Methods

A. Discussions With KIs

We selected KIs with expertise in one or more of the following areas: AS-ME; adult and pediatric asthma; environmental allergens and irritants; community-based interventions; and populations at high risk for significant asthma morbidity and poor outcomes. We interviewed either individually or collectively seven KIs located in the United States. We asked KIs about the content of AS-ME packages, ways design features of educational materials could be improved, challenges encountered when implementing AS-ME packages, and ways to facilitate the delivery of education. KIs also provided insight into how AS-ME should be evaluated, and how education interacts with other types of asthma interventions.

KI input helped inform GQs 1, 2, and 4. We also used KI input to refine the systematic literature search, identify grey literature resources, provide information about ongoing research, confirm evidence limitations, and recommend approaches to help fill these gaps. Table 1 presents the questions asked of the KIs.

B. Grey Literature Search

Grey literature was critical for identifying AS-ME packages (GQ 1 and GQ 2), and finding descriptions and evaluations of AS-ME packages implemented by individual hospitals, health systems, provider groups, community organizations, or public health agencies. ECRI-Penn Evidence-based Practice Center (EPC) Information Center searched multiple gray literature sources such as: Centers for Disease Control and Prevention (CDC), ClinicalTrials.gov, ECRI Guidelines Trust, Medscape, National Academy of Medicine, and the websites of organizations and agencies. We also searched for patient apps and other online AS-ME tools. Finally, input from the KIs helped to identify other grey literature sources. Complete lists of the resources searched and search concepts and strategies are available in Appendix A.

C. Published Literature Search

Evidence from the published literature search helped inform GQ 3. Medical librarians at ECRI-Penn EPC Information Center searched bibliographic databases including MEDLINE, PubMed (unprocessed records only), EMBASE, CINAHL, and the Cochrane Library using controlled vocabulary and text words. Searches covered the literature published from January 1, 2007, through December 15, 2019. Complete lists of the resources searched and search concepts and strategies are available in Appendix A. Reference lists from systematic reviews were reviewed and compared against our retrieved articles. If a systematic review contained references that appeared to meet our inclusion criteria but had not been captured by our initial search results, we reviewed the search strategy to determine if refinement of the search strategy was needed to include these articles.

Literature screening was performed in duplicate using the database Distiller SR (Evidence Partners, Ottawa, Canada). Literature search results were initially screened for relevancy based on predetermined eligibility criteria (Table 2). Full-text of relevant abstracts were requested and screened. All disagreements were resolved by consensus discussion among the two original screeners.

D. Inclusion of AS-ME Packages and Published Literature

Specific AS-ME packages were included if they contained an interactive component, described a methodology for use or implementation, and were available for use in the United States. Packages were excluded if they consisted only of paper materials, slides, checklists, or other materials without any interactive element, did not include guidance for how they should be used, or were used exclusively outside of the United States. Table 2 describes the inclusion and exclusion criteria.

Published studies were included if they presented postintervention data on asthma patients in the United States who used an AS-ME package, and were full-length English language publications. We did not require that studies have control groups. Studies were excluded if they were available only as abstracts or only examined patients outside the United States. If a study contained an AS-ME intervention used in multiple countries including the United States, we included it if at least 50 percent of patients were in the United States, or if patient data was stratified by country.

A. Information Management

We abstracted and tabled descriptive characteristics from AS-ME packages and published studies. Factors abstracted from AS-ME packages included, but were not limited to, the characteristics described in GQs 1 and 2. Factors abstracted from published studies included PICOTS categories (population, intervention, comparator, outcomes, timing, setting). We highlighted outcome measures that were used in these studies, and the applicability of the results to various populations. KI interviews helped refine which data points should be abstracted, and how they might be organized. KI interviews were documented during each call by a designated member of the project team.

B. Data Presentation

We designed an analytic framework that visually communicates the integration of AS-ME into patient care. This framework includes the roles and relationships of those who design, implement, use, and assess AS-ME packages, and the interaction of AS-ME with short- and long-term clinical and evaluative outcomes.

Characteristics of AS-ME packages and outcomes of published studies are presented in searchable evidence tables. We also developed figures and tables that synthesize the current state of knowledge regarding AS-ME evaluation and research and highlight evidence gaps that require further study and assessment. Finally, we narratively summarized significant perspectives and insights gathered from KIs.