38 Comments
In 2021, we reported that the age at which a researcher is designated on an NIH award for the first time had increased since 1995 and plateaued in the 2010s. We showed that the median age for PhDs first designated on an award remained around 41 in recent years while the median age of MDs and MD-PhDs was 44. There were no appreciable differences between researchers identifying as male or female. Today we are sharing an update on these data for fiscal years (FYs) 2021 – 2023, with the addition of information related to race, ethnicity, and disability status.
We have heard concerns about the rising age at which early career researchers are first supported on an NIH award. Some concerns have centered on the ways competitive funding and academic hiring impact early career scientists pursuing independent research careers.
For simplicity in this post, we refer to investigators receiving awards, but recognize that NIH makes R01-equivalent awards to institutions (not individual researchers).
Table 1 shows the age of investigators upon receiving their first R01-equivalent grant for FYs 2021 through 2023 disaggregated by their terminal degree. Between FYs 2021 and 2023, the median age for PhDs receiving their first award was 41 while the median age for MDs and MD-PhDs remained around 44. The difference between these groups may be due to the additional time spent by physicians in clinical training after receiving their degrees.
Table 1. Age at Receiving first R01-Equivalent for MDs, PhDs, and MD-PhDs for FYs 2021-2023
| Fiscal Year | Degree | Number of Investigators | Mean Age (Years) | Median Age (Years) | 10th Percentile | 25th Percentile | 75th Percentile | 90th Percentile |
| 2021 | MD | 373 | 46.0 | 44.0 | 39.0 | 41.0 | 49.0 | 56.0 |
| 2022 | MD | 325 | 45.5 | 44.0 | 39.0 | 41.0 | 48.0 | 56.0 |
| 2023 | MD | 381 | 46.0 | 45.0 | 39.0 | 41.0 | 48.0 | 56.0 |
| 2021 | PhD | 1,776 | 42.4 | 41.0 | 35.0 | 38.0 | 46.0 | 51.0 |
| 2022 | PhD | 1,872 | 42.4 | 41.0 | 35.0 | 38.0 | 46.0 | 52.0 |
| 2023 | PhD | 1,786 | 42.6 | 41.0 | 35.0 | 38.0 | 46.0 | 53.0 |
| 2021 | MD-PhD | 215 | 45.1 | 44.0 | 39.0 | 41.0 | 48.0 | 53.0 |
| 2022 | MD-PhD | 240 | 45.8 | 44.0 | 39.0 | 42.0 | 48.0 | 55.0 |
| 2023 | MD-PhD | 217 | 45.7 | 45.0 | 39.0 | 41.0 | 48.0 | 54.0 |
Table 2 shows the age of investigators identifying as male or female upon receiving their first R01-equivalent grant for FYs 2021 through 2023. In general, the median age at first award remained around 42 for both male and female researchers between FYs 2021 and 2023, consistent with prior recent data.
Table 2. Age at Receiving first R01-Equivalent for Investigators Identifying as Male or Female for FYs 2021 – 2023
| Fiscal Year | Gender | Number of Investigators | Mean Age (Years) | Median Age (Years) | 10th Percentile | 25th Percentile | 75th Percentile | 90th Percentile |
| 2021 | Male | 1,382 | 43.6 | 42.0 | 36.0 | 38.0 | 47.0 | 53.0 |
| 2022 | Male | 1,319 | 43.5 | 42.0 | 36.0 | 38.0 | 47.0 | 53.0 |
| 2023 | Male | 1,328 | 43.9 | 42.5 | 36.0 | 39.0 | 47.0 | 54.0 |
| 2021 | Female | 1,029 | 43.3 | 42.0 | 36.0 | 39.0 | 46.0 | 52.0 |
| 2022 | Female | 1,158 | 43.2 | 42.0 | 36.0 | 39.0 | 46.0 | 53.0 |
| 2023 | Female | 1,116 | 43.2 | 42.0 | 36.0 | 38.0 | 47.0 | 53.0 |
Responding to some feedback received on our last post, we are providing age at first R01-equivalent data broken down by additional demographic variables, specifically race/ethnicity and disability status.
Table 3 shows the age of Asian, underrepresented minority (URM), and white investigators receiving their first award between FYs 2021 and 2023. It is important to note that we use the White House’s Office of Management and Budget Minimum Standards for collecting and reporting race and ethnicity data (see this Nexus article for more). Race-ethnicity groups were combined to avoid potential privacy concerns.
In general, for FYs 2021 – 2023, the median age at first award is consistent between these groups at 42 years.
Table 3. Age at Receiving first R01-Equivalent for Investigators Identifying as Asian, URM, or White for FYs 2021 – 2023
| Fiscal Year | Race / Ethnicity | Number of Investigators | Mean Age (Years) | Median Age (Years) | 10th Percentile | 25th Percentile | 75th Percentile | 90th Percentile |
| 2021 | Asian | 641 | 42.8 | 42.0 | 36.0 | 38.0 | 46.0 | 51.0 |
| 2022 | Asian | 671 | 43.0 | 42.0 | 36.0 | 38.0 | 47.0 | 52.0 |
| 2023 | Asian | 672 | 43.7 | 43.0 | 36.0 | 39.0 | 48.0 | 53.0 |
| 2021 | URM1 | 251 | 44.2 | 42.0 | 37.0 | 39.0 | 47.0 | 54.0 |
| 2022 | URM | 304 | 43.1 | 42.0 | 36.0 | 39.0 | 46.0 | 51.0 |
| 2023 | URM | 312 | 42.8 | 41.5 | 36.0 | 38.0 | 46.0 | 52.0 |
| 2021 | White | 1,410 | 43.4 | 42.0 | 36.0 | 38.0 | 47.0 | 53.0 |
| 2022 | White | 1,394 | 43.4 | 42.0 | 36.0 | 38.0 | 47.0 | 54.0 |
| 2023 | White | 1,367 | 43.6 | 42.0 | 36.0 | 39.0 | 47.0 | 54.0 |
1 URM includes investigators identifying as American Indian or Alaska Native, Black or African American, Native Hawaiian or Other Pacific Islander, more than 1 race, and Hispanic or Latino – regardless of race.
In table 4, we focus on age disaggregated by disability status. In general, the median age for investigators reporting a disability remained around 40 between FYs 2021 and 2023.
Table 4. Age at Receiving first R01-Equivalent for Investigators Reporting a Disability Status for FYs 2021-2023
| Fiscal Year | Disability | Number of Investigators | Mean Age (Years) | Median Age (Years) | 10th Percentile | 25th Percentile | 75th Percentile | 90th Percentile |
| 2021 | Yes | 21 | 43.6 | 40.0 | 36.0 | 38.0 | 46.0 | 52.0 |
| 2022 | Yes | 24 | 44.3 | 41.5 | 36.0 | 37.5 | 49.5 | 61.0 |
| 2023 | Yes | 39 | 41.3 | 40.0 | 34.0 | 36.0 | 45.0 | 50.0 |
| 2021 | Unknown/ Withheld2 | 70 | 46.2 | 43.5 | 38.0 | 40.0 | 50.0 | 58.0 |
| 2022 | Unknown/ Withheld | 89 | 45.9 | 44.0 | 38.0 | 39.0 | 51.0 | 58.0 |
| 2023 | Unknown/ Withheld | 85 | 44.1 | 41.0 | 35.0 | 38.0 | 48.0 | 59.0 |
| 2021 | No | 2,386 | 43.3 | 42.0 | 36.0 | 38.0 | 47.0 | 52.0 |
| 2022 | No | 2,442 | 43.3 | 42.0 | 36.0 | 38.0 | 47.0 | 53.0 |
| 2023 | No | 2,395 | 43.6 | 42.0 | 36.0 | 39.0 | 47.0 | 54.0 |
2 “Unknown/Withheld” refers to a PI selecting the option “Do not wish to provide” or not checking any option on their eRA profile for disability status.
We are committed to ensuring a robust and vibrant future workforce. We will continue to assess age and other factors related to our supported workforce going forward.
I am grateful to my colleagues in the OER Division of Statistical Analysis and Reporting (DSAR) for their contributions to this blog.
Another NIH-independent selective mechanism that partially explain these data is that certain predatory PIs, especially those with influential networks, exploit their positions to impede the career development of postdocs and junior faculty. They may appropriate research ideas and intellectual contributions without due credit. Despite having limited understanding of the multidisciplinary research they oversee, these PIs receive funding and promotions, often publishing in high-impact journals even when their work lacks reproducibility.
Unless NIH-backed biomedical research funding strategies undergoes a structural overhaul, it risks losing its vital relevance to societal progress. This could lead to a talent drain, with future research and innovators opting for the “for-profit” sector, where a culture of true innovation flourishes.
I completely agree, the system needs a profound overhaul. To follow up on this, the reputation of investigators and their publication track record really influences one’s success in securing more funding. Senior investigators acknowledge all their grants in all their publications, regardless of the topic of the grant funded vs paper published, and whether it is their work or this of collaborators. And they have large teams working to support their publication records.
As a result, they appear super productive, and get their renewals funded for 30, 40 years in a row, depleting the funding pool while killing innovation, since their grants are so safe and more often than not, a repeat of the last. In reality, these groups contribute incremental advances to their respective field, even their high impact publications are sometimes a bit underwhelming. The resume should not be weighting so much in the funding decision. I feel that if we priced success at dollars per team staff or dollar per output from the lab (not from an extensive network), it would become very obvious that larger groups/ older PIs waste funding by the million. I am myself getting into that older age group soon, and I do NOT plan to stick around until my 70s, draining funding for younger researchers.
Perfectly said. Needs a real overall. Institutional support is also very important, which sometimes become one eyed dear.
The NIH should be commended for bringing to light the unpleasant inequities of NIH funding to aspiring investigators. It is lamentable as shown in Table 3 how the funding of “URM” investigators is less than half of the Asia investigators. By lumping Blacks, Hispanic, etc. together, shows that NIH is still bestraddled by its lack of concern for funding of URM investigators who will always serve as the workers for the mainstream investigators. They are often praised as being so productive. The NIH tried to improve on this number, but I believe the roadblock to the improvement lies at the cadre of grant reviewers who basically have the say on who is funded and which institution the “URM” applicant comes from. One other reason for this lethargic number of “URM” investigators lies with institutions’ hiring practices. Entry into the faculty/investigator position seems not to be protected, but it is protected and geared towards particular type of applicants.
Is it possible to examine age at first R01 for PIs conducting patient oriented versus basic science research? That might partially contribute to some of the gaps observed for MDs versus PhDs. Patient-oriented research typically takes much, much longer, so the length of time needed to obtain compelling preliminary data for the first R01 might lead to further delays in grant funding.
That is an excellent point. IRB regulations and approvals also slow down the processes. I would be helpful to consider simplifying the IRB process too. It seems to get more complex and slower every year.
It would be very helpful to see these data over a much longer period of time; at least 15 years, longer if possible.
I agree. I would really like to see the trend from say 2000 to today. As well as the number of years post-PhD individuals are spending (as a post-doc) before starting their own labs.
Our 2021 blog post on this topic covers fiscal years 1995 to 2020, however, please note that it only reports age, sex/gender, and degree. It does not cover race/ethnicity or disability status. Please see the post here: https://nexus.od.nih.gov/all/2021/11/18/long-term-trends-in-the-age-of-principal-investigators-supported-for-the-first-time-on-nih-r01-awards/
Is it possible to run age at first R01 comparing people who first received mentored career awards vs. people who did not?
It would be nice to see age and demographics for foreign (ie non-US) investigators not only for RO1s but NIH awards in general
It takes time to achieve the neccessary repution for awards under current NIH review climate. The recent change of NIH review policy may help. Ultimately, this problem and many other problems can be solved by anonymous scientific merit review followed by open qualification review.
In general “age at first R01” is not super informative as it would conflate folks who started graduate school at age 21 and did not get their first R01 until 20 years later due to a very long training (something that can be acknowledge as a toxic problem in the field) with folks who start their Ph.D.s at age 31 then get their R01 after five years of grad school and start a new assistant prof position at age 40 after four-five years of postdoc (a common scenario for most of my new hires I am aware of in my field).
In my experience, relatively few folks enter Ph.D. programs directly out of college (especially those receiving bachelors training outside of the USA), and seeing graduate students start their Ph.D.s in their late 20s/early 30s is not unusual at all. Anecdotally, folks with a Ph.D. at 26 years of age or so are the exception in the 21st century, not the rule.
I think age at first R01 equivalent is an important metric. I don’t see where this post claims it’s the only metric that matters or the one that most people might care about. However, for those who are planning life events, age matters. Knowing the age distribution at which investigators reach an important metric of stability in their careers can be an important factor for trainees making major life and career decisions. Time from terminal degree to first R01 equivalent would be a nice complementary metric. This post provides just one snapshot of the data and you can find a rather comprehensive data set on funding metrics in the NIH databook (https://report.nih.gov/nihdatabook/).
I think your second comment may be an inaccurate generalization based on your experience in a specific field/institution/department. The majority of PhD students in the program where I got my degree enter within 0-2 years of completing undergrad, so far from exceptional here.
Quick question/clarification – are you counting the R00 portion of a K99/R00 as “R01 equivalent”?
Over the past decade or more, early-career awards have become coin of the realm for junior faculty hiring decisions at many schools. Instead of going for an R01 or R21 straight out of post-doc, the K99/R00 route pushes the first R01 back by 4-5 years. Do you have a graph that shows the broader statistic of “age at first R-mechanism grant” (R00, R01, R21, R03, etc.) over an extended time? I would guess that with the wider adoption of K99/R00s, that age has fallen over time, even though age at first R01 has stayed high.
I agree with Dr Duncan’s comment. The average age at first R01 is less informative and taken out of context without knowing the time since graduation from a doctoral program. Further, publishing the standard deviation will also enable readers understand the level of variability and range of this spread.
Agree with the above comment that a more informative statistic would be time to first R01 from start of first graduate degree matriculation. Also, would be interested to see analysis including age when applying for R01. I postulate that older individuals are weeded out prior to obtaining R01 due to career uncertainty and more limited remaining years of employment.
It should also include when a junior faculty or Research Associate begins to apply for their first RO1 application and how long or how many attempts it takes to get an RO1 grant in general. Most PhDs start postdoc at 28-30 age.
This will provide an important data on how the new investigators or young investigators have tried to secure an RO1. Many do not succeed.
Mike, Thank you for sharing this data. I am 72-years-old and I am P.I. on a recently awarded SBIR/Fast Track grant. I have been asked whether I am the oldest first time PI. Do you have any information that can help answer this question?
It gives me confidence that diversity is equally distributed at the NIH platform. Besides being an important platform for clinical research, NIH supports so many of us with salaries, families, and stability. Fairness and immense support from your employees are remarkable. The astonishing fact is you talk about any aspect of science in a few weeks, you have the component supporting that research.
Age is not super helpful for most people. Please provide years after highest terminal degree, which would also help determine if the current 10 year limit for ESI is good or needs to be expanded! Thanks!
It would be nice to see this broken down by type of institution. For those of use working at small undergraduate institutions, getting an RO1 is extremely challenging.
I would be interested to see the demographic information for how many grants had to be submitted prior to success as well as how many investigators had to close their labs/stopped applying to NIH because they were never able to obtain NIH funding. My guess is that is where the disparities will show up. This data only tells the story of the successes (10-20% of applicants). How many researchers invest 20-30 years into NIH grant-writing, only to end up driving an Uber?
DO YOU WANT REAL PROGRESS?
MAKE THE REVIEW PROCESS BLIND and ULTIMATELY SCIENTIFIC MERIT BASED, SO REVIEWERS CAN NOT SEE WHICH GRANT IS THEIR BUDDIES.
THIS WILL TAKE CARE OF YOUR FUNDING ECOSYSTEM, ELIMINATING THE MONETARY BLACK HOLES 🙂
How about those PhDs who want a research career but never get an RO1? Are they too dim? What a waste of resources with regard to training. I hear so many PhD grads saying how they don’t want to bother with grants and go straight for teaching-oriented jobs or other careers. Sadly, I would not recommend obtaining a PhD in biology to anyone interested in STEM. Go for engineering or the clinic. Someplace where you will be in demand.
What is the African American/Black age for receiving their first R01-Equivalent? I.e.: Table 3. Age at Receiving first R01-Equivalent for Investigators Identifying as Asian, URM, or White for FYs 2021 – 2023
As noted in the blog, some racial/ethnic groups were combined to avoid potential privacy concerns. African American/Black investigators were included in the underrepresented minority group per the footnote on Table 3.
I agree with Paul Brookes’s comment. The change in funding opportunities (K99/R00) over the years needs also to be taken into account. Consequently, the trend would probably look differently.
Looking back, from the First Law of Thermodynamics (Helmholtz), to metabolic cycles (Krebs), to the dynamic state of body constituents, aka protein turnover (Schoenheimer), and to the double helix (Watson and Crick, Rosalind Franklin), in biology many of the major discoveries were made by critical minds in their 20s and 30s. Those investigators gained their independence early, and often under difficult circumstances. How much this “genius of youth” can be influenced by funding agencies (like the NIH) still remains to be determined. The critical role of the NIH in providing support for the acquisition of new knowledge of strategic value is unsurpassed.
The biggest problems I see are:
The budget pie is too small. The US spends $825 billion on military expenditures and only $47 billion on NIH funding. It should be doubled.
The grant system functions like a Taylorite piece system. It’s fixed budget and doesn’t really account for inflation. Nor does it account for unionization of lab workers and giving them livable salaries.
The private sector supplying goods for NIH research is price gouging and Congress doesn’t address it in a real way. Instead it creates paperwork to create the illusion of price control. All this for unique items. Why not limit suppliers profit rate?
The length of time Post Doctoral Research Scientists really need in an established lab to have a resume sufficient to get a job has increased. Grant periods should be lengthened.
Bravo! To all but supplier price limitations. Some of that would be solved by allowing purchases outside approved vendors to be easier. Competition.
It would be very helpful if NIH could publish this same data disaggregated for individuals with DVMs and DVM-PhD degrees. Both the NIH Physician-Scientist Workforce Report and the NIH ORIP One Health workshop highlighted the shortage of veterinarians entering the biomedical workforce and underscored a need to increase the number of these comparatively trained veterinarian scientists for translational research. Any data on R-level funding success would be very informative to our community and institutions.
One purpose of the very useful data like this is to demonstrate the lack of early funding opportunity for newly trained investigators. While the data can be more granular and informative, there is enough information to suggest that more funding is needed for early stage investigators. Especially since additional funding from Congress is unlikely to be forthcoming, redistribution of funding is the main option available. Limiting RO1 funding for all investigators by restricting the number of RO1s (2?) or the total annual RO1 funding ($1M?) could quickly accomplish this goal.
I think it would also be helpful to see these data presented by institutions – those who are research intensive vs non-research intensive. People are on different trajectories in these two settings and juggling different responsibilities. Some people are 100% research while others are in a 40:40:20 traditional model.
I wonder how the data would look like when ESI and new investigator non-ESI are separated. My bet is that the “flattening” observed is result of an artificial inflation by ESIs, and my hunch is that the age of the new investigator non-ESI is still increasing (congrats James Firman!). Nothing wrong about ESIs, but giving inherent problems in the system which is basically to support through review system older PIs past their prime absolving all the science and the innovation (and careers) of young investigators, artificially reducing the age does very little to fix fundamental problems, if anything is creating a new problem called “at risk” investigators – actually I wonder how many of those “at risk” were ESIs that were pushed too fast into R01 without enough training.
See data in this blog: https://nexus.od.nih.gov/all/2020/02/07/whats-happening-with-at-risk-investigators/
Based on hard data presented on the blog, two fixes come to mind: – all new investigators should have a 10 score advantage independent on how many years after PhD – there should be either a age cap or career length cap; because productivity is also a bell curve, and I think the problem now is too many people past their prime “just hanging on”; but they are resource draining nevertheless. And I think that if you have been getting R01s for 20-30 years you should be at the point that you can diversify your funding source (foundations, private donors, companies) or moving on to a new phase of your career.
Thank you for sharing this update. While age is certainly important, it would be at least as valuable to look into the years since the terminal degree.
A very relevant complementary parameter that is not considered is how many people give up. How many new investigator applicants apply for R01s and quit due to failure? Similarly, how many K awardees leave science either with or without trying for R01s?
Yes, true. Would love to see how many cycles and over what span of years it took to get that first R01. And how many people applied how many times and never applied again in the future.
I think the analysis on racial/gender groups should be further stratified by disciplines e.g., Basic science/population-based to gauge whether Asian/White/Underrepresented groups are over or under represented on R01 awards. Because it is of prime interest (the main issue on health disparities) to see whether the funds are going into the population target concordant with the PI’s race/ethnicity/gender identity/disability. E.g. if Asians are dying for receiving poorer care, the analysis will promote health disparities by shifting down R01 awarded to Asian psychologists working on suicide prevention among API populations. This is a form of structural harm based on partial data and decision on such categorization.