Patterns in the Evaluation of Hoarseness: Time to Presentation, Laryngeal Visualization, and Diagnostic Accuracy

The Laryngoscope VC 2014 The American Laryngological, Rhinological and Otological Society, Inc. Patterns in the Evaluation of Hoarseness: Time to Presentation, Laryngeal Visualization, and Diagnostic Accuracy Sarah E. Keesecker, MD; Thomas Murry, PhD; Lucian Sulica, MD Objectives/Hypothesis: Controversial recommendations regarding the evaluation of dysphonia have been made in the absence of evidence related to clinical practice. This study aims to describe existing patterns of care for dysphonia to generate data for potential systemic improvement and provide a baseline for dysphonia recommendations. Study Design: Retrospective review. Methods: Information regarding the current complaint, including duration of hoarseness; inciting factors; number and type of previous physicians seen; Voice Handicap Index-10; and details of prior evaluation, diagnosis, and treatment was collected from patient records. Results: A total of 259 patients complaining of hoarseness were evaluated. Of those, 35.1% presented directly to subspecialty care, whereas 61% were previously evaluated by another otolaryngologist. Median times (in months) from symptom onset to evaluation were as follows: initial evaluation, 3.0; laryngoscopy, 3.0; stroboscopic exam, 5.8; subspecialty evaluation, 6.6. A total of 64.5% of patients had at least one incoming diagnosis; 45% of all incoming diagnoses were revised on reevaluation. Diagnoses most commonly revised included no abnormality, edema or laryngopharyngeal reflux disease (LPR), infection or allergy, and muscle tension dysphonia (MTD) or behavioral disorders. Final diagnoses that most frequently differed from incoming diagnoses were paresis; MTD or behavioral disorders; malignancy; and sulcus, atrophy, or scar. Conclusions: Patients received prompt laryngeal visualization. However, we observed high rates of diagnostic error. Initial diagnoses of LPR, edema, infection, and allergy appear to be particularly likely to be revised on further evaluation; and scar, sulcus, atrophy, and paresis are likely to be overlooked. Key Words: Hoarseness, dysphonia, laryngoscopy, stroboscopy, diagnosis, guideline. Level of Evidence: 4. Laryngoscope, 125:667 673, 2015 INTRODUCTION Hoarseness (or dysphonia) affects nearly one-third of the adult population at some point. 1,2 Affected patients present to a variety of physicians including primary care doctors, otolaryngologists, gastroenterologists, pulmonologists, and even emergency rooms. Hoarseness is a symptom, not a diagnosis, and effective treatment recommendations cannot be made without determining the underlying cause. The range of such causes is vast; the more serious conditions require prompt treatment. Visualization of the larynx is essential for assessing and accurately diagnosing dysphonic patients. 3 Additional Supporting Information may be found in the online version of this article. From the College of Physicians & Surgeons (S.E.K.), Columbia University; and the Parker Institute for the Voice, Dept. of Otolaryngology Head & Neck Surgery (T.M., L.S.), Weill Cornell Medical College, New York, New York, U.S.A. Editor s Note: This Manuscript was accepted for publication September 8, 2014. Presented at the American Laryngological Association s 2014 Spring Meeting at Combined Otolaryngological Spring Meetings (COSM) in Las Vegas, Nevada on May 14 15, 2014. This work was supported by a grant from the Doris Duke Charitable Foundation to Columbia University Medical Center (S.E.K.). The authors have no other funding, financial relationships, or conflicts of interest to disclose. Send correspondence to Sarah E. Keesecker, MD, 180 Fort Washington Avenue, New York, NY 10032. E-mail: sek2157@columbia.edu DOI: 10.1002/lary.24955 Despite these considerations, there are no truly evidence-based guidelines regarding the evaluation of hoarseness or appropriate referral to a subspecialist. Meanwhile, controversial recommendations have been made without regard to clinical practice, some of which may negatively impact patient care. For example, the American Academy of Otolaryngology Clinical Practice Guidelines for hoarseness 4 allow for a 3-month wait period before patients with unresolved hoarseness undergo laryngoscopy. This 3-month window could delay appropriate diagnosis and treatment and allow malignancies to progress. 5 The recommendation appears to have no basis in clinical practice or expert opinion. A survey study of otolaryngologists found that the longest delay to laryngoscopy averaged 12.96 days. 6 A survey of members of the American Laryngological Association, the American Broncho-Esophagological Association, and the European Laryngology Society recommended a mandated time to laryngoscopy of 1 week to 1 month for the most acutely dysphonic patients, regardless of risk factors for malignancy, and no greater than 2 months in any situation. 7 Information regarding actual clinical practice may offer a basis from which to establish and formalize algorithms for the diagnosis and management of the hoarse patient. This type of data may provide insights into current patterns of evaluation and management and reveal inefficiencies and opportunities for systemic improvement. 667

The purpose of this study is to describe existing patterns of care in patients with a chief complaint of hoarseness presenting to a subspecialty care physician. The specific aims are to: 1. Evaluate median time from symptom onset to primary medical evaluation, specialty medical evaluation, initial laryngoscopy, and initial stroboscopic exam. 2. Investigate discrepancies in diagnosis between primary and subspecialty care. 3. Analyze patterns of diagnosis to highlight areas that require more specific attention. MATERIALS AND METHODS This study was approved by the institutional review board of Weill Cornell Medical College. A retrospective review was conducted on new patients with a chief complaint of hoarseness presenting to the laryngology service of a university medical center in an urban area between July 16, 2013, and November 19, 2013. All patients were evaluated by the senior author (L.S.), a fellowship-trained laryngologist, and received a final diagnosis based on history, physical exam, and stroboscopy. Data was primarily obtained from a standard intake form (Supp. App. S1; online only). Where necessary, the chart was reviewed to augment or clarify this data. Patients were excluded if intake forms were missing or the duration of the complaint could not be assessed. Demographic information was compiled, along with information regarding onset of complaint, characteristics and history of the hoarseness, alcohol and tobacco use, surgical history, medications, quality-of-life indicators, Voice Handicap Index-10 (VHI-10), and Singers Voice Handicap Index-10 (SVHI-10), where applicable. In addition, information regarding previous medical care and physician visits was obtained. The date of hoarseness onset and the dates of prior physician visits, procedures, and treatments were recorded. When exact dates were unavailable, they were approximated using a best estimate. For example, if the patient reported seeing a physician in October 2013, the date was approximated as October 1, 2013. If a patient reported the problem start date as many years ago, the duration of the complaint was estimated as 7 years (unless a better estimate could be obtained from the history). If a patient reported lifetime issues with hoarseness but had recently worsened, the period of worsening symptoms was considered to be the current complaint. All incoming diagnoses (those given by previous providers) and final diagnoses were recorded. We attempted to minimize the effect of the lack of uniform diagnostic terminology by creating broad, inclusive diagnostic categories: 1) cyst, polyp, or benign mass; 2) vocal fold paresis; 3) vocal fold paralysis; 4) neurological disorders other (other than paresis or paralysis; e.g., adductor spasmodic dysphonia, Parkinson s disease); 5) infection or allergy; 6) edema or laryngopharyngeal reflux disease (LPR); 7) sulcus, atrophy, or scar; 8) muscle tension dysphonia (MTD) or behavioral disorder; 9) malignancy; and 10) other (e.g., musculoskeletal injury, trauma). The dates for each physician visit were used to determine the order of doctors seen. For patients who had seen a prior physician, the mean number of prior visits was calculated, along with the mean number by diagnostic category. The complaint start date, dates of prior medical evaluation, current visit date, and procedural information were used to determine median times to primary medical evaluation, laryngoscopy, stroboscopic exam, and subspecialty evaluation. Median time to first physician evaluation by VHI-10 category was also TABLE I. Demographic Information Variable No. (%) Age < 18 8 (3.1) 18 35 69 (26.6) 36 55 79 (30.5) 56 70 67 (25.9) > 70 36 (13.9) Sex Male 112 (43.2) Female 147 (56.8) Voice Use History Performer 89 (34.4) Nonperformer 170 (65.6) VHI Category 0 9 63 (26.4) 10 19 73 (30.5) 20 29 54 (22.6) 30 40 49 (20.5) Number of Prior Physicians Seen 0 91 (35.1) 1 94 (36.3) 2 48 (18.6) 31 26 (10.0) VHI 5 Voice Handicap Index. calculated to examine the effect of hoarseness severity on time to presentation. Furthermore, categories for the total length of complaint time from symptom onset to subspecialty evaluation were created (0 to 3 months; > 3 months to 1 year; and > 1 year), and the distribution of final diagnoses in each group was evaluated. Diagnostic data was used to investigate discrepancies between incoming diagnoses and final diagnoses after subspecialty evaluation. Discrepancies between the incoming and final diagnoses were calculated for each diagnostic category. In addition, relationships between a variety of specific variables including medical history, demographic data, previous evaluative procedures, and prior treatment were investigated to answer specific questions. For example, data regarding prior treatment was used to investigate patterns of antibiotic prescription in the infectious category, and data regarding procedures was used to investigate the use of esophagoscopy. Descriptive statistics were used along with statistical tests, specifically analysis of variance (ANOVA), where applicable. RESULTS A total of 259 new patients with voice complaints presented for subspecialty evaluation between July 16, 2013, and November 19, 2013. Demographic information is summarized in Table I. Sixty-one percent of patients saw an otolaryngologist prior to subspecialty evaluation, and 9.7% initially presented to a primary care physician (PCP), emergency room physician, or other specialist. Sixty-one percent of patients had laryngoscopy, and 14.7% had stroboscopy prior to subspecialty evaluation. 668

Fig. 1. Average number of prior physicians seen (for patients seen by another physician, n 5 168) across the 10 diagnostic categories. Oneway ANOVA demonstrated statistically significant difference between groups (F [9, 158], 2.03; P value, 0.04). For the 64.9% (n 5 168) of patients who had seen a prior physician, the average number of previous doctor visits was 1.71 (standard deviation 5 1.06), with a range of 1 to 7. The mean number of prior physician visits was compared across final diagnosis categories (Fig. 1). There was a statistically significant difference between the groups, as determined by one-way ANOVA (F [9, 158] 5 2.03, P value 5 0.04), which may identify challenging diagnoses. Final diagnoses with the highest average number of prior physician visits were neurological disorders other, paresis, and malignancy. The median times from symptom onset to primary evaluation, diagnostic procedures, and specialty evaluation are summarized in Table II. Median time from symptom onset to first physician visit across the four different VHI-10 categories is also shown in Table II. Proportions of final diagnoses observed for each of the three length of complaint categories are shown in Figure 2. Several diagnoses demonstrated marked variability. The infection or allergy category made up 20% of total diagnoses in the 0 to 3 months category, 2% of the 3 months to 1 year category, and 3% of the greater than 1 year category. Sulcus, atrophy, or scar diagnoses made up 2%, 7%, and 28% of those categories, respectively, whereas neurological disorders other represented 1%, 5%, and 11%. There were a total of 209 incoming diagnoses for 168 patients who had seen a prior physician. Forty-five percent of incoming diagnoses were revised by the subspecialist (Supp. Fig. S1; online-only) and for those revised, the most common final diagnoses were: vocal fold paresis (16%); sulcus, atrophy, or scar (9%); cyst, polyp, or benign mass (8%); and neurological disorders other (4%). The final diagnoses with the highest proportion of discrepancy from incoming diagnoses (Table III) included: MTD or behavioral (100%; n 5 5); paresis (87%; n 5 38); malignancy (71%; n 5 7); and sulcus, atrophy, or scar (69%; n 5 29). Patients with a final diagnosis of malignancy had a total of seven instances of prior diagnoses. Of these, only two were diagnoses of possible malignancy. The other five included edema or LPR (3), infection or allergy (1), and benign mass (1). Of 38 final diagnoses of paresis, 37% were revised from an initial diagnosis of edema or LPR, 21% from infection or allergy, 11% from benign mass, 11% from paralysis, and 3% from no abnormality. The incoming diagnoses most commonly revised by the subspecialist (Table IV) included the following: no abnormality (100%; n 5 7); MTD or behavioral (100%; n 5 2); edema or LPR (93%; n 5 41); and infection or allergy (87%; n 5 31). Of 41 incoming diagnoses of reflux, 34% were revised to paresis; 22% to sulcus, atrophy, or scar; and 15% to a benign mass. Similarly, of 31 incoming diagnoses of infection or allergy, 26% were revised to benign mass; 26% to paresis; and 16% to sulcus, atrophy, or scar. In order to evaluate the cost of such discrepancies, we analyzed the use of antibiotic treatment and esophagoscopy in our cohort. Twenty-two patients with a prior diagnosis of infection presented to subspecialty care for an ongoing complaint of hoarseness. Of these, 64% 669

TABLE II. Median Times from Symptom Onset to First Physician Visit or Diagnostic Procedure Median Time to First Physician Visit or Diagnostic Procedure Median (m) Q1 25% (m) Q2 75% (m) Minimum (d) Maximum (y) First physician visit 3.0 0.9 12.0 0.0 20.0 First laryngoscopy 3.0 1.0 12.0 0.0 20.0 First stroboscopy 5.8 2.0 16.6 1.0 32.7 Subspecialty visit 6.6 2.3 21.2 1.0 32.7 Median Time to First Physician Visit by VHI-10 Category [1 9] [10 19] [20 29] [30 40] First physician visit 3.02 3.06 4.39 1.02 d 5 days; m 5 months; q 5 quartile; y 5 years. (n 5 14) had received at least one round of antibiotic treatment without resolution of their symptoms. However, upon subspecialty evaluation, only 14% (n 5 3) were given a final diagnosis of infection. In addition, a subset of seven patients out of a total of 39 previously diagnosed with reflux underwent esophagoscopy for this presumed diagnosis. However, none of these patients received a final diagnosis of reflux. In this subset, the final diagnoses included vocal fold paresis (n 5 1); adductor spasmodic dysphonia (n 5 1); and sulcus, atrophy, or scar (n 5 5). DISCUSSION In the studied cohort, most patients presented directly to an otolaryngologist (55.2% were initially evaluated by an otolaryngologist other than the senior author [L.S.]; 35.1% presented directly to subspecialty care). The median times from symptom onset to first presentation based on VHI-10 category (Table II) demonstrated that patients with more severe hoarseness (VHI-10 score of 30 40) presented to a physician sooner. The median time from symptom onset to first physician visit as well as symptom onset to first laryngoscopy was 3 months (Table II). This is consistent with a pattern of early specialty care and indicates that patients received prompt laryngeal visualization upon presentation, substantially sooner than the 3 months proposed by the American Academy of Otolaryngology Head and Neck Surgery guidelines. 4 Prior studies have demonstrated the importance of laryngoscopy in accurately diagnosing laryngeal disorders, given the inaccuracy of history and physical exam alone. 6,8 Fig. 2. Diagnoses based on total length of complaint categories. The distribution of diagnoses observed in three separate categories based on total length of complaint. Total length of complaint was calculated from symptom onset to evaluation by the subspecialist. 670

TABLE III. Final Diagnostic Category by Incoming Diagnostic Category. B mass 5 benign mass; E/LPR 5 edema or laryngopharyngeal reflux disease; Infx/All 5 infection or allergy; Malig 5 malignancy; MTD/B 5 muscle tension dysphonia or behavioral disorders; Neuro O 5 neurological disorders other than paresis or paralysis; No Ab 5 no abnormality; Prs 5 paresis; Pys 5 paralysis; S/A/S 5 sulcus, atrophy, or scar. More time elapsed to stroboscopy (median of 5.8 months from symptom onset) and to subspecialty evaluation (median of 6.6 months). The role of stroboscopy in hoarseness evaluation has not been precisely established, despite its clinical availability for decades. Studies suggest that stroboscopy is especially likely to be useful when continuous light laryngoscopy does not explain the severity of the patient s complaint, when hoarseness persists despite treatment, and when unexpected hoarseness occurs after microlaryngoscopy. 9 11 However, perceptions regarding its utility vary. Paul et al. 6 demonstrated that the majority of academic laryngologists consider stroboscopy very valuable, whereas private practice clinicians disagree. This perception may account in part for the observed delay in stroboscopy. In this study, several findings suggest which disorders present particular diagnostic challenges. Patients ultimately diagnosed with vocal fold paresis, neurological disorders other, and malignancy were evaluated, on average, by the highest number of prior physicians (Fig. 1). Additionally, several diagnoses sulcus, atrophy, scar, and neurological disorders other made up a larger proportion of the total diagnoses as the length of complaint increased (Fig. 2), indicating that these patients were still seeking care after a longer period of time. Although this could be attributed to the chronicity of certain diseases in these categories, it may also be driven by misdiagnosis and inappropriate or insufficient treatment. This interpretation is supported by the observation that most of the diagnoses which resulted in multiple physician visits and diagnostic delay do not involve gross motion abnormalities or mass lesions easily discernible on continuous-light laryngoscopy. According to Cohen et al., otolaryngologists tend to be less comfortable diagnosing disorders without obvious physical abnormalities, 12 which could partially account for the discrepancies. Almost one-half of all incoming diagnoses were revised upon re-evaluation (Supp. Fig. S1; online only). Final diagnoses that most frequently differed from incoming diagnosis included MTD or behavioral disorders; paresis; malignancy; and sulcus, atrophy, or scar (Table III). Once again, diagnoses without gross motion abnormality or mass lesion are prominently represented. Incoming diagnoses most commonly revised included no abnormality, MTD or behavioral disorders, edema or LPR, and infection or allergy (Table IV), conditions with laryngoscopic findings that are notably nonspecific and somewhat interchangeable. Other studies have demonstrated similar results. For example, a recent prospective, multi-institutional study found that 33% of patients with laryngological complaints had a final diagnosis after subspecialty evaluation that disagreed with the referring diagnosis. 13 The most commonly revised referring diagnoses included LPR (61% revised) and laryngeal dystonia (56% revised). Similarly, a longitudinal study found that the diagnosis of hoarseness commonly changed upon specialty evaluation and also within specialty evaluation. 14 Two groups were evaluated: 1) patients seen by a PCP followed by an otolaryngologist, and 2) patients seen at least twice 671

TABLE IV. Incoming Diagnostic Category by Final Diagnostic Category. B mass 5 benign mass; E/LPR 5 edema or laryngopharyngeal reflux disease; Infx/All 5 infection or allergy; Malig 5 malignancy; MTD/B 5 muscle tension dysphonia or behavioral disorders; Neuro O 5 neurological disorders other than paresis or paralysis; No Ab 5 no abnormality; Prs 5 paresis; Pys 5 paralysis; S/ A/S 5 sulcus, atrophy, or scar. by an otolaryngologist. Three-fourths of patients in the first group and one-half in the second group received a different diagnosis upon specialty evaluation. In both cohorts, acute laryngitis was the most commonly revised diagnosis. However, the authors could not differentiate between clinicians within a specialty (i.e., whether a patient saw the same otolaryngologist twice or saw two different otolaryngologists). Additionally, diagnostic data was obtained exclusively from ICD codes, which are notoriously vague for voice disorders. For example, scar, sulcus, atrophy, and even reflux laryngitis have no specific code and can be coded in more than one way. Furthermore, almost one-half of the subjects had to be classified as nonspecific dysphonia (presumably ICD 784.42). Finally, it was impossible to determine whether the diagnosis change was due to evolution of disease pathology, such as acute laryngitis resolving, or a difference in opinion between providers. Although the number of patients with malignancies in this study is small, it deserves attention because of the consequences of diagnostic delay. Based on our results, 60% of the cases referred for possible malignancy were actually due to benign pathology. More worrisome are cases that were misdiagnosed as benign and never referred to a subspecialist. When the final diagnosis was malignancy, 71% of initial diagnoses were not in agreement. These included benign mass, infection, allergy, edema, or LPR. It is concerning that several of these initial misdiagnoses represent nonspecific conditions. Cohen et al. 15 have shown that these are less likely to generate specialty referral. 672 An intelligent recommendation for specialty referral and perhaps subspecialty referral is clearly needed. The 2-week interval traditionally used is broadly consistent with the interval in which a self-limited condition such as viral laryngitis would be expected to resolve, which underlies its rationale. The 3-month interval proposed in the Academy guidelines 4 stretches well beyond that and may carry harm from delay in diagnosis. The only potential benefit is cost, assuming longer-lasting self-limited pathologies, but this is uncertain given the spectrum of possible diagnoses persisting beyond 2 weeks, as well as the cost of inappropriate testing and treatment. Therefore, this may be a false economy to the detriment of the patient. Although there is clear consensus that visualizing the larynx is necessary for diagnosis, 6 8 it appears that it is not always sufficient. A consistent group of diagnoses (paresis, sulcus, atrophy, scar, and neurologic disorders other than peripheral neuropathy) resulted in more physician visits and delay in diagnosis. Sulcus, atrophy, scar, and paresis are notoriously difficult to diagnose on continuous light examination. The subject of referral for stroboscopy has been virtually untouched in the literature, but data suggests that it should be considered in patients with hoarseness and no evident abnormality on laryngoscopy. Furthermore, practitioners should consider stroboscopy for patients diagnosed with inflammatory conditions marked by nonspecific findings such as LPR, infection, or allergy if they do not respond to treatment promptly. This is at odds with current recommendations for reflux, which encourage long-term

empiric treatment, but is consistent with other studies regarding reflux 13 and acute laryngitis. 14 Clinical common sense dictates that for hoarseness, examination of the sound source is likely to be more revealing than examination of the stomach and esophagus. As with diagnostic delay, inaccurate diagnosis has a dollar value and potential morbidity related to unnecessary procedures and treatments. Forty-one patients in this study received an initial diagnosis of LPR. Seven underwent esophagoscopy, none of which received a final diagnosis of reflux. Several of these patients had more extensive work-ups with other ultimately unhelpful procedures and treatments. For example, one patient a 50- year-old woman with hoarseness and cough since partial thyroidectomy in 2009 had been evaluated by three otolaryngologists, a gastroenterologist, an allergist, and a pulmonologist. Her prior work-up included allergy testing, bronchoprovocation challenge testing, esophagoscopy, barium swallow, ph monitoring, laryngoscopy, and stroboscopic exam. On presentation to the subspecialist, she was diagnosed with vocal fold paresis with neurogenic cough and has since reported near-complete relief with injection augmentation and tramadol. This study has limitations related to methodology, patient population, and diagnostic ambiguities. The study is a retrospective review of data from a single center. In addition, the data is largely patient-reported and therefore limited by the patients ability to understand and recall medical information. However, many patients continued to seek care for persistent symptoms, making a correct initial diagnosis less likely. Because of the nature of the data, we did not expect to capture the full picture and scope of evaluation for every patient. However, additional instances of physician visits, diagnosis, and management would probably further amplify the reported findings and further substantiate the demonstrated inefficiencies. An additional limitation is the fact that we combined certain diagnoses into diagnostic categories. Although this may cause some loss of detail in the analysis, the measure was necessary to minimize the effect of variability in nomenclature (e.g., nodule, midfold mass, and fibrovascular midfold lesion all referring to similar pathology). We believe that the categories are clinically valid and as detailed, or more detailed, than available in similar literature. In this project, we presumed that the final diagnosis given by the subspecialist was the most correct. However, excluding certain diagnoses for which pathologic correlation exists, such as malignancy, or that yield pathognomonic findings, such as sulcus and paralysis, there is no final standard for diagnosis. We recognize that many laryngological diagnoses rest on expert opinion, and some are notoriously subjective. Finally, this study examines only patients presenting to subspecialty care for persistent hoarseness. This may represent less than 10% of patients with hoarseness. 15 However, many of these patients represent individuals whose complaints have not been adequately addressed by prior physicians, and thus may be the most useful group to examine in order to address the efficiency of evaluation. CONCLUSION For patients presenting with hoarseness, median times from symptom onset to first physician visit and to laryngoscopy were equal, which indicates that patients received prompt laryngeal visualization. We observed high rates of discrepancy between incoming diagnoses and final diagnosis after subspecialty evaluation, which included stroboscopy. This was particularly true for patients initially diagnosed with inflammatory or infectious conditions with relatively nonspecific findings, such as reflux, allergy, or infection. Patients ultimately diagnosed with sulcus, atrophy, scar, paresis, and neurologic disorders other than paresis or paralysis sought care for longer and from more physicians, and account for a disproportionate number of discrepancies between initial and final diagnosis. Given the importance of laryngoscopy for diagnosis and the time-sensitive nature of treatment for malignancy, practitioners should continue to visualize the larynx without delay in the hoarse patient. Additionally, practitioners should be aware that certain laryngeal pathologies may be difficult to diagnose with continuous light endoscopy and that nonspecific inflammatory diagnoses may mask underlying pathology, incur unnecessary expense, and delay appropriate treatment. In patients who do not respond to treatment promptly, further laryngeal evaluation and/or subspecialty referral should be considered before embarking on prolonged empiric treatment or diagnostic investigations directed away from the larynx. BIBLIOGRAPHY 1. Roy N, Merrill RM, Gray SD, Smith EM. Voice disorders in the general population: prevalence, risk factors, and occupational impact. Laryngoscope 2005;115:1988 1995. 2. Cohen SM. Self-reported impact of dysphonia in a primary care population: an epidemiological study. Laryngoscope 2010;120:2022 2032. 3. Rosen CA, et al. Advances in office-based diagnosis and treatment in laryngology. Laryngoscope 2009;119(suppl 2):S185 S212. 4. Schwartz SR, et al. Clinical practice guideline: hoarseness (dysphonia). Otolaryngol Head Neck Surg 2009;141(suppl 2):S1 S31. 5. Hoare T, Thomson H, Proops D. Detection of laryngeal cancer the case for early specialist assessment. J R Soc Med 1993;86:390 392. 6. Paul BC, Branski RC, Amin MR. Diagnosis and management of new-onset hoarseness: a survey of the American Broncho-Esophagological Association. Ann Otol Rhinol Laryngol 2012;121:629 634. 7. Sadoughi B, Fried MP, Sulica L, Blitzer A. Hoarseness evaluation: a transatlantic survey of laryngeal experts. Laryngoscope 2014;124:221 226. 8. Paul BC, Chen S, Sridharan S, Fang Y, Amin MR, Branski RC. Diagnostic accuracy of history, laryngoscopy, and stroboscopy. Laryngoscope 2013; 123:215 219. 9. Casiano R, Zaveri V, Lundy D. Efficacy of videostroboscopy in the diagnosis of voice disorders. Otolaryngol Head Neck Surg 1992;107:95 100. 10. Sataloff R, Spiegel J, Hawkshaw M. Strobovideolaryngoscopy: results and clinical value. Ann Otol Rhinol Laryngol 1991;100:725 727. 11. Woo P, Casper J, Colton R, Brewer D. Diagnosis and treatment of persistent dysphonia after laryngeal surgery: a retrospective analysis of 62 patients. Laryngoscope 1994;104:1084 1091. 12. Cohen SM, Pitman MJ, Noordzij JP, Courey M. Evaluation of dysphonic patients by general otolaryngologists. J Voice 2012;26:772 778. 13. Solowski N, et al. Accuracy of referral diagnosis: a laryngological perspective. In: American Brochoesophagological Association Session of the 116th COSM meeting. Orlando, Florida, USA; April 10, 2013. 14. Cohen SM, Dinan MA, Roy N, Kim J, Courey M. Diagnosis change in voice-disordered patients evaluated by primary care and/or otolaryngology: a longitudinal study. Otolaryngol Head Neck Surg 2014;150: 95 102. 15. Cohen SM, Kim J, Roy N, Courey M. Factors influencing referral of patients with voice disorders from primary care to otolaryngology. Laryngoscope 2014;124:214 220. 673