Version of Record: https://www.sciencedirect.

com/science/article/pii/S2213219823005688
Manuscript_82d0ee5c3e16d999a81b123850ba4129

1 Pharmaceutical industry-sponsored meals and prescriptions of

22

© 2023 published by Elsevier. This manuscript is made available under the Elsevier user license
https://www.elsevier.com/open-access/userlicense/1.0/
23 Clinical Implications
24 Using the Medicare Part D and Open Payments databases, this study found there were
25 positive associations between meal payments to physicians by the healthcare industry
26 and increased prescriptions and healthcare costs for novel asthma biologics.
27
28 Key Words:
29 Asthma; industry payments; Open Payments Database; Sunshine Act; omalizumab;
30 mepolizumab; benralizumab; dupilumab; United States

31

2
32 Main body of the manuscript
33 Asthma is a long-term disease characterized by airway obstruction, inflammation, and
34 hyperresponsiveness resulting in cough, wheeze, breathlessness, and chest tightness. In
35 2020, asthma affected more than 25 million individuals in the United States (US),
36 constituting 7.8% of the population. Among those with asthma, an estimated 5-10%
37 have severe asthma and half of these patients remain uncontrolled.
38
39 Since 2015, five biologic drugs have been approved for the treatment of moderate to
40 severe asthma. The emergence of these new asthma biologics has brought several
41 treatment options for physicians and patients with moderate-severe asthma,1 as well as
42 extensive promotion to physicians by the pharmaceutical companies. The previous
43 study reported that the pharmaceutical companies spent more than $27 million to
44 allergists and clinical immunologists in marketing payments for these novel asthma
45 biologics.2 Numerous studies have shown that industry payments can influence
46 physician prescribing patterns, resulting in both benefits and harms for patients such as
47 increased opioid prescriptions3 and mortality due to opioid overdose4 and promoting
48 faster introduction of novel drugs with valid safety and efficacy5.
49
50 Given the competitive landscape of the asthma biologics market and the previous
51 findings, physician prescription patterns for the asthma biologics could be associated
52 with the industry payments. However, the extent of the influence of these industry
53 payments on physician clinical practice in the field of asthma remains largely unknown.
54
55 This cross-sectional analysis aimed to investigate the association between the industry
56 payments to physicians and the physician prescription patterns of the asthma biologics
57 in the US. The study linked four publicly accessible databases: the Centers for
58 Medicare & Medicaid Services (CMS) Medicare Part D, the Open Payments, the
59 National Plan and Provider Enumeration System (NPPES), and the Physician Compare
60 databases between 2017 and 2019. Of the five biologics approved for severe asthma by
61 the US Food and Drug Administration in 2020, the study considered four biologics
62 (omalizumab, mepolizumab, benralizumab, and dupilumab) that were prescribed by
63 more than 20 physicians between 2017 and 2019 in the Medicare Part D database. The
64 study sample was limited to physicians who prescribed at least one of the four
65 biologics between 2017 and 2019, and specialized in allergy and immunology,
66 pulmonology, pediatrics based on the NPPES database, because these specialists are
67 primarily responsible for prescribing the biologics. General payments for food and

3
68 beverages to the physicians were extracted from the Open Payments between 2017 and
69 2019, as food and beverage payments are significantly associated with the physician
70 prescribing patterns in other specialties.6
71
72 The payments and prescription data were descriptively analyzed. The associations
73 between meal payments for each drug and the prescriptions of that drug were evaluated
74 using a multivariable population-averaged logistic generalized estimating equation
75 (GEE) model with robust standard errors, adjusting the covariates including gender,
76 practice region, years in practice, the graduated medical schools, specialty, the annual
77 number of beneficiaries of each biologic (1 to 10, 11 to 20, and 21 or more), and
78 payment year. Additionally, associations between the number of meal payments and
79 total number of claims (30-day standardized, including refills) and total costs were
80 assessed using multivariable linear GEE models with the same covariables. As
81 dupilumab was approved for asthma in October 2018, the association between meal
82 payments and physician prescription patterns for dupilumab was analyzed between
83 2018 and 2019. No institutional board review and approval were required for this study
84 as all data used met the definition of freely available public data. Thus, the protocol of
85 this study was not registered or externally reviewed before the study initiation.
86
87 A total of 2024 physicians, including 1097 allergist-immunologists, 887
88 pulmonologists and 41 pediatricians, had the physician-level information in the
89 databases and reported more than 10 claims of at least one of the four biologics
90 between 2017 and 2019. Of them, 478 (23.6%) physicians prescribed more than one
91 biologic. The total amounts of claims were 62,047 for omalizumab, 12,974 for
92 mepolizumab, 1,927 for benralizumab, and 15,417 for dupilumab (2018-2019) over the
93 three years. Of 2024 physicians, 1716 (84.8%) received a total of $1,443,527 in food
94 and beverage payments related to the four biologics (Table 1). Median annual meal
95 payments per physician were $47.1 for omalizumab, $83.8 for mepolizumab, $110.7
96 for benralizumab, and $79.9 for dupilumab.
97
98 The physicians receiving the meal payments were significantly more likely to prescribe
99 omalizumab (odds ratio [OR]: 1.13 [95% CI: 1.01–1.27], p=0.03), benralizumab (OR:
100 1.96 [95% CI: 1.28–2.99], p=0.002), and dupilumab (OR: 2.40 [95% CI: 1.96–2.94],
101 p<0.001). However, there was no significant association between the receipt of meal
102 payments and the prescription of mepolizumab (OR: 1.17 [0.98–1.40], p=0.09).
103 Additionally, each increase in annual meal payments, the annual number of claims

4
104 increased by 0.29 (95% CI: 0.16–0.42, p<0.001) for omalizumab, 0.06 (95% CI: 0.02–
105 0.10, p=0.003) for mepolizumab, 0.02 (95% CI: 0.01–0.04, p=0.002) for benralizumab,
106 and 0.42 (95% CI: 0.33–0.52, p<0.001) for dupilumab (Table 2). The annual costs also
107 increased by $796.6 (95% CI: $313.4–$1,279.8, p=0.001) for omalizumab, $177.2
108 (95% CI: $46.9–$307.4, p=0.008) for mepolizumab, $81.5 (95% CI: $31.9–$131.1,
109 p=0.001) for benralizumab, and $1,302.6 (95% CI: $1,007.3–$1,597.7, p<0.001) for
110 dupilumab. Average pre-payment value was from $16.2 in benralizumab to $28.5 in
111 dupilumab.
112
113 Although this study demonstrated positive associations between the food and
114 beverages offered by pharmaceutical companies to physicians and the physicians’
115 prescription patterns for novel asthma biologics in the US, these findings do not
116 necessarily imply inappropriate prescriptions of the biologics or potential harm to
117 patients caused by the industry payments. Rather, given that only a limited number of
118 patients with severe asthma received biologic treatment in the US7, these findings
119 might suggest that industry-sponsored events where food and beverages were provided
120 to the physicians contributed to the rapid introduction of the novel, effective asthma
121 biologics for patients with severe asthma.
122
123 All the four biologics approved for severe asthma treatment have demonstrated
124 significant improvements in asthma control and patients' quality of life.8 Meanwhile,
125 there is no consensus so far on which biologic is the best to use in patients with severe
126 asthma. There is evidence of increased drug-related adverse events compared to
127 standard-of-care, and long-term safety profiles are not available for the majority of
128 biologics.8 The finding that payment for dupilumab was more associated with higher
129 prescriptions and costs than other biologics can be explained by the fact that
130 dupilumab was the only biologic approved for home-administration at initial approval.
131
132 However, it is important to note that financial conflicts of interest between physicians
133 and healthcare industry have sometimes led to inappropriate clinical practice and
134 negative consequences.6 At least, treatment choice should depend on patients'
135 preferences, and all physicians should pay attention to their financial relationships with
136 the healthcare industry and the consequent influence on patient care, as they can
137 influence physicians' clinical practice both positively and negatively.
138
139 This study has several limitations. First, this observational study does show

5
140 associations, and necessarily not causality between industry payments and physicians'
141 prescription patterns. Second, there are possibilities of inaccuracies in the CMS
142 databases.9 Third, this study only included prescription data from the CMS Medicare
143 Part D database, and prescription amounts would be underestimated, as the Part D
144 database contains only insurance claims for prescriptions that were filled more than 10.
145 Fourth, these findings might not be generalized to non-Medicare beneficiaries. Fifth,
146 there might be unmeasured confounding factors that could influence the associations,
147 such as the number of severe asthma patients managed by the physicians. However,
148 this information is not available from publicly accessible databases. Additionally, the
149 Open Payments Database does not provide information about physician attendance at
150 educational events where industry-sponsored meals were served, and the attendance at
151 such events could potentially impact physician prescribing behavior, independent of
152 the meals themselves. Finally, the protocol for this observational study was not
153 registered or externally reviewed prior to its initiation. Thus, there is a possibility of
154 unintentional bias in the study's hypotheses, design, setting, and interpretation.
155
156
157 Acknowledgments
158 I would like to thank Ms Megumi Aizawa for her dedicated support of my research.
159
160 References
161 1. Pepper AN, Hanania NA, Humbert M, Casale TB. How to Assess Effectiveness
162 of Biologics for Asthma and What Steps to Take When There Is Not Benefit.
163 The Journal of Allergy and Clinical Immunology: In Practice 2021; 9:1081-8.
164 2. Murayama A, Kamamoto S, Saito H, Tanimoto T, Ozaki A. Industry payments
165 to allergists and clinical immunologists in the United States during the
166 coronavirus disease 2019 pandemic. Ann Allergy Asthma Immunol 2022;
167 129:635-6.
168 3. Hadland SE, Cerdá M, Li Y, Krieger MS, Marshall BDL. Association of
169 Pharmaceutical Industry Marketing of Opioid Products to Physicians With
170 Subsequent Opioid Prescribing. JAMA Internal Medicine 2018; 178:861-3.
171 4. Hadland SE, Rivera-Aguirre A, Marshall BDL, Cerdá M. Association of
172 Pharmaceutical Industry Marketing of Opioid Products With Mortality From
173 Opioid-Related Overdoses. JAMA Network Open 2019; 2:e186007-e.
174 5. Inoue K, Figueroa JF, DeJong C, Tsugawa Y, Orav EJ, Shen C, et al.
175 Association Between Industry Marketing Payments and Prescriptions for

6
176 PCSK9 (Proprotein Convertase Subtilisin/Kexin Type 9) Inhibitors in the
177 United States. Circulation: Cardiovascular Quality and Outcomes 2021;
178 14:e007521.
179 6. Mitchell AP, Trivedi NU, Gennarelli RL, Chimonas S, Tabatabai SM, Goldberg
180 J, et al. Are Financial Payments From the Pharmaceutical Industry Associated
181 With Physician Prescribing? : A Systematic Review. Ann Intern Med 2021;
182 174:353-61.
183 7. Inselman JW, Jeffery MM, Maddux JT, Shah ND, Rank MA. Trends and
184 Disparities in Asthma Biologic Use in the United States. The Journal of Allergy
185 and Clinical Immunology: In Practice 2020; 8:549-54.e1.
186 8. Agache I, Beltran J, Akdis C, Akdis M, Canelo-Aybar C, Canonica GW, et al.
187 Efficacy and safety of treatment with biologicals (benralizumab, dupilumab,
188 mepolizumab, omalizumab and reslizumab) for severe eosinophilic asthma. A
189 systematic review for the EAACI Guidelines - recommendations on the use of
190 biologicals in severe asthma. Allergy 2020; 75:1023-42.
191 9. Murayama A, Nakano K, Kamamoto S, Sato M, Saito H, Tanimoto T, et al.
192 Trend in industry payments to infectious disease physicians in the United
193 States: a seven-year analysis of nonresearch payments from the Open Payments
194 Database between 2014 and 2020. Clin Microbiol Infect 2022; 28:1655 e1- e4.
195
196

7
197 Table 1. Summary of payments for food and beverages and prescription for omalizumab, mepolizumab, benralizumab, and dupilumab
198 between 2017 and 2019
199
Variables Omalizumab Mepolizumab Benralizumab Dupilumab Overall
Industry payments
Number of physicians 1270 (62.7) 1501 (74.1) 1307 (64.5) 1403 (69.3) 1716 (84.7)
receiving payments, n (%)
Total payment amounts
Monetary amounts, $ 254,603 416,172 441,950 330,801 1,443,527
Number of payments, n 12,297 22,839 27,348 11,608 74,092
Median annual payments 47.1 (20.6–110.7) 83.8 (38.8–145.0) 110.7 (49.3–196.1) 79.9 (27.9–191.9) 185.5 (78.2–399.2)
(IQR), $
Prescription amounts
Number of physicians 1400 (69.2) 482 (23.8) 102 (5.0) 621 (30.7) 2024
prescribing each drug, n
(%)
Total claims, n 62,047 12,974 1,927 15,417 92,365
Total costs, $ 207,207,652 40,532,409 9,554,404 48,756,346 306,050,811
200 IQR: interquartile range
201

8
202 Table 2. Associations between one increase in annual meal payments and change in
203 annual number of claims and costs for omalizumab, mepolizumab, benralizumab, and
204 dupilumab
205
One increase in Change in annual number Change in annual costs (95% CI), $
meal payments of claims (95% CI), n
Omalizumab 0.29 (0.16 – 0.42)*** 796.6 (313.4 – 1,279.8)**
Mepolizumab 0.06 (0.02 – 0.10)** 177.2 (46.9 – 307.4)**
Benralizumab 0.02 (0.01 – 0.04)** 81.5 (31.9 – 131.1)**
Dupilumab 0.42 (0.33 – 0.52)*** 1,302.6 (1,007.4 – 1,597.7)***
206 95% CI: 95% confidence interval; *p<0.05, **p<0.01, ***p<0.001
207 Legend: The estimations were adjusted for the physicians’ gender, practice region,
208 years in practice, graduated medical schools, specialty, the annual number of
209 beneficiaries of each biologic (1 to 10, 11 to 20, and 21 or more), and payment year.