Substantial work lies ahead to convert pediatric preventive care guidelines into computerized prompts for physicians, but the payoff has significant potential, according to a new report by researchers from the Indiana University School of Medicine and the Regenstrief Institute.
With the increasing use of electronic medical records (EMRs) and health information exchange (HIE), there is a growing demand for a computerized version of the preventive care guidelines pediatricians use across the U.S, but there is work to be done when it comes to computerizing the American Academy of Pediatrics' Bright Future's guidelines.
"In addition to covering an age range with wildly varying health needs, the Bright Futures guidelines cover a wide range of topics, from infant car seats to substance use," S. Maria Finnell, M.D., assistant professor of pediatrics and a Regenstrief Institute affiliated scientist who is the study's first author, said in an Indiana University news release. "A computerized Bright Futures would help pediatricians provide better care at the point of delivery.”
Bright Futures consists of a multitude of health supervision recommendations for children from birth through 21 years of age— from recommended shots for newborns to interventions for childhood bullying to risk assessments for sexually transmitted diseases for adolescents.
Currently, Bright Futures is not organized to easily translate into computerized prompts. Recommendations are listed according to what should happen at each visit, which assumes the child will be seen for health supervision at each age and that previous visits have been completed. However, children may miss or have delayed visits to their physician.
"Decidability—when I am supposed to take action—and executability—what action should be taken—are key to computer decision support," said Stephen M. Downs, M.D., senior author of the study. "So extensive work will be needed to prepare the Bright Future guidelines for electronic medical record systems.”
Downs added that many of the guidelines are vague, making them difficult to translate into computerized format. “For example, a statement like: 'inhaled steroids may be useful in severe asthma' is not instructive. The precise and clear language needed for conversion to electronic prompts would include the recommended dose type, amount and frequency, as well as a definition of what constitutes severe asthma,” he said.
By consolidating recommendations and vague constructions that were repeated across visits or consisted of many smaller actions, the researchers reduced the total number of recommendations from 2,161 to a more manageable 245. However, only one in five of these 245 was actionable and thus could be converted to prompts in an EMR system, the researchers found.