One of the biggest drivers of productivity growth in recent years has been “big data,” the analysis of large data sets to yield insights into how we might improve the operational efficiency of a wide array of systems. The McKinsey Global Institute provided a sketch of how this works back in 2011:
There are five broad ways in which using big data can create value. First, big data can unlock significant value by making information transparent and usable at much higher frequency. Second, as organizations create and store more transactional data in digital form, they can collect more accurate and detailed performance information on everything from product inventories to sick days, and therefore expose variability and boost performance. Leading companies are using data collection and analysis to conduct controlled experiments to make better management decisions; others are using data for basic low-frequency forecasting to high-frequency nowcasting to adjust their business levers just in time. Third, big data allows ever-narrower segmentation of customers and therefore much more precisely tailored products or services. Fourth, sophisticated analytics can substantially improve decision-making. Finally, big data can be used to improve the development of the next generation of products and services. For instance, manufacturers are using data obtained from sensors embedded in products to create innovative after-sales service offerings such as proactive maintenance (preventive measures that take place before a failure occurs or is even noticed).
And Paul Krugman recently riffed on how the big data revolution might make machines much smarter without actually making them think like humans. The self-driving car is perhaps the paradigmatic example: Google has not built machines with the judgment of human drivers; rather, it has harvested enormous amounts of data from exhaustively recording the behavior of human drivers to allow comparatively “dumb” machines to navigate a complex landscape.
It is easy to see how big data could contribute to improving health outcomes. The only problem is that health data tends to be highly fragmented. But as Peter Orszag explains, there is a seemingly minor provision of the cliff deal that creates an incentive for medical providers to expand their use of clinical data registries:
These registries collect information on patient characteristics, patterns of care and outcomes that can be crucial to evaluating what medical techniques and strategies work and which ones don’t. Unfortunately, registries are not as widespread as they should be — and the ones that exist often are limited to particular types of care.
Data from insurance-claim forms, in contrast, are easier to obtain and more comprehensive across episodes of care. However, these reveal only which tests and procedures were performed, and tell very little about how the patient fared.
What would be ideal for people who analyze health-care practices would be to have some way of combining registry and insurance-claim data.
This is where the cliff deal comes in:
The fiscal-cliff legislation takes a step in this direction by allowing doctors to opt out of other quality-reporting requirements if they participate in an approved registry.
If anything, I’m more enthusiastic about this development than Orszag, who includes an important cautionary note:
The Congressional Budget Office scored Section 601(b) as having virtually no effect on the federal budget, but partly because the office understandably struggles with evaluating the effects of provisions that are designed to further broaden structural changes.
This, in my view, is why those of us interested in streamlining government should think less about CBO scores and more about structural incentives.