Skip directly to search Skip directly to A to Z list Skip directly to navigation Skip directly to page options Skip directly to site content


For Healthcare Professionals

Implement at least one policy or practice to improve antibiotic prescribing, assess whether it is working and modify as needed.

Delayed Prescribing Practices or Watchful Waiting
Reference Interventions and Outcomes Methods, Participants, and Settings Intervention Conclusions
Chao JH, et al. Comparison of two approaches to observation therapy for acute otitis media in the emergency department. Pediatrics. 2008. 121(5):e1352–6. Intervention:

• Watchful waiting/observation therapy with no prescription or with a delayed antibiotic prescription


• Antibiotic use for AOM at 3 days (primary) and 7–10 days (secondary)

• Parental visit satisfaction


• Prospective randomized trial


• Children aged 2 to 12 years diagnosed with AOM and who met criteria for observation


• Pediatric emergency department of an urban public hospital in the United States (New York)


• 232 patients enrolled, 206 patients completed follow-up

• At 3 days: 87% parents of children in the observation group with no antibiotic prescription reported no antibiotic use versus 62% parents of children in the of children in the observation group with a delayed antibiotic prescription.

• At 7–10 days, 81% of the observation group with no antibiotic prescription reported no use of antibiotics compared with 53% in the group with a delayed antibiotic prescription.

• No differences in satisfaction were observed between the groups.

• Observation therapy was well accepted by parents of children with AOM.

• Observation without an antibiotic prescription led to lower antibiotic use for AOM than observation with a delayed antibiotic prescription without affecting visit satisfaction.


de la Poza A, et al. Prescription strategies in acute uncomplicated respiratory infections: A randomized clinical trial. JAMA Intern Med 2016. 176(1):21–9. Interventions:

4 antibiotic prescriptions strategies for acute uncomplicated respiratory tract infections.

o Delayed antibiotic prescription given to patients at the visit with instructions to wait to fill it unless not improving

o Delayed antibiotic prescription awaiting patient at clinic, patient to return and collect prescriptions if not improving

o Immediate antibiotic prescription issued at visit

o No antibiotic prescription issued at visit


• Primary: symptom duration and severity

• Secondary: antibiotic use, patient satisfaction, and belief about antibiotic effectiveness among patients complicated respiratory infections.


• Open-label, randomized clinical trial


• Adults with acute, uncomplicated respiratory infections


• 23 primary care clinics in Spain


• 405 adult patients with acute, uncomplicated respiratory infections

• Delayed prescription strategies led to lower antibiotic use:

o 91% of patients used antibiotics in the immediate prescription group;

o 33% of patients used antibiotics in the group with delayed prescription;

o 23% of patients used antibiotics in the group who had to collect the delayed prescription;

o 12% of patients used antibiotics in the no prescription group.

• Delayed and no prescription strategies led to “slightly greater” symptom burden.

• Similar satisfaction was observed among groups.

• Delayed prescription strategies for acute uncomplicated respiratory tract infections are effective in decreasing antibiotic use.
Francis NA, et al. Delayed antibiotic prescribing and associated antibiotic consumption in adults with acute cough. Br J Gen Pract 2012. 62(602):e639–46. Intervention

• No intervention; observational study


• Rates of delayed antibiotic prescribing in adults presenting with acute cough to primary care.

• Duration of advised delay

• Consumption of delayed antibiotic or another antibiotic at 28 days

• Factors associated with antibiotic consumption



• Prospective observational cohort study


• General practitioners

• Adult patients with acute cough


• 14 primary care networks in 13 European countries


• 3368 patients with acute cough

• About 6% (n = 210) were prescribed delayed antibiotics (median recommended delay 3 days).

• 44% (n = 75/169) with consumption data used the delayed prescription antibiotic course by 28 days

• 30% (n = 50/169) started on the day the prescription was written.

• 10% took another antibiotic by 28 days.

• 45% took no antibiotic by 28 days. Upper respiratory tract/viral infections diagnoses were associated with lower use of delayed prescription.

• Patients who wanted antibiotics were more likely to consume the antibiotics.

• Delayed antibiotic prescribing was not used often for adults presenting to primary care.

• Expanding delayed antibiotic prescribing and standardizing prescribing practices may improve antibiotic prescribing.

Little P, et al. Information leaflet and antibiotic prescribing strategies for acute lower respiratory tract infection: a randomized controlled trial. JAMA 2005. 22;293(24):3029–35.



• One of 3 prescribing strategies was used

• Immediate antibiotics

• No antibiotics

• Delayed antibiotics available by request after 14 days

• Information leaflet for acute lower respiratory tract infection


• Clinical signs and symptoms

• Reported antibiotic use

• Daily diary and satisfaction questionnaire



• Randomized controlled trial

• Factorial design involving 6 groups: leaflet or no leaflet and 1 of 3 prescribing strategies


• 37 English general practitioners

• Patients aged ≥3 years with acute uncomplicated lower respiratory infections


• Primary care clinics in England


• 807 patients recruited

• No implemented intervention altered cough duration or other clinical outcome.

• Cough lasted on average 11.7 days.

• The information leaflet did not have any impact on main outcome.

• Fewer patients in the delayed and control groups, compared with immediate antibiotic group, used antibiotics, were “very satisfied” with visit, and believed in the antibiotic effectiveness.


• Not prescribing antibiotics, or offering a delayed antibiotic prescribing is associated with minimal differences in symptom burden and may reduce antibiotic use.


Little P, et al. Delayed antibiotic prescribing strategies for respiratory tract infections in primary care: pragmatic, factorial, randomized controlled trial.

Brit Med J 2014. 348:g1606.



• Delayed antibiotic prescribing strategies

o Re-contact for a prescription (i.e., patient calls for the prescription)

o Post-dated prescription

o Post-visit collection of a prescription

• No antibiotic prescription


• Primary: Symptom severity at days 2–4

• Secondary: antibiotic use by 14 days and patient belief about antibiotic effectiveness


• Open, pragmatic, randomized controlled trial


• Patients aged ≥3 years with acute respiratory tract infections


• 25 primary care clinics in the United Kingdom


• 889 patients recruited

• No significant differences in symptom severity were observed between those who received no prescription and those receiving delayed prescription via any strategy.

• Symptom duration did not differ between groups, and no significant difference was observed for patient satisfaction.

• Those receiving antibiotics did not appear to benefit from them based on symptom severity scores.

• Interventions involving delayed antibiotic prescriptions or no prescription strategies resulted in fewer than 40% of prescribed antibiotics being used among patients.

• Interventions involving delayed prescriptions or no prescriptions were associated with less belief in antibiotic efficacy and similar symptom outcomes compared with immediate antibiotic prescriptions.


McCormick DP, et al. Nonsevere acute otitis media: a clinical trial comparing outcomes of watchful waiting versus immediate antibiotic treatment. Pediatrics 2005.115(6):1455–65.



• Watchful waiting (WW) versus immediate antibiotic prescription

• Educational intervention


• Patient satisfaction with care

• Resolution of symptoms

• Acute otitis media (AOM) failure/recurrence

• Nasopharyngeal colonization with antibiotic-resistant Streptococcus pneumoniae



• Single-blind, randomized controlled trial (investigators were blinded)


• Children aged 6 months to 12 years with nonsevere AOM


• Pediatric clinics in in the United States (Texas)


• 223 children recruited

• Parent satisfaction with care did not differ between treatment groups.

• Children treated with immediate antibiotics had faster symptom resolution.

• In the WW group, 66% of children did not take antibiotics by day 30.

• The WW group were reduced by 73% compared with the immediate antibiotic group.

• Immediate antibiotic treatment group had more antibiotic adverse drug events than WW group.

• Children in the immediate antibiotic group were more likely to have multi-drug resistant S. pneumoniae nasopharyngeal colonization at day 12.

• Immediate antibiotic treatment was associated with decreased treatment failures and improved symptom resolution compared with WW, but also higher adverse drug events and higher likelihood of carriage of multi-drug resistant S. pneumoniae.

• Classification of AOM severity, parent education, symptom management, followup care, and access to effective antibiotics when needed are all important in implementing watchful waiting for children with AOM.


Siegel R, et al. Treatment of otitis media with observation and a safety-net antibiotic prescription. Pediatrics 2003. 112(3):527–31.



• Delayed antibiotic prescription (“safety-net prescription”)


• Primary: parental willingness to treat AOM without antibiotics and with pain medicine alone

• Secondary: filling of antibiotic prescription, parents’ future plans to use antibiotics for AOM


• Cohort study


• Children aged 1 to 12 years with nonsevere AOM


• 11 pediatric clinics in the United States


• 194 children enrolled, 175 with complete follow-up

• At follow-up, 31% of parents had filled the antibiotic prescription.

• 63% of parents reported willingness in future to use pain medicine only without antibiotics for AOM.


• Safety-net prescriptions can decrease antibiotic use for non-severe AOM, and some parents find it an acceptable treatment strategy.


Spiro DM, et al. Wait-and-see prescription for the treatment of acute otitis media: a randomized controlled trial. JAMA. 2006. 296(10):1235–41.



• “Wait and see” (i.e., delayed) antibiotic prescription versus standard prescription for children with acute otitis media (AOM)


• Filling of the antibiotic prescription

• Clinical symptoms and symptoms resolution



• Randomized controlled trial


• Children aged 6 months to 12 years with AOM


• Emergency department in Northeastern United States


• 283 children

• More parents in the wait and see group did not fill the antibiotic prescription (62%) compared with the standard prescription group (13% did not fill antibiotic prescription, p<0.001).

• No differences between groups were observed for the frequency of fever, ear pain, or unscheduled medical visits.

• In the wait and see group, fever and ear pain were associated with filling the antibiotic prescription.

• Wait and see antibiotic prescriptions reduced antibiotic use in children with AOM.
Communication Skills Training
Reference Methods, Participants, and Settings Intervention Conclusions
Little P, et al. Effects of internet-based training on antibiotic prescribing rates for acute respiratory-tract infections: a multinational, cluster, randomized, factorial, controlled trial. Lancet. 2013. 382(9899):1175–82. Intervention

• Internet based training on communication skills, C-reactive protein (CRP) testing, or both versus standard care


• Changes in antibiotic prescribing for respiratory tract infections (RTIs)


• Cluster randomized controlled trial


• Primary care providers


• 246 primary care clinics in 6 European countries

• 4264 patients

• Training in CRP testing and communication skills independently led to reductions in antibiotic prescribing for RTIs, and combination of both trainings led to largest reduction.


• Internet training for CRP testing and communications skills led to reductions in antibiotic prescribing for RTIs.
Cals JW, et al. Enhanced communication skills and C-reactive protein point-of-care testing for respiratory tract infection: 3.5-year follow-up of a cluster randomized trial. Annals of Family Medicine. 2013. 11(2):157–64.



• Physician enhanced communication skills training

• Point-of-care C-reactive protein (CRP)


• Patient visits for respiratory tract infections (RTIs)

• Percent of RTI episodes treated with antibiotics



• Pragmatic, cluster-randomized controlled trial

• 3.5 years of follow-up


• Patients with family physician visits for RTIs


• 20 family practices in the Netherlands

• 379 patients

• No difference in number of patient visits for RTIs among groups.

• RTI episodes treated by physicians who received communications training were less likely to receive antibiotics in follow-up period (26% with communications training v. 39% control, p = 0.02).

• No difference in antibiotic treatment during follow-up for RTI episodes in CRP group.


• Communications training led to sustained reductions in the percent of RTIs leading to antibiotic prescriptions, while CRP testing did not.
Require Explicit Written Justification for Non-recommended Antibiotic Prescribing
Reference Interventions and Outcomes Methods, Participants, and Settings Results Conclusions
Meeker et al. Effect of behavioral interventions on inappropriate antibiotic prescribing among primary care practices: A randomized clinical trial. JAMA 2016. 315(6):562–70. Interventions: 3 behavioral interventions

• Suggested alternatives to antibiotics placed within electronic health records for these diagnoses

• Accountable justification required in medical record for non-recommended antibiotic prescribing

• Peer comparison to top-performing peers


• Rate of antibiotic prescribing for acute respiratory tract infections for which antibiotics are not indicated


• Cluster randomized clinical trial


• 248 primary care clinicians


• 47 primary care practices in the United States

• 31,712 visits for acute respiratory tract infections for which antibiotics are not indicated

o 14,753 during baseline

o 16,959 during intervention

• Antibiotic prescribing decreased from:

o Controls: 24.1% to 13.1%

o Suggested alternatives: 22.1% to 6.1% (p = 0.66 for differences compared with control group)

o Accountable justification: 23.2% to 5.2% (p<0.001)

o Peer comparison: 9.9% to 3.7 (p<0.001).

• Compared with the control group, no intervention showed significant diagnosis shifting.

• Accountable justification and peer comparison interventions reduced antibiotic prescribing for acute respiratory tract infections for which antibiotics are not indicated
Clinical Decision Support
Reference Interventions and Outcomes Methods, Participants, and Settings Results Conclusions
McGinn TG, et al. Efficacy of an evidence-based clinical decision support in primary care practices: A randomized clinical trial. JAMA Intern Med 2013. 173(17):1584–11. Intervention

• Clinical decision support involving integration of Walsh rule for streptococcal sore throat and Heckerling rule for pneumonia


• Frequency of antibiotic prescriptions and streptococcal tests in experimental versus control group

• Use of clinical prediction rule in EHR


• Randomized clinical trial


• Attending physicians, fellows, residents and nurse practitioners

• Patients with complaints consistent with pharyngitis or pneumonia


• Two large urban ambulatory care practices in the United States (New York)

• 168 primary care providers with 984 visits with clinical decision rule triggered

• Clinicians in the intervention group used the clinical prediction rules in 58% of visits.

• Intervention clinicians were less likely to prescribe antibiotics than control clinicians (RR = 0.75; 95% CI, 0.60–0.92).

• Number needed to treat to prevent one antibiotic prescription was 10.8.

• Intervention clinicians ordered rapid streptococcal tests for patients with pharyngitis less often than control clinicians (RR 0.75; 95% CI, 0.58–0.97).

• Clinical prediction rules integrated into EHRs can reduce inappropriate antibiotic prescribing.
Jenkins TC, et al. Effects of clinical pathways for common outpatient infections on antibiotic prescribing. Am J Med. 2013;126(4):327–35 e312. Intervention

• Clinical decision support targeting antibiotic prescribing for common conditions

• Patient education materials


• Change in antibiotic prescribing over time for non-pneumonia acute respiratory infections (ARIs)

• Change over time in broad-spectrum antibiotic prescriptions for ARIs


• Quasi-experimental study


• Clinicians working in primary care clinics


• Primary care clinics in the United States (Colorado), including adult and pediatric clinics; urban, suburban and rural clinics; academic and private providers

• 8 primary care clinics

• Antibiotic prescriptions for visits for non-pneumonia ARIs decreased from 42.7% to 37.9% (11.2% relative reduction) in the intervention group compared with 39.8% to 38.7% in the control group (2.8% relative reduction) during the intervention period.

• Use of broad-spectrum antibiotics decreased from 26.4% to 22.6% in the intervention group (14.4% relative reduction) compared with a 20.0% to 19.4% reduction in the control group (3.0% relative reduction).

• Clinical decision support was associated with reduced antibiotic prescriptions for non-pneumonia ARIs and reduced use of broad-spectrum antibiotics during one year of implementation.
Gonzales R, et al. A cluster randomized trial of decision support strategies for reducing antibiotic use in acute bronchitis. JAMA Intern Med 2013. 173(4):267–73. Interventions

• Clinical decision support, through the electronic medical record, or printed tools targeting antibiotic prescribing for acute bronchitis

• Clinician and patient education

• Audit and feedback

• Controls without interventions


• Reductions in antibiotic prescribing for acute uncomplicated bronchitis.


• Cluster randomized controlled trial


• Primary care clinicians


• 33 primary care practices in the United States (Pennsylvania)

• 12,776 visits for acute bronchitis

• Prescribing for acute bronchitis reduced by 11.7% in the print-based strategy and 13.7% in the EMR-based strategy.

• Prescribing at control sites increased slightly.

• Clinical decision support strategies for acute bronchitis can help reduce overuse of antibiotics in primary care.

• The observed effect in print-based versus computer-based interventions showed no significant differences.

Rattinger GB, et al. A sustainable strategy to prevent misuse of antibiotics for acute respiratory infections. PLoS One 2012. 7(12):e51147. Intervention

• Clinical decision support promoting adherence to clinical practice guidelines for acute respiratory infections (ARIs)


• Guideline concordance and proportion of inappropriate antibiotic prescribing

• Reductions in fluoroquinolone and azithromycin use



• Non-randomized retrospective controlled study


• Primary care providers for an outpatient veteran population


• Outpatient clinics in a veteran’s healthcare system in the United States


• 3831 patients

• Clinical decision support was associated with greater clinical practice guideline adherence (RR = 2.57 95% CI, 1.87 to 3.54).

• Inappropriate prescriptions for fluoroquinolones and azithromycin decreased from 22% to 3% (p<0.0001).


• A clinical decision support system decreased unwarranted use of fluoroquinolones and azithromycin for ARI and improved antibiotic use for ARI in an outpatient veterans’ healthcare system.
Linder JA, et al. Documentation-based clinical decision support to improve antibiotic prescribing for acute respiratory infections in primary care: A cluster randomized controlled trial. Inform Prim Care 2009. 17(4):231–40. Intervention

• Electronic health record-based clinical decision support for acute respiratory infection (ARI) — “ARI Smart Form” versus standard care


• Antibiotic prescribing for acute respiratory tract infections


• Randomized controlled trial


• Primary care providers


• 27 primary care clinics in the United States (Massachusetts)

• 21,961 visits for ARIs

• ARI Smart Form only used in 6% of eligible visits.

• Antibiotic prescribing for intervention clinics was not different compared with controls: odds ratio (OR) 0.8; 95% CI 0.6–1.2.

• When ARI Smart Form was used (per protocol analysis), ARI prescribing was modestly improved.

• A clinical decision support tool for ARIs, the ARI Smart Form, was rarely used by clinicians and thus did not improve antibiotic prescribing for ARIs.
Forrest, C. B., et al. Improving adherence to otitis media guidelines with clinical decision support and physician feedback. Pediatrics 2013. 131(4): e1071–1081.



• Clinical decision support (CDS) in an electronic health record system

• Audit and feedback to clinicians with peer comparison


• Physician guideline adherence for management of acute otitis media (AOM) and otitis media with effusion (OME)


• Factorial-design cluster randomized trial


• Primary care providers


• Primary care network in the United States (Pennsylvania, New Jersey, and Delaware)

• 24 practices with 139,305 visits for AOM and OME

• Guidelines were adhered to in 15% and 5% of AOM and OME cases, respectively during the baseline period.

• Improvements in guideline adherence was larger in visits with CDS and audit and feedback

• Audit and feedback combined with CDS did not improve guideline adherence beyond levels observed for audit and feedback alone.

• Both CDS and audit and feedback effectively increased adherence to guidelines for treatment of AOM and OME

• The effect of the individual interventions did not appear to be additive.

Call Centers, Nurse Hotlines, or Pharmacist Consultations
Reference Interventions and Outcomes Methods, Participants, and Settings Results Conclusions
Harper R, et al. Optimizing the use of telephone nursing advice for upper respiratory infection symptoms. Am J Manag Care 2015. 21(4): 264–270. Intervention

• Use of a nursing advice hotline to optimize self-care for upper respiratory infections


• Clinical outcomes associated with related cases

• Sufficiency of advice as evidence by no return calls within 7 days leading to a “higher” level of care, such as an in-person appointment.



• Retrospective observational study


• Adult patients 18 years and older who called into a self-care advice line for URI symptoms


• Large healthcare system in the United States (California)

• 279,625 calls

• For 88% of initial advice calls, self-care advice over the phone alone was sufficient.

• Most follow-up calls made by the patient were for additional advice or other information.

• URI symptoms can be effectively managed by nurses via a telephone advice line.


  1. Soumerai SB, Avorn J. Principles of educational outreach (‘academic detailing’) to improve clinical decision making. JAMA. 1990;263(4):549-56.

Top of Page