1 INTRODUCTION
When the COVID-19 pandemic emerged in late 2019, researchers across disciplines employed various techniques to estimate its impact. For example, some research estimates the expected number of infections (Jung et al., 2020), hospitalizations (Sussman, 2020), mortality (Jung et al., 2020; Sussman, 2020), and broader economic consequences (Borgonovi and Andrieu, 2020; Coibion et al., 2020a; 2020b; Cachanosky et al., 2020; Nicola et al., 2020). Hoping to implement effective policy responses to COVID-19, other literature examines the effectiveness of various voluntary and involuntary measures to prevent healthcare systems from becoming overwhelmed (Chudik et al., 2020; Gupta et al., 2020; Hornstein, 2020; Mongey et al., 2020; Piguillem and Shi, 2020).1
As the COVID-19 pandemic continued into 2020, health policy and medical literature began examining how the US healthcare sector might adjust to provide treatment for COVID-19 patients and treat patients with other conditions. This literature includes policy recommendations for technology adoption and expanded access to telemedicine (Hollander and Carr, 2020; Kandel et al., 2020; Keesara et al., 2020; Portnoy et al., 2020). However, many of the recommended policy changes face regulatory obstacles issued by the Food and Drug Administration (FDA). Consequently, much of the literature examining how to expand healthcare capacity during the pandemic notes how existing FDA regulations prevent or delay recommended policy changes (Atluri et al., 2020; Chen and Mao, 2020; March, 2020; Ravi et al., 2020).
Much of the analysis examining the FDA's impact during the COVID-19 pandemic is found in medical and public health journals. However, a considerable body of economics research finds that FDA regulations have hindered the U.S. healthcare sector's ability to treat patients long before COVID-19 (Higgs, 1995; Philipson and Sun, 2008). Previous literature specifically finds FDA regulations of approving pharmaceuticals, medical devices, and other forms of treatment prolong or prevent their adoption within the US to the detriment of patients and healthcare providers (Evans and Watson, 2015; Higgs, 1995; Peltzman, 1973; Ward, 1992).
Curiously, previous economic literature examining the FDA has not examined the agency's impact during public health crises, including pandemics. Given the impact of COVID-19 on the U.S. economy and healthcare sector, a more robust understanding of how FDA regulation impacted the United States' response to the COVID-19 pandemic is vital to developing effective policy responses to better help the healthcare sector serve patients.
This article examines this question by employing a political economy framework to examine how FDA regulations have impacted COVID-19 testing, the utilization of remdesivir, and COVID-19 vaccines. I specifically examine these components of addressing the pandemic because they were issued Emergency Use Authorization (EUA) by the FDA, allowing them to reach clinicians and patients without undergoing the agency's regulatory approval process. Comparing the healthcare sector's adoption of COVID testing and treatment before and after the EUAs provides insight into the impact of the FDA's regulations restricting access to testing and treatment. My analysis finds that the FDA's regulations enacted before the COVID-19 pandemic began strongly restricted clinician and patient access to COVID-19 testing, remdesivir, and vaccines. After the FDA issued EUAs, the healthcare sector quickly adopted COVID-19 testing and remdesivir with little evidence of negative consequences. These findings contribute to the economics literature examining the FDA and contemporary COVID-19 policy research.
This article proceeds as follows. Section 2 reviews the economic literature assessing the FDA's regulatory impact on the healthcare sector. Section 3 develops a political economy framework to examine the FDA's incentives when determining whether to approve goods. Section 4 applies this framework to examine the development of COVID-19 testing, remdesivir treatment, and vaccination before and after the FDA issued EUAs. Section 5 concludes and provides implications for future research.
2 LITERATURE REVIEW
Most economic analysis of the FDA occurs after the Kefauver-Harris Amendments of 1962.2 These legislative acts granted the FDA the authority to determine both efficacy and safety standards for drugs before entering the market. The amendments also allowed the agency to set clinical trial standards and designs (Hanson, 1995). In the first economic analysis of the FDA and the 1962 Amendments, Peltzman (1973) finds that the increased regulatory stringency resulted in an estimated 5–10% tax on drug purchases. The author also noted the high cost of compliance "engendered a marked reduction in drug innovation" (p. 1049).
Much of the literature following Petlzman's analysis concludes increased regulatory stringency into the drug market resulted in considerable increases in drug prices. Examining a sample of 93 randomly selected new chemical entities, DiMasi et al. (1991) estimate the cost of developing a new drug in 1991 was approximately $114 million ($260 million in 2020 dollars). Nearly a decade after, DiMasi et al. (2003) estimate the cost of new drug approval to be approximately $804 million in 2003 ($1.2 billion in 2020 dollars). Adams and Branter (2006) similarly estimated the cost to be between $500 and $2 billion in 2000 dollars (between $752 million and $3 billion in 2020 dollars).
FDA authority to set drug approval standards also lengthened the time required for drugs to reach patients. Grabowski et al. (1978) find the average time required to approve new drugs from 1962 to 1967 increased from an average of approximately 7–30 months. Pharmaceuticals' time to complete FDA approval continued to increase throughout the 1980s and mid-1990s. From 1980 to 1989, the average time from the synthesis of a new drug to marketing approval averaged 14.1 years, rising to 15.2 years from 1990 to 1996 (Miller, 1988). These long approval times exceeded those found in other developed nations. Over mostly the same period, Wardell (1973) notes the time required to have new drugs approved in the United States exceeded those in the United Kingdom by an average of 2 years. Kaitin et al. (1989) find a similar two-year difference between drug approval times between the United States and the United Kingdom from 1977 to 1987. From 1970 to 1993, approval times for drugs and medical devices in the United States trailed those in the United Kingdom, Germany, France, and Spain (Kaitin and Brown, 1995). Longer approval times for many drugs can be lethal. Gieringer (1985) estimates that a one-year delay in drug benefits can result in approximately 37,000–76,000 patient deaths.
Economic analysis of the FDA's drug approval process frequently finds increased costs to develop new drugs and longer times required to reach the market deters producers from introducing new drugs and engaging in pharmacological innovation. Faust (1985) writes, "since the passage of the New Drug Amendments of 1962, the influence of regulation on the drug development process and biomedical research, in general, has resulted in considerable negative fall-out highlighted by a decline in drug innovation" (p. 201). The author also finds that other advanced nations "where regulatory stringencies have not been as significant a deterrent to research productivity" did not experience a similar decrease in new drug introductions (p. 201). Case studies and empirical analyses support these claims. Grabowski et al. (2002) note that R&D for beta-blockers, anxiolytics, anti-depressants, and the first birth control drugs constituted a large component of drug producers' R&D before the 1962 amendments. Wiggins (1981) estimates the 1962 amendments reduced the introduction of new medicines by approximately 60%. More recently, Evans and Watson (2015) and Tabarrok (2017) find that the FDA's approval process has delayed genomic medicine advancement within the United States.3 Economic literature examining the FDA's medical device regulation through legislative acts in 1976, 1988, and 1990 (Munsey, 1995) reaches similar conclusions. From 1983 to 1990, the FDA received about 80 premarket approvals a year but only approved approximately 45 (Higgs, 1995).4 By 1992, the FDA had only approved 12 (Higgs, 1995). Miller (1988) similarly finds that the FDA's oversight of medical devices restricts new product development and innovation.
Consistent findings that the FDA regulatory barriers delay drug development, increase the cost of development, and prolong approval times have resulted in economists' widespread agreement that they are on net harmful to patients (Klein, 2000). With few exceptions, the FDA's regulatory scope continued to expand into the early 2000s (FDA, 2018a). Philipson et al. (2008) estimate the FDA's regulatory scope encompasses approximately 20% of all consumer spending in the United States. However, considerably less economic research has examined the FDA's impact on and its regulatory impact on the U.S. healthcare industry since the early 2000s (Philipson and Sun, 2008). What research has been produced continues to find that the agency's regulation continues to stifle innovation. Chorniy et al. (2021) find that three additional months of delayed approval time result in one less drug developed by drug producers.
Although a considerable body of research finds that FDA regulations are harmful, the agency does provide avenues for patients to experimental treatments under certain circumstances. For example, the FDA's expanded access allows patients with terminal illnesses to access experimental treatments with the agency's approval (FDA, 2020a). Dranvoe and Meltzer (1994) find that the FDA approves "more important drugs" comparatively more quickly.5 However, the author's estimates for drug importance do not account for medicinal urgency and broader public health considerations. Because addressing the impact of COVID-19 requires assessing tradeoffs of evaluating the safety and effectiveness of goods used to treat a novel virus with a need to quickly disseminate treatments due to medical urgency, further research into contemporary FDA regulatory actions are necessary to bolster our understanding of the agency's impact on the healthcare sector's ability to treat patients during pandemics.
3 THE FDA, POLITICAL ECONOMY, AND PANDEMICS
Although the FDA's regulatory scope includes a variety of products, Higgs (2004) notes the agency's primary authority is "banning existing products from the market until their manufacturers adduce satisfactory evidence of efficacy and safety" (p. 60). Because the FDA is a federal agency, it follows the incentive structure of a bureaucracy, mainly that of maximizing its revenue (Tullock, 1965; Mises, 1944). The FDA, as well as other regulatory bodies, typically obtains more revenue through expanding its number of operations, increasing the regulatory stringency over the products it oversees, and expanding its regulatory scope (Tullock, 1965).
The incentive to obtain more personnel, federal funding, and other resources while expanding its regulatory stringency and scope works to elongate and complicate the FDA's approval process for the goods it oversees (Higgs, 2004). The agency's incentives also influence its willingness to approve potentially harmful treatments and other goods.6 Because the FDA faces the threat of public backlash or political disapproval by approving a product which could potentially harm a patient, it has little incentive to approve the product despite its potential benefits. Instead, the agency either fails to approve the product or requires the producer to conduct more clinical research to further demonstrate it is safe (Higgs, 1995). The agency also faces little threat of repercussions from failing to approve a safe and effective product because the harms of denying patients a safe treatment option are rarely observable (Tabarrok, 2000).
Both considerations create a disincentive for the FDA to weigh the risks of harm against the benefits, providing overly stringent regulation (March, 2016; Tabarrok, 2000, 2017). A pandemic caused by a novel disease creates a demand to provide novel products to treat, detect, or deter a disease. Novelty combined with the FDA's motivation to approve products conservatively could motivate the agency to exercise harmfully stringent standards for goods created to address a pandemic. Recent analysis finds evidence for this. Using a Bayesian decision analysis, Isakov et al. (2019) find the FDA is overly conservative in designing clinical studies for drugs designed to treat terminal illnesses with no existing therapies.
Philipson and Sun (2008) note that the FDA's incentive to provide overly stringent regulation generates "dynamic considerations," which hinder the agency's ability to provide effective regulation (p. 96). Consequently, the agency's approval process results in higher costs to develop products and longer times to reach patients, but with potentially little additional assurance of product safety (p. 96).7 Higgs (2004) notes that side effects from FDA medications deemed safe for use constitute the fourth leading cause of death in the United States.8 While these findings could be interpreted as a need for more regulatory stringency, further research finds deregulatory processes occurring in the early 2000s provided faster approvals with sacrificing safety. Philipson et al. (2008) examine the Prescription Drug User Fee Acts, which allowed for prescription drugs to undergo an expedited review by the FDA for a fee.