Scott Schumacher | Another View: Identifying patients vs. patient identifiers
- By Scott Schumacher
- Sep 19, 2006
A range of government workers, academics and consultants have touted the benefits of a unique national health care identifier, the equivalent of a Social Security health care number for every United States resident.
With the goal of improving our national health care system, proponents suggest the National Health Information Infrastructure (NHII) be used as a mandate to fund a unique identifier. This, they say, may be the only surefire way of improving the availability and quality of patient information shared throughout our health care system.
Indeed, we are already improving the availability and quality of patient information shared throughout our health care system. In fact, we are improving it by using technology that exists today: technology that can work effectively with or without a national identifier and that can take advantage of existing patient identifiers.
There is a distinguishable difference between issuing a national patient identifier and establishing a system to successfully identify patients on a national level.
Technology that exists today can rapidly match and link data across state lines and disparate systems. It can go above and beyond meeting NHII goals. It is already improving our nation's health care system. And it links volumes of information in sub-second response times, without requiring a unique identifier of any sort.
Where? Here are just two examples:
A three-state health care data exchange prototype is currently under way, linking 20 million medical records associated with 500,000 patients across networks in Massachusetts, Indiana and California. As part of the Massachusetts SHARE (Simplifying Healthcare Among Regional Entities) initiative, data within the prototype is exchanged through a decentralized, federated network that protects patient privacy and does not require a centralized clinical database or a national health identifier. The actual clinical records are kept in the hands of the institutions that produce them or provide the patient care, which allows patient data to be securely requested from remote locations. And, keeping the information local keeps the information accurate, which is the most critical component.
The Canadian government, which has made national health care a priority, offers another example. Our northern neighbors are in the process of creating a pan-Canadian client registry. Spearheaded by Canada Health InfoWay, a federally funded nonprofit corporation, the registry will be able to link patient records from hospitals and doctors' offices throughout Canada's widespread provinces and territories. Currently up and running in two provinces, the registry today contains more than 150 million records from multiple, unrelated entities, where data is searched and integrated without a common identifier.
Enabling the accurate linking of patient demographic data on a national scale would be far easier than creating a new system of records based on current information that requires a unique identifier. In fact, given the inherent cleansing and classification capabilities in a linking approach, relying on today's technology'rather than creating a new patient identifier'would likely produce more accurate information.
That is, after all, the ultimate goal: to identify patients accurately so we can provide high-quality, consistent health care nationwide. This can be done now, and without the costs of creating a new system and of putting in place the people and processes necessary to issue and support a national identifier.Scott Schumacher, Ph.D., an expert in complex data algorithm analysis, is chief security officer and chief scientist at Initiate Systems Inc. of Chicago. He currently assists 'Connect for Health,' a collaborative promoting health care industry standards (email@example.com).