{"subscriber":false,"subscribedOffers":{}}

Cookies Notification

This site uses cookies. By continuing to browse the site you are agreeing to our use of cookies. Find out more.
×

Research Article

Pharmaceuticals & Medical Technology

FDA Global Drug Inspections: Surveillance Of Manufacturing Establishments Remains Well Below Pre-COVID-19 Levels

Affiliations
  1. Emily Cuddy, Duke University, Durham, North Carolina.
  2. Yun Peng Lu, University of Maryland, College Park, Maryland.
  3. David B. Ridley ([email protected]), Duke University.
PUBLISHED:Open Accesshttps://doi.org/10.1377/hlthaff.2023.00686

Abstract

During the initial phase of the COVID-19 pandemic, the Food and Drug Administration (FDA) halted inspections of most overseas drug manufacturing establishments. Looking at data from the period 2012–22, we observed steep declines in both foreign and domestic inspections in 2020. By 2022, numbers of inspections remained well below prepandemic levels, with a 79 percent decrease in foreign inspections and a 35 percent decline in domestic inspections compared with 2019. There was no corresponding reduction in drug manufacturing or imports. Also, the resources allocated per inspection surged, although the FDA’s overall budget and staffing remained steady. Finally, citations rose dramatically, despite all establishments being given advance notice of inspections. The findings of our study underscore the pressing need to explore alternative methods for ensuring drug safety.

TOPICS

At the start of the COVID-19 pandemic, the Food and Drug Administration (FDA) announced that it would drastically reduce its inspection activities, postponing all surveillance inspections (both domestic and foreign) and focusing on so-called mission-critical inspections—broadly defined as those with the highest public health benefit to US patients.1 By July 2020 the FDA resumed prioritized domestic inspections, including preapproval, prelicense, surveillance, and for-cause inspections, in low-risk areas based on county-level public health data; however, the FDA did not resume unprioritized domestic inspections and non-mission-critical foreign inspections until 2022.2

Foreign drug manufacturing plays a large role in the supply of medicines in the US. As of 2019, more than two-thirds of the establishments manufacturing active ingredients for pharmaceuticals sold in the US were located outside its borders.3 India, China, and Italy are the leading foreign producers of these active ingredients.4,5 Moreover, foreign establishments account for approximately half of the manufacturers producing finished dosage forms—the final drug product consumed by patients—in the US market.4,5

Inspecting foreign facilities has many challenges. The FDA faces jurisdictional issues, language barriers, and travel clearance requirements. To facilitate their visits, FDA investigators typically give advance notice to foreign establishments. The FDA also opened offices in strategic locations to increase its capacity for inspecting abroad. However, as of November 2019, FDA foreign offices in China and India were understaffed, with 30 percent vacancy rates. Furthermore, the FDA had eight vacant positions (out of twenty) in its dedicated foreign drug cadre of US-based investigators, who exclusively conduct foreign inspections.6

A tragic incident in 2008 underscored the importance of rigorous inspections. An uninspected Chinese manufacturer was identified as the source of adulterated heparin that caused eighty-one deaths in the US.7 Even so, nearly a decade later, the Government Accountability Office reported that approximately one-third of the 3,000 foreign establishments making drugs to sell in the US still had never undergone FDA inspections.8

The onset of the COVID-19 pandemic further complicated matters. The FDA drastically reduced physical inspections. In their place, the agency implemented alternative strategies to keep tabs on the safety of imported medicines. These included heightened scrutiny of drug products entering US ports, soliciting records in lieu of on-the-ground inspections, conducting remote interactive evaluations, and bolstering information exchanges with European counterparts through mutual recognition agreements.9

In this study we examined the consequences of these policy changes in both domestic and foreign drug inspections. We analyzed not only the frequency of inspections and their intensity—defined in terms of resource use—but also their outcomes (that is, classifications). We also examined whether changes in import volume, overall FDA budget and staffing, or travel restrictions could explain the reduction.

Study Data And Methods

Inspections

We used data on inspections performed by the FDA Office of Regulatory Affairs during the period 2012–22. To obtain this information, we filed a Freedom of Information Act request. These data include all inspections under the FDA’s Drug Quality Assurance program, including preapproval, postapproval, surveillance, and for-cause facility inspections.10

An observation in the inspection data corresponds to a unique inspector-inspection record in a city, with a start and end date. We aggregated observations to the monthly level. For inspections spanning multiple months, we used the month in which the inspection began.

We also aggregated foreign establishments into five regions—China, Europe, India, Latin America, and other foreign nations—corresponding to the FDA Office of Global Policy and Strategy’s foreign office locations (online appendix exhibit A.1).11 We note, however, that this classification is based solely on the location of the establishment itself and does not reflect the FDA office or investigator staff ultimately responsible for the inspection. Indeed, the FDA Office of Regulatory Affairs uses a combination of US-based and foreign staff to inspect establishments outside the US.12 Nevertheless, the inspection data do not reveal which office is ultimately responsible for each foreign inspection.

We created a measure of the intensity of inspections in terms of resources deployed. We reported the average number of inspector-days per inspection by region per month, defined as the total number of days required to complete an inspection by all assigned investigators (for example, a two-day inspection by two investigators would receive a value of four inspector-days).

In evaluating the inspection outcomes, we examined the proportion of inspections with an “official action indicated” (OAI) finding, which is the most severe classification. The FDA designates an establishment inspection as OAI when there is an unsatisfactory level of compliance with current good manufacturing practices. These violations can result in the FDA withholding approval of new applications.13

Establishments And Imports

We also calculated the average number of inspections per establishment. We used data from the FDA’s drug establishments current registration site.14 The registration data include all establishments involved in the manufacturing, preparation, propagation, compounding, or processing of drugs distributed in or imported to the US. We used data for each year from 2012 to 2022. We acquired data for each year from 2012 to 2020 from a Freedom of Information Act request, and we acquired data for 2021 and 2022 through captures of the online database.

The need for drug-quality inspections of foreign establishments depends on the extent of drug manufacturing occurring abroad. As such, we reported imports over time. We used imports summary data from the FDA data dashboard.15 We selected drugs and biologics for 2012–22. An observation in the data is an import line, meaning a separate entry in an import transaction.

Because import lines do not account for volumes of product, we supplemented the FDA data with US trade and tariff monthly and annual data from the US International Trade Commission from the period 2012–22. These data include measures of imports by weight. We reported first unit of quantity in millions of kilograms. We selected pharmaceutical products corresponding to Harmonized Tariff Schedule 30 from the category imports for consumption.

FDA Budget And Staff

Through a Freedom of Information Act request, we obtained data from the FDA on budget and full-time-equivalent staff allocated to the agency’s human drugs program. Values for 2012–18 and 2020–22 reflect the FDA’s actual spending and employment, whereas values for 2019 reflect what was provided in the enacted appropriation.

We used inspections data to determine whether the FDA added new inspectors after the start of the COVID-19 pandemic and where they inspected. We also used data on FDA full-time-equivalent staff for human drugs from the period 2012–22. The aggregated staff data included not only inspectors but also other functions such as drug evaluation and approval.

Travel Restrictions

To assess the effect of international travel controls on FDA inspections, we used data from the Oxford Covid-19 Government Response Tracker. The tracker features an index for assessing the level of restrictions imposed by governments. The index values are 0 for no restrictions on international travel, 1 for screening arrivals, 2 for quarantine for arrivals from some or all regions, 3 for a ban on arrivals from certain regions, and 4 for a ban on arrivals from all regions or total border closure.16

The tracker includes daily data for many countries. We selected the five countries that accounted for the highest quantity of pharmaceutical products imported to the US in 2022, in kilograms, as measured by the imports data described above.

Limitations

This study had several limitations. First, we did not observe the actual resources available for inspection activities—in terms of both financial and staff resources. We proxied these measures with aggregate data on both the budget and full-time-equivalent staffing of the FDA’s human drugs program and constructed measures from our inspection data. Second, we did not observe the full extent to which the COVID-19 pandemic created logistical challenges to conducting inspections. Travel restrictions are only one component of these logistical challenges. Third, we did not observe the types of inspections that were being performed in our inspection data—for example, whether an inspection was mission critical or prioritized. As a result, we could not make comparisons of inspections across inspection types. Fourth, we did not observe the would-be inspection classification for inspections that did not occur because of the COVID-19 pandemic. Therefore, although we could compare inspection outcomes before and after the pandemic, we could not determine whether any change reflected a change in average establishment quality or differential selection of establishments.

Study Results

Inspections

In 2019, before the onset of the COVID-19 pandemic, the number of foreign inspections was rising, albeit from a low baseline, and the number of total inspections was falling. For 2012 we observed 648 foreign and 1,392 domestic inspections in total, summing over all months. For 2019 the number of foreign inspections increased to 954, and the number of domestic inspections decreased to 738 (exhibit 1) (see also the interactive chart that accompanies this article online).17

Exhibit 1 Number of Food and Drug Administration (FDA) monthly drug quality assurance inspections, by world region, 2012–22

Exhibit 1
SOURCE Authors’ analysis using FDA data. NOTES This figure plots the total number of drug quality assurance inspections per month and region conducted by the FDA Office of Regulatory Affairs during the period 2012–22. The data in this exhibit are stacked; the maximum total shown is the total number of inspections for all regions combined. Regions correspond to the location of the establishment, not necessarily the location or office of the inspectors conducting the inspection. Latin America refers to Latin America and the Caribbean. China excludes Hong Kong, Macau, and Taiwan.

After the onset of the COVID-19 pandemic and the release of the FDA’s revised industry guidance, the number of inspections declined significantly and had not recovered as of 2022, especially for foreign establishments. In 2022, total annual foreign inspections were down approximately 79 percent compared with 2019, whereas total annual domestic inspections had decreased by around 35 percent (exhibit 1). Notably, the FDA conducted no inspections in China or India between April 2020 and March 2021.

Whereas the average inspection before 2020 used between five and eight inspector-days in countries abroad, the average postpandemic inspection used between seven and fifteen inspector-days in those countries (authors’ calculations based on data shown in exhibit 2). This was also the case among domestic inspections, where the average number of inspector-days for an inspection increased by more than 70 percent between 2012–19 and 2020–22. Only FDA inspections in Europe used similar levels of resources throughout the sample period.

Exhibit 2 Average number of annual inspector-days per drug quality assurance inspection by the Food and Drug Administration (FDA), by world region, 2012–22

Exhibit 2
SOURCE Authors’ analysis using FDA data. NOTES This figure plots the average number of inspector-days required per inspection conducted by the FDA Office of Regulatory Affairs per year and region during the period 2012–22. Inspector-days represent the total number of days required to complete an inspection by all assigned investigators. Regions correspond to the location of the establishment, not necessarily the location or office of the inspectors conducting the inspection. Latin America refers to Latin America and the Caribbean. China excludes Hong Kong, Macau, and Taiwan.

Finally, among the inspections that did occur, the share of inspections resulting in an OAI rose after the start of the COVID-19 pandemic. In 2019, about 14 percent of domestic and 9 percent of foreign inspections resulted in OAI findings; however, by 2021, both shares had doubled, although there was a partial reversal in this trend in 2022 (exhibit 3).

Exhibit 3 Share of Food and Drug Administration (FDA) drug quality assurance inspections with official action indicated, by domestic versus foreign establishment location, 2012–22

Exhibit 3
SOURCE Authors’ analysis using FDA data. NOTE This figure plots the share of inspections of domestic and foreign establishments conducted by the FDA Office of Regulatory Affairs that received a classification of official action indicated (OAI), the most severe classification, per year during the period 2012–22.

Establishments And Imports

The decrease in inspection activity did not correspond to a decrease in drug manufacturing activity. Although the number of inspections per establishment steadily declined between 2012 and 2019, it fell far more significantly at the start of the COVID-19 pandemic in 2020 and remained below prepandemic levels more than two years later (appendix exhibit A.2).11 International establishments also continued exporting their products to the US throughout the pandemic. FDA import data revealed an uptick in import lines from 2019 to 2021, albeit with a slight drop in 2022 (appendix exhibit A.3).11 Similarly, trade data, measured in kilograms of imported product, showed an initial dip at the pandemic’s onset, followed by consistent growth between 2021 and 2022 (appendix exhibit A.4).11

FDA Budget And Staff

Even with the FDA’s decision to delay many inspections during the pandemic, resources available for inspections remained stable. The budget for the FDA’s human drugs program continued to grow, albeit at a reduced pace (appendix exhibit A.5).11 Moreover, the total number of inspector-days had mostly recovered by the end of 2022: The FDA spent 15,412 days inspecting establishments (both internationally and domestically) in 2022 versus 17,133 in 2019 (a reduction of 10 percent) (exhibit 4). Even so, this recovery obscured a striking reallocation of resources to domestic inspections (appendix exhibit A.6).11

Exhibit 4 Number of Food and Drug Administration (FDA) drug quality assurance monthly inspector-days, by world region, 2012–22

Exhibit 4
SOURCE Authors’ analysis using FDA data. NOTES This figure plots the total number of inspector-days required per inspection conducted by the FDA Office of Regulatory Affairs per month and region during the period 2012–22. Inspector-days are defined in the exhibit 2 notes. The data in this exhibit are stacked; the maximum total shown is the total number of inspector-days for all regions combined. Regions correspond to the location of the establishment, not necessarily the location or office of the inspectors conducting the inspection. Latin America refers to Latin America and the Caribbean. China excludes Hong Kong, Macau, and Taiwan.

The investigator workforce during the pandemic was also stable. First, there was no observed decline in full-time-equivalent employment within the FDA’s human drugs program (appendix exhibit A.5).11 Second, no reports or anecdotal evidence indicate significant staffing changes within the inspection program. Third, staff turnover remained consistent with prepandemic rates (data not shown).

The investigator workforce also was not less experienced. Among those investigators active during the pandemic, approximately 30 percent were first-time investigators, whereas in prepandemic years, the rate ranged from 26 percent to 40 percent. Of the returning investigators, they were more experienced—on average, they had conducted forty-one inspections between 2012 and 2019 compared with ten among those investigators who stopped inspecting during the pandemic (appendix exhibit A.7).11

Discussion

After the introduction of revised FDA industry guidance, the FDA conducted fewer inspections, especially abroad, and the few inspections it did conduct used more resources and had worse outcomes during the COVID-19 pandemic. It is unclear what these changes imply for the future of the inspection program and drug quality.

A decrease in inspections during the pandemic might have been less consequential if manufacturing or importing also slowed, but the evidence suggests otherwise. Our analysis of FDA data on establishments revealed that the decline did not coincide with an underlying reduction in the number of registered establishments or, in the case of overseas establishments, a reduction in exports to the US.

Deferring inspections for operational establishments may have consequences. Increases in time elapsed since the last inspection are associated with decreases in regulatory adherence.18 As these inspections play a crucial role in upholding manufacturing standards, the delay could have exposed US consumers to lower-quality products.

And yet, scaling up inspection activity going forward to account for such deferred inspections may be difficult if the observed increases in per inspection costs during the pandemic are here to stay. This invites the following questions: What drove the increase in inspection costs during the pandemic? Is the increase only temporary?

Certainly the COVID-19 pandemic increased the logistical costs of conducting inspections. Travel restrictions imposed after the start of the pandemic made conducting foreign inspections more difficult. For instance, China continued to enforce travel bans and quarantines through 2022. Nevertheless, for many other major foreign sources of drugs, these restrictions ended earlier—Canada, India, and Italy had lifted their requirements by March 2022, and Mexico had done so more than a year earlier (appendix exhibit A.8).11 This raises doubt as to whether travel restrictions explain the more persistent increase in per inspection costs that we documented.

Instead, our results suggest that the FDA’s choice of which establishments to inspect during the pandemic was a key cost driver. That is, as FDA leaders prioritized certain inspections, the resources required to conduct those inspections also changed.19,20 About 70 percent of inspections during the pandemic were reinspections of establishments. Relative to establishments that were not inspected during the pandemic, reinspected establishments had more resource-intensive inspections prepandemic (appendix exhibit A.9).11 Even those first-time inspections conducted during the pandemic used more resources than their prepandemic counterparts.

The change in selection criteria of inspections may also explain the increased incidence of OAI findings after the start of the COVID-19 pandemic. Establishments inspected during the pandemic were nearly twice as likely to have had a history of objectionable conditions or practices than those that were not inspected. Therefore, to the extent that the FDA reverts to its previous selection criteria, inspection costs could be expected to fall back to their historical levels. However, if the underlying distribution of firm quality has deteriorated, then such cost reversals may be limited.

Indeed, concerns regarding drug quality in the market remain. Inspections conducted during the COVID-19 pandemic, both domestic and international, were preannounced,21 and historically, bad-actor firms have abused such notice to hide quality issues.3 In addition, if more frequent or randomly selected surveillance inspections improve quality, then the FDA’s decision to pause such inspections during the pandemic may have further deteriorated the quality of establishments in the industry. Both concerns suggest that our finding that OAI rates rose during the pandemic may indicate growing quality issues in the industry, even after accounting for selection, and they raise questions about the feasibility of clearing the Inspection backlog with currently available resources.

Policy Recommendations

In the face of an unprecedented situation, the FDA prioritized the most pressing inspections in its portfolio during the pandemic.

In the face of an unprecedented situation, the FDA prioritized the most pressing inspections in its portfolio during the COVID-19 pandemic. To the extent that travel restrictions and safety concerns limited the FDA’s ability to conduct investigations abroad, redirecting those resources to prioritized domestic inspections was wise. Our analysis, however, suggested that travel restrictions are no longer a first-order concern, and yet most foreign inspections remain on hold. It is hard to rationalize continuing to postpone foreign inspections, given the data available to us. As the world moves beyond the pandemic, it is crucial for the FDA to address the backlog of inspections and to consider new approaches for the future. We make five recommendations.

Increase The Investigator Workforce

First, the FDA should increase its investigator workforce. The 21st Century Cures Act of 2016 gives the FDA some power to hire for tough-to-fill roles. If that is not enough, Congress should give the FDA even more flexibility to hire. More staff can enable more inspections, shorter lags between inspections, and investigator rotation across different establishments, which is essential to counteract complacency arising from repeat visits by the same inspector.22

Prioritize Foreign Inspections

Prioritizing foreign inspections going forward would close the foreign-domestic inspection disparity that COVID-19 reopened.

Second, because expanding the workforce takes time, the FDA should prioritize foreign inspections. Since March 2020, the FDA has postponed nonessential foreign inspections, creating a significant backlog. Prioritizing foreign inspections going forward would close the foreign-domestic inspection disparity that COVID-19 reopened.

Increase International Cooperation

Third, the FDA should continue to use mutual recognition agreements that recognize inspection reports and product sampling programs by regulators from a country other than the US and other than the country in which the establishment is located. Furthermore, the FDA should continue to ask for records before or instead of conducting on-site inspections and to use technology such as video calls and screen sharing for reviewing data from inaccessible establishments. Nevertheless, these efforts should supplement, and not be substitutes for, in-person inspections abroad going forward.

Spread The Drug Quality Assurance Burden

Fourth, others should share the burden of ensuring drug quality. For instance, both Kaiser Permanente and the Department of Defense are now working with a laboratory to test drugs and choose their producers based on quality.23 Other intermediaries, such as wholesalers and retailers, should follow suit. Although third parties will not be able to determine quality for a given manufacturing establishment, they can give insight into average quality for a given company. Initiatives that identify companies selling high-quality products could reward them with larger market shares and higher prices.24 As Janet Woodcock, then director of the FDA’s Center for Drug Evaluation and Research, said to a journalist, “Inspections are not a panacea. …It really is important to make this industry responsible for quality itself.”3

Increase Patients’ Awareness Of Drug Quality

Fifth, patients should be made aware of drug quality issues. Few know that foreign drug production sites are rarely inspected by the FDA and receive advance warning before an inspection. If patients were made aware of average quality differences across manufacturers of generic drugs, then they could make quality-informed purchase decisions.

Conclusion

Addressing the challenges imposed by the COVID-19 pandemic and the FDA’s response will require combined action by the FDA, Congress, manufacturers, and buyers. Past collaborations in other areas of FDA responsibility have been successful, including speeding important drugs to market,25,26 encouraging development of drugs and vaccines for rare and neglected diseases,27,28 and reducing drug shortages.29 We hope that implementing the recommendations outlined above will promote the use of safe and effective medicines beyond the COVID-19 era.

ACKNOWLEDGMENTS

The authors thank M. Usaid Awan and Yue Zhang for helpful insights. The authors did not receive external funding for the research. This is an open access article distributed in accordance with the terms of the Creative Commons Attribution (CC BY-NC-ND 4.0) license, which permits others to distribute this work provided the original work is properly cited, not altered, and not used for commercial purposes. See https://creativecommons.org/licenses/by-nc-nd/4.0/. To access the authors’ disclosures, click on the Details tab of the article online.

NOTES

   
Loading Comments...