Clinicians improvised on validated speech-perception testing or used non-speech evaluation for CI candidacy in LEP patients (59 and 52 percent)

Description #

In the interviews (n = 29), providers reported two non-exclusive workarounds for cochlear-implant candidacy evaluation in LEP patients: improvising on a validated speech-perception testing battery (59%; n = 17) and relying on non-speech perceptive evaluation (52%; n = 15) (Figure 2A). Improvisation included live interpretation of an English test battery (38%; n = 11) and use of validated English pediatric word lists in adult LEP patients (24%; n = 7), and one-quarter (24%; n = 7) reported family reports influencing their disability assessments. These adaptations substitute for validated language-matched testing and can compromise the validity of the assessment.

"Providers reported employing 2 nonexclusive approaches: improvisation on validated speech perception testing battery (59%; n = 17) and reliance on nonspeech perceptive evaluation (52%; n = 15) (Figure 2A)." (Govindan, 2025, p. 1234)

"Examples of improvisation on validated speech perception testing battery included live interpretation of an English test battery (38%; n = 11) (eg, “we rely on having a translator there to assist us in trying to translate sentences”), and utilization of validated English pediatric word lists in adult LEP patients (24%; n = 7)." (Govindan, 2025, p. 1234)

Methods Context #

What? #ⓘ

The observable: clinician-reported methods used to evaluate cochlear-implant candidacy in LEP patients, coded as frequencies of improvisation vs non-speech evaluation (and their subtypes).

"Providers reported employing 2 nonexclusive approaches: improvisation on validated speech perception testing battery (59%; n = 17) and reliance on nonspeech perceptive evaluation (52%; n = 15) (Figure 2A)." (Govindan, 2025, p. 1234)

How? #ⓘ

Structured Zoom interviews using an IRB-approved script, coded via directed content analysis guided by modified grounded theory (iterative axial coding, third-reviewer adjudication); subtheme frequencies reported as a percentage of responses in the domain.

"Two authors (A.G. and A.S.) conducted interviews to maintain reproducibility and standardization of the conversation." (Govindan, 2025, p. 1233)

Who? #ⓘ

29 US hearing-health clinicians (16 otologists/neurotologists, 13 audiologists) recruited from 12 cities with the highest immigrant populations and top-10 non-English household languages.

"Licensed hearing health clinicians, including audiologists and neurotologists were recruited from 12 cities encompassing the top ten highest immigrant populations and the top 10 non‐English languages spoken in households." (Govindan, 2025, p. 1233)

Other Notes #

A measurement-practice finding: in the absence of validated language-specific batteries, clinicians improvise (interpreted English tests, pediatric word lists) or fall back on non-speech/family-report measures, threatening the validity of CI-candidacy assessment in LEP patients.

Caveats #

• Small non-uniform sample skewed to Spanish-Chinese languages and away from the Midwest limits generalizability (Govindan 2025) The survey and interview samples each represent only a fraction of US providers serving LEP hearing-health patients, the geographic distribution is non-uniform (notably few Midwest respondents), and the LEP populations sampled are skewed toward Spanish and Chinese languages. The authors state these factors collectively limit the generalizability of the results — so the confidence, effort, and assessment-practice findings may not transfer to providers or LEP language groups outside the sampled distribution.
• Interview responses subject to recall bias personal bias and self-estimated surgical and LEP volumes (Govindan 2025) The interview-derived findings rest on clinician self-report and are subject to recall bias, the respondents' personal bias or experience with multiculturalism, and self-estimated surgical and LEP patient volumes. The reported frequencies of assessment practices and barriers therefore reflect provider recollection and perception rather than verified clinical behavior, which the authors flag explicitly.