Background: Total laryngectomy (TL) results in unique challenges to communication. Traditional methods of assessing speech and voice after TL include objective measures (e.g., acoustics), listener-ratings of speech intelligibility and acceptability, as well as patient-reported measures, such as voice-related quality of life scales. Recently, a patient-reported scale that measures communication in everyday contexts, or “communicative participation”, was validated in head and neck cancer survivors. Consequently, we examined how this new tool, the Communicative Participation Item Bank (CPIB), relates to traditional post-laryngectomy speech and voice outcomes.

Objective: To determine relationships between communicative participation with traditional post-laryngectomy speech and voice outcomes including: a) speech intelligibility; b) speech acceptability; and c) voice handicap.

Methods: Thirty-six individuals (29 males, 7 females) who had undergone TL (n=11 electrolaryngeal speakers; n=23 tracheoesophageal speakers; n=2 esophageal speakers) completed patient-reported outcomes including the CPIB (Baylor et al., 2013) and the Voice Handicap Index-10 (VHI-10; Rosen et al., 2004). Individuals also recorded 6 sentences (5-15 words) from the Sentence Intelligibility Test (SIT; Yorkston et al., 1996) and a standard reading passage (Rainbow Passage; Fairbanks, 1960). Forty-eight inexperienced listeners each transcribed 6 sentences for 3 speakers using the SIT protocol. Each speaker’s intelligibility was based on the average score across 3 listeners. Eighteen listeners also judged speech acceptability for the 36 speakers using 100 mm visual analog scales. Twenty percent of the samples were repeated to determine intra-rater reliability (r = .71); intraclass correlation coefficients were calculated as a measure of interrater reliability (ICC = .96, speech acceptability). To determine the relationships between the CPIB and speech and voice outcomes (VHI-10, speech intelligibility, and speech acceptability), correlational analyses were performed.

Results: Listeners judged tracheoesophageal speakers significantly more intelligible and acceptable than electrolaryngeal speakers (p < .01).  However, patient-reported outcomes such as voice handicap and communicative participation were not significantly differentiated by speaker type. In addition, listeners’ ratings of speech acceptability were only moderately related to intelligibility (r = .43). Relationships were weak between ratings of speech and CPIB, with speech acceptability tending to be a stronger predictor than intelligibility (see Table 1). The strongest relationship was observed between the two patient-report measures; decreased voice handicap related to increased communicative participation (r = -.756).

Table 1. Correlations between ratings of post-laryngectomy speech outcomes with CPIB.

Conclusions: Alaryngeal speakers exhibited a wide range of speech acceptability and intelligibility, although these dimensions were not strongly related. These results suggest that while listeners may be able to understand some alaryngeal speakers, being understood is not sufficient to render the speech “acceptable” to the listener. Results also revealed that listeners’ judgments of speech are not necessarily predictive of the speakers’ own perceptions of communication in everyday contexts. As hypothesized, the patient-reported measures (VHI-10; CPIB) were strongly related. Listener-rated and patient-reported measures are complementary after TL. Therefore, both should be adopted as part of an assessment battery.