Control 65% OR = 1.50 (CI 0.71 to 3.17) |
Little et al. [35] (delayed vs. immediate) | Intervention 93% Control 96% OR = 0.61 (CI 0.25 to 1.49) | Intervention 60% Control 87% OR = 0.22 (CI 0.13 to 0.36) | Intervention 57% Control 79% OR = 0.35 (CI 0.22 to 0.55) |
Little et al. [36] (delayed vs. immediate) | Intervention 77% Control 91% OR = 0.32 (CI 0.16 to 0.65) | Intervention 46% Control 76% OR = 0.24 (CI 0.14 to 0.40) | Intervention 63% Control 83% OR = 0.35 (CI 0.20 to 0.62) |
Little et al. [37]4 (different variants of delayed prescriptions vs. no antibiotic) | Delayed re-contact 74% Post-dated 80% Delayed collection 88% Delayed patient-led 89% Likelihood ratio test X2 = 2.38 (P value = 0.67) | Delayed re-contact 74% Post-dated 73% Delayed collection 72% Delayed patient-led 66% Likelihood ratio test X2 = 1.62 (P value = 0.80) | Delayed re-contact 18% Post-dated 10% Delayed collection 14% Delayed patient-led 14% Likelihood ratio test X2 = 2.97 (P value = 0.56) |
Poza Abad5 et al. [39] (delayed collection vs. immediate and delayed patient-led vs. immediate) | P value6 = 0.14 | Delayed collection 84.4% Immediate 91.8% OR = 0.37 (CI 0.14 to 0.99) Delayed patient-led 81% Immediate 91.8% OR = 0.37 (CI 0.14 to 0.98) | Delayed collection 69.1% Immediate 85.7% OR = 0.62 (CI 0.19 to 2.06) Delayed patient-led 69.0% Immediate 85.7% OR = 0.35(CI 0.1 to 1.30) |
Patient/public information and intervention | |||
Francis et al. [42] (interactive booklet vs. control) | Intervention 90.2% Control 93.5% OR = 0.64 (CI 0.33 to 1.22) | Intervention 55.3% Control 76.4% OR = 0.34 (CI 0.20 to 0.57) | |
Little et al. [45] (online educational program vs. control) | Intervention 19.3% Control 19.3% OR = 0.93 (CI 0.73 to 1.16) |