Why the problems at CIHR won’t be solved by quick money

I previously noted that Canadian health researchers are facing a precipice of funding success rates in the near future due to the Canadian Institutes of Health Research (CIHR) having to fund three competitions in FY2017 compared to only 2 this year. Moreover, the most recent competitions experienced the lowest success rates ever for the agency (~13%). I also suggested a couple of scenarios for mitigating the effects, one of which was for the government to inject a substantive amount of new funding into the agency. Now is the time when government ministries prepare for the Spring federal budget and it is also high season in lobbying of various agencies and interested parties to influence that process. Hence, I suggested that the government would need to add something in the order of $150 million to the current CIHR base budget to rescue success rates into the double-digit range. It’s widely recognized that adjudication via scientific peer review is unable to distinguish the best/highest potential grant applications below a funding success rate of 20% or so and CIHR’s disastrous attempt to squeeze out increased precision via virtual peer review utterly back-fired this Summer.

Jeremy Berg, the ex-Director of the National Institute of General Medical Sciences (NIH) and newly installed editor-in-chief of Science Magazine, has a long record of providing analysis of science funding behaviours through his blog “DataHound”. He has maintained that effort in his new role and recently published a fascinating article on modeling grant success rates by looking at agency appropriations (budget) histories. Using data going back before the historic doubling of the NIH budget, he shows that success rates took a dive shortly after the doubling period. The entire blog is well worth a read (explaining the fact that committed funds in any year are far greater than funds available due to the 3–5 year tems of grants) but it is also a timely reminder of the negative effect of bolus funding. While he doesn’t go into the reasons (a future blog?), it is a well understood facet of supply and demand. As funding increases more grants are initially funded but there is a commensurate increase in applications. In part, this is like the employment market where, after period of unemployment, some people give up looking and fall off the unemployed list. As the economy improves more jobs become filled but more people are drawn back into the market.

Jeremy’s analysis suggests that gentle increases in funding are more efficient than a short bursts of funding investment. He makes this material by counting the lost time/effort associated with preparation of unfunded grants. In the boom/bust scenario where changes in funding are erratic and subject to political whim, application “overshoot” is much higher than when there is a more consistent increase — and this translates to fewer grants that are rejected and therefore less wasted effort.

Similar forces are likely at play in Canada and this begs the policy question of how best to support the CIHR (and other agency) budgets? Would attempts at immediate relief lead to subsequent pain? Perhaps the most effective approach would be to declare a target budget amount over the next 8-10 years and commit to reaching it via a predictable ramp. The current CIHR budget (aside indirects and flow-throughs) is around $850 million. Why not set this to double by 2027? About $200 million of that would be inflation (at 2%) so the real increase would be ~$650 million. This would be a bitter pill in the first couple of years but would provide stability over time and allow adjustment/recalibration of the health funding system as well as the capacity for long term planning. More importantly, it would send a clear message to younger researchers that there is a future trajectory for them that doesn't involve feast and famine.

The Naylor committee has been tasked with evaluating the entire federal portfolio in fundamental research investments and is projected to report by the end of this year. If they set a roadmap for the next decade they will positively influence Canada’s scientific future for much longer.

One clap, two clap, three clap, forty?

By clapping more or less, you can signal to us which stories really stand out.