OBSERVATIONS - october 12, 2009
Can Your Doctor Trust Your Electronic Health Data?
As U.S. health care moves slowly into the world of electronic health data portability and exchange, we confront a dilemma between choosing “inclusive” technology that can aggregate a wide range of health data and “exclusive” technology that relies on intact clinical records data. This dilemma arises because the inclusive choice aggregates data that may not be as reliable. By contrast, the exclusive technology, which involves exchanging only intact clinical data, though seemingly less consumer-centric, may ultimately bring us closer to realizing the benefits of information technology in improving clinical quality.
At issue is whether physicians and other care providers will trust and use information from so-called composite or dynamic electronic health records that compile information from disparate sources, including patients, providers, and third-party data repositories. While either the inclusive or exclusive approaches can give patients control over what content of their health records may be exchanged, inclusive composite records more frequently offer the seeming advantage of building trust among consumers by giving them greater control. However, some inclusive records put providers in the position of not knowing whether that information is accurate.
Not all electronic data are equal, and the application of health information technology to health data may result in unintended disparities of care and lower quality. The potential for problems arises when composite electronic records and dynamic documents are compiled by combining disparate data elements of personal health information into data sets for exchange, instead of only exchanging complete electronic clinical documents and intact electronic clinical document sections that are unchanged and have a transparent path back to their original source. Third-party personal health records (PHRs), health record banks, or health information exchange (HIE) repositories may aggregate data elements from many clinicians and other sources to create these composite records.
A mooted benefit of composite records is that they will increase consumer control over personal health information and build consumer trust in the electronic exchange of private health data. Individuals who are the subject of these records may control the selection and sharing of data elements. However, links to original records are severed and the important clinical context of each data element is lost. While consumers or their agents may attest to the integrity and original source of these data, other users of these records, notably clinicians, have no objective way to ascertain data integrity or source. The largest of such systems today have no ability to track the original source of data at all.
Consequently, some PHRs and HIEs compare unfavorably to standard electronic clinical records in the ability of recipients to rely on the data they contain. Unable to ascertain the provenance and integrity of such data, health providers may limit or refuse to apply information that they believe may be invalid, unreliable or incomplete. There is poor or no ability to transfer liability for false or misleading data to the party that supplied it, and there is evidence that some consumers alter health data when they can. The trust equation has two sides and clinicians are likely to use only health data they feel they can rely on. At the same time these data are materially less useful for research and quality measurement purposes.
A two-track model of health information technology and electronic health records could emerge as a result. Consumers whose health information is of proven authenticity and integrity are likely to see their data used to inform clinical decisions, based on the most advanced information tools in modern medicine. But consumers whose PHR or HIE provides health information of questionable reliability or quality may not receive the same benefits of evidence-based practice tools; these patients end up on the bottom track and a disparity is created. Ironically, patients who demand greater control of their personal health data through some PHRs and HIEs may inadvertently disadvantage themselves and their care teams.
By contrast, information about the data in original electronic clinical documentation, such as the original source of clinical observations and findings, can remain intact when data are shared or exchanged in other PHRs or HIEs that follow specific standards. Current US DHHS interoperability standards allow the integrity of health records to be proven from their source to any recipient, and the provenance of information to be audited at the data element level. Such methods could be used in PHRs and HIEs today instead of questionable composite data.
Several policy alternatives could mitigate the potential HIT disparities related to data reliability. Incentive differentials, grants or other economic tools could help organizations that use dynamic documents today to adopt standards for better data (this would require reworking their systems). State and federal governments could mandate that electronic clinical data be able to prove provenance and integrity when such data may be available (this could be part of state and federal HIE programs). Medical liability reform could reduce the risk to clinicians of relying on these data (but would not make lower-quality data equal for clinical decisions).
Because of differentials in data quality, a new kind of disparity in care and quality is appearing on the horizon. The worst alternative would be to do nothing and watch preventable disparities increase. At the same time, the easiest path for policymakers would be to enforce current federal standards more broadly; this would help stakeholders converge upon HIT that would not create new disparities in care. Doubtless, some valuable new uses will be found for patient-controlled data. Let’s not let that promise compromise clinical care, research and quality.
- Jamie Ferguson, Executive Director, Health IT Strategy & Policy,
Kaiser Permanente, Senior Fellow, KP-IHP
|
About
