Assessment of Trends in Utilization of Nasal Endoscopy in the Medicare Population, 2000-2016

Key Points

Question  What trends in nasal endoscopy utilization have been observed in the Medicare population?

Findings  In this study of data from Medicare patients undergoing nasal endoscopy from 2000 to 2016, the total number of nasal endoscopies increased 313% and nationwide reimbursement from Medicare increased from $10 199 497 to $86 301 850. In 2015, the percentage of Medicare beneficiaries undergoing nasal endoscopy varied from 0.1% to 1.7% among the states.

Meaning  The findings suggest that the utilization of nasal endoscopy has increased substantially in the United States and varies among states.

Importance  Nasal endoscopy is the most highly reimbursed routine clinical procedure in otolaryngology by Medicare. To our knowledge, the economic and demographic characteristics of nasal endoscopy have not been characterized at a population level.

Objective  To quantify trends in use of nasal endoscopy among Medicare recipients at a national and state level.

Design, Setting, and Participants  Cross-sectional study of procedure and beneficiary data from January 1, 2000, to December 31, 2016, and available health care provider data from 2015 obtained from the Centers for Medicare & Medicaid Services for all diagnostic nasal endoscopies categorized as Current Procedural Terminology code 31231.

Exposures  Diagnostic nasal endoscopy.

Main Outcomes and Measures  Nasal endoscopies were analyzed by state, medical or surgical specialty, mean reimbursement, health care provider density, and type of health care practitioner performing the procedure.

Results  The Centers for Medicare & Medicaid Services reimbursed $86.3 million for 559 547 nasal endoscopies in 2016. The total number of nasal endoscopies increased 313%, from 135 494 in 2000 to 559 547 in 2016 (9.3% average annual rate of increase). The mean reimbursement rate per nasal endoscopy varied by state, from $114.25 in Puerto Rico to $189.53 in New York. The percentage of Medicare beneficiaries receiving nasal endoscopy per state ranged from 0.1% in Alaska to 1.7% in New York. Almost all nasal endoscopies were performed by physicians (97.3%), with otolaryngologists (97.2%) being the most common specialty. Reimbursement rates (r = 0.60) and density of health care providers (r = –0.56) were correlated with higher utilization in a state's Medicare population.

Conclusions and Relevance  Utilization of nasal endoscopy by otolaryngologists in the Medicare population appears to have increased substantially over the past 2 decades. Practice patterns and reimbursement appeared to vary across the United States.

In 1901, the first nasal endoscopy was performed with a modified cystoscope to visualize the maxillary sinus through an oroantral fistula. However, it was not until the 1960s that the nasal endoscope was developed by Harold Hopkins, leading to the development of endoscopic sinus surgery.^1,2 During the 1990s, the nasal endoscope was popularized in the United States and recommended for use in office examinations of patients.³

Diagnostic nasal endoscopy is classified under Current Procedural Terminology (CPT) code 31231 and is defined by the American Rhinologic Society as a “procedure performed to better characterize the anatomy of the nasal cavity and/or paranasal sinuses and to identify sinonasal pathology not afforded by anterior rhinoscopy”; it can be performed “using rigid or flexible endoscopes.”⁴ In a typical nasal endoscopy in the physician office setting, the nose is first topically anesthetized and decongested. Then, a 0° or 30° telescope is passed along the floor of the nose into the nasopharynx, examining the inferior turbinate, sphenoethmoidal recess, roof of the nasal cavity, middle turbinate, and middle meatus.^3,5 To our knowledge, the consequences of this procedure within the US health care system since it was introduced decades ago has not been studied. However, based on an analysis of Medicare data from 2012, nasal endoscopy is the most reimbursed procedure performed by otolaryngologists in the United States.⁶

The main objective of this study was to quantify diagnostic nasal endoscopy utilization in the Medicare population across the United States at a national and state level. Additional objectives included identifying the role of different health care professionals in performing nasal endoscopies.

Annual procedure data from January 1, 2000, through December 31, 2016, were obtained from the Part B National Summary Data File. Detailed health care provider and Medicare payment data from 2015 were obtained from the Medicare Provider Utilization and Payment Data: Physician and Other Supplier Public Use File (PUF) from the Centers for Medicare & Medicaid Services. Detailed Medicare data from 2016 were not available at the time of this study. The total number of beneficiaries per state from 2015 was obtained. All data were accessed online at https://www.cms.gov/ and http://www.kff.org/ between April 1, 2018, and May 2, 2018. The study was determined to be exempt from review by the University of Southern California Institutional Review Board because the study analyzed existing data that were publicly available and had no identifiers.

The total number of procedures performed and the total amount paid by Medicare nationwide for diagnostic nasal endoscopy (CPT 31231) were extracted from the Part B National Summary Data File. The total number of ethmoidectomy (CPT 31254 and CPT 31255) procedures was also extracted to serve as a comparison.

All health care professionals in the Physician and Other Supplier PUF who billed for nasal endoscopy (CPT 31231) during 2015 were identified. To protect the identity of patients, the database does not include data for services that were performed for 10 or fewer beneficiaries. Microsoft Excel (Microsoft Corporation) and pivot tables were used to analyze the database.

Health care professionals were categorized into 4 types of health care providers based on the specialty documented in the database: physician (MD or DO), physician assistant, nurse practitioner, and other. Physician assistant and nurse practitioner were already labeled categories in the database. All physician specialties were consolidated into the MD category. Other included health care providers labeled as surgical centers or osteopathic manipulative medicine providers. A subset analysis of physician specialties was performed to investigate the number of procedures performed by each physician specialty.

Nasal endoscopy rates were analyzed by state. For each state, the total number of nasal endoscopies performed, the total amount reimbursed by Medicare, the percentage of Medicare beneficiaries per state receiving nasal endoscopy, and the mean nonfacility reimbursement per nasal endoscopy were calculated. The mean reimbursement rate was the mean Medicare payment for nasal endoscopy performed by a health care provider in that state. The total amount reimbursed by Medicare was based on a summation of the mean Medicare payment amounts multiplied by the number of services performed for each health care provider. To calculate the percentage of Medicare beneficiaries per state receiving nasal endoscopy, data on the total number of Medicare beneficiaries was obtained from the “Total Number of Medicare Beneficiaries” table from the Kaiser Family Foundation for 2015. All health care providers were classified as submitting claims from a facility or nonfacility setting. Nonfacilities are generally considered to be office settings. In the PUF database, the Medicare payment amount represented only a physician’s professional fee and did not include the facility payment. For services delivered in a nonfacility setting, the Medicare payment amount represented the complete payment for the service. Therefore, health care providers who submitted claims from a facility were excluded from calculating mean reimbursement per nasal endoscopy to prevent underestimation.

The percentage of Medicare beneficiaries receiving nasal endoscopy by state was geospatially mapped using the online mapping tool Viz (https://viz.socialcops.com/). This map has not been published elsewhere.

Scatterplots of the percentage of Medicare beneficiaries in that state undergoing nasal endoscopy vs the mean reimbursement per nasal endoscopy and density of health care providers performing nasal endoscopy in that state were generated. The density was calculated by dividing that state’s total number of Medicare beneficiaries by the number of health care providers who performed nasal endoscopies in that state. Correlations were tested at a 95% CI using the Pearson correlation test.

In 2015, the otolaryngology procedure with the highest total allowed charges by Medicare was diagnostic nasal endoscopy at $94 472 734, followed by diagnostic laryngoscopy ($58 471 364) and removal of impacted ear wax ($37 124 601) (Table 1). The most commonly billed procedure was antigen therapy services, with 2 072 867 services performed.

From 2000 through 2016, the total number of nasal endoscopies increased 313%, from 135 494 to 559 547 (9.3% average annual rate of increase), and the nationwide reimbursement from Medicare increased 746%, from $10 199 497 to $86 301 850. During the same period, the total number of ethmoidectomies increased 50%, from 25 699 to 38 609 (2.6% average annual rate of increase) and the nationwide reimbursement from Medicare increased 55%, from $9 509 852 to $14 773 413 (Figure 1).

Of the 507 962 nasal endoscopies performed in 2015, 494 093 (97.3%) were performed by physicians (Table 2). Of all physicians, otolaryngologists performed 480 457 (97.2%) of all nasal endoscopies, followed by ophthalmologists (0.7%) and allergists and immunologists (0.7%).

When diagnostic nasal endoscopies were analyzed by state, practitioners in New York (78 994), Florida (70 193), California (48 899), and New Jersey (30 285) performed the greatest number of diagnostic nasal endoscopies (eAppendix in the Supplement). New York also had the highest percentage of Medicare patients undergoing diagnostic nasal endoscopies at 1.7% followed by New Jersey (1.3%), Florida (1.2%), and Washington, DC (1.2%). Alaska, Idaho, and Puerto Rico had the lowest percentage of Medicare beneficiaries undergoing nasal endoscopy at 0.1% (Figure 2). The mean reimbursement per nasal endoscopy was highest in New York ($189.53), followed by Washington, DC ($187.40), New Jersey ($184.00), and California ($181.21). Washington, DC, had the greatest concentration of health care providers performing nasal endoscopy (4654 Medicare beneficiaries per provider), followed by the US Virgin Islands (4823 Medicare beneficiaries per provider) and New York (6013 Medicare beneficiaries per provider).

A correlation (Pearson r = 0.60; 95% CI, 0.39-0.75) was noted between the percentage of Medicare beneficiaries undergoing diagnostic nasal endoscopy and the mean nonfacility Medicare reimbursement for nasal endoscopy within each state (Figure 3A). A correlation (Pearson r = –0.56; 95% CI, –0.70 to –0.34) was also present between the percentage of Medicare beneficiaries undergoing diagnostic nasal endoscopy and the density of potential Medicare beneficiaries per health care provider in the state (Figure 3B).

To our knowledge, this was the first study to report economic results and the distribution of diagnostic endoscopies in the United States. We found geographic disparities in the utilization and cost of nasal endoscopy nationally that, to our knowledge, have not been described in the literature. The findings suggest that a majority of nasal endoscopies continue to be performed by otolaryngologists, with very few ancillary health care providers or other physicians performing nasal endoscopies.

Overall, we observed a 4-fold growth in nasal endoscopies (9.3% annual growth rate) performed from 2000 through 2016, whereas the number of ethmoidectomies, used as a measure of traditional sinus surgeries performed, had a significantly slower annual growth rate at 2.6%. Previous studies have reported similar slow increases in other sinus procedures in the same period except for balloon sinuplasties.⁷Current Procedural Terminology codes for balloon sinuplasty were introduced in 2011, and the number of balloon sinuplasties performed in the Medicare population has increased at a greater rate annually compared with other sinus procedures.^7,8 The increase in popularity of balloon sinuplasty may indicate the increasing utilization of office-based rhinologic procedures, which avoid the costs associated with an operating room.⁹ However, the indications for and efficacy of balloon sinuplasty compared with endoscopic sinus surgery are still debated, with limited high-quality studies comparing the 2 surgical interventions.¹⁰ Also, the increase in balloon sinuplasty is seen on a scale of thousands of procedures, which is miniscule compared with the growth in the number of nasal endoscopies performed (on a scale of hundreds of thousands of procedures).

Health care professionals performing sinus procedures increased by 30.9% from 2012 to 2014, with most of those individuals performing balloon sinuplasty.⁷ As more health care professionals incorporate sinus procedures such as balloon sinuplasty into their practice, diagnostic nasal endoscopy may become more frequently used to identify candidates for balloon sinuplasty. The increase in nasal endoscopies would be greater than the increase in the number of sinus procedures performed because not every patient would be a candidate for an intervention, but a diagnostic endoscopy would be necessary to rule out or identify candidates. However, the reported regional variations in balloon sinuplasty are different from those identified in nasal endoscopy. Calixto et al⁷ reported that Kansas, Texas, and Louisiana had the greatest number of balloon sinuplasties performed per 10 000 beneficiaries. Those results are consistent with reports that a greater proportion of endoscopic sinus procedures involving balloon dilations were performed in the southern United States.⁸ These findings differ from those for the states in our study with the highest percentage of Medicare beneficiaries undergoing nasal endoscopy: New York, New Jersey, and Florida. The difference may be related to the indications for each procedure. Venkatraman et al¹¹ analyzed a 20% sample of total Medicare beneficiaries in 2006 and reported a 5-fold variation in per capita endoscopic sinus surgery rates among hospital referral regions. Higher rates of chronic rhinosinusitis and more beneficiaries were not associated with an increase in endoscopic sinus surgery within the hospital referral regions. Their findings suggest that endoscopic sinus surgery rates in most areas of the United States were associated with practice patterns and physician and patient expectations rather than disease prevalence and clinical guidelines.

The cause of the increase in nasal endoscopy utilization in the United States is unclear, and we can only speculate. Higher reimbursement for nasal endoscopy and a greater density of health care professionals to perform nasal endoscopies for Medicare beneficiaries were correlated with increased utilization of nasal endoscopy among the Medicare population in our study. At present, diagnostic nasal endoscopy is the most highly reimbursed otolaryngology procedure in the Medicare population. This financial incentive to perform nasal endoscopy during a clinic visit may contribute to an increased or possible overutilization of this procedure. Another possible explanation is that the nasal endoscope is a relatively new instrument that was introduced in the 1990s, and its adoption into otolaryngology clinics across the country has been a gradual process as older otolaryngologists not trained in nasal endoscopy during their residency retire and recent graduates of residency and fellowship programs who are more comfortable with nasal endoscopes integrate the technology into their practices. The advancements in technology for office-based nasal endoscopy have progressed markedly in the past 2 decades with the continued development of brighter light sources and high-definition camera systems that provide better visualization, improving the diagnostic capability of nasal endoscopy. In addition, the cost of nasal endoscopy systems has decreased gradually with time, which has enabled more otolaryngologists to afford multiple towers in their clinics. Therefore, the increased availability of affordable nasal endoscopy equipment in otolaryngology clinics across time may be a possible contributing factor to the increased use of this diagnostic procedure. The increase in nasal endoscopy utilization may be a reflection of a combination of health care provider familiarity and technological advancement.

We found a correlation between a higher concentration of health care providers in states and a higher percentage of Medicare beneficiaries undergoing diagnostic nasal endoscopy (Figure 3B). The areas with the highest utilization of nasal endoscopy appeared to be concentrated along the eastern United States (Figure 2), specifically in the New York area and surrounding states as well as Florida. These areas also are known to have the highest concentrations of otolaryngologists per county in the United States according to an analysis of all board-certified otolaryngologists in 2013.¹² Puerto Rico, Alaska, and Idaho had the lowest concentration of health care providers performing nasal endoscopy and consequently had the lowest percentage of Medicare beneficiaries receiving nasal endoscopy. The uneven national distribution of otolaryngologists most likely contributes to the geographic variation in the rates of nasal endoscopy.

In our study, there was also a correlation between the amount reimbursed by Medicare and the percentage of Medicare beneficiaries receiving nasal endoscopy per state. The relative range in Medicare reimbursement between states ($114.25 to $189.53 per procedure in Puerto Rico and New York, respectively) was minimal compared with the relative range in percentage of Medicare beneficiaries undergoing nasal endoscopy (0.1% to 1.7%). These geographic variations are likely multifactorial and complex. Increased reimbursement is a contributing factor, but its degree of influence is unknown. Most previous studies evaluating otolaryngology practice patterns have relied on surveys of otolaryngology society members, which are limited by possible recall and social desirability bias because participants for these types of surveys tend to agree with widely accepted opinions rather than express their own opinions.^13-15 Also, survey completion rates are often low. However, the health services literature has identified several factors that may influence a physician’s practice. These include organizational factors, such as financial incentives and patient expectations; social factors, such as how colleagues practice and what opinion leaders say; and professional factors, such as sense of competence or clinical certainty.¹⁶ A thorough discussion of all possible factors accounting for variations in nasal endoscopy services performed is outside the scope of this study, but further research is necessary to delineate the potential factors that are associated with these trends in otolaryngology practice patterns.

There are limitations to this study that should be considered when interpreting the results. The most important limitation is that the Medicare patient population is older than 65 years; thus, the results are representative of older patients more than the general population. Patients undergoing sinus surgery are usually younger than 65 years; thus, diagnostic nasal endoscopy may be more commonly used in the general population than in the Medicare population.¹⁷ Second, diagnosis codes are not available in the Medicare database. Therefore, we can only speculate on why the procedure was performed. Third, Medicare payments can vary based on modifiers such as disease severity and geography. These possible confounding variables were not reported in the database, and therefore we were unable to adjust for them.

The findings suggest that the number of diagnostic nasal endoscopies performed in the Medicare population increased substantially from 2000 through 2016 compared with other sinus procedures. Geographic variations in practice patterns and reimbursements appeared to be present across the United States. Further study is necessary to investigate the factors associated with the current trends observed in nasal endoscopy utilization.

Accepted for Publication: November 18, 2018.

Corresponding Author: Kevin Hur, MD, Caruso Department of Otolaryngology–Head and Neck Surgery, Keck School of Medicine, University of Southern California, 1540 Alcazar St, Ste 204M, Los Angeles, CA 90033 (kevin.hur@med.usc.edu).

Published Online: January 31, 2019. doi:10.1001/jamaoto.2018.4003

Author Contributions: Drs Hur and Liang had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Hur, Wrobel, Liang.

Acquisition, analysis, or interpretation of data: Hur, Ference, Liang.

Drafting of the manuscript: Hur.

Critical revision of the manuscript for important intellectual content: All authors.

Statistical analysis: Hur.

Supervision: Wrobel, Liang.

Conflict of Interest Disclosures: None reported.