8 ways that we’re incentivizing scientists to stop innovating
By 
Shan-ul-Hai •
April 29, 2008
In today’s world, a culture of innovation is integral to advancement at the same rate that we’ve seen in the past century, especially in the medical field. Unfortunately, some scientists have done a few unethical things in the past; because of this, we’ve decided to severely restrict all scientific research. Here is how we’re giving our scientists incentive to restrict themselves:
- The members of Institutional Research Boards (IRBs), which supervise all human-related research, are often (but not always) lawyers and not scientists. This gives them an incentive to reject anything that they think might become a legal liability, without any knowledge of the scientific background. For example, one of my projects was recently rejected because nobody had ever done it before, so they couldn’t be sure that it was safe (despite the fact that any knowledgeable doctor/scientist could tell you that there were no risks involved). In summary: they wouldn’t let me do it because nobody has done it before… if that’s our attitude, then how will we ever move forward?
- The IRBs are deathly afraid of interfering with patient care. They once rejected my research proposal because I needed to analyze the patients’ snot (yes, their snot) . Normally, the nurses just throw it away… but I couldn’t get IRB approval to analyze the trash.
- As I wrote previously, the current peer-review system, because it is run by private journals instead of a public library (which the National Institute of Health is working on, but they don’t have enough political support), takes months or years to make “novel” research available to the scientific community. This slows down every process, since my research (which takes a year to publish) is usually based on somebody else’s research (which takes a year to publish), which is based on somebody else’s research (which takes a year to publish), and so on.
- Because of this slowdown, researchers have a strong incentive to wait until they have a lot of data before they publish any of it. This means that a very simple experiment must be repeated many times before it can be shared with other scientists; I have my name on one important paper, for example, that was written in 1997 but still hasn’t been published because the lab doesn’t have the resources to repeat one insignificant (but expensive) experiment. If we could just put our data into a public library of science, then other scientists could use it in their research… but the private journals want control over all information.
- This slowdown also causes many different scientists to perform the exact same experiment because they don’t know that anybody else is doing it.
- The National Institute of Health and the National Science Foundation are always low on manpower and money. This is because to train a good scientist, it takes about 12-16 years (4 years of undergrad, 4-6 years of grad school, 4-6 years of postdoctoral fellowship) and they end up getting paid less than a person with an MBA, which is much easier to get. If we raised scientists’ salaries, we’d have much more incentive for people to go down that track… but that would require more tax money.
- Pursuit of tenure is the main incentive for young scientists to succeed in their research. Unfortunately, most universities only offer tenure to people who publish a lot of papers and the quality of the research is less of a factor. This causes scientists to spend their effort on easy, quick research instead of slower, more important projects.
I’ll cite the example of my boss: after he became a professor, he came up with a great idea and decided to pursue it. He published a few groundbreaking papers and filed a patent, but was denied tenure because he failed to publish the minimum of 12 papers in his first 7 years. Now, his patent has brought the university $10 million in the last three years (no other patents even brought $500,000 last year) and will bring at least another $40 million over the next 10 years. If he’d pursued the expected plan instead of trying to get his invention off the ground, that $50 million would have ended up going to a drug company.
Because of those first 7 years, he is no longer eligible for tenure. - Most researchers are always competing with each other for grants. For this reason, they have a strong incentive not to share their research with other scientists until it’s published… they’re afraid of their ideas being stolen.
Most of these problems are caused by rules and regulations. The rules are written like that because they want to avoid any possible legal/ethical problems and because they want to save money wherever possible. Tomorrow I’ll write about my proposed solutions (I promise that tomorrow’s post will be shorter).
EDIT: Click here for the post with the proposed solutions.
Popularity: 24% [?]
| 2.5 |
If you enjoyed this post, make sure you subscribe to my RSS feed!
[…] liberalbaiter wrote an interesting post today onHere’s a quick excerptUnfortunately, some scientists have done a few unethical things in the past; because of this, we’ve decided to severely restrict all scientific research. Here is how we’re giving our scientists incentive to restrict themselves: … […]
[…] thelibraryspace wrote an interesting post today onHere’s a quick excerptIf we could just put our data into a public library of science, then other scientists could use it in their research… but the private journals want control over all information. This slowdown also causes many different scientists to … […]
And I thought that the IRB would have a scientist who can judge the ethicality (if thats a word) of the study. but wow. and yeah about scientists competing for grants ahh that makes it much worse for the scientists. As one scientist said ” that they can be using their mind to figure out a solution instead of how to get money for the research” if there was no money…
So much for “If I suggest something, I’ll back it up with facts”–
I will comment on this point by point:
1. Are you only referring to a specific IRB? Have you seen a list of members of an IRB? The vast majority of the members are PhDs/MDs/Masters/etc. Look, for example, at Stanford, Louisville, UC-Irvine, and UT-Austin, who publish their membership online.
2. Sounds like this personal matter has biased your overall opinion. For example, look at this exciting counterpoint to your example that is now underway http://www.mc.vanderbilt.edu/reporter/index.html?ID=4103
3. As I said in a previous comment, if your research is truly ground-breaking or life-saving, it can be published rather promptly
4. Again, a rather rare and personal situation. In my experience in several diverse settings, I have found that most experiments do not require lots of unnecessary repetition to validate the results
5. Most research these days is so narrow that only a few labs worldwide may be looking at the exact same question. Furthermore, the truly exciting research occurs when questions are approached in a completely new way (i.e., no one else has thought about it)
6. I generally agree with this, although I meet quite a few newly minted MBAs from non-brand name schools these days who are having a tough time finding good work.
7. I disagree with this statement. From my experience, the researchers who are leading the way are those who are truly in it for the science and curiosity, not for the fame or money or tenure. They know that if they can perform excellent research, the increase their chances of tenure, but that is not their primary motivation. Perhaps your boss is more suited for industry than academia. Generally speaking, it’s the academics who publish papers and companies that file patents.
8. Again, most labs’ research strongly overlaps with only a few other labs in the world. (In fact, a good way to get a grant is to expose the need for more research resources to examine a specific question.) Most research is extremely focused, such as studying fruit flies to investigate odorant receptors (Nobel Prize winning, at that).
Thanks for the suggestion… I always thought that personal experience counted as “facts”, but I guess that’s a bit of a gray area.
1. I’ll concede that point… since I’m speaking from personal experience, I can’t speak for all IRBs. I’m referring to a specific one, and I’m sure there are plenty of examples on both sides of the fact.
2. Again, I’ll concede that I was generalizing… but just because you can find one counterexample doesn’t make it a “counterpoint”. I think we can find plenty of examples on both sides of the issue, which means that it’s a problem that needs to be addressed.
3. Not all important research is groundbreaking. It sounds like you are a scientist, in which case you would know that every groundbreaking paper cites at least 20 smaller papers. Sure, the biggest news will get into the journal quickly… but what about when I am looking for an obscure piece of information that is important to my study or to one of my patients?
4. It seems curious that all of my personal situations are “rare”. I was just pointing out an incentive with this point… any reasonable person would agree that the incentive is there, whether or not you choose to use it. You’re right, “most experiments” do not require unnecessary repetition… but the problem is that since “most” of them don’t, that means that some of them do.
If this specific paper had been published in 1997, it would have saved a lot of other researchers from doing other pointless research (since the paper disproves a fact that was previously accepted by researchers in the field).
5. Yes, as you said, “only a few labs worldwide are looking at the exact same question.” You don’t see a problem with the fact that a few different labs are looking at the exact same question?
6. Yes, I also know plenty of MBAs who have trouble finding jobs. But a person who has to choose between a PhD and an MBA is usually pretty well-qualified… which means that they could probably get into a “name-brand” school. I know that I could have easily gotten into many different name-brand MBA schools, but I had trouble getting into many graduate science schools.
7. I don’t see where “fame” and “money” came in. Tenure is a major motivator for many qualified researchers, and although science is usually the first priority, the tenure system is very important to a lot of people. My boss did perform excellent research, but he didn’t get tenure.
Yes, he’s more suited for industry… but the fact is that his research has made a big change in the field. Had he done what his supervisors pressured him to do, he never would have made those discoveries.
Yes, academics usually publish papers and drug companies usually file patents. Personally, I think that’s another problem… I’d rather have more money going to academics instead of drug companies. He’s not in it for the “fame or money”, but the money that he brought in was substantial and the university can use it to buy better facilities and hire more professors.
8. I’ve noticed a lot of competition for grant funding. Again, this isn’t the norm, but it happens more often than I’d like to see.
[…] Shan-ul-Hai • April 30, 2008 Yesterday, I wrote about how we’re incentivizing scientists to stop innovating. In that post, I promised to suggest some solutions today. Here are some things we can do to […]
Well your goal of providing a “logical evidence-based” perspective ought to mean that you back things up with more than limited personal experience when making broad generalizations about complex issues. Citing statistics, studies, analysis, or experts would be more helpful evidence or support than personal anecdotes. How long have you been doing full-time research? A few years? Are you a graduate student or a PhD?
1. I am sure that there are many more examples on my side than on your side. I’m not 100% sure about the medical school, but if you look at MU’s campus IRB, it is composed of faculty members of the school’s departments as well as community laypersons, not lawyers. http://research.missouri.edu/cirb/membership.htm
It would be very strange if one MU IRB has this composition and the other one is dominated by lawyers. IRBs are required to have members who have the “experience, expertise, and diversity” to make decisions about research studies. Every IRB I have ever dealt with is composed mostly of scientists who can evaluate the ethical aspects of research proposals. http://en.wikipedia.org/wiki/Institutional_review_board
2. If you want to argue semantics by putting a lot of my words in quotes, go ahead. If you look up the definition of counterpoint, it is almost synonymous with counterexample. In any case, my example contrasted directly with yours–you cite evidence of lack of IRB support for a student proposal to analyze a few patient’s snot whereas I cite an IRB condoning analysis of hundreds of thousands of blood samples in a very well-designed and interesting proposal. Several other medical centers are going to start projects similar to Vanderbilt’s. And several institutions will begin sequencing thousands of genomes in the next few years. Places like the NIH, the Broad Institute, etc. are already doing it or planning on doing it for things like genome-wide association studies.
http://www.bloomberg.com/apps/news?pid=20601082&sid=a9FTNggspOLs&refer=canada
http://ccr.cancer.gov/staff/staff.asp?profileid=5588
I would not expect tremendous variation among IRBs of large academic centers with respect to the proposals that they accept–odds are either your IRB is evil or there is another reason your proposal was rejected. I could continue to spend time finding examples of large, exciting studies that received IRB approval, but I think I’ve made my point. Almost every issue of JAMA and NEJM these days has a GWA study. An IRB somewhere has to be approving these. It is hard to believe that IRBs “are deathly afraid of patient care” when thousands of studies have been approved to analyze patient’s blood, tissue, medical records, etc.
3. I can only go on personal experience on this one, but I have never needed to rely on other people’s obscure results for my projects. Also, a lot of the more obscure results that do not get published quickly (or at all) are presented at conferences which are often attended by many researchers. Like I stated in an earlier for your other post, it will be difficult to vastly improved time-to-publication given how busy reviewers are and the amount of time it takes for several people to review and edit the manuscript. This is the major holdup when it comes to publishing. I also wish it was faster, and I’m glad the NIH and some institutions are bucking the status quo with respect to journal publication.
4. If your point is that a small minority of research experiments require repetition for validation, then I agree; but I do not see how this would be a major incentive to not be innovative if it impacts a rather small proportion of research. Again, if your paper had been as important as you suggest, why did no one submit to a different journal? I have seen papers submitted to 3 journals before finally being accepted.
5. In my opinion, with the complexity of modern science, only a few labs ought to be looking at the exact same question with the same approach.
6. It is not as easy to be accepted to a top MBA program as you suggest. The name of the school holds much more clout in business and law than in medicine/science. For example, you’re much more likely to see a PhD or MD from Penn State out-earning a PhD or MD from Penn than a MBA from Penn State outearning an MBA from Penn. The GPA requirements are certainly not as high as for medical or law school, but business schools often look at what you’ve accomplished in your work after graduation. Therefore there is competition for admission to the top business schools.
The vast majority of MBA students have several years of work experience following undergrad. Average age at entry is 27. So 4 years undergrad +4 years work experience + 3 years MBA=11 years; for a general internist, you have 4 undergrad + 4 med school + 3 residency= 11 years. But I do agree that graduate school training in the sciences is a bit prolonged, though. Schools like having cheap labor to carry out experiments.
7. Tenure is simply a word–what is it about tenure that academics seek? It mainly involves autonomy and job security and entails more money and prestige/fame/academic standing.
8. There is certainly tremendous competition for funding. But your original assertion that this is a major reason for lack of sharing among researchers is what I disagreed with. For example, many labs may be looking at a specific biochemical pathway implicated in renal failure, but one group may use mass spec, one group bioinformatics, one group molecular biology, one group proteomics, one group therapeutic gene therapy etc. There is little disincentive to share data among these groups for funding since they are taking very different approaches. Most researchers find their own unique niche, and I therefore wouldn’t think that lack of data sharing is a major result of the intense competition for funding.
Other people have pointed out that you need to do some fact checking so I won’t reiterate too much, but the claim that IRBs ‘are made up of lawyers and not scientists’ is simply false. The rules governing the composition of an IRB preclude that circumstance. First, at least one scientist must be on the board and second the board cannot be all of the same discipline– that is, the board by definition cannot be all lawyers.
[…] Siddiqi presents 8 ways that we?re incentivizing scientists to stop innovating posted at Globally Rational, saying, “How government involvement in scientific research is […]
I guess I should rephrase. I was just speaking from personal experience; the people reviewing my IRB projects are usually lawyers and paper-pushers rather than actual scientists who understand the scope of the work.