Correlation between the Voice Handicap Index and voice laboratory measurements after phonosurgery.
Phonosurgery is an effective treatment for some vocal fold pathologies, and the Voice Handicap Index (VHI) survey has been shown to be a useful instrument for evaluating treatment effectiveness. We conducted a nonrandomized, prospective study of 21 patients who underwent phonosurgery for the treatment of non-neoplastic vocal fold lesions at our academic tertiary-care referral center. Our goals were to compare pre- and postoperative VIII scores (subjective assessments) and pre- and postoperative results of acoustic and aerodynamic tests (objective assessments). We sought to determine if there was any correlation between the subjective and objective findings. We looked for differences between professional voice users (n = 10) and nonprofessional voice users (n = 11) in both subjective and objective measures. We found statistically significant differences between pre- and postoperative values in three of four VHI parameters, but in only one of 13 objective measures. There was no correlation between preoperative VIII scores and preoperative acoustic and aerodynamic test results. The professional voice users expressed greater postoperative improvement as reflected by lower VIII scores than did the nonprofessional voice users, confirming that the former are more negatively affected by a voice disability.
Analyses of non-neoplastic voice disorders and quantification of their effects on patients have been conducted for quite some time, yet there is still much we do not understand about the physiology and mechanisms of these disorders. In analyzing a patient's voice, clinicians must consider subjective as well as objective measurements. While objective assessments of voice obtained by computer-assisted analyses are useful, they may not capture the global function of a patient's voice. Subjective assessments, therefore, can be helpful in attempting to overcome some of the limitations of objective testing.
Subjective assessments are also useful in evaluating a patient's response to treatment. A patient's satisfaction with a perceived improvement in voice following an intervention can have a positive effect on his or her emotional and functional status. Conversely, dissatisfaction can have the opposite effect. Cohen et al conducted a meta-analysis of studies on the impact of non-neoplastic voice disorders on patients' well-being and concluded that the management of patients with voice complaints should include a quality-of-life assessment. (1)
Jacobson and colleagues developed the Voice Handicap Index (VHI) as a means of subjectively quantifying the psychosocial effects of voice disorders. (2) The VHI was subsequently used in clinical research to measure voice changes after therapy. (3) However, there is no consensus in the literature as to whether subjective VHI surveys and objective data measurements should be used together. Therefore, we conducted a study (1) to compare pre- and postoperative VHI scores and pre- and postoperative objective measurements, (2) to determine if there was any correlation between VHI scores and objective measures, and (3) to analyze differences between nonprofessional and professional voice users.
Patients and methods
This nonrandomized, prospective study was undertaken to investigate subjective and objective outcomes measures in patients who underwent phonosurgery for non-neoplastic vocal fold lesions. Institutional review board approval was obtained prior to the initiation of the study protocol, and informed consent was obtained prior to the enrollment of patients.
Patients were eligible for this study if (1) they demonstrated a voice impairment that affected their daily life, (2) they agreed to undergo phonosurgery in an attempt to imp rove their voice, (3) they agreed to complete a VHI survey preoperatively and 6 weeks postoperatively, and (4) they agreed to undergo acoustic and aerodynamic testing preoperatively and 6 weeks postoperatively. Patients were recruited by the senior author (P.W.) at the Grabscheid Voice Center of the Mount Sinai Medical Center in New York City from January 2007 through June 2008.
Initially, 23 patients met the eligibility criteria and were enrolled in the study. Two patients did not complete the study, and thus our final study population was made up of 21 patients--12 men and 9 women (mean age: 48.4 [+ or -] 15.9 yr). All patients provided a medical history and underwent a physical examination and laryngovideostroboscopy. Each patient's particular vocal fold pathology was diagnosed by the senior author. Seventeen patients (81.0%) were diagnosed with a vocal fold polyp and 1 patient each (4.8 %) was diagnosed with a vocal fold cyst, a rheumatoid nodule, papillomata, and a scar.
Patients were asked to rate their voice use on a scale of 1 (low use) to 5 ("elite" use). A score of 1 indicated that a patient's voice use was not a concern with respect to daily activities, while a score of 5 indicated that the patient was a professional or an a vocational voice user (e.g., a singer at religious services). Patients with a voice use rating of 1, 2, or 3 were classified as nonprofessional voice users, and those with a rating of 4 or 5 were classified as professional voice users. The mean score among the 21 patients was 3.48 ([+ or -]); 10 patients qualified as professional voice users (figure).
Subjective assessments. The VHI is a self-administered survey in which patients rate on a scale of 0 (never) to 4 (always) the frequency with which they experience various negative situations related to their voice impairment (e.g., "My voice makes it difficult for people to hear me."). The 30 survey items are evenly arranged in three domains: functional, physical, and emotional. Possible subscale scores range from 0 (no perceived handicap) to 40 (worst possible handicap), and the overall score ranges from 0 to 120. Within each domain, disability was classified as mild (a score of [less than or equal to] 20), moderate (21 to 30), and severe (31 to 40). For the total scores, disability was classified as mild ([less than or equal to] 30), moderate (31 to 60), severe (61 to 90), and very severe (91 to 120). The VIII has been validated as a reliable tool for measuring an individual patient's voice. (2)
Objective assessments. During the same office visit, 13 acoustic and aerodynamic parameters were measured. The acoustic tests were performed with Computer Speech Lab equipment (Model 4300B; KayPENTAX; Lincoln Park, N.J.), and aerodynamic tests were conducted on the Phonatory Aerodynamic System (KayPENTAX). The sound-treated audiologic booth met or exceeded the American National Standards Institute (ANSI) 1977 requirements for noise level. The eight acoustic parameters included fundamental frequency ([F.sub.0]), physiologic [F.sub.0] range of phonation, lower and upper pitch limits, jitter, shimmer, the noise-harmonic ratio (NHR), and percent voiceless. The five aerodynamic parameters included maximum phonation time, target flow, efficiency, resistance, and phonation threshold pressure. Each of the specific measurements was taken three times during the session, and the mean of the samples was used for our analysis.
Voice Handicap Distress Index. In order to correlate the VIII scores with objective findings in the nonprofessional and professional voice user groups, we created the Voice Handicap Distress Index (VHDI), with normal values adjusted for sex. The VHDI is calculated thus:
Statistical analysis. Statistical analysis was performed with VassarStats, an Internet-based statistical computation program. (4) A correlated paired t test was used to analyze the pre- and postoperative total and subscale VHI scores, to compare pre- and postoperative voice laboratory measurements, and to compare subjective and objective results in nonprofessional and professional voice users. We also used a Pearson rcorrelation analysis to determine if there was any correlation between preoperative VHI scores and preoperative voice laboratory measurements. A p value of [less than or equal to] 0.05 was considered to be statistically significant.
Overall, we observed statistically significant differences between pre- and postoperative VHI scores, but not in objective measurements.
All patients. Among all patients, the mean total preoperative VHI score was 47 and the mean postoperative score was 28, which represented an improvement from moderate to mild dysphonia (table 1). This difference was statistically significant (p = 0.002). The mean preoperative VHI subscale scores for the functional, physical, and emotional domains were 12.2, 20.8, and 14.0, respectively. Postoperatively, the corresponding scores were 9.5, 11.0, and 7.5. The decrease in the physical domain represented an improvement from moderate to mild disability, while the scores in the functional and emotional domains were mild to begin with. Statistically, the decreases were significant for the physical and emotional domains (p < 0.001 for both), but not for the functional domain (p = 0.11).
With regard to acoustic and aerodynamic measures, the only statistically significant difference between pre- and postoperative values in the group as a whole was in the efficiency parameter (p = 0.01) (table 2).
Although we found statistically significant correlations between a few preoperative VHI scores and preoperative laboratory results (i.e., upper pitch limit with total VHI score and physical and emotional subscale scores, and efficiency with total VHI score), we did not find an overall correlation (table 3).
Nonprofessional voice users. Preoperatively, the mean total VHI score among the nonprofessional voice users was 36 (table 1). The postoperative score was 25, representing an improvement from moderate to mild dysphonia, although not a statistically significant one (p = 0.12). For the functional, physical, and emotional domains, the mean preoperative subscale scores were 9.8, 15.8, and 10.0, respectively, and the corresponding postoperative scores were 8.6, 10.4, and 5.9. All scores were in the mild disability category both pre- and postoperatively, but the decrease in the emotional domain was statistically significant (p = 0.05).
An analysis of the 13 laboratory parameters demonstrated only one statistically significant difference between pre- and postoperative values in the nonprofessional group (table 2); the target flow rate was significantly lower following phonosurgery (0.30 vs. 0.23 L/sec; p = 0.03).
Professional voice users. Preoperatively, the professional voice users had a mean total VHI score of 60 (table 1). Postoperatively, the score fell to 32, representing an improvement from the high end of moderate dysphonia to the low end; the difference was statistically significant (p = 0.007). The mean preoperative subscale scores for the functional, physical, and emotional domains were 14.9, 26.3, and 18.5, respectively, and the corresponding postoperative scores were 10.5, 11.8, and 9.2, representing an improvement in the physical domain from moderate to mild disability. The changes were statistically significant for the physical (p = 0.002) and emotional (p = 0.01) domains, but not for the functional domain (p = 0.10).
No significant difference was seen between any pre- and postoperative objective measure in the professional group (table 2).
VHDI. The professional voice users had a higher mean VHDI (13.63 [+ or -] 5.40) than did the nonprofessional voice users (7.22 [+ or -] 6.26); the difference was statistically significant (p = 0.02) (table 4). This indicates that the professional voice users perceived their preoperative voice handicap to be more serious than did the nonprofessionals when normalized against their objective measures.
When patients with voice complaints provide a history, they often use qualitative descriptions and examples of how their voice dysfunction affects their everyday activities that are difficult to quantify. The VHI has been validated as a reliable means of quantifying the functional, physical, and emotional aspects of voice dysfunction. (2) An accurate assessment of voice complaints also depends on the clinician's diagnostic acumen and ability to discern intricacies and details in a patient's voice. Interobserver differences may easily arise. Therefore, acoustic and aerodynamic testing is conducted to obtain an objective assessment of vocal dysfunction.
In theory, a subjective measurement of a particular patient's voice dysfunction should correlate with an objective assessment. However, we found no correlation in our study. Indeed, as a profession, we do not yet understand how physiologic acoustic analyses should correlate to VHI scores. This dissociation is one of the barriers to developing a generalizable paradigm regarding how individual patients are differently affected by voice disability. (5)
Wheeler et al studied 50 patients and found that the results of acoustic analyses were not predictive of overall VHI scores and that the individual components of the VHI survey did not consistently and significantly correlate with acoustic findings. (6) They surmised that the reason for this inconsistency was that there is a nonlinear relationship between perceptions of handicap and individual patient circumstances, such as occupation, social status, previous experiences with vocal dysfunction, overall personality, etc. Earlier, Hsiung et al undertook a related investigation and came to a similar conclusion] Using a Pearson rcorrelation analysis, they too found that VHI overall and subscale scores were poorly associated with voice laboratory measurements. As a result, they were unable to identify any definitive prognostic indicator in their cohort of dysphonic patients. Woisard et al were able to demonstrate a fair correlation between total and subscale VHI scores and the minimum frequency as measured by voice laboratory measurements in 58 patients; they also used a Pearson r correlation analysis. (8) They also found that vocal range was fairly correlated with the physical domain of the VHI. Overall, however, they concluded that VHI scores were independent of laboratory findings.
We believe that voice specialists should not be discouraged by these negative findings and that we should continue to investigate ways to improve diagnosis and prognosis. Phonosurgeryhas been demonstrated to be an effective treatment for benign vocal fold pathologies, and the VHI has been shown to be useful in assessing treatment efficacy. Patients with vocal fold cysts and polyps in particular seem to benefit from microlaryngoscopy surgery, and their postoperative VHI scores were shown by Rosen et al to reflect significant improvement. (9) Cohen et al were also able to show a statistically significant reduction in postoperative VHI scores among patients with vocal fold polyps and cysts who underwent surgery and voice therapy. (10) Our study results appear to echo those findings.
Several circumstances might explain why our study found so little correlation between VHI scores and acoustic and aerodynamic test results. One possible explanation is the small number of patients in our study. Another might be the disparities in the voice qualities among our patients. Yet another might be the fact that acoustic parameters in patients with vocal fold polyps and cysts (which accounted for >85% of the lesions in our study) are usually only slightly disordered, so any acoustic improvement from surgery may not be reflected in measurements of jitter, shimmer, and NHR. Another factor may be the timing of laboratory testing; inflammation, scarring, and healing take place over a period of weeks to months after phonosurgical intervention, so early findings might not be reliable.
Based on our experience, we believe that subjective and objective evaluations of voice dysfunction should probably be interpreted independently. Because voice quality is multidimensional, it might not be prudent to rely on one instrument to evaluate voice dysfunction and postoperative improvement. We believe the VHI should be used as an addition to objective laboratory measurements. Also, we believe that larger studies are needed to better elucidate the test-retest validity of objective measurements.
To determine whether perceptions of voice handicaps are greater in professional voice users than in nonprofessional voice users, we created the VHDI. Our results suggest that all aspects of the VHI were more significantly affected in professional voice users. Their total and subscale perceptions of handicap were significantly higher, but their perceived postoperative improvement was also greater (p = 0.02). Since the results of subjective tests of voice disability may not be consistent among different patient populations, we suggest that more refined and focused measures be used for professional voice users. For example, the validated Singing Voice Handicap Index has been shown to be useful in monitoring treatment-related changes in singers. (10)
Our findings confirm those of other authors that objective measurements are less sensitive in identifying voice changes than are subjective measures. Also, even in our small cohort, the VHI survey was significantly more sensitive in professional voice users than in nonprofessional voice users. Such a disparity may affect other comparisons of voice outcome measures based solely on the VHI across different population groups.
In conclusion, our study showed that the VHI can be useful in monitoring the efficacy of treatments for voice disorders. Acoustic and aerodynamic voice laboratory measurements did not significantly correlate with VHI scores, and they were relatively less sensitive in measuring differences between pre- and postoperative status. Our study serves as a further illustration of how difficult it is (1) to integrate subjective and objective voice disability measurements and (2) to make meaningful interpretations of the multiple and varied aspects of voice with a single instrument.
We thank Chandler Thompson, MA, Linda Carroll, PhD, Chandra Ivey, MD, and Melissa Mortensen, MD, for their contributions to the collection of patient data.
(1.) Cohen SM, Dupont WD, Courey MS. Quality-of-life impact of non-neoplastic voice disorders: A meta-analysis. Ann Otol Rhinol Laryngol 2006;115(2):128-34.
(2.) Jacobson BH, Johnson A, Grywalski C, et al. The voice handicap index (VHI): Development and validation. Am J Speech Lang Pathol 1997;6(3):66-70.
(3.) Bogaardt HC, Hakkesteegt MM, Grolman W, Lindeboom R. Validation of the voice handicap index using Rasch analysis. J Voice 2007;21 (3):337-44.
(4.) Lowry R. Concepts & Applications of Inferential Statistics 2008.
http://faculty.vassar.edu/lowry/webtext.html. Accessed Feb. 25, 2010.
(5.) Deary IJ, Wilson JA, Carding PN, Mackenzie K. The dysphonic voice heard by me, you and it: Differential associations with personality and psychological distress. Clin Otolaryngol Mlied Sci 2003;28(4):374-8.
(6.) Wheeler KM, Collins SP, Sapienza CM. The relationship between VHI scores and specific acoustic measures of mildly disordered voice production. J Voice 2006;20(2):308-17.
(7.) Hsiung MW, Pal L, Wang HW. Correlation between voice handicap index and voice laboratory measurements in dysphonic patients. Eur Arch Otorhinolaryngol 2002;259(2):97-9.
(8.) Woisard V, Bodin S, Yardeni E, Puech M. The voice handicap index: Correlation between subjective patient response and quantitative assessment of voice. J Voice 2007;21 (5):623-31.
(9.) Rosen CA, Murry T, Zinn A, et al. Voice handicap index change following treatment of voice disorders. J Voice 2000;14(4):619-23.
(10.) Cohen SM, Witsell DL, Scearce L, et al. Treatment responsiveness of the Singing Voice Handicap Index. Laryngoscope 2008;118(9):1705-8. Jeffrey Cheng, MD; Peak Woo, MD, FACS
From the Eugen Grabscheid, M.D., Voice Center, Department of Otolaryngology, Mount Sinai Medical Center, New York City.
Corresponding author: Jeffrey Cheng, MD, Department of Otolaryngology, Mount Sinai Medical Center, One Gustave L. Levy Place, New York, NY 10029. E-mail: firstname.lastname@example.org
Table 1. Comparisons of mean pre-and postoperative VHI scores in all patients and by subgroups Domain Preop Postop All patients Functional 12.2 [+ or -] 7.2 9.5 [+ or -] 7.2 Physical 20.8 [+ or -] 8.7 11.0 [+ or -] 8.1 Emotional 14.0 [+ or -] 9.3 7.5 [+ or -] 7.2 Total ([dagger]) 47 [+ or -] 22.0 28 [+ or -] 21.0 Nonprofessional voice users Functional 9.8 [+ or -] 6.5 8.6 [+ or -] 7.7 Physical 15.8 [+ or -] 7.6 10.4 [+ or -] 6.4 Emotional 10.0 [+ or -] 7.1 5.9 [+ or -] 6.2 Total ([dagger]) 36 [+ or -] 16 25 [+ or -] 20 Professional voice users Functional 14.9 [+ or -] 7.3 10.5 [+ or -] 6.8 Physical 26.3 [+ or -] 6.4 11.8 [+ or -] 9.9 Emotional 18.5 [+ or -] 9.8 9.2 [+ or -] 8.2 Total ([dagger]) 60 [+ or -] 20 32 [+ or -] 24 Domain Change p Value All patients Functional 2.7 0.11 Physical 9.8 0.001 Emotional 6.5 0.001 Total ([dagger]) 19 0.002 * Nonprofessional voice users Functional 1.2 0.61 Physical 5.4 0.13 Emotional 4.1 0.05 * Total ([dagger]) 11 0.12 Professional voice users Functional 4.4 0.10 Physical 14.5 0.002 * Emotional 9.3 0.01 * Total ([dagger]) 28 0.007 * * Statistically significant, paired t test. ([dagger]) Total VHI whole numbers are rounded off. Table 2. Mean pre-and postoperative voice laboratory results in the entire group (All) and in the nonprofessional (Non) and professional (Pro) voice user subgroups Preoperative All Non Pro Acoustic parameters Fundamental 140.27 135.96 145.00 frequency [+ or -] 41.87 [+ or -] 41.92 [+ or -] 43.54 ([F.sub.0]) (Hz) Physiologic 32.87 30.67 35.29 [F.sub.0] range [+ or -] 10.3 [+ or -] 10.16 [+ or -] 10.43 of phonation Lower pitch 96.35 100.02 92.33 limit (Hz) [+ or -] 29.89 [+ or -] 35.05 [+ or -] 24.22 Upper pitch 713.29 604.16 833.33 limit (Hz) [+ or -] 281.17 [+ or -] 277.01 [+ or -] 244.65 Jitter % 2.17 2.99 1.27 [+ or -] 2.92 [+ or -] 3.91 [+ or -] 0.47 Shimmer % 5.58 8.02 2.91 [+ or -] 9.95 [+ or -] 13.6 [+ or -] 0.80 Noise-harmonic 0.16 0.18 0.13 ratio [+ or -] 0.10 [+ or -] 0.14 [+ or -] 0.02 Voiceless 6.45 11.41 0.99 [+ or -] 19.9 [+ or -] 27.12 [+ or -] 1.89 Aerodynamic parameters Maximum 15.50 13.54 17.65 phonation [+ or -] 8.31 [+ or -] 7.49 [+ or -] 9.02 time (sec) Target flow 0.26 0.30 0.23 (L/sec) [+ or -] 0.11 [+ or -] 0.13 [+ or -] 0.08 Efficiency 50.98 46.31 56.12 [+ or -] 40.59 [+ or -] 38.29 [+ or -] 44.30 Resistance 50.52 46.79 54.62 [+ or -] 31.30 [+ or -] 36.40 [+ or -] 25.88 Phonation 5.42 5.12 5.76 threshold [+ or -] 1.90 [+ or -] 1.66 [+ or -] 2.18 pressure Postoperative All Non Pro Acoustic parameters Fundamental 142.98 139.52 146.80 frequency [+ or -] 38.28 [+ or -] 31.00 [+ or -] 46.45 ([F.sub.0]) (Hz) Physiologic 34.40 32.37 36.63 [F.sub.0] range [+ or -] 5.47 [+ or -] 3.86 [+ or -] 6.28 of phonation Lower pitch 97.14 97.22 97.05 limit (Hz) [+ or -] 24.09 [+ or -] 21.58 [+ or -] 27.79 Upper pitch 724.80 633.88 824.80 limit (Hz) [+ or -] 234.24 [+ or -] 158.45 [+ or -] 269.85 Jitter % 1.54 1.56 1.54 [+ or -] 1.09 [+ or -] 1.10 [+ or -] 1.13 Shimmer % 3.24 3.42 3.05 [+ or -] 2.29 [+ or -] 2.19 [+ or -] 2.49 Noise-harmonic 0.14 0.12 0.15 ratio [+ or -] 0.06 [+ or -] 0.03 [+ or -] 0.08 Voiceless 2.84 2.42 3.30 [+ or -] 6.01 [+ or -] 4.41 [+ or -] 7.63 Aerodynamic parameters Maximum 15.56 15.57 15.55 phonation [+ or -] 6.63 [+ or -] 7.01 [+ or -] 6.56 time (sec) Target flow 0.23 0.23 0.22 (L/sec) [+ or -] 0.10 [+ or -] 0.08 [+ or -] 0.13 Efficiency 108.34 110.77 105.67 [+ or -] 100.96 [+ or -] 119.80 [+ or -] 81.78 Resistance 63.89 62.88 65.01 [+ or -] 51.15 [+ or -] 60.23 [+ or -] 42.21 Phonation 5.14 4.83 5.47 threshold [+ or -] 2.26 [+ or -] 1.75 [+ or -] 2.78 pressure p Value All Non Pro Acoustic parameters Fundamental 0.62 0.71 0.71 frequency ([F.sub.0]) (Hz) Physiologic 0.50 0.55 0.73 [F.sub.0] range of phonation Lower pitch 0.87 0.77 0.13 limit (Hz) Upper pitch 0.85 0.75 0.92 limit (Hz) Jitter % 0.29 0.18 0.51 Shimmer % 0.27 0.25 0.86 Noise-harmonic 0.43 0.23 0.52 ratio Voiceless 0.41 0.28 0.28 Aerodynamic parameters Maximum 0.96 0.06 0.44 phonation time (sec) Target flow 0.13 0.03 0.88 (L/sec) Efficiency 0.01 * 0.09 0.10 Resistance 0.08 0.07 0.43 Phonation 0.61 0.74 0.70 threshold pressure * Statistically significant, paired t test. Table 3. Correlation of preoperative VHI subscale and total scores with voice laboratory results (Pearson r correlation analysis) VHI domain VHI Functional Physical Emotional total Acoustic parameters Fundamental frequency ([F.sub.0]) 0.07 0.11 0.35 0.10 Physiologic [F.sub.0] range of phonation 1.00 0.66 0.08 0.36 Lower pitch limit (Hz) 0.50 0.29 0.36 0.08 Upper pitch limit (Hz) 0.06 0.04 * 0.01 * 0.01 Jitter % 0.93 0.52 0.82 0.85 Shimmer % 1.00 0.45 0.66 0.91 Noise-harmonic ratio 0.55 0.63 0.47 0.76 % Voiceless 0.98 0.40 0.77 0.82 Aerodynamic parameters Maximum phonation time (sec) 0.92 0.76 0.84 0.94 Target flow (L/sec) 0.88 0.82 0.77 0.93 Efficiency 0.13 0.07 0.09 0.05 * Resistance 0.20 0.96 0.29 0.39 Phonation threshold pressure 0.78 0.25 0.21 0.99 * Statistically significant. Table 4. Comparison of the calculated Voice Handicap Distress Index in the two subgroups Subgroup Mean score * 95% Confidence interval Nonprofessional 7.22 [+ or -] 6.26 3.00 to 11.42 Professional 13.63 [+ or -] 5.40 9.77 to 17.49 * Statistically significant difference (p =0.02). Figure. Chart shows the breakdown of voice users by category. Patients in categories 4 and 5 are classified as professional voice users. Voice user classification Category 2: Low to average voice user, such as a person in a 1- or 2-person household (n = 4; 19%) Category 3: Average to heavy voice user, such as a mother with small children (n = 7; 33%) Category 4: Professional voice user, such as a teacher, politician, barrister (n = 6; 29%) Category 5: Elite voice such as a professional or an o vocational singer (n = 4; 19%) Note: Table made from pie chart.
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||ORIGINAL ARTICLE|
|Author:||Cheng, Jeffrey; Woo, Peak|
|Publication:||Ear, Nose and Throat Journal|
|Date:||Apr 1, 2010|
|Previous Article:||Sensitivity and specificity of rapid antigen detection testing for diagnosing pharyngitis in the emergency department.|
|Next Article:||Laryngeal cancer in patients younger than 30 years: a review of 99 cases.|