Medical data as an underprovided public good

Evidence based medicine requires data that are often difficult to assemble. Since drugs and medical devices are regulated, regulators have the option of requiring data to be collected. But data collection costs money (particularly in light of stringent laws that require the privacy of individual patients to be protected), and so it is often difficult to study the effect of medical interventions by following up on the health of patients.

A recent story summarized the issue succinctly: Heart Device Dispute Renews Push for User Registry

"Conflicting data this week about the failure rate of a critical and widely used Medtronic heart device has set off a debate among researchers who want to understand the discrepancies and the implications for patient care.
But some experts say that debate would not be occurring if federal officials, medical device makers and more doctors had thrown their weight more fully behind efforts to develop a national database of patients who get heart devices.
The Obama administration has announced plans to pour hundreds of millions of dollars into studies to compare the effectiveness of competing medical treatments and devices. Monitoring patients’ outcomes through registries could be part of that process.
But setting up such registries in this country has proved difficult so far. "

It seems that Medicare mandated the establishment of such a registry for defibrillators, but didn't fund it.

"And while defibrillator makers did help support the $3 million annual cost of operating the registry in its first year, they have since cut back that financing... Dr. Alan Kadish ... said he did not think that manufacturers believed that they would “be fulfilling their fiduciary obligations to shareholders by funding” studies that compare the effectiveness of their devices to those of competitors. " "

HT Scott Kominers