A Simple Way to Meet the “Preponderance” Standard: Break it Into Steps

By Sandip Buch, MD (Guest Blog)

When physicians serve as medical expert witnesses, we often face a tension: medicine lives in shades of gray, while the legal system demands firm answers. Courts ask us to apply the preponderance of the evidence standard—essentially, to decide whether an outcome was more likely than not (>50%) had some condition been different. How can we, with a “reasonable degree of medical certainty,” predict a hypothetical outcome? The credibility of an expert hinges on grounding opinions in clear, logical reasoning. Otherwise, opposing counsel can easily portray the testimony as guesswork.

One reliable way to simplify causation analysis—and make it understandable for judges and juries—is to break the problem into a sequence of steps. Each step must be true for the outcome to follow.

A Case Example

As a psychiatric expert witness, I used this approach in a case involving a detained man with a psychiatric history who refused psychiatric evaluation and medications several times. While in custody, he expressed suicidal thoughts and displayed psychotic symptoms. He was seen by mental health staff, but not referred to a psychiatrist. He ultimately died by suicide.

The key legal question:
Would a psychiatrist referral, more likely than not, have prevented the suicide?

At first glance, this seemed like a very difficult question to conceptualize and render an opinion.

But breaking it into steps simplified the complexity and clarified the answer for me:

For the psychiatric referral to have prevented the patient’s death, all of the following would have needed to occur:

1. The patient would have needed to agree to see a psychiatrist (despite multiple refusals).
   
2. He would have needed to accept medication (separate from agreeing to an evaluation).
   
3. He would have needed to actually take his medication daily.
   
4. His suicide would have needed to stem from a condition treatable by that medication.
   
5. The medication would have needed to be effective in his case.
   

Every link must hold for causation to be established.

Breaking the analysis down clarified my opinion, and I was able to see that the >50% likelihood threshold was clearly not met.

Step 1: The patient had already refused psychiatric consultation multiple times. In my experience, this makes simply cooperating with a psychiatric evaluation questionable.  

Step 2: Even if he agreed to evaluation, he had refused medications repeatedly. Acceptance of medication was improbable.

Step 3: Even if he accepted a prescription, daily compliance (actually swallowing the medicine every time it was offered) is a separate hurdle, especially given his prior history of non-adherence.

Step 4: Suicide can result from conditions treatable with medication (e.g., psychosis, bipolar disorder) but also from factors medication does not address (e.g., personality disorder, impulsivity). This patient exhibited both psychosis and impulsivity. It is unclear what actually caused the patient to have committed suicide.

Step 5: No medication is 100% effective. Even correct diagnosis and compliance do not guarantee prevention.

Explaining It to a Jury

An analogy helps: the chance of flipping three heads in a row isn’t close to 50%; it’s ½ × ½ × ½ = 1/8. When multiple conditions all must occur, the combined likelihood drops sharply.

For this case, even if opposing counsel assigned high probabilities—say 85% at each step, which is wildly optimistic and unrealistic—compounding the steps (0.85)⁵ results in only a 44% likelihood that the patient would not have committed suicide, which fails the legal standard. 

This highlights an important point- if a series of steps need to happen- even if several of the steps individually have a high probability of happening- when combined, still lower the overall likelihood significantly and can lead to a less than 50% liklihood of some result occurring (thus failing the preponderance standard). This is why it is important to break down any question into simple steps, even ones that by themselves have a good chance of happening but are not actually 100%. 

General Method

Here’s a simple way to apply this approach in any case:

1. Define the exact outcome you’re testing.
   
2. List each step that would have to happen in a chain for that outcome to happen.
   
3. Look at each step on its own. Ask: how likely is this to happen?
   
4. Step back and look at the chain. If every hurdle must be cleared, the overall chance of reaching the outcome gets smaller with each step.

The Core Insight

By breaking down a complex causal question into these simple, logical steps, you can clarify your thinking and present a clear argument that the preponderance standard has not been met. This approach is not only medically sound but also persuasive in a courtroom setting.

Need an Expert in Causation?

If you are evaluating a case involving causation, High Rock Experts can help. Start your custom match here.

About the Author

Sandip Buch, MD is a board-certified psychiatrist with over 20 years of clinical experience. He is the founder of a psychiatric practice in New York overseeing a team of psychiatrists and therapists. Dr. Buch has worked across diverse settings, including hospitals, prisons, and community programs, and has extensive expertise in workers’ compensation evaluations. He also teaches, testifies in legal cases, and communicates psychiatric concepts to professional and lay audiences alike.

Disclaimer: The views and opinions expressed in this article are those of the author and do not necessarily reflect those of High Rock Experts®. This content is for educational and informational purposes only and does not constitute medical or legal advice.