Let's Pool Our Medical Data
4 likes2,511 views
The document presents three hypotheses about how cheap and ubiquitous data could change fields like health. Hypothesis A suggests that cheap data may change what constitutes justified beliefs in a field by prioritizing data-driven metrics over narrative models. Hypothesis B argues that data is becoming cheap and democratized enough to drive epistemic changes in health. Hypothesis C posits that costs may not automatically decline with more data and that certain experiments should be tried to see if data sharing works to benefit health. The document then outlines plans for portable legal consent and experiments with open data sharing groups in health research.
Let's Pool Our Medical Data
- 4. cheap data changes the epistemology of fields.
- 5. what makes justified beliefs justified? http://plato.stanford.edu/entries/epistemology/
- 7. player x is a good hitter because
- 8. OPS = AB(H + BB + HBP) + TB(AB + BB + SF + HBP) / AB(AB + BB + SF + HBP)
- 9. fields based on narrative, but tested in reality, are very vulnerable to data-driven change.
- 11. information > gut models and metrics low transaction costs pissed off pundits
- 13. hypothesis B:
- 14. data is (becoming) cheap enough to drive epistemic change in health.
- 16. not just cheap, but potentially democratized.
- 20. not just democratized, but ubiquitous.
- 23. hypothesis C:
- 24. costs will not magically decline thanks to data.
- 25. 25
- 26. 26
- 27. 27
- 28. 28
- 29. 29
- 30. assume 1,000,000 downloads assume 10% false positive rate 100,000 doctor visits $1000 per biopsy
- 33. that s the setup.
- 35. sharing by small but coherent groups can create asymmetrically valuable resources.
- 38. small group sharing
- 39. coherentism from: standard legal tools common infrastructure norms metadata
- 41. proven to work in: software content
- 42. the experiment:
- 43. let s a small but coherent group to share data and see if it works in health.
- 50. Also there is no suggestion of consanguinity in your pedigree. http://www.ianlogan.co.uk/
- 53. 53
- 56. terms in blue pop up definitions
- 58. all boxes must be checked volunteer must click to proceed
- 59. volunteer cannot skip video
- 64. http://weconsent.us/informed- consent/ 64
- 65. moving PLC to an open consent standard.
- 66. email patient #XYZ123 researcher patient
- 67. Public Portable Legal Sage Bionetworks Sage Bionetworks Recruitment Consent CGRE Synapse Founda3on Advocacy CGRE Database Synapse Database 1 2 Portable Legal Consent 3 4 5 Individual Level De-‐iden3fied Hospital Data Dataset Researchers Founda3on Advocacy Par3cipants
- 68. 3 Portable Legal Consent 2 Advocacy Advocacy Founda3on Hospital Founda3on Par3cipants 1
- 71. 71
- 72. five clinical studies on Bridge this year under PLC.
- 74. it s likely that we will end up with a sharing monopoly of some sort.
- 75. once entrenched, hard to move.
- 76. once entrenched, hard to move.
- 82. a. just like now, but moreso
- 84. b. the cartel.
- 86. c. the commons.
- 88. only the last choice allows all three choices to exist.
- 90. thank you @wilbanks john.wilbanks@sagebase.org