In a time when health care providers and patients have hundreds of diagnostic tools at their disposal, containing costs and preventing unnecessary or duplicate testing is crucial.
The United States’ health care spending has grown recently to nearly $5 trillion a year, and some of this spending is due to unnecessary testing.
Unnecessary testing isn’t always the result of an overzealous provider or a worried patient though. It can also occur because of fragmentation and miscommunication between providers and health care systems. Without an ability to connect, fragmentation can increase the risk of adverse health outcomes.
The most common way to connect disparate providers and health systems is through health information exchanges, also known as electronic medical record interfaces. These platforms enable providers to communicate and see prior testing, imaging, and other medical encounters. But, in practice, does provider access to these interfaces actually reduce duplicate testing for Veterans?
Recent Study
Evaluators at the Partnered Evidence-based Policy Resource Center set out to answer that question in a study published in JMIR Medical Informatics. The authors sought to estimate the impact of provider usage of the Joint Longitudinal Viewer (JLV) on the ordering of duplicate imaging across the Department of Veterans Affairs (VA) and the Department of Defense (DoD).
The JLV is a joint electronic health system utilized by both VA and DoD and allows providers across both enterprises read-only access to their patients’ health records from the other enterprise. For this study, authors examined data from fiscal year 2018 and conducted a retrospective cross-sectional analysis.
Authors looked at 892 unique medical encounters involving recently separated Veterans with at least one primary care visit at VA within 90 days of an imaging study at DoD. There were a number of exclusionary criteria, such as Veterans who had a primary care visit as part of a compensation and benefits screening or those diagnosed with cancer which may require frequent testing to monitor their illness.
To estimate the relationship between use of the JLV during the primary care visit and duplicate imaging, authors used a logistic regression model. The model controlled for potential confounders, including Veteran age and sex and provider imaging rate in the past six months. To test the results of the model for robustness, evaluators used 2-stage ordinary least squares models and other specifications.
Findings and Limitations
Overall, authors found that JLV use by VA providers increased since fiscal year 2015 when the system was first introduced. Monthly queries in JLV grew to over 1.4 million by 2018.
During this year of peak JLV use, for providers in the study cohort, use of the system was associated with a significant reduction in the likelihood of ordering duplicate imaging. VA providers who did not use the JLV ordered duplicate imaging 11.2% of the time, compared to 6.1% of the time for JLV users.
Additionally, evaluators found differences in duplicate image ordering between providers with historical JLV use and those without. VA primary care providers with a history of using the JLV at least once in the six months before the study period were five percentage points less likely than their counterparts to order duplicate imaging.
However, the authors also acknowledged some study limitations.
First, they only examined primary care visits within 90 days of imaging, so it’s possible that duplicate imaging still occurred, either outside this window or in non-primary care settings.
Secondly, authors noted that they could not delineate between necessary or unnecessary duplicate testing due to data source limitations. Put simply, some seemingly duplicative imaging may have in fact been necessary for monitoring disease. Though authors tried to mitigate this by excluding Veterans with cancer, other conditions may also necessitate frequent testing.
Lastly, authors noted that the results lost robustness with adjustments to the definition of a provider with historical JLV use. According to authors, this suggested “heterogeneity of JLV benefits by frequency of use.”
Takeaway
Study findings suggest that using the JLV and similar longitudinal health information exchanges may reduce duplicate imaging and, therefore, patient burden and unnecessary spending. More research is necessary to understand how electronic medical record features could impact other testing and care in the future.
