Abstract
The aims of this study were to evaluate the cross-cultural adaptation of the Speech Handicap Index (SHI) for Korean subjects and to determine its reliability and utility in patients with oral cavity cancer. The Korean version of the SHI was administered to 50 healthy subjects and 56 patients with speech problems resulting from treatment for oral cavity cancers. The content and construct validity, internal consistency, and test–retest reliability were examined. Healthy subject and patient group scores were compared, and the Mann–Whitney U -test was used to determine discriminatory ability. The Korean version of the SHI had high internal consistency (Cronbach’s alpha = 0.99) and test–retest reliability for the total and subscales: total (T) 0.98, speech (S) 0.99, and psychosocial (P) 0.97. Mean scores in the healthy group were 0.5 (T), 0.2 (S), and 0.2 (P), whereas those in the patient group were 34.3 (T), 16.6 (S), and 15.5 (P). The scores differed significantly between the groups ( P < 0.05). The Korean version of the SHI can be a useful tool to evaluate a patient’s self-perception of their speech dysfunction in daily life and to better understand postoperative speech disorders in patients with oral cavity cancer.
Treatments for head and neck cancer may have a negative effect on the patient’s speech, swallowing, appearance, social relationships, and psychosocial health in daily life. For better treatment outcomes, patient quality of life (QOL), as well as survival outcomes after treatments, should be considered prior to the selection of the treatment modality. Significant impairments of speech function can be caused by oral and oropharyngeal cancer surgery and also by radiotherapy.
Speech function has been considered one of the most essential factors influencing the determination of QOL for head and neck cancer patients. Speech can be evaluated with either objective instrument measures or subjective methods using self-rating after treatment. In previous studies, speech has typically been evaluated in these patients through patient-reported speech outcomes (e.g., speech subscale of the European Organization for Research and Treatment of Cancer Quality of Life Questionnaire-Head and Neck 35 module, EORTC QLQ-H&N35) and the assessment of speech itself (e.g., communicative suitability, speech intelligibility, articulation, nasal resonance, consonant errors, and aerodynamic measurements). Although a number of assessments – both objective and subjective – are speech-related, specific Head and Neck cancer self-rating QOL questionnaires provide important information on speech disturbances and the problems patients feel they face in daily life. In addition, the subjective experience of speech-specific functions is not always correlated with the severity of the speech disorder. Furthermore, a patient’s perception and subjective complaints of speech function might be more important than the objective tests used to measure the effectiveness of treatment in the scope of their limitations in activity or restrictions in participation. Thus several specific Head and Neck cancer self-rating questionnaires have been developed to measure patient-reported experiences with speech disorders.
Currently, the EORTC QLQ-H&N35, Functional Assessment of Cancer Therapy Head–Neck (FACT-HN), Health-Related Quality of Life instrument (HRQOL), and University of Washington Quality of Life questionnaire (UWQOL) are available for the measurement of QOL in patients with head and neck cancer. However, all head and neck cancer QOL questionnaires are limited in their ability to measure speech and speech-specific outcomes adequately. Other methods – the Voice Handicap Index (VHI) and voice-related QOL (VRQOL) – have also been used in patients with laryngeal cancer. However, these self-rating questionnaires are focused on voice and not the speech-specific dimension. It is important to distinguish the evaluation of voice (laryngeal function) from that of speech (vocal tract, articulator function). The Speech Handicap Index (SHI) was recently developed to resolve these issues. This index measures the patient-reported impact of speech on his/her daily life. The SHI was first validated in the Dutch language.
The SHI consists of 30 statements and is scored on a Likert scale ranging from 0 (never) to 4 (always). The objective of the SHI is to measure the severity of speech problems, and it includes items covering speech (14 items) and psychosocial factors (14 items). This evaluation tool has been translated into different languages, including English and French.
Although the SHI has been used to measure the patient’s perspective of their speech, a validation of the Korean version of the SHI has not yet been performed. The aims of this study were to validate the cross-cultural adaptation of the SHI for Korean subjects and to determine its reliability and utility in patients with oral and oropharyngeal cancer.
Materials and methods
Participants
The study was approved by the institutional review board (IRB) of Seoul National University Hospital (Seoul, Korea). The study included healthy subjects randomly selected from the general population who were not directly related to the study researchers. Fifty healthy Korean-speaking subjects without a past or present history of speech disorder, hearing loss, or disability affecting their speech, were recruited. This group consisted of 25 males and 25 females, ranging in age from 31 to 69 years (mean 51.9 years, standard deviation (SD) 9.3 years). Fifty-six Korean-speaking patients with oral and oropharyngeal cancer attending Seoul National University Hospital were also recruited. This group comprised 34 males and 22 females, ranging in age from 25 to 83 years (mean 56.8 years, SD 13.7 years). Patients with other associated malignancies, any diagnosed neuromuscular disease known to affect speech function, or a cognitive problem were excluded from the study. These cancer patients underwent various modalities of treatment, such as surgery only, surgery and postoperative radiotherapy, and concurrent chemoradiotherapy ( Table 1 ). Thus a total 106 subjects participated in this study at the hospital.
| Characteristics | No. (%) |
|---|---|
| Age, years, mean ± SD | 56.8 ± 13.7 |
| Male to female ratio | 34/22 |
| Tumour location | |
| Tongue | 48 (86) |
| Base of tongue | 8 (14) |
| T classification | |
| T1 | 28 (50) |
| T2 | 17 (30) |
| T3 | 2 (4) |
| T4 | 9 (16) |
| N classification | |
| N0 | 33 (59) |
| N1 | 10 (18) |
| N2 | 13 (23) |
| Clinical stage | |
| I | 24 (43) |
| II | 9 (16) |
| III | 0 (0) |
| IV | 23 (41) |
| Treatment | |
| Surgery only | 28 (50) |
| CCRT | 10 (18) |
| Surgery + RT | 13 (23) |
| Surgery + CCRT | 5 (9) |
| Glossectomy | |
| Yes | 46 (82) |
| Partial glossectomy | 37 (66) |
| Hemi-glossectomy | 6 (11) |
| Total glossectomy | 3 (5) |
| No | 10 (18) |
| Follow-up, months, median (range) | 14.6 (3–79) |
Cross-cultural adaptation of the SHI for Korean subjects
For the cross-cultural adaptation of the SHI for Korean subjects, the English version of the SHI was first translated into Korean, as recommended in the guidelines of Guillemin et al. The translation was completed by two bilingual (Korean/English) teachers. The translated Korean version questionnaire was then back-translated into English by a native speaker of the original language to check for any discrepancies. Two qualified bilingual translators (bilingual (Korean/English) speech-language pathologists) reviewed the back translation. Each participant completed the translated version of the SHI autonomously and discussed the wording and meaning of each item on the questionnaire to identify any practical problems and assess the relevance and clarity of language. Following this procedure, the final Korean version was generated. The consensus translated SHI was also forwarded to the investigators for review, and the final consensus version was prepared (see Appendix A ). This final translation was pilot-tested on 16 Korean patients with oral cavity cancer (nine males and seven females) and 26 healthy subjects (11 males and 15 females) in 2013.
Speech evaluation
The 56 participants with oral cavity cancer underwent evaluations of articulation, including the percentage of consonants correct (PCC) at the word and sentence levels, and a speech intelligibility test. The Korean standardized articulation test, the Urimal Test of Articulation and Phonology (U-TAP), was administered by a speech-language pathologist and the PCC was calculated. Additionally, speech intelligibility of a standardized reading passage was assessed on a 7-point Allison scale (ranging from 1 to 7) by two experienced speech-language pathologists (see Appendix B ). After 2 weeks, inter-rater reliability for the speech intelligibility test and the PCC at the word and sentence level was assessed; the speech-language pathologists were blinded to the participants in the patient group during recording. The correlation coefficients were 0.86, 0.97, and 0.93, respectively.
Validation instruments
Content validity was measured with a 5-point validity index (CVI) (1, poor; 3, good; 5, excellent) by five experts; each item was rated based on relevance, clarity, simplicity, and ambiguity. The final consensus version of the Korean SHI was administered to oral cancer patients and age- and sex-matched healthy people. In addition, the validated and standardized Korean version of the UWQOL was used to assess the construct validity of the Korean version of the SHI.
Statistical analysis
IBM SPSS Statistics for Windows version 19.0 software (IBM Corp., Armonk, NY, USA) was used to perform the statistical analysis. The content validity was assessed using Cronbach’s alpha. The construct validity was assessed by calculating the correlations between the Korean SHI scores and the Korean UWQOL questionnaire scores.
Within a period of 2 weeks, 50% of the participants again completed the Korean version of the SHI questionnaire by telephone. The reliability analysis of the Korean version of the SHI was evaluated by calculating the internal consistency of the index items using Cronbach’s alpha. Spearman’s correlation coefficients (rho) were calculated for the assessment of test–retest reliability.
Comparisons of the mean SHI scores between the healthy subjects and the patients were performed using the independent t -test. Spearman’s correlation coefficients were used to assess the relationships between the Korean version of the SHI and the PCC (word and sentence level) of the U-TAP and speech intelligibility. The level of significance was set at 0.05.
Results
The content validity score (Cronbach’s alpha) of the Korean version of the SHI indicated good to excellent content validity: Cronbach’s alpha was 0.97 for the total SHI score, and the domain scores were 0.85 for the psychosocial domain and 0.88 for the speech domain ( Table 2 ).
| SHI domains | Number of items | Content validity (Cronbach’s alpha) |
|---|---|---|
| Total | 30 | 0.97 a |
| Speech | 14 | 0.88 a |
| Psychosocial | 14 | 0.85 |
Construct validity was assessed by correlating the speech scores from the Korean version of the UWQOL questionnaire with the Korean SHI scores. The correlation coefficients between the total, speech, and psychosocial domains of the Korean version of the SHI and the speech domain of the Korean version of the UWQOL were 0.64, 0.71, and 0.57, respectively. Spearman’s rho correlation was 0.71, demonstrating a high correlation in speech items. The correlation coefficients between total, speech, and psychosocial domains of the Korean version of the SHI and the social domain of the Korean version of the UWQOL were 0.59, 0.44, and 0.42, respectively. Cronbach’s alpha coefficients for internal consistency were 0.99, 0.99, and 0.98 for total, speech, and psychosocial domains, respectively, of the Korean version of the SHI ( Table 3 ). The correlation coefficients for test–retest reliability were 0.98 for total domains, 0.99 for the speech domain, and 0.97 for the psychosocial domain, which indicated good test–retest reliability ( Table 3 ).
