Conversational Information Technology (IT) for Better, Safer Pediatric Primary Care

Throughout the health care system there are gaps between recommended care and the care that is actually provided to adults and children. One way to help close this gap for children is to better understand how families manage their medications at home. The use of interactive telephony technologies can be used in a patient-centric manner to help guide parents through interactive discussions to gather information while reinforcing recommendations and treatments for children.

This project evaluated whether an interactive voice response (IVR) system called the Personal Health Partner (PHP) used by parents prior to routine health maintenance visits could improve parental activation, the comprehensiveness of care provided, and medication safety. The PHP system addresses three core areas of parental activation: 1) having the confidence and knowledge necessary to take action; 2) actually taking action to maintain and improve their children’s health; and 3) staying the course even under stress. The PHP is a fully automated system that uses synthetic text-to-speech and automatic speech recognition to gather health data and counsel parents. 

The main objectives of this project were to:

• Develop an automated telephony system that uses fully automated conversations to perform previsit pediatric primary care assessments, offer parental counseling (including appropriate medication use), and support clinician decisionmaking by incorporating the PHP child assessments into their electronic health record (EHR) at the point-of-care.
• Conduct a randomized clinical trial to determine: 1) whether PHP assessment alone (no counseling) with EHR data exchange leads to higher-quality preventive care and medication management; and 2) whether the addition of PHP counseling to PHP child assessments (before and after visits) is associated with increased quality and healthier parental behaviors. 

A randomized trial was used to evaluate the PHP in a busy pediatric primary care office at Boston Medical Center. Parents who had scheduled an annual well-child exam were contacted via a mailed brochure and invited to participate. Those who were interested were asked to call the PHP anytime before the day of the appointment. At the time of their call they were randomized to PHP use, PHP use without counseling, or to usual care consisting of a safety assessment. Content for the calls included information on medication safety, diet and activity, parental smoking and depression, tuberculosis screening, developmental screening, and home safety. A week after the appointment, parents were asked to complete a followup assessment over the phone. At the end of the study providers completed an assessment via email.

PHP successfully identified and counseled parents. Those parents who used PHP were found to be more likely to discuss issues such as depression and medication use with their clinicians than those who did not use PHP. PHP parents also reported that they felt better prepared for visits with their children’s providers, and the majority indicated they would recommend PHP to other parents. The project team concluded that these systems have the potential to improve health-related behaviors, detect concerning patient safety situations, and enhance patients’ experience and engagement with care.