Decreasing unnecessary antibiotic prescribing due to caregiver pressure for antibiotics: protocol for development of a text message intervention

Abstract Corresponding Author: Nicole M. Poole, MD MPH, 13123 E 16th Ave, Aurora, CO 80045, 720-777-7507, [email protected] Alternate Corresponding Author: Sean O’Leary, 1890 N Revere Ct, 3rd Floor, Aurora, CO 80045, [email protected] Funding source: Research Scholar Award, Child Health Research Enterprise (CHRE) and Children’s Hospital Colorado No conflicts of interest. Background Over 30% of antibiotics prescribed for acute respiratory tract infections (ARTI) in children are unnecessary. Clinicians identify caregiver or parent pressure for antibiotics as a primary driver of the evidence-practice gap, defined as the gap between how clinicians know antibiotics should be prescribed and how they are actually prescribed. This study aims to fill this gap by developing a text message intervention for caregivers that can be disseminated widely across diverse clinical settings. The text message intervention will be developed with the intent to (1) address multifaceted drivers of caregiver expectations about antibiotics for ARTI and (2) improve the quality of clinician-caregiver interactions surrounding antibiotics during ARTI visits. Methods Caregivers will be the primary users of the text message intervention. An iterative design process will be guided by the Practical, Robust, Implementation and Sustainability Model (PRISM), an implementation science model that identifies contextual factors to facilitate intervention adaptations to prevent intervention failure and increase intervention uptake across diverse settings. PRISM domains and its application to the intervention design are outlined in Figure 1. PRISM includes patient-level implementation outcomes of the RE-AIM framework. The format and content of the intervention will be a text message series developed from results of clinician and caregiver focus groups (FGs). Clinicians from 3 clinics within a primary care practice-based research network in the Denver-metro area will be included (n=3 FGs, 6-8 clinicians each). To ensure perspectives of historically marginalized minority groups are included, caregiver FGs will be conducted with four different race/ethnicity categories (2-3 FGs for each category, 8-12 FGs total, 6-8 participants each). After content development, the text message series will be developed in English and Spanish, using a secure, modifiable, mHealth text message platform with the capability to use “smart forms” for interactive messaging and queries, along with automated responses via pre-programmed algorithms. Interaction with messages will be measured through data dashboards provided by the text messaging platform. After prototype development, caregivers (n=15) and content experts (n=6-10) will conduct face validity testing followed by content validity testing using the Content Validity Ratio to assess the ‘essential’ and valid elements of the text message series. Response process validity will be assessed by conducting cognitive interviews with caregivers (n=35). The text message series will be iteratively modified and revised based on each step of the design process. At the completion of this study, a text message intervention will be developed and ready to pilot test its impact on clinician-caregiver interactions and antibiotic prescribing for ARTI. This study has received university IRB approval. Discussion This study will develop a text message intervention using a robust implementation science framework to address multifaceted drivers of caregiver pressure for antibiotics. Ultimately, an intervention to improve the quality of clinician-caregiver interactions as a primary outcome will be a novel way to address the real-time behavioral factors that contribute to unnecessary antibiotic prescribing to children with ARTI. Figure 1: PRISM domains and its application to the text message series intervention design.