Skip to main content

Academia (vs. Industry)

Clinical Trials: NIH and the Drug Industry

JAMA has an article out on who funds and conducts clinical trials. It shouldn’t surprise anyone  (although it does seem to be news to some headline writers) that industry funds the great majority of them. What does stand out is that the proportion funded by industry is increasing, although that shouldn’t necessarily be news, either. (If you don’t have access, this article has some of the data).

That change is happening in both the numerator and the denominator. From 2006 to 2014 (the period covered by the article), the number of trials conducted by the NIH went down from 1376 to 1048, a decrease of about 24%. During that same time, though, industry-sponsored trials went up from 4585 to 6550 (up 43%). The first thing to note is that this is definitely not a zero-sum situation – the total number of clinical trials is up about 27%, but all of that rise is due to the biopharma industry. But the concern is that NIH-sponsored trials can (should) be more disinterested. An industry-sponsored trial is more likely to be answering the question “Does our new drug X actually work?”, and that’s certainly a good question to ask. The NIH is better positioned to ask questions like “What seems to work better, Drug X or lifestyle intervention Z?”, or “Which of these generic-drug-based therapeutic regimens works the best?”

The usual gripe about industry trials is that they’re against placebo, and thus harder to compare to each other. But there are a lot of trials against “standard of care” – you don’t run a placebo-controlled trial in diabetes, obviously, and you don’t run one in a cancer indication if there’s any sort of therapy at all. But when you have several different mechanistic choices in a given disease area (diabetes and hypertension are two notable examples), then good head-to-head comparisons may be harder to find, and that’s where NIH trials can be very useful as well.

So why is the NIH running fewer trials? Well, their budget has been drifting downward over that period. The NIH had some of its biggest funding increases ever under the first term of the Bush-43 administration, peaking in 2003 (a fact that you could probably win some bar bets with, if you’re in the right sort of bar). The only comparable period was the late 1950s and early 1960s, in percentage terms. The funding wandered down after that, but was still far higher by 2008 than it had been in 2000 (under the Obama administration, it’s been flat or down every year (see figure 2 in this link). So starting in 2006 would mean that you might well be catching the agency at or near its all-time clinical trial peak – that’s enough time for the peak funding years to have worked their way into the clinic. It’s important to not look back at 2003-2006 as a period representative of how things had always been, because that was an era when the agency had just gone through its biggest expansion ever.

I should note, as an aside, that the Figure 2 that I linked to has a deeply stupid trend line attached to it, labeled “pre-austerity growth trend”. That steeply ascending line should be called “growth trend if the historic Bush-era NIH budget increases had continued every single year thereafter”. If you pick the 1980-2000 budget line, though, and extrapolate out, you actually get to where the NIH budget is today, although I’m sure that this would be a deeply unpopular line of argument at the agency and with those that it funds. Looked at this way, the Bush-era increase has turned out to be an outlier – the second Bush term and both of Obama’s have, in effect, let the NIH budget settle back to where it would have been by now had that rise never taken place. (The problem, though, is that the agency’s funding does not look set to begin rising again at even its former rate, much less the rocket launch of 2000-2003).

What would be interesting would be a chart of the total number of NIH-sponsored clinical trials going back to the 1990s, but I’ve been unable to find those numbers so far. My guess is that these also experienced a big jump, a couple of years delayed from the budget increases, and have been trending downward ever since (the latter part of which fits the JAMA numbers). Keep in mind, though, that clinical trials themselves are (and have always been) a small part of the agency’s activities.

So overall, I’d guess that the NIH’s clinical presence has probably tracked its budget pretty closely, whereas industry is definitely spending more. We should be glad about the number of clinical trials going on, absent more information about just how well-thought-out all of them are.

11 comments on “Clinical Trials: NIH and the Drug Industry”

  1. Hap says:

    Aren’t the average costs of trials increasing as well (even above the level of inflation that you would expect)? This would imply that NIH would be losing trial strength in any case, excepting the doubling years. (It also is part of what doesn’t make sense about the “whack R+D” theory of pharma cost cutting – unless you know that you can’t get any useful information or ways to lower trial costs or failure rates from R+D, you need information to make trials less costly, and since they’re the biggest expense, that’s where the money is to cut, mainly.)

  2. Glen says:

    Is there any evidence that the major insurance carriers are sponsoring comparative studies? This is an area they have an interest, and a vast amount of data that can be usefully mined in order to help the studies.

    Glen in Texas

    1. MTK says:

      Good point, Glen.

      I’ve always been somewhat surprised that the insurance industry doesn’t sponsor comparative trials more.

      Maybe they have their own ways of data mining through meta analyses or whatever which they’d rather believe than to spend money on clinical trials.

      1. Drug Developer says:

        The “insurance industry” sponsoring comparative trials? It would seem to be in their collective interest, but I don’t know of a vehicle for this. And individual insurance companies don’t seem inclined to foot the bill for expensive Phase 4 studies themselves.

  3. ab says:

    Do we have any data on the numbers of patients or dollars spent in each set of trials? I suspect if we factored in the size of average industry-sponsored trial vs. NIH-sponsored trial we’d see an even more dramatic skewing of the numbers. I’m not certain that’s true though. There are a lot of small pharmas trying to get some P1 data. But I’d still argue the dollars spent or number of patients data is more interesting.

  4. Small Biotech says:

    While your discussion here focused on budget, there are also changes to the kind of trials being run. Recently there was a strategic shift at the NHLBI to fund more larger, landmark-ish trials (i.e. SPRINT), in lieu of smaller trials. So this may decrease the trial count number but perhaps elucidate more interesting biology.

  5. Vaudaux says:

    Worth remembering that there are government agencies other than the NIH running clinical trials. The one I’m thinking about is BARDA, an agency that (like NIH) is part of Health & Human Services – they are currently running Phase II and Phase III studies of new antibiotics.

  6. Mike P says:

    Looking at the data in Table 1 in the article, what I find interesting is the dramatic increase in the number of clinical trials under the “All others” category which seems to include individuals, universities, and private organizations. From 2006 to 2014, the number tripled from 3240 to 10597. What accounts for this increase? I don’t have access to the full article, so I don’t know if it is discussed in there or not.

  7. zero says:

    More and more, it seems to make sense for a central clearinghouse for clinical trials. An organization that recruits and compensates volunteers, executes the trials, reports all data to all stakeholders and is not under the same potential bias as industry studies. Stakeholders like pharma, biotech, insurance companies, academic institutions and the federal government would contribute funds and in some cases manpower and facilities. The methods would be peer-reviewed and statistically sound from the beginning, plus transparent enough that any deviation would be caught. Negative results would be published rather than buried. A stable source of jobs in the field would be maintained.
    Nothing would prevent a group from doing their own trial, but they would have a significantly higher bar to clear if their design was so bad that it was denied a free review.

  8. DrJimbo says:

    Trials comparing different active treatments (rather than comparison to standard of care) would typically need to be larger – perhaps much larger – to generate meaningful results. If, as Small Biotech suggests, the NIH is shifting to this sort of study, the same budget will fund many fewer trials.

  9. gippgig says:

    Are there any citizen science based trials? This might be a good way to compare approved drugs or test an off label use (there was at least one do-it-yourself trial of existing antivirals back in the early days of the AIDS epidemic).

Comments are closed.