Skip to main content
The Early Auditory Referral-Primary Care Study aimed to improve identification and referral rates of patients with hearing loss by using an Epic electronic medical record Best Practice Alert (BPA). The article states that over 30% of those aged 55 years and older are affected by hearing loss, yet 75% of those with hearing loss remain underdiagnosed and undertreated. Eligible participants completed the Hearing Handicap Inventory for the Elderly (HHI) and a BPA prompted their physician to ask, “Do you have difficulty with your hearing?” Physicians were then tasked to refer their patients as needed. Audiologists who received referrals were asked to rate the necessity of the referral by level of appropriateness, percent hearing loss, and whether they would recommend a hearing aid for the participant. The researchers concluded that a BPA reminder to ask patients aged ≥55 years about hearing loss significantly increased audiology referrals of patients at-risk , which was supported by audiologic and audiogram data.
This study used a multiple baseline design (MBD) with sites brought into the study approximately 3-4 months apart within two health systems. The group thought an MBD was beneficial as the study’s intervention would result in a difficult-to-reverse change to the hospital networks’ electronic medical record (EMR). Participating sites were Family Medicine clinics within either the University of Michigan (UM) or Beaumont Hospital (BH) care networks. Physicians would receive the BPA when they met with an eligible patient who was not already diagnosed with hearing loss. The physician was then given 5 options to respond to the BPA, including not addressing it. If that occurred, the alert would reappear at subsequent visits. We found the reappearance of the alert to be a valuable part of the study. The authors reported that if the alert was addressed, on average 1.7 encounters occurred before the doctor screened that patient. The group discussed that Family Medicine doctors typically have full schedules and are thus limited in how many patient concerns can be appropriately addressed during each visit. By having the alert recur, physicians would be able to plan to discuss any potential hearing loss at a patient’s subsequent visit. Nevertheless, physicians did not address the alert for 26.5% of patients even with an average of 2.6 encounters to do so throughout the study period.
Another aspect the group examined was the Hearing Handicap Inventory for the Elderly. Participants were asked to complete this 10-question survey, which currently serves as the gold standard to identify at-risk patients. The group thought the survey may have impacted the participants’ role in the study by priming and potentially influencing their answer to the BPA question. For example, completing the HHI before seeing their physician may have caused the participant to be more open to or seek a referral for hearing loss. However, other participants may have minimized their hearing loss symptoms as significant stigma exists around hearing impairment. The authors said that physicians were blinded to the results of the HHI, but as the patient knew their results, it could have easily been shared or influenced either party’s approach to the study.
A major barrier to this study was physician buy-in. The authors stated that the mental model around hearing loss may have had a role in physicians feeling uncomfortable discussing hearing loss, thus resulting in “not addressed” being the most common BPA response. Though the study included an 11-minute informational video reviewing hearing loss, hearing aids, and patient counseling for this diagnosis, only 28% of UM and 52% of BH clinicians viewed it. The group considered whether physicians would have been more open to addressing the BPA, if other treatment options besides hearing aids had been included in this video. Additionally, our group was particularly intrigued by the study’s finding that only 50% of those recommended to use hearing aids ultimately purchased them, due to cost constraints. Therefore, we thought this suggested that limited preventative options and the lack of affordable treatments may also negatively impact a busy physician’s desire to screen patients for hearing loss.
The inclusion of numerous sites within two different health systems was an important aspect to this study and its results. As discussed above, more doctors at Beaumont Health clinics attended the information session at the start of the investigation. Both health systems saw a significant increase in referral rates for at-risk patients. However, there was a significant increase in documentation of hearing loss on the patient’s problem list as compared to baseline by BH clinicians. Furthermore, only 6.4% of participants from a BH clinic never discussed hearing loss during an encounter, compared to 32.4% at UM. The group agreed with the authors that the comparatively higher number of BPAs at UM prior to the start of this study may have caused UM physicians to experience alert fatigue and thus possibly played a role in discrepancies between response to the alerts and problem list documentation.
An additional variable that impacted the study was the inclusion of hearing loss screenings in Medicare wellness visits for persons 65 years. The group felt this was well addressed by the authors and an important piece of the study results. This unexpected potential confounder allowed the authors to use UM general medicine sites as a control group. With the new Medicare guidelines alone, UM general medicine sites saw a non-clinically significant increase in referral rates for hearing loss. Therefore, the group saw this as good support for the study’s BPA prompt as a clinically meaningful way to increase appropriate audiology referrals.
The authors designed the study to also prove the necessity of referrals based on the opinions of audiologists and participants. The audiologists reported that 93.3% of the referrals were appropriate and that 85% of participants seen had hearing loss. The participants themselves felt their referrals were appropriate 61% of the time and over half were glad their physician brought the subject up. While physicians may not want to screen due to lack of treatments, time, or any number of reasons, the group thought these results showed that patients and audiologists believe increased referrals are needed. Additionally, the group was surprised that 72% of participants with a positive HHI screen (scores ≥10) were not referred. As this is a self-report form filled out by the participant, the HHI score to referral ratio furthers the importance of investigations of interventions like the Epic BPA.
Finally, our group includes medical students receiving additional training in rural medicine. They added their opinion on both the location of and the participant demographics in this study. It would be difficult to reproduce the results of this study in a rural setting as fewer clinics have Epic or even computers in their exam rooms. A BPA might be seen during pre-charting. Conversely, rural hospitals and clinics are smaller, which may make changes like instituting hearing loss screening guidelines easier. Additionally, the group wondered about the generalizability of the study as over 85% of the participants were white and all spoke English. Language differences are a common barrier to the diagnosis and treatment of any condition. A non-English speaking patient’s hearing loss may be dismissed as a lack of understanding rather than diminished ability.
In conclusion, the Early Auditory Referral-Primary Care Study was able to increase the referral rates of patients at-risk for hearing loss to a hearing specialist in two health systems. Our group discussed the benefits and challenges of introducing a Best Practice Alert to electronic healthcare documentation in order to reach this goal. We wondered if this intervention would have lasting impacts, especially on medical students rotating through these clinics. Those students who receive training on hearing loss and experience EMR alerts to ask patients about their symptoms may refer future patients at a greater rate. Additionally, the group wanted more information on interventions that may come from early detection of hearing loss. This study brought light to an area that is underreported, but until there are affordable and accessible treatment options, screening for hearing loss may not become a priority for Family Medicine or other physicians as the authors hoped.