Abstract
Objective
To examine neurotologists' 2013 to 2016 Medicare Part‐D data and evaluate commonly prescribed medications, longitudinal changes in prescribing patterns, presumed associated pathologies, and cost distribution across United States.
Methods
Comprehensive prescription data of Part‐D‐participating neurotologists was quiered from the 2013 to 2016 Medicare Part‐D database. Outcome variables consisted of the 25 most commonly prescribed + refilled medications, cost distribution per medication, presumed associated pathologies, and standardized prescription cost across United States.
Results
Of the 594 available U.S. neurotologists, 336 (57%) were found in the Medicare Part‐D database. In 2016, total prescription costs were $4 483 268 with an averaged $13 343 ± $18 698 per neurotologist. The three most frequently filled drugs were fluticasone propionate, ciprofloxacin, and triamterene‐hydrochlorothiazide. From 2013 to 2016, the greatest change in prescription pattern was observed with azelastine (+188%), montelukast sodium (+104%), mupirocin (+63%), and mometasone (−91%), whereas the greatest change in relative drug cost distribution was seen in ofloxacin, (+695.7%) neomycin‐polymyxin‐hydrocortisone (+262.1%), and mometasone (−83%). Triamterene‐hydrochlorothiazide, prednisone, montelukast, amoxicillin‐clavulanate, azelastine, spironolactone, and mupirocin had statistically significant increases in average number of prescriptions per physician, whereas ofloxacin and mometasone had significant decreases. Medications presumably treating Eustachian tube dysfunction, Meniere's disease, and vestibular migraine had the greatest percent changes across years. Cost distribution of four drugs increased upwards of 100%. Geographic analysis demonstrated that Southern and Midwest regions had higher standardized prescription costs.
Conclusions
This study is the first to analyze neurotologists' trends in prescribing patterns, regional prescription cost distributions, and commonly treated pathologies. This can lead to better standardization of prescribing patterns and cost in the future.
Keywords: drug benefit program, drug cost, Medicare, neurotology, otolaryngology, Part D, pharmaceutical, prescription
1. INTRODUCTION
Understanding the prevailing trends in prescribing patterns and the cost of medications over time may help to identify interventions that can improve the delivery of cost‐efficient health care. Until recently, data on these variables were limited to pharmaceutical companies that purchased them from pharmacies or pharmacy benefits administrators. Recently, Center for Medicare and Medicaid services (CMS) has released large amounts of data on the Medicare prescription drug benefit program, also known as Medicare Part‐D. 1 This has facilitated efforts in promoting transparency in medical payments and transactions. 2 Insights into Medicare Part‐D prescription data, which concern individuals 65 or older or receiving social security disability insurance, has led to important findings in the field of otolaryngology. Previous studies have utilized Part‐D data to evaluate patterns of opioid prescription, 3 , 4 reflux medications, 5 and the association between industry payments and brand‐name prescriptions. 6 Though previous studies have examined specific classes of prescription drugs, to date, no study has analyzed a large cohort of subspecialists to evaluate the longitudinal trends in that field's overall prescribed medications and treated pathologies over the years.
Neurotologists medically treat a variety of disorders including vertigo, Meniere's disease (MD), vestibular migraine (VM), chronic otitis, hearing loss, and tinnitus among others. To date, no study has analyzed Medicare Part‐D data to evaluate the prescribing patterns of neurotologists. This cannot only show prescription trends of certain medications and their corresponding cost distributions, but it can also demonstrate possible etiologies most commonly being treated. Furthermore, such nationwide data can be analyzed regionally to determine the neurotologist‐prescribed Part‐D cost distributions across states, many of which may suffer from large hospital referral regions without neurotologists. 7
2. METHODS
This study did not require Institutional Review Board approval because of the use of publicly available data and lack of direct patient involvement or identifying information. We obtained a list of U.S. registered neurotologists by using the CMS 1 and National Plan and Provider Enumeration System (NPPES) database 8 of providers and querying for the term “neurotology” in their taxonomy descriptions. Only physicians that had “otology” or “neurotology” keywords under their National Provider Number (NPI) taxonomy specialty were considered for subsequent analyses. Year‐specific lists of prescription drugs prescribed by individual physicians and paid for under the Medicare Part‐D program were obtained from Data.CMS.gov Part‐D Prescriber Public Use Files (PUFs) for years 2013 to 2016. By using the list of neurotologist NPIs obtained from the CMS NPPES Registry, we filtered these PUFs to include only prescription data from neurotologists. Using an in‐house C# application, we organized the filtered PUFs into the following neurotologist‐specific variables: number of new prescriptions and refills per drug, total cost of prescriptions per drug, number of patients receiving each drug, and state of practice. Brand and generic names of equivalent drugs were combined under one prescription and reported as the generic name. The 25 most commonly prescribed and refilled drugs were obtained based on the 2013 data, and the same list of drugs were used for data extraction in the 2014 to 2016 years. For each physician, the annual total number of prescriptions with refills of a specific drug had to be more than 10 in order for it to be reported by the CMS database, and anything below 10 were reported as zero. The presumed pathologies associated with each drug were extrapolated from the standard of practice of the senior author.
Annual distribution of each drug's cost (in percentage) was calculated via dividing that prescription's cumulative cost by the total cost of all neurotologist‐prescribed Part‐D drugs. In creating a U.S. heat map of neurotologist‐prescribed Part‐D drug cost distributions across the states, we utilized the 2016 state aggregated data for neurotologists and standardized each state's total monetary value by dividing it by the state's total number of Medicare Part‐D Eligible residents. In comparing each drug's total number of prescriptions between 2013 and 2016, two‐tailed unpaired t‐test analyses were used with a threshold of P < .05 for significance. The PASW Statistics 18.0 software (SPSS Inc., Chicago, Illinois) was used for statistical analysis.
3. RESULTS
Of the 594 neurotologists obtained by the CMS NPPES Registry, 336 (56.6%) were identified in Medicare's Part‐D Prescriber PUFs by their NPIs. A total of 114 768 Part‐D prescriptions were filled in 2016, 17.1% greater than the number of prescription fulfillments in 2013, whereas the number of neurotologists only increased by 3% in this time period (328‐337). The total price of prescriptions by all neurotologists and average price by each physician in 2016 were $4 483 268 and $13 343 ± $18 698 (median $7401; range $280‐$142 578), respectively. These prescriptions consisted of over 200 distinct drugs. Analysis of the top 25 Part‐D prescribed and refilled drugs during 2013 to 2016 years, along with their presumed indications, are demonstrated in Table 1. Fluticasone propionate, ciprofloxacin, and triamterene‐hydrochlorothiazide were found to consistently be the three most frequently filled prescriptions. The greatest relative increases in prescription fulfillment from 2013 to 2016 were by azelastine (+187.7%) and montelukast sodium (+104.0%), whereas mometasone had the greatest decrease in prescription (−90.6%) (Table 2). Evaluating changes in the average number of prescriptions by neurotologists over time revealed that triamterene‐hydrochlorothiazide, prednisone, montelukast, amoxicillin clavulanate, azelastine, spironolactone, and mupirocin had statistically significant increases, whereas ofloxacin and mometasone significantly decreased in number of prescriptions (Table 2 ).
TABLE 1.
Medications prescribed by Medicare's Part‐D‐participating neurotologists
| Drug | Annual total number of prescriptions filled with refills (% of total drugs prescribed) | % Change of number from 2013 to 2016 (% change of total drugs) | |||
|---|---|---|---|---|---|
| 2013 | 2014 | 2015 | 2016 | ||
| Fluticasone propionate | 13 130 (13.4) | 15 353 (14.5) | 15 745 (14.2) | 14 976 (13.0) | 14.1 (−3.0) |
| Ciprofloxacin and dexamethasone | 7576 (7.8) | 7293 (6.9) | 7272 (6.6) | 8260 (7.2) | 9.0 (−7.7) |
| Triamterene and Hydrochlorothiazide | 5786 (5.9) | 6718 (6.3) | 7222 (6.5) | 7415 (6.5) | 28.2 (10.2) |
| Omeprazole | 5601 (5.7) | 5597 (5.3) | 5404 (4.9) | 4857 (4.2) | −13.3 (−26.3) |
| Diazepam | 4057 (4.1) | 4729 (4.5) | 4916 (4.4) | 4549 (4.0) | 12.1 (−2.4) |
| Hydrocodone and acetaminophen | 3649 (3.7) | 3553 (3.3) | 3163 (2.8) | 3332 (2.9) | −8.7 (−21.6) |
| Ofloxacin | 3310 (3.4) | 3801 (3.6) | 4544 (4.1) | 3262 (2.8) | −1.5 (−17.6) |
| Mometasone | 2938 (3.0) | 2005 (1.9) | 1056 (1.0) | 277 (0.2) | −90.6 (−93.3) |
| Prednisone | 2289 (2.3) | 2552 (2.4) | 3045 (2.7) | 3465 (3.0) | 51.4 (30.4) |
| Montelukast sodium | 2215 (2.3) | 2915 (2.7) | 3732 (3.4) | 4519 (3.9) | 104.0 (69.6) |
| Ipratropium bromide | 2055 (2.1) | 2335 (2.2) | 2604 (2.3) | 2863 (2.5) | 39.3 (19.0) |
| Clonazepam | 2046 (2.1) | 2394 (2.3) | 2394 (2.2) | 2216 (1.9) | 8.3 (−9.5) |
| Meclizine hydrochloride | 1885 (1.9) | 2094 (2.0) | 2083 (1.9) | 2062 (1.8) | 9.4 (−5.3) |
| Fluocinolone acetonide oil | 1540 (1.6) | 1472 (1.4) | 1295 (1.2) | 1299 (1.1) | −15.6 (−31.3) |
| Amoxicillin and clavulanate potassium | 1489 (1.5) | 1711 (1.6) | 1743 (1.6) | 2051 (1.8) | 37.7 (20.0) |
| Hydrochlorothiazide | 1484 (1.5) | 1666 (1.6) | 2049 (1.8) | 1999 (1.7) | 34.7 (13.3) |
| Methylprednisolone | 1400 (1.4) | 1446 (1.4) | 1684 (1.5) | 1854 (1.6) | 32.4 (14.3) |
| Azelastine | 1392 (1.4) | 2112 (2.0) | 2677 (2.4) | 4005 (3.5) | 187.7 (150.0) |
| Cephalexin | 1256 (1.3) | 1290 (1.2) | 1466 (1.3) | 1375 (1.2) | 9.5 (−7.7) |
| Spironolactone | 1213 (1.2) | 1288 (1.2) | 1484 (1.3) | 1715 (1.5) | 41.4 (25.0) |
| Neomycin and polymyxin and hydrocortisone otic | 1115 (1.1) | 1081 (1.0) | 1288 (1.2) | 1680 (1.5) | 50.7 (36.4) |
| Acyclovir | 1077 (1.1) | 1123 (1.1) | 954 (0.9) | 1012 (0.9) | −6.0 (−18.2) |
| Mupirocin | 1055 (1.1) | 1395 (1.3) | 1540 (1.4) | 1723 (1.5) | 63.3 (36.4) |
| Nortriptyline | 983 (1.0) | 1156 (1.1) | 1275 (1.1) | 1407 (1.2) | 43.1 (20.0) |
Note: Percentages in a column do not add up to 100% because this table contains only the top 25 medications.
Abbreviations: ETD, Eustachian tube dysfunction; FN, facial nerve paralysis; LPR, laryngopharyngeal reflux; MD, Meniere's disease; SHL, sudden hearing loss. STI, soft tissue infection; VM, vestibular migraine.
TABLE 2.
Longitudinal trends in prescribed drugs by Part‐D‐participating neurotologists
| Drug | Presumed pathology | Means and SD per physicians with at least 10 prescriptions a (2013 vs 2016) | P‐value |
|---|---|---|---|
| Fluticasone propionate | ETD/Allergy | 56.6 ± 75.3 vs 64.2 ± 85.4 | .319 |
| Ciprofloxacin and dexamethasone | Otitis | 38.1 ± 43.3 vs 44.5 ± 49.1 | .139 |
| Triamterene and Hydrochlorothiazide | MD | 33.1 ± 38.0 vs 42.4 ± 47.5 | .047 |
| Omeprazole | LPR | 61.1 ± 91.6 vs 51.2 ± 70.3 | .420 |
| Diazepam | Vertigo/Tinnitus | 33.8 ± 66.5 vs 39.9 ± 87.1 | .552 |
| Hydrocodone and acetaminophen | Pain | 23.0 ± 19.0 vs 20.3 ± 18.5 | .220 |
| Ofloxacin | Otitis | 18.6 ± 18.9 vs 22.8 ± 19.6 | .027 |
| Mometasone | ETD/Allergy | 33.6 ± 38.5 vs 3.2 ± 8.0 | <.001 |
| Prednisone | Multiple a | 17.2 ± 25.1 vs 26.4 ± 39.6 | .027 |
| Montelukast sodium | ETD/Allergy | 28.3 ± 47.4 vs 58.7 ± 86.1 | .007 |
| Ipratropium bromide | ETD/Allergy | 29.6 ± 44.6 vs 41.1 ± 50.2 | .157 |
| Clonazepam | Vertigo/Tinnitus | 52.1 ± 122.4 vs 56.0 ± 104.4 | .879 |
| Meclizine hydrochloride | Vertigo | 23.4 ± 33.2 vs 25.8 ± 25.1 | .613 |
| Fluocinolone acetonide oil | Itching of ear | 20.8 ± 21.7 vs 17.0 ± 20.0 | .269 |
| Amoxicillin and clavulanate potassium | Otitis/sinusitis/STI | 14.4 ± 15.4 vs 20.2 ± 15.3 | .009 |
| Hydrochlorothiazide | MD | 31.9 ± 47.0 vs 42.3 ± 66.1 | .386 |
| Methylprednisolone | Multiple a | 15.8 ± 19.8 vs 21.35 ± 23.6 | .103 |
| Azelastine | ETD | 16.8 ± 27.4 vs 47.7 ± 61.9 | <.001 |
| Cephalexin | STI | 17.4 ± 12.2 vs 18.5 ± 15.6 | .645 |
| Spironolactone | MD | 18.5 ± 20.7 vs 54.6 ± 112.5 | .022 |
| Neomycin and polymyxin and hydrocortisone otic | Otitis | 18.5 ± 20.7 vs 29.4 ± 24.3 | .012 |
| Acyclovir | FNP | 99.3 ± 181.5 vs 89.1 ± 145.3 | .886 |
| Mupirocin | STI | 16.3 ± 20.2 vs 26.9 ± 27.8 | .014 |
| Nortriptyline | VM | 28.1 ± 30.4 vs 38.4 ± 45.3 | .276 |
Note: Physicians with <10 prescriptions and refills for the respective medications were not included for this analysis. Percentages in a column do not add up to 100% because this table contains only the top 25 medications.
Multiple: oral steroids could be used for treating a variety of conditions such as SHL, vertigo, FNP, ETD, and so forth.
Bold values represent statistical significance (p < 0.05).
A compilation of the number of Part‐D prescription‐fulfilled drugs based on the presumed treated pathologies demonstrated that the prescribed medications most commonly addressed Eustachian tube dysfunction (ETD), allergy, otitis, and MD (Table 3). Oral steroids were also in the top five most commonly prescribed medications. Between 2013 and 2016, there was a statistically significant increase in physician‐averaged prescriptions for ototopical drops. Additionally, medications potentially addressing MD and VM had noticeable overall percent increases of 43% and 31%, respectively. The 2013 to 2016 longitudinal trends in Part‐D prescription‐treated pathologies is depicted in Figure 1.
TABLE 3.
Longitudinal trends in presumed pathologies prescribed by Medicare Part‐D‐participating neurotologists
| Pathology | Annual total number of medications filled with refills | % Change from 2013 to 2016 | Means and SD per physician a (2013 vs 16) | P‐value | |||
|---|---|---|---|---|---|---|---|
| 2013 | 2014 | 2015 | 2016 | ||||
| ETD | 21 730 | 24 720 | 25 814 | 26 640 | 22.6 | 70.9 ± 120.0 vs 86.8 ± 153.4 | .157 |
| Allergy | 20 338 | 22 608 | 23 137 | 22 635 | 11.3 | 66.3 ± 110.0 vs 73.7 ± 128.2 | .447 |
| Otitis | 13 490 | 13 886 | 14 847 | 15 253 | 13.1 | 50.0 ± 58.1 vs 61.6 ± 68.8 | .025 |
| MD | 8483 | 9672 | 10 755 | 11 129 | 31.2 | 27.6 ± 63.2 vs 34.2 ± 78.4 | .253 |
| Vertigo | 7988 | 9217 | 9393 | 8827 | 10.5 | 25.4 ± 78.4 vs 28.8 ± 82.5 | .601 |
| LPR | 5601 | 5597 | 5404 | 4857 | −13.3 | 18.2 ± 57.1 vs 15.3 ± 44.9 | .483 |
| Pain | 3649 | 3553 | 3163 | 3332 | −8.7 | 11.7 ± 17.8 vs 10.4 ± 16.7 | .363 |
| FNP | 1077 | 1123 | 954 | 1012 | −6.0 | 3.6 ± 38.0 vs 3.2 ± 31.3 | .896 |
| VM | 983 | 1156 | 1275 | 1407 | 43.1 | 3.2 ± 13.4 vs 4.3 ± 19.3 | .392 |
Abbreviations: ETD, Eustachian tube dysfunction; FNP, facial nerve paralysis; LPR, laryngopharyngeal reflux; MD: Meniere's disease; VM, vestibular migraine.
All physicians included even those who prescribed <10 (recorded as 0) of the respective medications.
FIGURE 1.
Graph of trends in treated pathologies by neurotologist‐prescribed for Part‐D patients. Numeric values are calculated via adding all drugs for pathology as shown in Table 1
Comparing 2013 and 2016 drug cost distributions showed that ofloxacin (+695.7%) and neomycin‐polymyxin‐hydrocortisone (+262.1%), both primarily used to manage otitis, had the greatest relative increases in percent cost per total cost of all drugs (Table 4 ). On the other hand, the greatest relative decreases in drug cost distributions were found among mometasone (−83.1%), fluticasone propionate (−50.6%), and omeprazole (−47.9%), which are primarily used to treat allergic rhinitis or ETD, and laryngopharyngeal reflux, respectively. Figure 2 demonstrates a U.S. heat map of neurotologist‐prescribed Part‐D cost distributions, where Florida ($2.14), Missouri ($1.65), Texas ($1.60), Kentucky ($1.55), and New York ($1.39) had the highest adjusted costs per Part‐D eligible resident.
TABLE 4.
Trends in cost distribution of drugs prescribed by Part‐D‐participating neurotologists
| Drug | Distribution of drug costs (%) over time | % Change 2013 to 2016 | Presumed Pathology | |||
|---|---|---|---|---|---|---|
| 2013 | 2014 | 2015 | 2016 | |||
| Fluticasone propionate | 8.26 | 6.07 | 4.27 | 4.08 | −50.6 | ETD/Allergy |
| Ciprofloxacin and dexamethasone | 21.32 | 20.90 | 23.10 | 24.73 | 16.0 | Otitis |
| Triamterene and Hydrochlorothiazide | 1.68 | 1.80 | 1.87 | 2.04 | 21.4 | MD |
| Omeprazole | 3.05 | 2.06 | 1.67 | 1.59 | −47.9 | LPR |
| Diazepam | 0.47 | 0.55 | 0.58 | 0.56 | 19.1 | Vertigo |
| Hydrocodone and acetaminophen | 0.75 | 0.97 | 0.81 | 0.90 | 20.0 | Pain |
| Ofloxacin | 0.37 | 0.31 | 0.35 | 2.92 | 689.2 | Otitis |
| Mometasone | 11.10 | 8.85 | 5.99 | 1.88 | −83.1 | ETD/Allergy |
| Prednisone | 0.25 | 0.39 | 0.47 | 0.50 | 100.0 | Multiple a |
| Montelukast sodium | 2.12 | 1.80 | 1.32 | 1.63 | −23.1 | ETD/Allergy |
| Ipratropium bromide | 1.14 | 1.49 | 1.60 | 1.71 | 50.0 | ETD/Allergy |
| Clonazepam | 0.47 | 0.47 | 0.48 | 0.55 | 17.0 | Vertigo/Tinnitus |
| Meclizine hydrochloride | 0.94 | 0.92 | 0.87 | 0.79 | −16.0 | Vertigo |
| Fluocinolone acetonide oil | 2.18 | 5.55 | 4.87 | 4.39 | 101.4 | Itching of Ear |
| Amoxicillin and clavulanate potassium | 0.84 | 0.70 | 0.59 | 0.69 | −17.9 | Otitis/Sinusitis/ STI |
| Hydrochlorothiazide | 0.23 | 0.23 | 0.27 | 0.27 | 17.4 | MD |
| Methylprednisolone | 0.73 | 0.65 | 0.70 | 0.65 | −11.0 | Multiple a |
| Azelastine | 2.40 | 3.08 | 3.69 | 4.45 | 85.4 | ETD |
| Cephalexin | 0.19 | 0.17 | 0.17 | 0.18 | −5.3 | STI |
| Spironolactone | 0.38 | 0.32 | 0.30 | 0.36 | −5.3 | MD |
| Neomycin and polymyxin and hydrocortisone otic | 0.58 | 0.58 | 0.82 | 1.77 | 262.1 | Otitis |
| Acyclovir | 0.97 | 1.03 | 0.92 | 0.90 | −7.2 | FNP |
| Mupirocin | 0.35 | 0.39 | 0.43 | 0.35 | 0 | STI |
| Nortriptyline | 0.26 | 0.29 | 0.32 | 0.41 | 57.7 | VM |
Abbreviations: ETD, Eustachian tube dysfunction; FN, facial nerve paralysis; LPR, laryngopharyngeal reflux; MD, Meniere's disease; SHL, sudden hearing loss. STI, soft tissue infection; VM, vestibular migraine.
Multiple: oral steroids could be used for treating a variety of conditions such as SHL, vertigo, FNP, ETD, and so forth.
FIGURE 2.
U.S. heat map of 2016 Part‐D neurotologist‐prescribed cost distributions. The cost was standardized by dividing the state's total cost by the number of Part‐D eligible residents in that state. Of the states with neurotologists data, this cost ranged between $0.10 and $2.14 per eligible resident
4. DISCUSSION
This is the first study to utilize the CMS Part‐D database to analyze the longitudinal patterns and trends in drug prescriptions, cost distributions, and medical conditions treated by an otolaryngology subspecialty. Our results demonstrate that although fluticasone, ciprofloxacin, triamterene‐hydrochlorothiazide, and omeprazole have been consistently the most commonly prescribed medications overall, there exists a significant increase in oral steroids as well as medications presumably addressing otitis between 2013 and 2016. Also, medications potentially utilized for VM, MD, and ETD had nonsignificant physician‐averaged, yet noticeable overall increases. We further demonstrated a statistically significant increase in prescribing patterns of medications such as triamterene‐hydrochlorothiazide, prednisone, montelukast, amoxicillin clavulanate, and azelastine, whereas prescribing patterns for ofloxacin and mometasone had significantly decreased over the years. It was observed that the cost distribution of many drugs, such as ofloxacin, neomycin, fluocinolone acetonide, mometasone, and fluticasone, had changed noticeably over time. This bird's eye view of the prescription practice and temporal changes among U.S. neurotologists can help otolaryngologists who prescribe ear‐ and balance‐related medications to compare their current prescribing records with these nationwide results. This has potential to identify opportunities to reduce unnecessary medical cost, or consider whether they may be under‐ or over‐treating certain disease processes compared to a cross‐sectional nationwide overview.
According to 2016 CMS reports, 39.5 million people consisting of 58% female and 17% under‐65 disabled individuals were enrolled in the Medicare Part‐D program. 9 This approximate 15% increase in number of part‐D beneficiaries since 2013 is in‐line with our demonstrated 17% increase in overall prescriptions, though many specific drugs' percent change in number and price are vastly different. According to ProPublica's reports of recent Part‐D data, 1.4 million providers accounted for the $137.4 billion total cost for all prescriptions. 10 The evaluated 337 neurotologists with a relatively modest upper range of $142 000 and a mean prescription price of $13 300, constituted around 0.02% of overall providers in the database (physicians, nurse practitioners, and physician assistants) while accounting for 0.003% of total drug costs. This is in‐line with the logical assumption that neurotologists do not prescribe many medications compared to most other medical specialties, and what they do prescribe is relatively inexpensive. There were however a few anomalies in our price analyses that encourage future discussions and speculations, such as ofloxacin's price increase by almost seven times or Ciprofloxacin and dexamethasone's substantial proportion of the overall annual medications (more than 20% consistently). Previous research on the ratio and regional variation of U.S. neurotologists has suggested a scarcity of neurotologists in certain hospital referral regions. 7 Though the presented results are not meant to reverse‐engineer disease epidemiologies, practical information regarding prescription patterns, drug costs, and potential trending prevalence of medical pathologies can be deduced. Such approach can play a role in better resource and expert allocation by evaluating potentially over‐ or under‐utilized medications, underserved regions, and the prevailing pathologies most commonly presented to neurotologists by Medicare patients.
In this study, the most common medically treated diseases by neurotologists, presumed by the classes of prescribed medications, were described and potential longitudinal trends were explored. Prior research has suggested that some of the most commonly treated neurotological diseases, such as otitis and MD, are less frequently being researched. 11 , 12 , 13 Unfortunately, epidemiological data regarding many of the mentioned conditions such as ETD, VM, MD, and vertigo have been limited. 14 , 15 , 16 , 17 , 18 We made significant assumptions regarding the presumed pathologies treated by various medications. These assumptions were made based on the most common conditions for which the medications would be prescribed. However, some medications are used for other purposes. For instance, some clinicians use acyclovir for SHL or MD despite lack of evidence, whereas we assumed most neurotologists use acyclovir for acute facial paralysis. Therefore, caution is required when looking at trends for medications that could be used for multiple conditions. Other medications such as eardrops are categorized with more certainty since they are most likely used to treat chronic ottis media or otitis externa.
After per capita standardization, our data demonstrated distinct geographic differences among prescription costs across U.S. regions. Namely, we observed a higher standardized cost allocated in the Southern (Florida, Texas, Kentucky) and Midwest (Michigan and Missouri) states. A similar trend has been reported by other CMS studies in otolaryngology or other medical specialties. Svider et al reported that otolaryngologists in the Midwest had the greatest opioid prescription, vs Northeast's otolaryngologists with the fewest opioid prescriptions. 4 An increase in opioid prescribing amongst sinus surgeons practicing in the South has also been observed. 3 These regional differences have also been reported in other classes of drugs, as greater quantities of oral antibiotics and proton pump inhibitors have been prescribed to residents in the Midwest or Southern U.S. states. 19 , 20 , 21 These data signal that the volume and cost per capita of some medications, including those prescribed by neurotologists, may be more concentrated in Central and Southern U.S. regions, warranting further investigation into the possible rationale.
Though we took great care in appropriate extraction and analysis of data, this study contains a number of important limitations. First, the data pertains to a specific population (Medicare Part‐D beneficiaries) and generalizing the results for the overall population is thus cautioned. Second, the prescription‐associated pathologies were based on the senior author's standard of care practice, some of which may be different from other neurotologists' preferences. In that case, the detailed quantitative data are provided for readers to make other subjective deductions. Third, this research did not provide physician characteristics such as academic position or number of years in practice, which may influence prescribing trends. 3 Furthermore, there has been evidence that physician prescribing patterns may be influenced by government regulations, 22 brand‐names, 23 and industry incentives, 24 , 25 which was beyond the scope of this study. Fourth, some neurotologists employ mid‐level providers or work with residents who write prescriptions under their own name and NPI number, which may not be reflected in the Medicare database as a prescription given by a neurotologist. Furthermore, we were limited by the years that Part‐D data have become public, thus future longitudinal studies with more available years of data are warranted. Lastly, it is unclear whether all the included physicians were fellowship‐trained neurotologists who exclusively treated otology & neurotology disorders, or whether it also included physicians that treated otology and neurotology disorders as part of a more comprehensive practice. Despite these limitations, we believe this study still provides valuable information for the field to consider in terms of nationwide prescription trends, practice patterns, and regional distributions over a four‐year period.
5. CONCLUSION
In this examination of U.S. neurotologists' 2013 to 2016 Medicare Part‐D prescription data, we demonstrated an increase in prescribing triamterene‐hydrochlorothiazide, prednisone, montelukast, amoxicillin clavulanate, and azelastine, and a decrease in oflaxacin and mometasone. Physicians had a statistically significant increase in averaged prescriptions of oral steroids and medications presumably treating otitis, where those presumably treating VM, MD, and ETD had noticeable overall percent increases as well without reaching statistical significance. The greatest increase in relative price was observed with ofloxacin and neomycin‐polymyxin‐hydrocortisone, whereas the greatest relative price decreases were seen in mometasone and fluticasone propionate. With an overall prescription cost of $4.5 million among Part‐D beneficiaries in 2016, a higher standardized cost allocation was observed in Southern and Midwest regions.
CONFLICT OF INTEREST
Hamid R. Djalilian holds equity in MindSet Technologies and Cactus Medical LLC, and is on the advisory board of Novus Therapeutics.
ACKNOWLEDGMENTS
Mehdi Abouzari is supported by the National Center for Research Resources and the National Center for Advancing Translational Sciences, National Institutes of Health, through Grant TL1TR001415. Arash Abiri is supported by the National Institute of General Medical Sciences of the National Institutes of Health under award number T32GM008620. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.
Goshtasbi K, Abouzari M, Abiri A, et al. Trends and patterns of neurotology drug prescriptions on a nationwide insurance database. Laryngoscope Investigative Otolaryngology. 2021;6(5):1096‐1103. 10.1002/lio2.617
Khodayar Goshtasbi, Mehdi Abouzari, and Arash Abiri contributed equally to this study.
Portions of this work have been presented at the 2019 Combined Otolaryngology Spring Meetings for the Triological Society, Austin, TX.
Funding information National Center for Advancing Translational Sciences, Grant/Award Number: TL1TR001415; National Center for Research Resources, Grant/Award Number: TL1TR001415; National Institute of General Medical Sciences of the National Institutes of Health, Grant/Award Number: T32GM008620
BIBLIOGRAPHY
- 1. Centers for Medicare & Medicaid Services. Part D Prescriber Data CY 2013‐2016 : Medicare Provider Utilization and Payment Data. https://www.cms.gov/Research-Statistics-Data-and-Systems/Statistics-Trends-and-Reports/Medicare-Provider-Charge-Data/Part-D-Prescriber.html. Accessed March 14, 2019.
- 2. Agrawal S, Brown D. The physician payments sunshine act—two years of the open payments program. New Engl J Med. 2016;374(10):906‐909. [DOI] [PubMed] [Google Scholar]
- 3. Arianpour K, Nguyen B, Yuhan B, Svider PF, Eloy JA, Folbe AJ. Opioid prescription among sinus surgeons. Am J Rhinol Allergy. 2018;32(4):323‐329. [DOI] [PubMed] [Google Scholar]
- 4. Svider PF, Arianpour K, Guo E, et al. Opioid prescribing patterns among otolaryngologists: crucial insights among the Medicare population. Laryngoscope. 2018;128(7):1576‐1581. [DOI] [PubMed] [Google Scholar]
- 5. Luetzenberg FS, Jiang N. Practice patterns of reflux medication prescriptions in otolaryngology compared to other specialties. Laryngoscope. 2020;130(2):321‐327. [DOI] [PubMed] [Google Scholar]
- 6. Morse E, Fujiwara RJ, Mehra S. Industry payments to physicians and prescriptions of brand‐name proton‐pump inhibitors. Otolaryngol Head Neck Surg. 2019;160(1):70‐76. [DOI] [PubMed] [Google Scholar]
- 7. Vrabec JT. Workforce analysis of neurotologists in the United States. Otol Neurotol. 2013;34(4):755‐761. [DOI] [PubMed] [Google Scholar]
- 8. National Plan and Provider Enumeration System (NPPES) National Provider Identifier (NPI) Registry. https://npiregistry.cms.hhs.gov. Accessed March 14, 2019.
- 9. Centers for Medicare & Medicaid Services Statistics Reference Booklet 2016. Edition. https://www.cms.gov/Research-Statistics-Data-and-Systems/Statistics-Trends-and-Reports/CMS-Statistics-Reference-Booklet/2016.html. Accessed July 20, 2019.
- 10. Jones RG, Groeger LV, Ornstein C. ProPublica Prescriber Checkup. Updated February 2019. https://projects.propublica.org/checkup/. Accessed July 20, 2019.
- 11. Muelleman T, Shew M, Subbarayan R, et al. Epidemiology of dizzy patient population in a neurotology clinic and predictors of peripheral etiology. Otol Neurotol. 2017;38(6):870‐875. [DOI] [PubMed] [Google Scholar]
- 12. Boerner R, Hatch JL, Harruff E, et al. Publishing trends in otology and Neurotology. Otol Neurotol. 2018;39(1):127‐132. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Strupp M. Challenges in neuro‐otology. Front Neurol. 2010;1:121. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Hsu LC, Wang SJ, Fuh JL. Prevalence and impact of migrainous vertigo in mid‐life women: a community‐based study. Cephalalgia. 2011;31(1):77‐83. [DOI] [PubMed] [Google Scholar]
- 15. Lempert T, Neuhauser H. Epidemiology of vertigo, migraine and vestibular migraine. J Neurol. 2009;256(3):333‐338. [DOI] [PubMed] [Google Scholar]
- 16. Vila PM, Thomas T, Liu C, Poe D, Shin JJ. The burden and epidemiology of eustachian tube dysfunction in adults. Otolaryngol Head Neck Surg. 2017;156(2):278‐284. [DOI] [PubMed] [Google Scholar]
- 17. Monasta L, Ronfani L, Marchetti F, et al. Burden of disease caused by otitis media: systematic review and global estimates. PLoS One. 2012;7(4):e36226. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Lai YT, Wang TC, Chuang LJ, Chen MH, Wang PC. Epidemiology of vertigo: a National Survey. Otolaryngol Head Neck Surg. 2011;145(1):110‐116. [DOI] [PubMed] [Google Scholar]
- 19. Nothelle SK, Sharma R, Oakes AH, Jackson M, Segal JB. Determinants of potentially inappropriate medication use in long‐term and acute care settings: a systematic review. J Am Med Dir Assoc. 2017;18(9):806.e1‐806.e17. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Bustillos H, Leer K, Kitten A, Reveles KR. A cross‐sectional study of national outpatient gastric acid suppressant prescribing in the United States between 2009 and 2015. PLoS One. 2018;13(11):e0208461. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Kakpovbia E, Feng H, Feng PW, Cohen JM. Antibiotic prescribing trends among US dermatologists in Medicare from 2013‐2016. J Dermatol Treat. 2021;32(1):70‐72. [DOI] [PubMed] [Google Scholar]
- 22. Studdert DM, Mello MM, Brennan TA. Financial conflicts of interest in physicians' relationships with the pharmaceutical industry—self‐regulation in the shadow of federal prosecution. N Engl J Med. 2004;351(18):1891‐1900. [DOI] [PubMed] [Google Scholar]
- 23. Kesselheim AS, Misono AS, Lee JL, et al. Clinical equivalence of generic and brand‐name drugs used in cardiovascular disease: a systematic review and meta‐analysis. Jama. 2008;300(21):2514‐2526. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Morse E, Fujiwara RJ, Mehra S. Increasing industry involvement in otolaryngology: insights from 3 years of the open payments database. Otolaryngol Head Neck Surg. 2018;159(3):501‐507. [DOI] [PubMed] [Google Scholar]
- 25. Campbell EG, Gruen RL, Mountford J, Miller LG, Cleary PD, Blumenthal D. A national survey of physician–industry relationships. N Engl J Med. 2007;356(17):1742‐1750. [DOI] [PubMed] [Google Scholar]