Impact of Direct-to-Consumer Drug Advertising during the Super Bowl on Drug Utilization

Abstract

Background

Direct-to-consumer advertising (DTCA) of prescription drugs impacts patients’ requests for medications, and clinician prescribing. However, the impact of DTCA during the Super Bowl has not been previously described.

Objective

Evaluate the impact of prescription drug DTCA during the Super Bowl on drug utilization using 2014–2016 Medicare data.

Methods

Efinaconazole was advertised during Super Bowls XLIX (02/01/2015) and L (02/07/2016). The number of prescriptions for efinaconazole and for a comparator drug, tavaborole, were calculated in 31-day intervals from July 2014–December 2016. Interrupted time-series analysis models were created to test changes in trends of prescriptions for efinaconazole and tavaborole.

Results

Following Super Bowl XLIX, the number of prescriptions per 100,000 Medicare beneficiaries increased by 91% for efinaconazole, and 275% for tavaborole. After Super Bowl L, prescriptions increased significantly for efinaconazole (p-value<0.001), but not tavaborole (p=0.70). Interrupted time-series analyses estimated that, in the absence of DTCA during Super Bowl XLIX, prescriptions for efinaconazole would have increased by 40%, instead of the observed 91%. For tavaborole, prescriptions would have increased by 90% instead of 275%.

Conclusions

DTCA during the Super Bowl resulted in sharp increases in utilization of the drug advertised, which supports further regulation of DTCA.

INTRODUCTION

Direct-to-consumer advertising (DTCA) of drugs became popular in the US after 1997, when the Food and Drug Administration (FDA) specified manners on how pharmaceutical advertisements could meet the regulatory requirements of informing patients about drugs’ risks and benefits.¹ Since then, DTCA has rapidly expanded across multiple platforms such as television advertising, which has become the most proliferative and expensive form of prescription drug DTCA.²

DTCA of prescription drugs increases drug exposure and utilization, and remains a subject of heated debate.³ While proponents of prescription drug DTCA argue that the advertisements are educational, encourage adherence, and prevent underdiagnosis; those opposed claim that the advertisements often underemphasize risk and lead to overprescribing and overtreatment.⁴ Prior research has shown that DTCA of prescription drugs impact patients’ requests for medications, physician prescribing, and subsequently, drug utilization.^5,6,7 The impact of DTCA on drug prescribing and utilization has been observed not only for beneficial medications with appropriate indications, but also for those with no evidence of benefit for the patient, and even for clinically inappropriate medications. For instance, DTCA has been linked to higher utilization of lipid-lowering treatment and higher compliance with clinical goals for cholesterol levels.⁸ DTCA for cancer products has led to increases in beneficial prescribing, with minimal effects on inappropriate prescribing.⁹ Similarly, DTCA has also been associated with increases in the use of guideline-supported first-line treatments for benign prostatic hyperplasia. However, exposure to DTCA has been associated with an increased use non-preferred agents, which may be clinically inferior and more expensive compared to preferred therapies.^10,11

Prior research has noted a significant association between viewership and exposure to DTCA and its effects on drug utilization.¹² For example, Chang et al. used Nielsen television ratings, health-system claims, and sales data to estimate that each 100-unit increase in viewership of DTCA for rosuvastatin and atorvastatin were directly associated with a 2.2% increase in statin sales.¹³ This relationship between increased viewership and pharmacy sales was also corroborated in another study using similar methods applied to asthma medications.¹⁴ Additionally, repeated exposure to DTCA has been associated with increases in prescribing.¹⁵ Regardless of this prior research on the association between viewership and effects of DTCA, to our knowledge, no study has assessed the impact of prescription drug DTCA during the Super Bowl, one of the most watched annual television broadcasts which receives major media coverage.

To address this evidence gap, Medicare Part D claims data was used to evaluate the impact of prescription drug DTCA during the Super Bowl on drug utilization using 2014–2016 Medicare data.

METHODS

Drug Selection

Using an online repository, all drug advertisements that aired during each Super Bowl in 2009–2018 were identified.¹⁶ From 2009–2018, there were only three advertisements for specific prescription drug products broadcast during any Super Bowl, including two for efinaconazole during Super Bowls XLIX(02/01/2015) and L(02/07/2016) and one for rifaximin, which was advertised during Super Bowl XLVIII (02/02/2014). The analysis was restricted to efinaconazole because of the lack of a comparator drug for rifaximin. Tavaborole was selected as a comparator for efinaconazole because it was not advertised during the Super Bowl in either year analyzed, and it shares the same singular FDA indication (onychomycosis due to dermatophyte) and route of administration with efinaconazole. Additionally, both drugs were exclusively available as branded products during the study period, and entered the US market in the same year (2014).

Data Source and Statistical Analyses

Using 2014–2016 claims data from a 5% random sample of Medicare beneficiaries, the number of prescriptions per 100,000 Medicare beneficiaries for efinaconazole and tavaborole at intervals of 31 days between July 2014 and December 2016 were quantified.

To determine the impact of each Super Bowl advertisement on utilization, the study period was divided into 5 periods, and interrupted time-series analyses with linear regression models were constructed. Interrupted time-series analyses have been previously used to study DTCA timing and subsequent changes in drug utilization.^9,17–19 The first period was the time between the start of the study (7/15/14) and Super Bowl XLIX (1/16/15), the second corresponded to the two 31-day intervals following Super Bowl XLIX (1/17/15–3/19/15), the third corresponded to the ten 31-day intervals before Super Bowl L (3/20/15–1/23/16), the fourth to the two 31-day intervals after Super Bowl L (1/24/16–3/25/16), and the final to the remainder of 2016 (3/26/16–12/29/16). Since previous research has shown that the impact of DTCA exhibits is largest over the following month, the two 31-day periods after each advertisement were defined to best capture any potential differences in prescription rates.¹² Interrupted time-series analyses regressed the number of prescriptions against a continuous variable for interval period, four indicator variables for periods of interest, the second-order interactions between them, an indicator variable for the comparator drug (tavaborole), and the second and third-order interactions between the time variables and this indicator. Using estimates from these models, predictions for the number of prescriptions that would have been expected in the absence of advertisements were created.

RESULTS

Following Super Bowl XLIX, the number of efinaconazole prescriptions per 100,000 Medicare beneficiaries increased by 91% (Figure 1), from 3.7 in the two 31-day periods before to 7.1 in the two 31-day periods after (p-value for trend change<0.001, Figure 2). Following Super Bowl XLIX, the number of tavaborole prescriptions per 100,000 Medicare beneficiaries increased by 275% (p-value <0.05). After Super Bowl L, number of prescriptions also increased significantly for efinaconazole, from 6.5 to 7.8 (p-value <0.001), but not for tavaborole, from 1.9 to 2.2 (p=0.70).

Figure 1.

Observed Trend in Number of Prescriptions on Efinaconazole and Tavaborole per 100,000 Medicare Beneficiaries, July 2014 to December 2016.

Figure 2.

Expected Trend for the Number of Prescriptions on Efinaconazole and Tavaborole per 100,000 Medicare Beneficiaries

NOTES:

Predicted with Interrupted Time-Series Analyses from dates July 2014 to December 2016. Solid lines indicate observed trends with Super Bowl advertisements while dotted lines indicate trends that would have been expected if the slope had not changed following the Super Bowl advertisements.

Interrupted time-series analyses estimated that, in the absence of DTCA during Super Bowl XLIX, the number of prescriptions on efinaconazole would have increased by 40%, instead of the observed 91% increase (Figure 1). For tavaborole, prescriptions would have increased by 90% instead of observed 275%. In the absence of DTCA during Super Bowl L, number of prescriptions on efinaconazole would have increased by 1%, instead of 21% (Figure 2).

DISCUSSION

Using efinaconazole as a case study, prescription drug DTCA during the Super Bowl was associated with an immediate and sharp increase in utilization of the drug advertised. Additionally, there was a spillover effect on a comparator drug following the Super Bowl advertisement in the first year, but not in the following year. This latter finding may not be generalizable outside DTCA during Super Bowl given its particularly large viewership and media coverage, but is consistent with previous research displaying spillover effects of DTCA on competing brands.⁷

The results are consistent with prior research that demonstrated a significant association between prescription drug DTCA viewership and drug utilization. Nevertheless, the results in this study are an important contribution to the existing literature because the analysis specifically evaluated changes in drug utilization following DTCA during the Super Bowl, which is the most watched television broadcast in the US. With over 110 million estimated viewers, efinaconazole advertisements during Super Bowls XLIX and L potentially reached more than one third of the US population.²⁰ Advertising during the Super Bowl is unlike any other form of advertising not only because of the near national coverage that it achieves, but also because of the attention that Super Bowl advertisements receive due to their originality.²¹ The preferred therapy for treatment of onychomycosis due to dermatophyte is not well described²², and the increased utilization of efinaconazole seen cannot be asserted to represent beneficial increases in prescribing. Because of this, these findings support calls for DTCA to better communicate expected benefit information that can be used empower consumers and physicians in making informed treatment decisions.^4,23

The analysis is subject to several limitations: First, it did not evaluate whether the increased use of efinaconazole and tavaborole represented increased patient awareness for a potentially underdiagnosed problem or conversely inappropriate use of these drugs. Second, due to the difficulty in capturing DTCA for efinaconazole and tavaborole over time and across different media, the analyses did not account for DTCA outside of the Super Bowl. This includes other forms of DTCA such as marketing to physicians, and other television advertisements for efinaconazole during this time period. It is possible that additional DTCA for these drugs also influenced prescribing rates. Nevertheless, the significant increase in efinaconazole utilization following Super Bowl L evidences the large impact of DTCA during this broadcast, since the trend in utilization was relatively stable before then. Third, both efinaconazole and tavaborole were approved by the FDA in 2014, so this analysis is also affected by trends of uptake of these new drugs.²⁴ This may actually be the reason why there was a significant increase in the utilization of tavaborole after the first Super Bowl DTCA, but not after the second. Fourth, one could argue that seasonality in the utilization of these drugs could affect the results; however, seasonality for onychomycosis diagnosis has been found to be minimal.²⁵ Fifth, potential changes in the formulary placement of efinaconazole and tavaborole across the study period would have confounded the results. Also, the analysis focused only on television DTCA and are not generalizable to other forms of DTCA. Finally, this analysis is unable to determine whether or not the increases seen in prescribing represent overprescribing or a beneficial increase in awareness and treatment, which is a major consideration in discussions regarding the impact of prescription drug DTCA.

CONCLUSIONS

Using efinaconazole as a case study, prescription drug DTCA during the Super Bowl resulted in sharp increases in utilization of the drug advertised. Additionally, there was an observed spillover effect on a comparator drug during the first year. These results add to the ongoing debate surrounding prescription drug DTCA, and support the proposed regulations improving the relaying of expected benefit information in DTCA for prescription drugs.