Value-Based Medicine

Decision Analysis: From Military to Medical

By Melissa M. Brown, MD, MN, MBA

I have been trying to talk myself out of writing on the topic of decision analysis and how we can utilize what was originally a military concept in the practice of everyday medicine. However, decision analysis can play a key role in value-based medicine, so I am going to devote the next couple of columns to this important tool.

The problem is that the term "decision analysis" (DA) sounds so non-clinical. But it is really all about clinical relevance. As a clinician, you likely will not actually perform DA with the computer model, but you should begin to learn to understand the process, know what it is and how it can be used in medicine.

Why Use Decision Analysis

The why is easy: because it is the best way of looking at all of the key variables which we use to decide treatment. And if clinicians don't participate in this process most appropriately, other stakeholders in healthcare will do it in our absence.

We have worked over the past few months to learn the key essentials to Value-Based Medicine and how we gather the right clinical evidence and measure quality of life to guide us in making sound medical judgments. We know from the clinical trials, metaanalyses and other best evidence which treatments are possible options for care in a particular situation.

A National Security Concept

Decision analysis was developed years ago in the military to assist in decisions concerning national security or defense. Eventually it became clear that the process would be useful in medicine, as treatments usually involve uncertainty in outcomes while having the potential for adverse events to variable degrees.

"Decision analysis is a method to provide structure and a systematic framework for making hard decisions. Decisions, by their nature, are complex, contain uncertainty and may have more than one objective. Three of the key terms to define in decision analysis are (1) decision, (2) risk and (3) uncertain events. A decision is defined as an "irrevocable allocation of resources." Risk "is the probability and severity of loss linked to hazards." Uncertain events are defined as events "where the outcome is unknown."

The passage cited above refers to methodology as described by Major Rudolph Butler, USAF et al.¹ in describing the process as it may be applied in the Armed Services — such as in training and exercises in preparation for real military operations. Pretty cool. So much for the history… let's take it and look at how we apply the methodology in medicine.

Transferring DA to Medicine

Using the military analogy for a bit longer, the "decision" would identify the medical intervention, such as cataract surgery. The "uncertain events" would apply to the potential outcomes, such as an uncomplicated outcome with excellent vision, as well as potential for adverse events, such as vitreous loss, retinal detachment, endophthalmitis, dislocated IOL, etc. The risk would identify the chances of such events, as measured by our evidence of study as well as the severity of loss. We would define the "severity of loss" as quality-of-life alterations measured in utilities.

Additionally, decision analysis could help the clinician decide between two pharmacologic agents. At first look, they may seem to have the same clinical benefit, but upon closer scrutiny, the clinician recognizes that the incidence and severity of potential adverse events differs between the drugs. By evaluating all aspects of the clinical benefits and all potential side effects, one treatment can then emerge as the superior agent.

Decision analysis incorporates elements of evidence-based medicine, but values them in terms that are relevant to patients — the desirability of different outcomes. Utilizing the method, various treatment alternatives can be compared to one another. It is particularly valuable for the fact that DA can demonstrate that one intervention is preferable to another when clinical judgment and primary evidence-based outcomes cannot clearly differentiate between the value conferred by two interventions that appear similar.

Next month we will look at a simple decision-analysis model, likely using the intervention of cataract surgery, and see how the various parameters fit together. Remember, DA masquerades as esoteric, but it is really all about clinical care and the effect our treatments have on our patients' quality of life. OM

Reference

1. Butler, Rudolph et al. Using Decision Analysis to Increase Commanders' Confidence for Employment of Computer Network Operations. IOSphere Joint Information Operations Center, Fall, 2005, p.33-38.

Value-Based Medicine is a registered trademark of the Center for Value-Based Medicine.

Melissa M. Brown, MD, MN, MBA, is president and CEO of the Center for Value-Based Medicine in Philadelphia, She can be reached via e-mail at mbrown@valuebasedmedicine.com