Blood Pressure Screening in Children - Back to the Drawing Board
BudWiederman, MD, MA, Evidence eMended Editor, Grand Rounds
The authors of this study did a great job of explaining their results and potential
study limitations. Frontline pediatric providers would do well to compare their own
practices to those of this large managed-care practice.
The study originates from the Kaiser Permanente of Southern California health system
which, because of its large patient population and electronic health record system,
has the ability to measure practice and health outcomes with large study sample size.
In this instance, investigators looked at 4 years' worth of blood pressure measurements
in children between 3 and 17 years of age. After excluding various conditions that
could be associated with hypertension or high blood pressure recordings, they focused
on 186,732 children who had at least 1 BP reading at the 95th percentile or above,
looking mainly to see how well practitioners adhered to screening guidelines* and how this might affect eventual health outcomes. Their discussion section alone
is worth the price of reading the article.
The study's key findings were that only about half of the children had correct BP
classifications based on the initial visit, with the remainder classified as a false-positive
high BP, false-negative low BP, or unknown. Eighty percent of the group with initial
high measurements did not even have a repeat measurement performed at the same visit;
this occurred even though the Kaiser system automatically translates the readings
into percentiles, which I (and the authors) would have thought would be a great help
to the providers. That failure to repeat the measurement at the same visit accounted
for half of the false-positive results in the study. That would certainly cause problems
by requiring unnecessary follow-up visits to repeat BP measurement, though perhaps
sadly not as big a problem as expected with this group since only one-third of those
children completed the follow-up within 3 months of the initial measurement.
I mentioned some helpful take-home messages for primary care providers.
1. Take time now to review the current AAP blood pressure screening guidelines, linked
2. Ensure office staff are trained and re-certified annually in blood pressure measurement.
(However, the Kaiser staff did have this condition in place and still ended up with
3. Assess your office sphygmomanometers. Automated digital devices are widely used
now, including in this study, but do have a tendency to overestimate true BP. I'm
not suggesting we return to those manual sphygmomanometers with glass tubes filled
with toxic liquid mercury, but aneroid (from the Greek meaning "without water") devices
require frequent recalibration due to being dropped or other trauma.
4. Watch out for implicit bias in BP screening. (Note that implicit bias applies to any form of stereotyping; it
is any bias that we are not aware of in a conscious manner.) I mention this because
the authors found that children with high BP were more likely to have repeat measurements
performed if they were male, older, and obese. This meant that younger, non-obese
female patients had poorer care for this particular screening, which suggests implicit
bias was in play.