My academic training is in Operations Research with a large dosage of lateral thinking (Edward de Bono).

A classic story is below:

One famous story of operations research success during the war involved an analysis of Allied bombers returning from bombing missions over Europe. The military analyzed the location of shrapnel damage and bullet holes in returning bombers, to identify where to place additional armor on aircraft. Operations researchers were brought in at the last minute to do a “confirmatory” analysis, but they recommended that additional armor be placed on bombers everywhere except the places with damage or bullet holes! The operations researchers realized that analyzing damage to returning bombers involved a sampling error. It was the bombers that did not return that needed extra protection—and they needed it in the most vulnerable places (the places not damaged on the returning bombers).

In the past, most medical work has focused on bacterial shifts statistically linked to individual symptoms. Think of the “bullet holes” problem: the literature usually looks at one symptom at a time, while real patients often present with dozens.

Recently, I started using odds ratios instead. Most people know odds ratios from smoking and cancer risk, but they can be applied broadly—for example, the odds of working at Microsoft if you graduated from a particular university.

Using about 5,500 samples and roughly 350 symptoms, I built odds tables and then tested those odds ratios against a reference set of healthy individuals. To my surprise, sample after sample showed the highest odds for being asymptomatic, far more often than I expected.

On reflection, this implies we now have a well-defined, statistically grounded model of a healthy (asymptomatic) microbiome. That was the “lightbulb” moment.

Instead of hunting for individual “holes” and trying to patch them, we should look at all the shifts away from this asymptomatic model. Once those shifts and their contributions (odds ratios) are identified, we can use published research to determine what is most likely to normalize the microbiome. A long list of symptoms (bullet holes) stops being the target; the real target becomes making the microbiome asymptomatic.

Some past links:

• Significance Ordered list of Bacteria Associated with Autism using Odds Ratio
• Mathematically Derived Healthy Microbiome
• Generalized Odds Ratio to predict Symptoms from microbiome reports
• Open Access Odds Tables

A walk through on applying the above methodology pattern.