Meaningful Translational Research

Today’s issue of JAMA has a commentary on the meaning of translational research by Steve Woolf, one of my favorite people and someone I respect tremendously. He is concerned at how few resources are devoted to T2 (bedside to community) translation. He knows that T1 (bench to bedside) translational research is more expensive due the the infrastructure that must be supported and the crap shoot involved in such research. T1 comprises everything from histone ubiquitylation to Phase III clinical trials plus all in vitro and in vivo work in every conceivable species between. Without the T1 part of the equation, there is no T2. Plus, we need the hope made possible through and the excitement generated by T1 research.

But I work both sides of the aisle, so I understand Steve’s frustration on the T2 side. Personalized medicine to a family doc does not mean taking blood to run a genomic profile (see commentary on why not in NEJM) … it might mean automated alerts in a patient’s electronic health record regarding the need to ask about dietary salt/sodium intake, side effects from a recently prescribed drug, or Hemoglobin A1c screening; for the pharmacist, perhaps an alert that a prescription has not been filled on schedule. More difficult is how, upon learning the patient with diabetes does not routinely check blood sugar, to motivate such behavior. How to get a morbidly obese 12 year old to eat sensibly and become more active when nothing in the social environment supports adherence to this advice. How to convince politicians to ban smoking in public spaces (Viva La France!). The solutions seem so common sense that funding them does not.

T1 research keeps getting funded because it generates measurable, reproducible, statistically significant, and scientifically valid results. It succeeds, which in turn allows NIH ICs to justify their budget requests. T2 efforts all too often … do not. And they’re hard to distill into an attention-grabbing sound bite (“Impact of Medicare Part D and Racial Disparities in Treatment and Outcomes for Hypertension”). Sometimes the T2 side of the equation does come out ahead, such as NYC’s impressive success in reducing smoking rates among teens and adults (and commentary as to why Connecticut does not do as well). As the disease-of-the-month lobbies have learned, the key is to capture headlines, hearts, and then pocketbooks and budget line items. This is easier when you can say you’re funding work to cure disease XYZ than when you say you’re funding a Markov decision processes model for optimizing statin start times for patients with Type II diabetes.



  1. whimple said

    This one is easy. T2 IS NOT RESEARCH (not biomedical research anyway). T2 is social engineering. We already know what to do, T2 is all about implementation. Calling T2 research only serves to confuse the issue of what research is and does a disservice to both T1 and T2.

    I’m also going to quibble with your definition of T1. To my way of thinking (which may not be the consensus view), unless the work directly involves a (possibly until-recently) living breathing human being, it’s not translational research. I am totally unimpressed with arguments along the lines of, “if only we can repair this damage and/or cure this disease in this critter, then we can also fix people”. What I see is a whole pile of critters, a bigger pile of cash, and ZERO work on humans. If the work isn’t actually being moved directly into humans (no, not just used-to-be-human-50-years-ago cell lines), it’s not translational.

    While I’m at it, let me bitch about the first sentence in the JAMA article:
    Translational research means different things to different people, but it seems important to almost everyone.
    I would add, “everyone, that is, except the R01 study sections at the NIH.”

  2. drugmonkey said

    “We already know what to do, T2 is all about implementation. Calling T2 research only serves to confuse the issue of what research is and does a disservice to both T1 and T2.”

    Ok, whimple, a few questions for you. Feel free to answer with how you know the answers, not just your seat of the pants assumption.

    Do “Just Say No”, “DARE” or cutesy anti-drug jingles work to keep adolescents from experiencing adverse consequences of drugs of abuse? Which one works best and deserves broad scale implementation? What best practices can we derive that could guide any approach to avoiding problems associated with drugs of abuse in adolescents?

    What is the best community level strategy for keeping acutely detoxified addicts from relapsing to drug abuse? Or is this a fool’s errand and we are best off spending our money on pharmacotherapeutic development?

    What information is best provided to the general public to keep adverse consequences of recreational drug use low? Prohibitory messages? Realistic facts on “addictive potential”? “Scared straight” type messages? Libertarian ones? what works?

  3. whimple said

    None of those are biomedical research questions. They’re all sociological in nature. That some of them may have superficial connections to biomedical research is not relevant. They’re largely equivalent to asking “how can we get people to wear seatbelts?” and “what can we do to reduce shoplifting?”. These are questions worthy of research, but certainly they are not biomedical research and they shouldn’t pretend that they are. Sociological research is a valid field of study, but there is no added utility to calling it “translational”.

  4. whimple said

    Actually, I take it back. I think these are translational, but I’m not sure they fit the definition of “bedside to community”. Isn’t “bedside to community” about distribution/implementation of results? It’s not obvious to me. Maybe these examples are really T1. Part of the problem is a definition of “translational research”. I’d go with something like:

    1) involves living breathing humans


    2) asks the question, “if we do X, is there a net benefit to those humans?”

    Translational research is hard to do because both of these get whammied. 1) requires enormous administrative overhead. IRB’s need to be approved, participants need to be recruited and consented etc. 2) is hard to get funded. If the answer to the question of benefit turns out to be “no”, then a case can be made (erroneously I feel) that resources have been wasted.

    But, look at what happens when 1) and 2) both succeed: there is a direct benefit to living, breathing humans. That’s what the taxpayer wants. This doesn’t happen if the research doesn’t involve both factors 1) and 2).

    The question is, since research meeting criteria 1) and 2) is harder to do than research that does not meet these criteria, what can be done by way of compensation, to convince people to actually bother attempting translational research?

  5. drugmonkey said

    I think you can boil the definitions down to :

    there is a direct benefit to living, breathing humans. That’s what the taxpayer wants.


    although we might extend “wants” to “understands”. Precise definitions of what “translational” means are a bit of a distraction.

    I see the whole T2 thing as just being research intended to move beyond the individual case-by-case patient. Very broadly speaking.

    In drug abuse in particular there is a clear role for this sort of thing. In fact, one might argue that it is our best success story in terms of public health application of all drug abuse science to date.

    you seem to feel that learning about individual behavior, motivations, compulsions, mental disorder, decision making and the like is “sociology”. To me it is basic behavioral neuroscience. and once we know something about this we can generate hypotheses about population-level interventions to shift behavior. that requires what I would describe as T2 type efforts.

  6. PhysioProf said

    “In fact, one might argue that it is our best success story in terms of public health application of all drug abuse science to date.”

    Is there really anything that one can do from a “marketing” standpoint (“just say no”, “DARE”, “try to scare the shit out of kids by making them think they are going to kill their friends with guns if they smoke a joint”) that has any effect on how much teenagers do drugs?

    Ronaldo Martinez scared a lot of people, including kids, in NYC into giving up tobacco products. – writedit

  7. writedit said

    Can’t leave yuns alone for a minute, I see. I assume Whimple won’t be enrolling in a PhD program in Translational Science (or Research) at a CTSA site or wannabe institution any time soon. And indeed, our SVC says everyone at all the health sciences schools is now a translational researcher by default – ready, willing or not. Sort of makes the term meaningless and not worth quibbling – or wibbling – over.

    What’s missing from this discussion, and Steve’s commentary, is the backward translation, or perhaps reverse engineeering. The T2 community-based researcher (bear with me Whimple) or epidemiologist who sees a pattern of behavior, symptoms, side effects, etc. at a population level and reports this back to the T1 folks … whether a clinician investigator who can study the phenomenon in a controlled research setting (GCRC, eg) or the biomedical scientist who recreates what is seen in society in an animal model or cell culture to tease out the underlying mechanisms and come up with some answers for the referring clinicians to offer patients (or communities, if it turns out to be an environmental toxin … or an especially addictive/lethal drug combo interaction).

    I see this happen here, and it is really cool to watch unfold … and all the participants get excited by each player’s contribution (including the patients & referring MDs/NPs/et al.). As an example, an MD-PhD recognized atypical patterns of cardiomyopathy in clinic (routine cardiology – not research center or study), traced family trees characterized by this odd pathology, got DNA from said trees, worked through the statistical & molecular genetics to identify the candidate genes (primary causative & modifier), looked at what proteins were encoded, created knockout mouse models, nailed the pathogenesis, and is directing the development of highly targeted therapeutics (since nothing currently exists) to take back to the patients (locally & across the US) who set off the light bulb in the first place. All this and an incredibly pleasant, well spoken, & modest physician scientist.

  8. writedit said

    And I’m waiting for bikemonkey to jump in with a cynical commentary on the gigundous biomedical research economy that needs (is addicted to) T1ers to keep developing patentable lucrative new drugs & tests & devices, the profits of which in turn fund the marketing … T2 … economy.

  9. drugmonkey said

    pp, not really my area so i’m not steeped in the lit. however.

    point the first: len bias. a massive jump in perceived risk and decrease in incidence for cocaine (MtF data) thereafter. trends of a magnitude not observed for any other drug, except perhaps…

    point the second: the MDMA effect right around 2002 when perceived risk shot up (high schoolers) with an associated dropoff in incidence. i’ve hypothesized that the hoopla surrounding the Ricaurte affair is at root. on the any-press is good-press rather than accuracy of his findings. despite this, i will say that i still get neurosci undergrads (and grads!) asking “didn’t michael j. fox get parkinson’s from ecstasy” so one cannot discount inaccurate (if “useful”) communication…

    point the third: general downward trend for most drugs of abuse starting in mid eighties along with spool up of just say no, DARE and the like. indirect to say the least as there are economic and other trends to consider. as i said, not my area

    point the fourth: I have the slightest brush on those who do research this stuff. meaning i view a poster or six during a relevant meeting. and they all do small-ish scale interventions that the data suggest have effect. shrug. these T2-ers think they are doing science so there you have it.

  10. PhysioProf said

    I’m with you in concluding that it is “biomedical science”. My definition of “experimental biomedical science” is simply, “Doing something to a biological system and seeing what effect it has”.

  11. whimple said

    Per the JAMA article:
    US patients receive only half of recommended services. The second area of translational research seeks to close that gap and improve quality by improving access, reorganizing and coordinating systems of care, helping clinicians and patients to change behaviors and make more informed choices, providing reminders and point-of-care decision support tools, and strengthening the patient-clinician relationship.

    This is biomedical science? This is what they’re talking about with T2 research.

  12. drugmonkey said

    “helping clinicians and patients to change behaviors and make more informed choices”

    yep. biomedical science.

    but perhaps to put a finer point on it, it is not the U.S. “Biomedical Science” Service nor the National Institutes of Biomedical Science. US Public Health Service and National Institutes of Health last I checked.

    so in some senses hair splitting over what is/is not “biomedical” (is it “bio+medical” or “bio/medical” by the bye?) is beside the point is it not?

  13. Steph said

    Hmmm… Add T1 to T2 and you’ll get the biomedical formula to the marketing standpoint of most of the scientific community. It’s science people!

RSS feed for comments on this post · TrackBack URI

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: