As the keynote speaker at this year’s BioPharm America 2010 conference, Dr. Isaac Kohane, the Henderson Professor of Pediatrics and Health Sciences and Technology at Harvard Medical School and Chair of the Informatics Program at Children’s Hospital Boston made a provocative point: We as consumers spend much more energy reporting on crash testing in the automotive industry than we do evaluating the predictive value of our diagnostic tests. Decisions that can affect the course of our lives are made based on these tests, and yet there is no equivalent of “Consumer Reports” for the diagnostics industry.
He noted that:
- Many existing diagnostic technologies serve as gold standards of practice but are not helpful in improving patient outcome. Tests may have as little as 60% specificity, leaving room for error and misinterpretation. He mentioned the Prostate Specific Antigen (PSA) test for prostate cancer, which is notorious for false positives.
- Analysis of large aggregated data sets will reveal patterns and make biomarkers more robust. If the industry processed its large sets of patient data with informatics algorithms, the predictive power of biomarkers would be easily apparent. Also, phenotypic data already being recorded as part of standard medical procedure are poorly used. If data were collected across institutions in large numbers it could be used to identify statistically significant trends. The tools to do these analysis already exist (see intelligent histories and i2b2). For example, in a previous study, Dr. Kohane showed that the seemingly unrelated diagnoses of an individual over several ER visits could be processed with an algorithm to identify domestic abuse two years in advance of detection by a physician.
Would a third party “Consumer Reports” style evaluation group generate the urgency required to make the changes Dr. Kohane deemed necessary? I agree that awareness offered by a consumer report function might be an appropriate solution. Already patients are highly influencing physicians’ choices of both diagnostic tests and therapies. A consumer reporting group might even further inform the requests patients make of their doctors in a way leads to a profound change. Could existing stakeholders, such as American Medical Association or Medco be mobilized to serve this purpose? Researchers play this role to some extent: they evaluate technologies in peer reviewed journals. But somehow the broad overview information does not always transition effectively to the consciousness of the patients.
There has never been a time more ripe for thoughts leaders to report on biomarkers. Children’s Hospital Boston hopes to play a part in the greater task of getting the right information to the right people. From our vantage point, straddling both research and clinical aspects of diagnosis, we are in a position to reject and advance technologies. We hope to use vehicles such as Vector and Thrive blogs to disseminate hospital ‘know how’ to other institutions. But the need is bigger then what we alone can fill. I applaud new mechanisms of reporting: the brokenness of healthcare in particular as it relates to diagnosis needs to be exposed in a way that makes solutions apparent.
Join the conversation! What industries do consumer reporting well? What could be borrowed from those industries? What technologies still need to be invented in order to be able to effectively report on biomarkers? Where’s the bottleneck?