Summary

Biomedical researchers investigate how the human body works with the aim of finding new ways to improve health. Biomedical research has likely produced large returns to society per researcher in the past; we expect it to continue to be a high-impact area in the future; and it appears to be constrained by good researchers. Its drawbacks are that it takes a long time to train, has high drop-out rates and leaves you relatively fixed in the biomedical field. You should strongly consider this path if you have an undergraduate degree from a top university with high grades (GPA 3.8+); you’ve tested your fit for research by doing a placement in a lab; and you have a place at a top 10 PhD program for your specialty. However, success in this path is very hard to predict, and so we encourage you to have a back-up plan.

Pros

  • Biomedical research is promising cause area.
  • The field seems to be constrained by good researchers.
  • Highly interesting work for the intellectually curious.

Cons

  • Long time to train (4-12 years).
  • Highly competitive; people drop out even in their late thirties and forties.
  • Relatively narrow exit options.

Ratings

Career capital: 

Direct impact: 

Earnings: 

Advocacy potential: 

Ease of competition: 

Job satisfaction:

Our reasoning for these ratings is explained below. You might also like to read about our approach to rating careers.

Key facts on fit  

Very high intelligence, intense intellectual curiosity and interest in biomedical research, grit, programming and statistics in demand.

Next steps

Contact lab managers during your sophomore (2nd) or junior (3rd) year at university to get a job as a research assistant in a lab to test your fit. Then read this advice on applying to graduate school.

What is this career path?

Biomedical researchers investigate how the human body works with the aim of finding new ways to improve health. They work in academia and industry, but in this profile we focus on those who work in academia. In general, academia tends to focus on improving tools and techniques, studying healthy biological processes and the causes and progress of diseases, whereas industry tends to focus on generating and evaluating possible treatments.2

Source: GiveWell

Training is usually done either by doing a PhD in biomedical science, or by doing a medical degree. As we explain below, it’s also possible to enter without a PhD in biomedical science if you have strong quantitative skills.

Note that you don’t need to study Biology undergraduate to enter many biomedical PhD programs – it’s possible to enter with other science or mathematical subjects.3 There are also masters programs that aim to help science students transfer into biomedical research, such as CoMPLEX at UCL.

After training you usually start as a postdoc, and work your way up to professor level. Junior researchers spend most of their time in the lab, while senior researchers spend most of their time managing, writing papers and speaking to other researchers. More on what it’s like at the different stages.

How did we research this career?

We interviewed three leading senior biomedical researchers in the UK: Prof. McMichael and Prof. Townsend1 (both work on vaccines for infectious diseases) and Prof. Todd (who works on the genetics of disease). We also interviewed Dr. Ewer, a mid-career researcher working on vaccines, read GiveWell’s interviews on biomedical research, interviewed Holden, the co-Executive Director of GiveWell, by email, read the most relevant challenge paper from the Copenhagen Consensus, and did a web search for existing advice.

Why should you do biomedical research?

Promising cause area

Overall, our view is that biomedical research is a promising cause.

Large historical returns

There have been some economic analyses attempting to estimate the returns to biomedical research.4 Many of these conclude that the historical returns have been very high compared to costs, and that future returns are also likely to be very high. According to one estimate, the prevention and treatment of cardiovascular disease in the US in the 1970’s and 1980’s alone had $31 trillion of associated gains.5 This is on the order of 60 times as large as all spending on medical research over the period.6 A recent review concludes that:

“Overall, there is strong evidence that new biomedical technologies have created significant value, as measured through the economic value of health improvements.”

However, these estimates have been criticised and so shouldn’t be taken at face value, and may not reflect returns from future work.

Potentially high returns at the margin

Our guess is that biomedical research offers large potential returns from future work. It seems like very valuable improvements in health are possible with comparatively small amounts of investment. For instance, GiveWell has done a first pass forward-looking cost-effectiveness analysis of cancer research concluding that:

“The estimate is ~$2800 per life-year, which is substantially worse than our estimate of ~$80/DALY for LLIN [malaria bed net] distribution, but not so much worse as to make it implausible that well-directed scientific research (as opposed to “the average dollar spent on cancer research”) could have greater (even substantially greater) benefits.”

Another analysis estimates that a 1% reduction in cancer mortality in the US would be worth $500 billion (the current research budget is around 5 billion).7
Biomedical research also contains some very exciting areas of research that could offer enormous upside, for instance anti-ageing research and synthetic biology.

Other groups in global priorities research agree

Biomedical research is also a priority area being investigated by the Open Philanthropy Project, which we think is one off the best groups that looks for causes with opportunities to accomplish as much good as possible – suggesting there may be opportunities to have outsized impact here.

Constrained by good researchers

Senior researchers we’ve spoken to have said that really good researchers are rare, have vastly more impact than average researchers, and that they’d be willing to turn down large amounts of money if they could get a good researcher for their lab. For example:

“The best people are the biggest struggle. The funding isn’t a problem. It’s getting really special people. I call them the one percenters…If you have a good person, it’s easy to get the grants for them.”
“One good guy can cover the ground of five, and I’m not exaggerating”
John Todd, Professor of Medical Genetics at Cambridge

The fact that lab leaders would be willing to forego large amounts of grant money for a single good researcher suggests that (i) more grant money isn’t a very effective way to attract top researchers, and (ii) top researchers are more valuable than the other things that lab leaders can buy with grant money (like lab equipment). Although we expect this to overestimate the extent to which the field as a whole is constrained by good researchers,8 it suggests that if you have the potential to be a top researcher, contributing to this cause through direct work is likely to lead to more impact than contributing to the cause by donating.

Satisfying work

Being a biomedical researcher can be highly satisfying because you get to:

  • Work on interesting, challenging problems with intelligent people
  • Have high autonomy. Through most of your career, the work you do is largely self-directed. In the earlier stages, you’ll be put on projects by your supervisor, but even then you’ll be highly independent day to day and probably able to have a substantial say over what you do (though this varies by lab).
  • Have a lot of variety. Biomedical researchers investigate many different questions, use different techniques and technologies, and collaborate with people around the world.

However, it’s weaker for job satisfaction in that many projects take years to bear fruit, which can be less motivating. We’ve also heard that during your PhD the work can be quite monotonous (like counting fruit flies).

Drawbacks of biomedical research careers

High drop-out rates

Biomedical research is highly competitive, and most people don’t make it to tenured professor roles. In the UK, only 3.5% of people with a science PhD make it to permanent research positions in academia, and only around 20% end up in any sort of research roles.

Source: The Royal Society Note that “permenant research staff” in the UK is similar to “professor” in the US – the proportion of professors in the UK is much lower.

In the US the picture is similar. The percentage of people with tenure track jobs five or six years after their PhD has been steadily falling:

Source: The Atlantic

This trend is likely due to the increase in the number of PhDs being produced, without a corresponding increase in the number of tenure track positions.

The field is also notorious for long hours, especially early in the career where there is fierce competition for jobs.

Moreover, biomedical researchers can get pushed out even in their forties:

“It can leave people a bit stranded mid career. You start out well, but you don’t quite make it to the top. You’re on a 3-5 year contract. You find it doesn’t get renewed. You’re 45 and stranded.” Prof. Sir Andrew McMichael

Long time to train

There are two main routes into biomedical research, both of which take many years:

  • Medical degree – you first train as a doctor, and then you transition into research, often by doing a PhD after medical school whilst doing your junior medic training. If you take this route it can take 10-12 years before you’re fully qualified in medicine and able to run a research program.
  • PhD in biomedical sciences – you go straight into a PhD, which takes 4-6 years, and then move into academic research roles.

Relatively narrow exit options

If you’ve taken the PhD entry route, this path is not particularly good at broadening your career options, because the career capital you gain is mainly relevant to careers in medicine and biology. The most common exit options are academic roles focusing more on teaching and/or administration than research, roles in allocating research funding and public health. If you’ve built quantitative analysis skills, there is also the option going into data science.

If you’ve taken the medical degree route, then you’re in a much better position, because you can go back to medicine, as well as the options above.

Another potential positive is that biomedical researchers may be able to use their expertise to reduce the risks from natural pandemics and dual use research.

Who should consider this option?

You should consider this option if you think you have a high chance of being a top researcher. This is both because of the funding landscape (the senior researchers we interviewed said that good researchers can generally get funding, but it’s significantly harder for mid-range ones) and because good researchers are likely to have a lot more impact. This is because:

  • Some researchers publish scientific papers at a rate at least fifty times greater than others, and the distribution appears to be log normal.9 This is what we would expect if publishing papers was a multiplicative function of several independent skills.
  • The distribution of citations is very peaked: the top 0.1% of papers have 1000 citations, compared to ~1 citation per paper at the median. The citation distribution may be more skewed than the true distribution of impact per paper, because it’s exaggerated by feedback effects in which one paper becomes the standard paper that everyone cites. Nevertheless, we suspect the basic point that some papers are vastly more influential than others remains.
  • The senior researchers we interviewed all said that a ‘good researcher’ was rare and valuable. Prof. Townsend used the phrase “worth their weight in gold.” Prof. Todd said “One good guy can cover the ground of 5, and I’m not exaggerating.”

What traits are important for success?

All our interviewees mentioned that having intense intellectual curiosity is crucial for succeeding.10 Having a lot of grit was also frequently mentioned – you will likely face many setbacks in research, over many years, while working in the highly competitive environment of academia. Many of our interviewees also mentioned that you have a better chance of doing well if you know programming, maths or statistics, because these skills are highly in-demand.11

Another key trait is being willing and able to be highly strategic in positioning yourself for career progression, not just excelling at research. Successful researchers optimise heavily around building strong collaborations to make it easier to get funding, publishing large numbers of papers in top journals and working at the most prestigious labs to establish their credentials.12

The interviewees also mentioned intelligence, which is important in most roles and we’d expect to be especially important in a complex research role. Indeed, studies have shown that IQ is a predictor of success even at the highest levels of science.13

How can you work out whether you’ll fit this career?

  1. Assess yourself on the traits above. In particular, try to think of concrete situations in which you have demonstrated these traits before. For instance, when doing science experiments in school, did you enjoy the times when things went wrong or were difficult and unclear? Did you find yourself motivated by curiosity (rather than simply by success) and unfazed by uncertainty?
  2. Speak to several people in the field and read extensively about the nature of the work, particularly in your areas of interest.
  3. Do a summer placement in a lab. This was suggested by several people as one of the best indicators.
  4. After your PhD, reassess your chances of success, including on the concrete signs looked for by lab managers below.

What concrete signs are looked for when entering the field after your PhD?

The senior researchers we interviewed look for:

  1. A good degree class from a top university
  2. Several publications: First author on a Nature paper would be very good, second or third author on a Nature paper and lead author on a few others would be more common.
  3. A strong reference: This is a useful differentiating factor, especially because they normally know the referee.
  4. Statistics, programming, maths and technical skills that are especially needed by the lab (e.g. imaging at the moment), or other relevant hands-on experimental experience.

Should you do a medical degree or a PhD?

Among the people we interviewed, there was no clear consensus. Medicine offers better backup prospects and is more motivating to some, but the PhD route is faster, allowing you to get into research more quickly.

MedicineBiology PhD
Time to qualify10-12 years4-6 years
Exit optionsCan exit into practicing medicine and public health; easier to exit into jobs in pharmaceutical industryMain exit options are academic roles focusing on teaching and/or administration, and allocating research funding
BreadthBroader training may give you more ideas for research topicsMore focused on research
FitWorking with patients may increase motivation for researchBetter suited for those who want to focus entirely on research and prefer not to work with patients
PayPotential for higher salaries if you practice medicine whilst doing researchAcademic salaries

The more committed you are to research, the more the PhD route makes sense compared to the medical path. If you think you’d find either path motivating, then we’re tempted to lean towards the medical path for the increased flexibility. In the future, we want to investigate the prospects of entering from PhD’s in other sciences and applied mathematics.

Do you need to do a PhD at all?

We’ve been told it’s possible to go into research roles without a PhD if you have programming or statistics skills, because these skills are highly in demand. Depending on your background, it may be possible to enter directly into informatics and data analysis roles. Otherwise, something like a Masters in Bioinformatics could be a good path.

If you’re interested in biomedical research, therefore, learning to program seems like a good step, since it keeps your options open in lots of other areas too (learn more in our software engineering profile). The same goes for data science.

Further reading

Further issues to investigate

  • How good are the opportunities for biomedical research outside of academia?
  • How promising is entering biomedical research from PhD’s in other sciences?

Notes and references

  1. Our interview with Prof. Townsend is not published.
  2. “(A) – (C) are generally associated with academia, while (D) – (F) are generally associated with industry.“ GiveWell Blog - The Path to Biomedical Progress
  3. For example MIT’s admission page: “Does it matter if I'm a chemist, physicist, or mathematician rather than a biologist? Yes, it means that we're really interested in you! Our program is designed for students with diverse backgrounds, and students who have majored in chemistry, physics, and mathematics as undergraduates have done extremely well here.” MIT Department of Biology: FAQ About Applying.
    Also Harvard’s admission page: “To qualify for admission, applicants must demonstrate strong enthusiasm and ability for the vigorous pursuit of scientific knowledge. Minimal requirements include a bachelor’s degree and undergraduate preparation in the sciences.” Harvard Division of Medical Sciences: Admissions
    UCL’s CoMPLEX is a centre for interdisciplinary research in the medical and life sciences, which accepts students with backgrounds in mathematics, physics, computer science and engineering.
  4. “Increases in life expectancy in just the decades of the 1970’s and 1980’s were worth $57 trillion to Americans – a figure six times larger than the entire output of tangible good and services last year. The gains associated with the prevention and treatment of cardiovascular disease alone totaled $31 trillion.” Exceptional Returns: The Economic Value of America's Investment in Medical Research
  5. “Murphy and Topel estimate that the total economic value to Americans of reductions in mortality from cardiovascular disease averaged $1.5 trillion annually in the 1970-1990 period. So if just one-third of the gain came from medical research, the return on the investment averaged $500 billion a year. That’s on the order of 20 times as large as average annual spending on medical research – by any benchmark an astonishing return for the investment.” Exceptional Returns: The Economic Value of America's Investment in Medical Research
  6. Estimates of Funding for Various Research, Condition, and Disease Categories
  7. One complication is that top researchers are also better at attracting grant money than average researchers. This means that an individual lab leader would be willing to forego large amounts of grant money for a top researcher for at least two reasons:
    1. A top researcher would lead to more and better research than the grant money.
    2. A top researcher would attract more grants for the lab in the future.
    But because individual lab leaders are competing with other labs for a fixed amount of grant money, this means that the extra money a lab gets as a result of having a top researcher is money that would have gone to other labs instead. This is a cost that lab leaders are likely not taking into consideration, and so the amount of grant money they’d be willing to forego for a top researcher is likely an overestimate of their true value.
  8. Shockley, William. "On the statistics of individual variations of productivity in research laboratories." Proceedings of the IRE 45.3 (1957): 279-290.
  9. For example, Dr Ewer said: “Unless you have an absolute obsession with your subject, it’s very hard to persevere and not become demoralized.”
  10. For example, Prof. Todd said: “The MD and programming/statistics combo is lethal. Top of the world. There’s major demand.” “All the kids need to learn to program and understand statistics. The data sets are getting bigger and bigger, and better. You need to learn to move data around. You need to avoid being fooled by randomness.”
  11. “...you want to go to the best labs in your general area – neuroscience, immunology, cancer, cell biology. You just need to go where the best science is – the most Nature papers, the most Cell papers.” Interview with leading HIV vaccine researcher – Prof. Sir Andrew McMichael ““Get into a really really good lab (in a major centre). (e.g. LMB in Cambridge) Then work on a really difficult important problem for 10 years. Try to get long-term funding.” “To get the first step, you’ll need to initially focus on publishing in a couple of top journals. There’s a lot of hard work and luck involved; always be asking the question, if I get the answer to my big question, how many people in the world will care?” Interview with a Cambridge Professor of Medical Genetics on research careers - Prof. Todd
  12. One study of 64 eminent scientists found that their median IQ scores were above 150. See also Lubinski, David, et al. "Top 1 in 10,000: a 10-year follow-up of the profoundly gifted." Journal of applied Psychology 86.4 (2001): 718.