The hidden power of incentives in Healthcare.

Charlie Munger once famously said the under recognition of the power of incentives is one of the main sources of psychological misjudgment. He also went on to say (perhaps facetiously) that he sometimes crosses off preventative investigations ordered by his doctor to ‘not allow him a chance to make a mistake’. What he meant was if a pathology is discovered, then the doctor has an overwhelming incentive to treat that pathology. Charlie clearly prefers sitting on his arse and doing nothing unless there’s an obvious problem he needs a solution for.

To know or to not?

My wife is in oncology rehab and we often speak about theubiquity of investigations ordered. Mass screenings and a strong push for earlydiagnosis and intervention seem to be the trend presently.

Preventative investigations, however, can be a double-edged sword. If a pathology is discovered, the diagnosing clinician will often have a bias to intervene. In medicine the radiologist will refer to a surgeon to minimize medicolegal risk to themselves. The surgeon will have incentive to perform the recommended surgery both for monetary value and mitigate medicolegal risk.  In dentistry the incentive misalignment is more direct where dentist is both diagnostician AND surgeon, leading to a hairdresser selling shampoo affect.

What would have happened if we did nothing?

A specialist colleague of mine and I were debating naïve intervention in dentistry. The example we used was a radiolucency (shadows) in the lower jaw. Nearly all shadows are treated via surgery. Therefore, the data for what happens if we did nothing is not known. It’s feasible to conclude that out of the treated cases some may have been stable for many years. The patient may be out of pocket and have a worse outcome than if nothing happened. The difficulty are the cases where the intervention helped, but it’s almost impossible to differentiate these from a single data point. Data driven medicine, where multiple data points over time are analyzed for trends are a promising alternative. Strict outcome measures would also be ideal, but these are presently not commonplace in dentistry.

Availability heuristic.

Pianists want to play Piano. When a pianist sees a piano, what comes most easily to mind is a vision of themselves playing a piece they really enjoy. When a dentist sees a small shadow on an X-ray they think of all the times they’ve fixed a similar shadow with a beautiful composite filling. Add to this the monetary reward and targets to hit, and many fillings that would have been fine having a “watch” placed on it, gets an invasive filling.

Complexity and compounding trouble over time.

The problem with “early intervention” in dentistry is the long term effects this may have. A natural, unfilled tooth can last a lifetime. Not so for a filling. A small “preventive’ filling will often set the tooth spiraling into the restorative cycle which goes like this; Small filling ==> large filling ==> root canal ==> crown ==> extraction ==> implant. Doing a small filling, especially if it’s due to a single data point or poor incentives will lead to long term harm to the patient and will  drop them in to the the restorative cycle.  

What advice would I give a patient?

Understand that dentistry/medicine isn’t a perfect science. Whenever subjective human judgment and monetary rewards are at play costly mistakes can be made. But what to do?

  1. Get a second opinion and see if there is consensus.
  2. Get multiple photos (or x-rays etc) 6-12 months apart to see what the trend is and whether it's stable.
  3. If in doubt, do nothing!

Further reading