Discovered Correlations in Databases: What To Do? Who Pays?

Systems biology, including biomarkers, is developing information, particularly correlations with disease states. Why? To date, the most commonly cited scenario is that the large databases will be curated for interrogation by researchers, physicians, and perhaps patients. There will be costs associated with maintaining the database, including electricity and technical support. Today this is a research or overhead expense, but going forward, anticipate that eventually fees will be charged for accessing the annotated database. In the U.S.A., I expect that charges will be paid by patients via responsible insurers. In most other countries, the charges will probably be buried in the overhead of the country’s health-care scheme, much as libraries are today.

But what about diseases or correlates that are discovered in existing research databases? Should a biomarker be discovered and validated for a disease that will appear later in life, will the database be interrogated to find other individuals already included in the database who have not yet presented with the symptoms? It seems that someone should. But could they? HIPAA requires that the subfiles from patients be irreversibly de-identified.

Another scenario: Suppose the patient is receiving a specific therapy that is contraindicated by recent experience of patients with the same biomarkers. Who reports? How to report negative results? Albert Einstein is reported to have defined insanity as expecting a different outcome from repeating an experiment.1

One real scenario is genetic disorders. Several institutions are curating databases including whole genome sequences (WGS) of hundreds of thousands of people. In time, if whole gene sequencing continues to be useful, the databases could have millions and potentially billions of people. After all, Prof. Eric Topol2 advocates digitizing every patient with a complete “Snyderome.” Certainly, as the data are accumulated and correlations are developed, developmental disorders will be found for individual members of the database.

Is someone responsible for reaching out or back? Is it even possible for them to reach back? If they could, who is contacted—the individual patient? The health-care provider? What if it is 20 years since the sample was collected? Contact their progeny? Who keeps the records? Most significantly, who pays for this?

This is a serious problem in modern health care. Since the need will certainly grow in frequency, it must be considered by technical and legislative bodies.


  1. Einstein, A.
  2. Topol, E. The Creative Destruction of Medicine How the Digital Revolution Will Create Better Health Care; Basic Books: New York, NY, 2013; p 257.

Robert L. Stevenson, Ph.D., is Editor, American Laboratory; e-mail: