Dr David McGrath

Spine Physician

MB BS (Hons) FAFOM, RACP, FAFMM
Master of Pain Medicine

Ideal Treatment

An ideal treatment is one that has
1.An observable positive effect WITHOUT any probability of a negative effect.
2.The positive effect can reproduced a number of times until cure.
3.Less interventions are better. eg one successful surgery.

Very few treatments achieve this ideal. Nonetheless this is the goal of medical interventions. We can table the situation as follows.

TreatmentOutcome	Good Outcome	Neutral Outcome	Bad Outcome
Medical Intervention	a	e	b
No Medical Intervention	c	f	d

The numbers, a,b,c,d,e,f are probabilities of each outcome. With such a table we are a position to calculate the usefulness or utility of the treatment intervention and compare with competing strategies.
Putting in some average values.

TreatmentOutcome	Good Outcome	Neutral Outcome	Bad Outcome
Medical Intervention	40	40	20
No Medical Intervention	0	100	0

At first sight,this looks like a reasonable result. The trick is to give values to the good,nuetral,bad outcomes. Let's say neutral=0, good=1 , and bad= -3. That is, we may value going backwards much more than going foward. The total utility of the treatment as the sum of probability times nominated value would add up to -20 compared to the non-treatment total value of zero. In this instance it is better to do nothing.
Coming back to an ideal treatment.

TreatmentOutcome	Good Outcome	Neutral Outcome	Bad Outcome
Medical Intervention	10	90	0
No Medical Intervention	0	100	0

Using the same, assigned values to positive, neutral, bad the total utility adds up to 10; that is 30 utility points better than the first treatment which had a possibility of a bad outcome.
These examples, illustrate the value of safe treatments even if the expected good result is small.
With treatments such as Recursive Modulation the downside is indeed small (virtually zero) while the small upside can be recursively added to a large significant result.

Article Index