Saturday, May 11, 2013

Price variations in US healthcare market and the power of data visualization

Competitive markets deliver cost-effective health care? No, if evidence presented in the graphic from New Jersey on treatment of chronic obstructive pulmonary disease is to be believed.

In the same New York city area, a joint replacement surgery costs anywhere between $15000 and $155000! The variation in prices, over such a small area, is truly stunning. This variation is equally stark when procedure rates in the US are compared with those in other countries. A few weeks back Time carried a well researched story on the high health care costs in the US, which had this graphic.

The data on the variations in prices for the 100 commonest diagnosis related groups (DRGs), spanning 163,065 patients, charged by 3337 hospitals in 306 metropolitan areas in 2011 was recently released by the US Government's Centers for Medicare and Medicaid. The data is available here. Is there any more compelling evidence of a market failure? Health care is clearly not broccoli.

The Times has rendered the database in a superbly informative interactive infographic. Its cognitive appeal and user-friendliness is striking. It is also an excellent example of how appropriate rendition of "big data" can be transformational in bridging the "last mile gap" in awareness creation efforts. Such graphics present critical decision-support information (here, a very good representative metric, fee charged by the hospital for each DRG for both Medicare and non-Medicare patients, both compared with respective national averages) in a manner that resonates cognitively (it is mentally easier to compare hospitals when information is so presented, on a geographical plane) with its audience.

In this context, considerable academic research has been done in recent years on the impact of information dissemination on increasing the responsiveness of elected representatives, quality of learning outcomes in primary schools, girl child literacy, prices of farm products, and so on. In many respects, all these studies are merely regurgitating conventional wisdom. At a conceptual level, even before any such research, people at the cutting edge in each of these areas already knew that bridging information asymmetry can be transformative. But the challenge, and they also knew it all along, is in presenting it in a manner that would facilitate easy use by its targeted audience. The academic research, beyond validating conventional wisdom, does nothing to address that challenge.

Imagine having similar visualization in developing countries to help people shop for school admissions or buy insurance policies and savings instruments, parents know the learning trajectory of their child, farmers know the optimal farm-gate price for their produce, and so on. And also if this information can be made available on a mobile phone interface.

I strongly believe that the search for such solutions require a smart intersection of statistical analysis, data visualization techniques, communication technologies, and behavioral psychology. Given the inherently complex nature of these issues and co-ordination problems, the market is not likely to resolve this challenge. It requires a partnership between governments, large private foundations, non-government organizations, and private service providers. But governments have to take the lead in facilitating this partnership.

PS: In the US, even with the available information, in the absence of some form of regulation, the local market power of hospitals, the presence of an insurance intermediary, and the very nature of demand for health care services, will conspire to maintain the diverse price distribution. 

No comments: