Menu Close

Dr Caleb Stowell Interview

By Dr Tom Foley, Dr Fergus Fairmichael.


Caleb Stowell is Vice President, Research and Development, at the International Consortium for Health Outcomes Measurement (ICHOM); and Senior Researcher at Harvard Business School. His role involves overseeing the development of ICHOM’s Standard Sets, developed in collaboration with international physician and registry leaders and patient advocates. In the past, he worked under Michael Porter, Bishop William Lawrence University Professor at Harvard Business School, to develop and expand the concepts of value-based health care. He holds a medical degree from Harvard Medical School.

Interview Synopsis

Standard sets and Comparative Effectiveness Research

ICHOMs Standard Sets are not focussed per se on comparative effectiveness research (CER) though they could easily be used for this end. To do so would require more detail, specifically the interventions of interest to test.  For example the ICHOM Low Back Pain Standard Set would provide an effective set of outcome and case-mix indicators to study the comparative effectiveness of instrumented versus non-instrumented fusion for spondylolisthesis, but one would need to clarify in detail and document what is meant by instrumented versus non-instrumented. It is possible that one could also conduct comparative research by merging the ICHOM Standard Set with administrative data (ICD codes), though ICD codes don’t always provide a sufficient granularity to distinguish one type of intervention from another.

We see a much different angle for learning and improvement than traditional CER. The data is pretty compelling that adoption for traditional CER is incredibly slow (estimates range around 15 years). From my perspective this is because people need the experience of that innovation in practice, to be able to try it, and then take it back to their own institution. So our approach is rather to compare the outcomes of institution A vs institution B rather than treatment A versus treatment B. If we give clinicians back outcomes data compared to their peers, they will say, “hey, why is this place doing so much better than us?” That motivates them to visit those institutions and pull into their practices the innovations developed there.

In my opinion, we have huge untapped potential for healthy competition based on outcomes. Clinician are very bright (and at least in the US, very competitive!). If we can get people striving to drive the frontier of performance, we will unleash a powerful motivator that today is dormant.

Translation into practice

Basically, the old model of quality is to organize a “god-like’ committee that bestows its great knowledge into guidelines, which then get repurposed into quality indicators. But this process takes a decade just to develop a single quality indicator, and most recent work shows it simply hasn’t worked – process measures typically affect only a small amount of the outcome variation. The literature is starting to show that organizational factors (culture, management attention focused on quality, communication) is much more important to driving good outcomes than simply applying established guidelines. There is simply so much guidelines can’t cover!

Data collection

There are a lot of new vendors providing services to minimise the burden of collecting outcomes, and in particular, patient-reported outcomes.  There have also been examples of this functionality being built into the electronic medical records.  This is becoming increasing cheap to get up and running.  However it is not only about the financial investment.  It also takes a strategic investment to get it done – the institution that has to want this and see it as a priority. 

Our view is the primary customer of outcome data is the patient and provider relationship.  It is important that they are on the same page about results that matter.

There is also an opportunity for payors to take a more proactive role in patient tracking over time, though most haven’t focused on this in the past. Often it is only the payer that is connected to the patient over a long period of time, the provider often does not have such a longitudinal relationship. 

Adoption of outcome measurement

Right now, we are still at the early adopter phase on the adoption curve. With ICHOM we are working with those early adopters to align on a common standard. If these early institutions can demonstrate the path towards transparency, we feel confident we can push adoption across the chasm into the early majority


We believe our first Standard Sets such as that for prostate cancer will lead to transparent outcome data within the next 5 years. I think this is achievable but that the real impact will be between the 5 and 10 year period.  We have an ambitious goal of covering half of all medical care with transparent medical data in 10. 

We also have a goal of continuing to refine our standards. They are not static and will need to continually improve.  

We are also engaging the technology industry to make this data more easily collected in practice. As patient-generated data through smartphones and other connected devices becomes more common, this should get easier.