Dialogue Around Respiratory Illness Treatment (DART)

Children's Hospital of Philadelphia, University of California, Los Angeles, Portland State University, NorthShore University HealthSystem, American Academy of Pediatrics, University of Washington

Unwarranted use of antibiotics for pediatric acute respiratory tract infections (ARTIs) and use of second-line, broad spectrum antibiotics for bacterial ARTIs has contributed to the rapid development of resistance in many strains of bacteria. Provider-parent communication during pediatric visits for ARTIs strongly influence antibiotic prescribing rates. The overall goal of this study is to develop and test a distance learning quality improvement (QI) program called Dialogue Around Respiratory Illness Treatment - DART. The DART program aims to improve provider communication practices and treatment decisions during pediatric ARTI visits, with the ultimate goal being to decrease rates of antibiotic prescribing for these illnesses in children.

The specific aims of the study are:1) to test the DART program's effectiveness on a) overall antibiotic prescribing rates for all pediatric ARTIs and b) first-line antibiotic prescribing rates for bacterial ARTIs (acute otitis media, Group A Streptococcal pharyngitis, and sinusitis), 2) to test the DART program's effectiveness in changing providers' communication practices during ARTI visits, 3) to determine visit-specific satisfaction levels for parents of children seen by study providers and assess how satisfaction changes as a function of exposure to the DART program, and 4) to assess the cost of implementing the DART program and its impact on health care expenditures. To accomplish these aims, we propose a quasi-experimental study utilizing a stepped wedge design. In collaboration with two practice-based research networks: the electronic Pediatric Research in Office Settings (ePROS) network and the NorthShore Pediatric network, we will recruit 20 practices to participate. Over a 20 month period, the intervention will be sequentially deployed to 4 groups of practices (5 practices/group). Each practice will function as part of the control group until they receive the intervention. The DART program's effectiveness will be assessed by examining provider antibiotic prescribing rates for ARTIs (both overall and first-line), communication practices, and parent satisfaction both pre- and post-intervention exposure using survey and electronic health record data. If the DART program is effective for both reducing ARTI antibiotic prescribing to levels consistent with bacterial prevalence rates and increasing use of first-line antibiotics for bacterial ARTIs, we will have an innovative, highly disseminable QI intervention program to further address this critical public health problem.

Physicians, pediatric nurse practitioners and physician assistants (referred to as providers from here forward) will be recruited from 20 practices, with a maximum of four providers participating from a single practice, for a maximum of 80 providers. Providers will receive an online tutorials, interactive group webinars, simulated booster sessions as well as feedback reports.

The number of parents who participate will depend on the number of providers who agree to participate at each of the 20 practices. The total could range from a minimum of 1800 parents to a maximum of 7200 parents. Throughout the study, parents at participating practice sites will be offered the opportunity to complete a DART Parent Survey after their child's visit.

For each group of practices, during the first month of intervention participation, study staff at Seattle Children's Research Institute will send each participating provider within a group/wedge, via email, a password to access the online tutorial. Participants will be asked to view the 40-minute tutorial at a time that is convenient for them but prior to the first webinar. Investigators will be able to track whether and how often participants access the online tutorial as a measure of engagement in the intervention. PROS will follow up, via email or phone, with providers who have not completed the tutorial within a three week timeframe.

The first webinar, focused on making appropriate antibiotic choices for bacterial ARTIs, will be held at the beginning of month 2 of intervention participation, and the second webinar, focused on recommended communication practices during visits for ARTIs, will be held at the beginning of month 3 of intervention participation. Due to webinar faculty time constraints, we will establish dates/times for the webinar prior to recruitment. Providers will be given information on webinar dates in both the randomization email, as well as include the dates in the study box letter. The webinars will be recorded to facilitate attendance for those who are unable to make the scheduled date/time. Included in the recording will be the presentation and any questions and answers from participants.

An online link to the first booster video vignette will be sent to practice participants via email at the beginning of month 5 of intervention participation for each group/wedge. The link will lead to a website in which the provider will need to create a user name and password to access the booster videos. Two additional booster vignettes will be made available for viewing at the beginning of months 7 and 9 of intervention participation. For each of the three booster sessions, up to three reminders to complete viewing the video vignette will be sent over a 2-week period. The study team will be able to monitor whether participants access and complete the questions at the end of each vignette.

During the intervention phase, providers will also receive 6 audit and feedback reports given by parents. The first report will have two parts: The first part will focus on the provider's overall antibiotic prescribing rates for all ARTIs (viral and bacterial) and rates of using second-line antibiotics for bacterial ARTIs. The second part will report the rates at which the provider uses the targeted communication practices as well as their average satisfaction scores.

Throughout the study, parents at participating practice sites will be offered the opportunity to complete a DART Parent Survey after their child's visit.

This outcome will determine whether rates of prescribing antibiotics during pediatric ARTI visits change as a function of the intervention. A baseline period will be compared to periods both during the intervention and post-intervention. For each time-period, the investigators will assess the number of ARTI visits (denominator) where antibiotics were prescribed (numerator).

This outcome will determine whether rates of prescribing antibiotics during pediatric visits for viral ARTI (bronchitis, non-streptococcal pharyngitis, or viral upper respiratory infection) change as a function of the intervention. A baseline period will be compared to periods both during the intervention and post-intervention. For each time-period, the investigators will assess the number of viral ARTI visits (denominator) where antibiotics were prescribed (numerator).

This outcome will determine whether rates of prescribing antibiotics during pediatric visits for pharyngitis (both non-streptococcal and streptococcal) change as a function of the intervention. A baseline period will be compared to periods both during the intervention and post-intervention. For each time-period, the investigators will assess the number of visits for pharyngitis (denominator) where antibiotics were prescribed (numerator).

These outcomes will determine whether rates of prescribing second-line antibiotics during pediatric visits for streptococcal pharyngitis, sinusitis, and acute otitis media change as a function of the intervention. A baseline period will be compared to periods both during the intervention and post-intervention. For each time-period assessed, the investigators will assess the number of visits for streptococcal pharyngitis, sinusitis, or acute otitis media (denominators) where second-line antibiotics were prescribed (numerators).

During exposure to the intervention, provider use of combined negative and positive treatment recommendations will increase during visits for ARTI compared to baseline. Investigators will pool parent survey data and calculate rates of positive and negative treatment recommendations for each eligible visit for each period.

During exposure to the intervention, provider use of contingency plans will increase during visits for ARTI compared to baseline. Investigators will pool parent survey data and calculate rates of contingency plan use for each eligible visit for each period.

Providers exposed to the intervention will have higher parent-reported visit-specific satisfaction scores during exposure to the intervention compared to baseline. To measure satisfaction the investigators will use an adapted version of the 3-item communication composite and the 1-item global visit satisfaction scale from the Consumer Assessment of Healthcare Providers and Systems Clinician and Groups (CG-CAHPS®) Survey. The survey items were adapted so that they are addressed to parents of patients rather than patients themselves. Both the 3-item composite and the 1-item global visit satisfaction scale range from 0-100 with higher scores indicating higher levels of satisfaction.

The cost of implementing the DART intervention and the cost of any resulting utilization changes will be offset by decreased expenditures on antibiotic prescriptions.

The cost of implementing the DART intervention and the cost of any resulting utilization changes will be offset by decreased expenditures on antibiotic prescriptions.

Inclusion Criteria: Visits to enrolled providers by children aged 6 months to 10 years with an International Classification of Diseases-10 (ICD-10) diagnosis code for AOM, bronchitis, pharyngitis, sinusitis, or upper respiratory infection (URI) will be included in the study. Only systemic (oral) antibiotic prescriptions on the same date as the study visit will be included in prescribing measures for each ARTI. Exclusion Criteria: ARTI visits will be excluded from measures of prescribing if there are any competing non-ARTI bacterial diagnoses (e.g. urinary tract infection) or any antibiotic prescriptions during the 30-days prior to the index visit. Visits by children with allergies to penicillin or cephalosporin antibiotics will be excluded from measures of second-line prescribing.

If requested and criteria for receipt are met, data will be shared with investigators outside the DART project team when all planned data collection and analyses are completed. The data shared will be limited to visit-based prescribing rates, patient characteristics, and dummy variables for clinic site (de-identified). No protected health information for study participants will be shared. Data will be provided as a CSV file with a data dictionary defining all variables included in the file. Additional analytic tools will not be provided. Data will be transferred using a secure file transfer protocol.

A data use agreement will be required between Seattle Children's Research Institute and any entity requesting access to the data. Those requesting access to the data will be required to provide a detailed plan for their use of the data. This agreement will also require that the requesting investigator/entity obtain permission to publish results based on these data from Seattle Children's Research Institute and the Principal Investigator prior to submission for any such publication.

Mangione-Smith R, Robinson JD, Zhou C, Stout JW, Fiks AG, Shalowitz M, Gerber JS, Burges D, Hedrick B, Warren L, Grundmeier RW, Kronman MP, Shone LP, Steffes J, Wright M, Heritage J. Fidelity evaluation of the dialogue around respiratory illness treatment (DART) program communication training. Patient Educ Couns. 2022 Jul;105(7):2611-2616. doi: 10.1016/j.pec.2022.03.011. Epub 2022 Mar 14.

Kronman MP, Gerber JS, Grundmeier RW, Zhou C, Robinson JD, Heritage J, Stout J, Burges D, Hedrick B, Warren L, Shalowitz M, Shone LP, Steffes J, Wright M, Fiks AG, Mangione-Smith R. Reducing Antibiotic Prescribing in Primary Care for Respiratory Illness. Pediatrics. 2020 Sep;146(3):e20200038. doi: 10.1542/peds.2020-0038. Epub 2020 Aug 3.