Early Auditory Referral in Primary Care

The Best Practice Alert (BPA) was developed via an iterative process during a two year pilot phase at one site at each health system. First, a "standard" BPA was developed to be launched at each site during the pilot phase. Similarly, a pilot 10-minute education system was developed to be presented just before the above BPA was launched. Pre and post surveys of clinicians were conducted to evaluate the effectiveness of the education program. After clinicians had experience with the BPA, Cognitive Task Analyses (CTAs) with clinicians were conducted to identify ways to make the BPA more effective, and then the BPA was iteratively improved. Similarly, the CTAs helped the investigators understand why clinicians do or do not use BPAs in general. Using this process, a highly rated, easy to use BPA was created and will be utilized in the actual study. The investigators also revamped the education program to a 10 minute video that focused on clinicians mental models of HL, to be used in the actual study. The bigger study, i.e., the R33 phase, will launch at several Family Medicine practices within each of the two different health systems. One of these systems is a traditional academic institution with only academic faculty (physicians and audiologists), midlevels and residents, and the other is a newer academic system, based within a private health system, that includes non-academic physicians and private audiologists. Patients who trigger the BPA will be 55 years or older, do not have an open referral in their chart to audiology, and do not have a known hearing loss already on their problem summary list (PSL). These patients are asked to complete a Hearing Handicap Inventory (HHI) at check-in, a common hearing loss screening tool (score of 10+ indicates probable hearing loss); the results of this will not be shared with clinicians. If the physician chooses to address the prompt during the patient encounter, the BPA design allows them to, 1) Indicate that the patient declines hearing screening (BPA is dismissed for 1 year), 2) Indicate that the patient already has a known hearing loss (HL) and add HL to the PSL (The BPA will be permanently dismissed), 3) Indicate that the patient does not have any HL at this time (The BPA is dismissed for one year), or 4) refer the patient to audiology for hearing screening. The data generated by the BPA is extracted from the electronic health record (EHR) and analyzed to determine if HL screening rates improve when compared to baseline data, whether or not providers are interacting with the BPA, and whether HL is being added to the PSL. The HHI results are used as the "gold standard," i.e., when compared to the data generated by the BPA, indicate whether or not patients with probable HL are getting appropriate care. When an enrolled patient shows up in Audiology the audiologist is asked to complete a three question survey to verify that 1) the referral was appropriate, 2) what is the severity of hearing loss if any, and 3) were hearing aids recommended. This step is to study whether the BPA is generating appropriate referrals. 20% of patients that score 10 or above on their HHI and/or were referred to audiology are contacted by phone and researchers ask them questions about any conversation they may recall related to HL at their appointment as well as their experience in audiology. Iterative improvements will be made to the BPA based on Cognitive Task Analysis (CTA) interviews with randomly selected providers (mostly family physicians). Finally, the implementation of the BPA into clinical practice is observed using the principles of Normalization Process Theory (NPT) to study whether there are other potential issues that may impact whether patients at risk for HL are being screened and referred. The revamped educational video will be shown a week prior to the BPA going live at each site, and repeating the pre and post evaluations. The investigators will be studying whether the education video increases identification and referral of patients at high risk for HL.