The goal of the PedsBP CDS research project is to adapt a previously tested web-based clinical decision support tool that appropriately identifies high blood pressure in youth for use in a primarily rural health system and compare approaches to CDS implementation in 45 primary care clinics treating children in 3 upper Midwest states. This project will advance implementation science and address a critical need for youth at risk for cardiovascular disease and with limited access to pediatric subspecialty care.

Hypertension (HT) in youth tracks into adulthood, contributing to adult cardiovascular morbidity and mortality. National guidelines for the diagnosis and treatment of HT in children and adolescents were last updated in 2017, with definitions for HT that vary by age. To date, most children and adolescents with elevated blood pressure (BP) or HT are not diagnosed or inadequately treated. Factors that contribute to these deficits in care include: the need to translate pediatric BP measures into BP percentiles, lack of clinician familiarity with pediatric HT guidelines, and competing demands at clinical encounters.

Electronic health record (EHR)-linked clinical decision support (CDS) can be used to address these barriers and improve the identification and management of elevated BP and HT in children and adolescents. In a previous study, the investigators developed, implemented, and evaluated a sophisticated web-based, EHR-linked CDS to provide patient-specific clinical care recommendations in real time and in accordance with national guidelines for BP management in youth. In a 2-year cluster randomized trial in 20 urban and suburban primary care clinics in an integrated health system in Minnesota, the investigators demonstrated that the CDS increased repeat measurement of elevated BP during a visit and more than doubled clinician recognition of HT, while promoting dietitian referrals and additional next steps in care consistent with national guidelines. The CDS system was well accepted by providers and as such, is now standard of care in 55 primary care and 17 subspecialty clinics serving children across our health system. Implementation of this CDS in a new health system is a logical next step, yet optimal strategies for adaptation and implementation of CDS in clinics serving rural populations have not been well described.

In this study, the investigators will implement PedsBP CDS in a large health system with many clinics located in rural regions of Minnesota, Wisconsin and North Dakota. In order to compare approaches to implementation of PedsBP CDS to usual care, the investigators will randomly assign 15 primary care clinics to receive high-intensity implementation (CDS with online and in-person training, and audit-feedback), 15 clinics to receive low-intensity implementation (CDS with online training only), and 15 clinics will continue with usual care (no CDS).

Inclusion Criteria:

Patients will be eligible for the PedsBP CDS if

1. 6-17 years of age
2. BP measured and entered in the vital sign section during an ambulatory primary care visit in a randomized primary care clinic
3. not pregnant or postpartum

Patients must meet these eligibility criteria to be included into study analyses:

1. have at least one index visit with a primary care provider at a randomized clinic in the intervention period
2. meet eligibility for PedsBP CDS at index visit (a-c above)
3. no previous HT diagnosis prior to index visit
4. not taking antihypertensive medication prior to index visit
5. not opted out of use of their data for research via general consent prior to performing analyses

Primary care providers must meet these eligibility criteria to participate in this study:

1. practice at a randomized primary care clinic
2. be a pediatric or family medicine care provider (pediatrician, family physician, nurse practitioner or physician assistant), and
3. provide ongoing clinical care for children and adolescents

Exclusion Criteria:

Patients will be excluded from analyses if the following criteria are met:

1. outside of the inclusion age range (<6 years and ≥18 years) at index visit
2. pregnant or postpartum adolescents during study period
3. known HT diagnosis or taking antihypertensive medication at index visit
4. opted out of use of their data for research via general consent prior to performing analyses

Baseline data was not collected on the providers. We report provider actions at the patient level in the outcomes section.

Safety events were passively monitored every 6 months during the 18-month intervention period and 12-month follow up period. Patients were monitored for events occurring within 6 months of index visit newly meeting stage 1 and 2 hypertension criteria.

We conducted passive safety surveillance for patients in control and intervention sites using routinely collected electronic health record data to monitor for potential safety events. We did not gather safety information on providers.

Safety data was gathered on all 596 patients who newly met criteria for hypertension.

The study was conducted in a large, primarily rural health system. Study results may not be generalizable to other health systems, especially those with different structures or urban settings.