17 years from research to impact – is this true?

By David Phipps (York University; @mobilemobilizer)

We often say that it takes on average 17 years for health research to turn into a new or improved clinical practice. Is this true? And what about in other disciplines?

Morris, Wooding and Grant[1] reviewed the literature on time lags between research and its use in clinical practice and health interventions. Their paper was more about the specific questions asked of the previous estimates, rather than the estimates themselves but generally yes, it takes years depending on the question being asked.

But what about other disciplines?

Well, that’s where recent data from Hong Kong’s 2020 Research Assessment Exercise (RAE) can help. The RAE produced 264 research impact case studies that were reviewed by panels of experts and assigned a score of 3* (internationally excellent) or 4* (world leading). These 264 were then analyzed by Vertigo Ventures[2] for:

  1. What types of impact outcomes have been submitted, and what pathways have been taken?
  2. What are the time lags between underpinning research and impact outcomes exampled in the impact case studies?
  3. What quantitative data (for inputs and/or outcomes) can be extracted and synthesised from the impact case studies e.g. commercialisation revenue, visitor numbers?
  4. According to the impact case studies, what types of research users benefit from the research, and to what extent?

There is a ton of good data in all these sections but the one I am interested in is #2 about the time lags as this presents data from all 41 Units of Assessment (UofA) capturing everything from creative arts, humanities, social sciences, health sciences, environmental sciences, and engineering with some unusual ones for those of us for familiar with the UK Research Excellence Framework (REF). Things like hotel management & tourism, Chinese language & literature as well as Chinese medicine (this is Hong Kong after all) are new UofAs to me.

The interesting data on time lags start on page 67.

From the report (because they say it better than I could)

  • “For the average ranges, the UoAs that took the longest time for developed researches to realise impacts within the allowable time window include UoA 4 (Clinical dentistry, 19 years), UoA 36 (Philosophy, 16.5 years), UoA 1 (Biological Sciences, 15.4 years), and UoA 37 (Religious studies, 15.33 years), in descending order. This indicates that researchers whose research fall into these units can expect at least 15 years to develop impact case studies with potential of attaining 4* and 3* quality”
  • “For the average ranges, the UoAs that took shortest time for developed researches to realise 4* and 3* impacts include UoA 18 (Planning and surveying, 7 years), UoA 38 (Visual Arts, 7.28 years) and Units of Assessment (Communications & Media, 7.6 years. This indicates that impact case studies in these UoAs take 7 years on average to attain 4* and 3* quality for the timeframe within which impact activity could be claimed.”

Everything else falls in the middle and is detailed in the report.

This is the first analysis I have seen of time lags from research to impact in disciplines other than health. And specifically in health UofA

UofA #3 (clinical medicine): lag = 12.23 years

UofA #4 (clinical dentistry): lag = 19 years

UofA #5 (Nursing, rehab etc): lag = 13.86 years

UofA #6 (Chinese medicine): lag = 11 years

The average for health research to result in impact on health interventions was 14 years. So close but not quite 17 years.

Read the rest of the report for interesting data on beneficiaries of Hong Kong research.


[1] Morris ZS, Wooding S, Grant J. The answer is 17 years, what is the question: understanding time lags in translational research. J R Soc Med. 2011 Dec;104(12):510-20. doi: 10.1258/jrsm.2011.110180.

[2] Vertigo Ventures (2022) RAE 2020 Research Analysis. https://8102620.fs1.hubspotusercontent-na1.net/hubfs/8102620/Campaigns/2022%20RAE%20Report/RAE%202020%20Research%20Analysis.pdf