Leveraging the Best EHR Data
Tracking and leveraging the best available data
are key components in identifying occurrences
of low-value care and accurately measuring
unnecessary use of healthcare services.
Many organizational efforts and studies on
this topic rely solely on administrative claims
data to measure low-value care. However, these
data have limitations and frequently won’t quite
capture clinical nuances of whether or not a
test or procedure is appropriate for a patient.
Using supplemental data in the EMR can reveal
that a patient received a test because of risk
factors that were not captured in administrative
data. Therefore, healthcare overuse reports that
only rely on claims data run the risk of misclassifying appropriate testing or imaging.
In a study published in the American Journal of
Managed Care last year, we found that EMRs can
capture detailed clinical information and context
behind unnecessary or inappropriate healthcare
use that are not available in claims data alone.
(See “Choosing Research Wisely.”)
Finding these nuances gave us context
in measuring overuse and proved to play
an important role in accurately capturing
healthcare waste. For example, in cases
of DEXAs (screenings for osteoporosis) in
women younger than 65 years, we identified
129 instances of explanatory risk factors in
the EHR that made the procedure clinically
necessary, along with another six risk factors
found after manual chart reviews of the sample population of 200 patients. Because more
than half of those procedures would have been
deemed clinically inappropriate by only analyzing claims data, this significantly adjusted
our measurements for overuse rate of DEXA
screenings for this population—from 57% in
2013 to an adjusted overuse rate of 18.5%.
painting a rich portrait of a patients’ clinical
needs through both their claims data and EHR
histories, we closely tracked how we performed
on our Choosing Wisely recommendations for
In addition to education and accurately
measuring low-value testing, we innovated
within our EMR to support our clinicians with
decision-making around these issues.
First, we judiciously use best practice alerts,
which pop up on the clinician’s EMR screen when
certain low-value treatments are ordered. We
use this approach to reduce low-value care for
bronchitis. In most cases, prescribing antibiotics
for bronchitis or a simple cough is unnecessary—
unless the patient is in a compromised state or
has certain comorbidities.
We also built out “Smart Sets” in our EMR,
which provide a menu of standard lab and imaging ordering options for common diagnoses. This
helps clinicians order appropriate workups for
specific diagnoses and reduces low-value testing.
Finally, we provide easy links to evidence-based guidelines in our EMR for the best uses
of labs and tests. For example, we have links
to guidelines for appropriate use of DEXA and
imaging that are available should our clinicians
need these resources.
As health systems continue to take on risk, it is
important that we think about all the ways we
can deliver the highest quality of care to patients
while also managing total medical expense. An
important contributor to this goal is reducing
the occurrence of low-value testing and screenings while improving use of high-value tests. By
educating clinicians on the most updated and
evidence-based guidelines around these issues,
accurately measuring low-value care across the
organization, and supporting our colleagues as
they make care decisions, we can make impressive strides in ensuring the best possible care
for our patients.
Thomas Isaac, M.D.,M.B. A, M.P.H., is senior medical
director of quality and safety, performance excellence
at Atrius Health.
1. Massachusetts Health Policy Commission. 2018. Health Policy Commission
Board Meeting, December 13, 2018. Page 36. Accessed February 8, 2018
2. American Board of Internal Medicine Foundation. 2019. Choosing Wisely:
Promoting Conversations Between Patients and Clinicians. Accessed June 20,
2019 at choosingwisely.org.
3. Massachusetts Health Policy Commission. 2018. Op cit. Page 37.
4. T. Isaac, M.B. Rosenthal, C.H. Colla, et al. Jan.
17, 2018. Measuring Overuse with
Electronic Health Records Data. American Journal of Managed Care,
24( 1): 294–300.
6. Ibid. Page 298, Table 3.
that only rely
on claims data
run the risk of