The Impact of the COVID-19 Pandemic on the Extramural Scientific Workforce – Outcomes from an NIH-Led Survey



Photo of Dr. Marie Bernard
Marie A. Bernard, M.D., Deputy Director of the National Institute on Aging, Acting Chief Officer for Scientific Workforce Diversity

One year later, the COVID-19 pandemic has drastically affected our individual lives and communities. We have observed disproportionate effects observed in underserved populations, leaving them vulnerable to higher infection and mortality risk. These effects have led to an increased reliance on biomedical researchers and clinicians to offer public health solutions to this crisis. Within the research workforce, early-career scientists may bear the brunt of pandemic-related mitigation measures at institutions and limitations due to inability to be in the physical workspace.

At NIH, we recognized the many ways the COVID-19 pandemic could adversely affect the biomedical workforce, particularly members of underrepresented groups and vulnerable populations. In October 2020, NIH fielded two online surveys to objectively document COVID-19’s impact on extramural research. One survey assessed the perspective of individual research administration leaders at extramural institutions, and the other survey assessed the perspective of the researchers themselves. In this post, we offer a high-level overview of general trends noted within both surveys. This infographic here also describes the outcomes from the surveys.


The former NIH Chief Officer of Scientific Workforce Diversity, Dr. Hannah A. Valantine, spearheaded the development of the survey questionnaires in close collaboration with the Office of Extramural Research, and several other NIH offices. For the Extramural Institutions Survey, which was fielded from October 7, 2020 – November 6, 2020, a research administration leader (vice president for research or equivalent position) was identified from each of the following types of institutions:

  • The 1,000 top-funded domestic institutions based on FY2019 NIH awards
  • Schools that are part of the Association of American Medical Colleges
  • Minority-serving institutions that received grant awards in FY2019

There were 224 participants out of 705 invites (a response rate of 32%).

For the Extramural Researchers Survey, which was fielded from October 14, 2020 – November 13, 2020, the eRA Commons system was used to generate a list that included individual researchers at domestic institutions who logged into eRA Commons within two years prior to the survey, and who identified as having a scientific role (e.g., principal investigators, trainees, sponsors, undergraduate students, graduate students, postdoctoral researchers, scientists, and project personnel). There were 45,348 participants out of 234,254 invites (a response rate of 19%).


Figure 1 shows the main findings from the institutional survey. The majority of respondents noted concerns about research functions, research productivity, and financial status. Most were implementing COVID-19 monitoring measures, but only a minority were providing or expanding facilities for childcare.

Figure 1 displays institutional high-level findings. The graphic shows columns with key questions around Research Functions being Jeopardized, Moderate Major Impacts in Research Productivity, concerns with Institution’s Financial Status, Substantial Impact from Loss in Endowment, Testing Available to Anyone, Implementing COVID19 Monitoring Measures, and Facilities for Childcare. The rows represent All Respondents, Doctorate-granting universities without a Professional School, Independent Research institutions, Special Focus/Other Institutions, Minority Serving Institutions, and Non-Minority Serving Institutions. Finally, boxes represent more negatively (orange) or less negatively (blue) impacted than overall average, or on par (no color) with overall average.
Marie A. Bernard, M.D., Deputy Director of the National Institute on Aging, Acting Chief Officer for Scientific Workforce Diversity

Figure 2 presents an Executive Summary of the extramural researchers’ survey. We focused on three outcome measures: belief that the pandemic will negatively affect career trajectory, concerns about mental health and external stressors, and concerns about research productivity and institutional support.

Figure 2 is a graphic showing the executive summary as described in the text
Figure 2

Figure 3 shows the main findings according to race/ethnicity, gender, and career stage. The left column shows the composition of the respondents (e.g. 69% were White, 53% were women, and 53% were early-career). Early-career scientists and Asians were most likely to report concerns about career trajectory. The vast majority of nearly all groups reported lower job productivity. Women, Hispanics, and early-career scientists were most likely to report concerns about mental health and external stressors. Not quite half of respondents reported that caretaking responsibilities made it substantially more difficult to be productive; we take a closer look at this a bit later. Finally, less than half of respondents felt that their organization was supportive in helping them to remain productive.

Figure 3 shows high-Level Findings from the survey. The columns describe responses to key questions such as the Pandemic has Negative Impacts on Career Trajectory, Lower Job Productivity, Societal/Political Events, Mental and physical Health, and Productivity. The rows represent all Respondents or broken down by Race/Ethnicity, Gender Identity, and Career Stage. Finally, boxes represent more negatively (orange) or less negatively (blue) impacted than overall average, or on par (no color) with overall average.
Figure 3

We now take a closer look at concerns about career trajectory. Respondents who were most concerned were in earlier career stages (Figure 4) and conducting laboratory-based research (Figure 5). We used a machine learning algorithm, specifically gradient boosting machines, to determine which of 47 variables were most highly associated with career trajectory concerns. The top variables were by far impact on ability to apply for grants, followed by decreased research productivity, career stage, race, effects of caretaking, and the primary type of research.

Figure 4 shows impact on Career Trajectory by Career Stage. The X axis is career level, such as Postdoc, Faculty (0-6 years or 7-14 years), Student, Researcher (0-6 years or 7-14 years), Faculty (15+ years). The Y axis is the % Negative Impact by Career Stage. Each group’s column is a different color, ranging from orange to gray to blue
Figure 4
Figure 5 shows the Impact on Career Trajectory by Type of Research. The X axis is the type of laboratory from Computational, Clinical, Sociological, or Epidemiologic. The Y axis is the % Negative Impact by Type of Research.
Figure 5

Figure 6 shows the decreased ability to apply for grants were caretaking responsibilities, lost access to research staff, and lost access to collaborators, which together may have adversely affected the ability to generate preliminary data.

Figure 6 is the Impact on Career Trajectory assessed by the Ability to Apply for Grants. The X Axis represents groups such as Having caretaking responsibilities, providing care for young children, losing access to research staff, and losing access to collaborations. The Y axis is the Impact on respondents with difficulty applying for grants (%). The dark orange columns are for respondents with difficulty applying for grants, while the light orange columns are for all respondents.
Figure 6

We now move to research productivity. Figure 7 shows that, as might be expected, decreased access to laboratory facilities was among the top predictors of decreased productivity. Figure 8 shows that earlier career stage was also predictive. In a machine learning model, the strongest correlates of decreased productivity were lost access to laboratory facilities, decreased ability to apply for grants, and caretaking responsibilities.

Figure 7 is the Top Factors that Negatively Impacted Productivity. The X axis (from highest to lowest percentage) is being Virtual instead of in-person interactions, cancellation of conferences, changes in lab or animal facility access, uncertainty about timeline for return to workplace, research was put on hold, changes in life priorities, and personal mental/physical health. The Y axis is the percentage of respondents
Figure 7
Figure 8 is the impact on Research Productivity by Career Stage. The X axis (from highest to lowest) is faculty (7-14 years), faculty (0-6 years), student, postdoc, faculty (15+ years), researcher (0-6 years), Researcher (7-14 years), researcher (15+ years). The Y axis is the Percentage of respondents.
Figure 8

Finally, we look at mental health and external stressors. Figure 9 shows top correlates of mental health concerns. These included societal and political events and physical and/or social isolation.

Figure 9 are the factors that Negatively Impact Mental Health. The X axis (from highest to lowest) are societal/political events, physical/social isolation, disruption of promotion/tenure timeline, expectations of the supervisor, visa considerations, access to programs focusing on stress during pandemic, and access to programs assisting with applications to professional schools. The Y Axis is the percentage of respondents
Figure 9

Figure 10 shows more details on caretaking responsibilities. Parents with young children reported the greatest decreases in research productivity, while women were more likely than men to report that caretaking made it substantially more difficult to complete their work responsibilities.

Figure 10 is the Impact of Caretaking by Gender & Type of Dependent. The X axis: orange columns for women with young children, light orange columns for women with other dependents, blue columns for men with young children, and light blue columns for men with other dependents. The key factors are lower productivity, negative outlook on career trajectory, and substantially more difficult to complete work responsibilities. The Y axis is the percentage of respondents.
Figure 10

Figure 11 shows data on civil unrest; Blacks were most likely to report that civil unrest tied to racism adversely affected research productivity.

Figure 11 is the Impact of Civil Unrest. The Orange columns represent negative impacts and blue columns represent somewhat negative impact. The respondents are listed by key demographic groups , including by gender, race/ethnicity, disability status, and sexual orientation.
Figure 11


The effects of the COVID-19 pandemic have been far-reaching. Our survey findings show that the scientific workforce has not been immune to its effects. It is clear the NIH-funded community of extramural researchers has experienced inequities in several domains, with early-career researchers and those with caregiving responsibilities most affected.

NIH has already begun taking preemptive actions for investigators early in their careers, recently issuing a guide notice detailing opportunities for extension of Fellowship (F) and Career Development (K) awards impacted by COVID-19. NIH has also allowed investigators affected by COVID-19 (e.g., university closures) to submit requests for an extension of Early Stage Investigator (ESI) status through eRA Commons.

We will continue to analyze these data and high-level findings will be shared with the extramural research community in the coming months. Using these and other data as they become available, NIH will maintain its focus on evidence-based actions to foster inclusive excellence within the scientific workforce, to better support the health of our entire population.


  1. I feel like the one issue not addressed in this survey was the direct financial impact of the pandemic. Being in Ohio, we were effectively shut down from mid-March to mid-August, although some COVID work was allowed from mid-June on. However, we still had to pay all our personnel for that entire period, without any institutional support whatsoever and there has been no recognition of that by NIH – losing 5 months of a 4-year R01 is a huge hit. I am not complaining about paying my people as it was the right thing to do, but NIH might have helped cover those costs.

    1. The financial impact is particularly painful for small labs, which may also contribute to gender disparity.

  2. Strongly concur with Ian. The non-science world received various forms of compensation to make up for lost jobs and wages, etc. While we had the capacity to maintain employment, the end suffering from insufficient funds to complete objectives and aims is fundamentally a major loss of investment by funding agencies, and a potentially huge blow to individual investigators’ research and career objectives.

Leave a Reply

Your email address will not be published. Required fields are marked *