SummaryPrimary care practices affiliated with Children's Hospital of Philadelphia receive real-time decision support through an electronic health record that uses popup, color-coded alerts during visits with pediatric patients who have asthma. The alerts notify physicians of the severity of the condition, and of instances when broadly accepted guidelines call for an inhaled corticosteroid to be prescribed, a spirometry test to be performed, or an asthma care plan filed. The system also makes it easy for physicians to take necessary action steps, such as clicking a link to order a prescription or access patient and family education materials. Practices also receive support to encourage appropriate spirometry testing and interpretation. The alerts moderately increased adherence to the guidelines, as compared with a control group of practices that did not have access to the alerts but could make use of embedded asthma tools if they proactively entered and searched the electronic record.Strong: The evidence consists of a cluster-randomized trial that examined trends in adherence to key asthma guideline recommendations in two urban and four suburban primary care practices that implemented the program, compared with the same trends in two similar urban and four similar suburban practices not using the alerts.
Developing OrganizationsChildren's Hospital of Philadelphia
Date First Implemented2007
The initial trial testing ran from April 16, 2007 to April 15, 2008.
Vulnerable Populations > Children
Problem AddressedPediatric asthma is a common condition that leads to diminished quality of life and costly emergency department (ED) visits and inpatient admissions. Managing pediatric asthma according to broadly accepted guidelines can minimize the potential for negative outcomes, but primary care physicians, including pediatricians, often do not adhere to such guidelines.
- A common condition, leading to poor quality of life and need for costly care: Asthma affects more than 22 million Americans each year, about one-third of whom are children.1 When poorly managed, asthma has a negative impact on quality of life and often leads to acute exacerbations requiring ED or inpatient care.2
- Poor adherence to guidelines: Existing guidelines can help practitioners effectively manage pediatric asthma and hence reduce the risk of exacerbations, but adherence to such guidelines remains low among practitioners in a variety of settings, including primary care.2
Description of the Innovative ActivityPrimary care practices affiliated with Children's Hospital of Philadelphia receive real-time decision support through an electronic health record (EHR) that uses color-coded alerts to notify physicians seeing pediatric patients with asthma of the severity of the condition, and of instances when accepted guidelines call for an inhaled corticosteroid to be prescribed, a spirometry test to be performed, or an asthma care plan filed. The system also makes it easy for physicians to take necessary action steps, such as clicking a link to order a prescription or to access educational materials for patients and families. Practices also receive support to encourage appropriate spirometry testing and interpretation. Key elements of the program include the following:
- EHR with embedded asthma-related tools: All primary care practices owned by the hospital operate in a paperless environment, with clinicians routinely using an EHR. In 12 practices that care for many pediatric patients with asthma, the EHR has embedded within it an internally developed and validated pediatric asthma control tool that guides the nurse or physician through eight questions related to the patient's asthma (focusing on changes since the last visit), with answers entered into the EHR. The EHR also has standardized documentation templates to facilitate severity classification, order sets to facilitate ordering medications and spirometry, and an asthma care plan that can be supplied to families. The tools are based on established guidelines developed by the National Asthma Education and Prevention Program (NAEPP) published by the National Heart, Lung, and Blood Institute.
- Color-coded, popup alerts on needed care and interventions: Based on information entered into the tools, the EHR provides color-coded popup alerts to physicians during the office visit to notify them of the severity of the patient's asthma (red if uncontrolled, green if controlled) and of the current need, if any, for the following interventions: a prescription for an inhaled corticosteroid, a spirometry test, and filing of an annual asthma care plan. The alert appears in red if the intervention is needed, and in green if not (i.e., the patient has already received the intervention as recommended in the guidelines). During the initial trial, physicians in 6 of the 12 practices received these alerts. Physicians in the other practices could access the same information, but to do so had to manually use the previously described embedded tools within the EHR.
- Easy ordering: To facilitate adherence to the alerts, the EHR makes it as easy as possible for clinicians to order the recommended services. For example, by clicking on one easily identified button, physicians can order inhaled corticosteroids or spirometry, or print out a care plan for the patient and family.
- Easy access to educational materials: A separate section of the popup screen provides links to asthma education materials that can be printed out and given to patients and families.
- Enhanced capacity to provide and interpret spirometry testing: Each clinic received a spirometer, as the practices had not historically offered this test (instead referring patients to a pulmonologist). As additional support for spirometry, physicians can participate (over the phone or in person) in monthly case discussions of spirometry results with a pediatric pulmonologist, and have on-call access to the pulmonologist who can assist with interpretation as needed.
- Between-visit discussions of asthma control: The decision-support tool prompts between-visit discussions of asthma control among primary care providers.
Context of the InnovationStarted in 1855 as the nation's first hospital devoted exclusively to caring for children, Children's Hospital of Philadelphia provides a full range of inpatient and outpatient services to pediatric patients. The hospital owns 31 primary care offices that collectively employ 675 providers and handle 675,000 visits a year. These offices care for approximately 185,000 children under the age of 18 years old, a meaningful portion of whom have asthma. The hospital operates a Pediatric Research Consortium that includes a number of these clinics, 12 of which participated in the trial of this program. The impetus for the program came from an Agency for Healthcare Research and Quality (AHRQ) request for proposals issued to its Practice-Based Research Networks. Leaders at the hospital, which operates a research network, felt the project offered a good opportunity to build upon earlier work the organization had done in embedding alerts and prompts related to immunizations in its EHR. The hospital has a group of talented physician bioinformaticians who have the skills to program the EHR to provide such alerts within the existing workflow of physicians. Because asthma represented a common condition, particularly among practices located in the heavily blue-collar, African-American West Philadelphia area, program leaders felt that automated alerts could improve adherence to asthma guidelines, and hence decided to pursue the AHRQ grant and develop and test the program.
ResultsThe program moderately increased adherence to the guidelines, as compared with a control group of similar practices that could access the same information only by manually using the tools in the EHR.
Strong: The evidence consists of a cluster-randomized trial that examined trends in adherence to key asthma guideline recommendations in two urban and four suburban primary care practices that implemented the program, compared with the same trends in two similar urban and four similar suburban practices not using the alerts.
- More controller medications prescribed: In two urban practices using the alerts, the proportion of eligible patients with a controller medication prescribed rose by 7 percentage points (from 71 to 78 percent), compared with a 1 percent rise (from 79 to 80 percent) in two similar urban practices that did not implement the alerts; this difference was statistically significant. Although the four suburban practices implementing the program also experienced a 7 percentage point rise (from 67 to 74 percent), this increase was not statistically different than the 3 percentage point increase (from 48 to 51 percent) experienced in the control group of four similar suburban practices.2
- More asthma care plans filed: In the four suburban practices implementing the program, the percentage of patients with persistent asthma who had an annual plan filed rose by 14 percentage points (from 39 to 53 percent), compared with an 11 point decline (from 47 to 36 percent) in the comparable suburban practices not using the alerts; this difference was statistically significant. No meaningful difference for this measure was observed in the urban practices.2
- More spirometry tests: The percentage of patients with persistent asthma who had a spirometry performed in the primary care office rose by 9 percentage points (from 15 to 24 percent) in the two urban practices implementing the program, compared with a 6 percentage point increase (from 16 to 22 percent) in the control group of urban practices. In the suburban practices, the percentage of patients having a spirometry test rose by 6 percentage points (from 8 to 14 percent) for those implementing the program, compared to a 7 percentage point decline (from 8 to 1 percent) in the suburban practices not implementing the alerts. For both sets of practices, these findings were statistically significant.2 Program developers attribute the very low rates of adherence (even after implementation of the alerts) to the fact that spirometry testing represented a new task for most of the clinics, particularly those in suburban settings. In fact, most practices had only recently received spirometers and training on how to perform and interpret the test. (See Planning and Development Process section for more details.)
Planning and Development ProcessKey elements in the planning and development process included the following:
- Forming team to develop plan, secure grant: The hospital formed a team, led by the head of its research network, to plan the program and write the proposal to AHRQ. The team included a primary care physician with experience in bioinformatics, a biostatistician to assist in study design, an ED physician with experience in health services research who understood the role of the ED in managing pediatric asthma patients, an expert in clinical decision support tool development, and several community-based pediatricians. The team met at least weekly (and sometimes daily) to plan the program and write the proposal.
- Designing alerts with clinician input: Team members, including physicians who worked in the hospital-owned clinics, visited with each of the practices, explaining the program and inviting them to participate. As part of these conversations, the team solicited input from doctors and nurses about the best way to design the computerized decision support. Their input emphasized the need for the alerts to come at the appropriate time and to not disrupt existing workflow.
- Integrating tools into EHR: To prepare for implementation, programmers embedded the previously described asthma management tools within the EHR at each of the 12 practices participating in the study. In addition, six of the practices had the real-time decision support programmed into the EHR, thus providing physicians in these practices with the automated alerts described earlier.
- Educating and training practices on new features: Two team members visited each of the 12 practices to teach physicians, nurses, and other staff about asthma and about the new features within the EHR (including the alerts at those practices that had access to them). Training consisted of a 4-hour session at each practice.
- Conducting initial trial: As described earlier, the alert system was tested over a 12-month period, with six practices participating and six similar practices serving as a control group.
- Considering program expansion: Since the trial ended in April 2008, the 6 practices that tested the alerts have continued to use them, and all 12 practices continue to use the embedded tools. A decision as to whether to roll out the alerts and embedded tools to all 31 owned practices has not yet been made, but will likely be based on physician input along with an analysis of other decision support tools available now or in the near future.
Resources Used and Skills Needed
- Staffing: Program development required intense work by the entire team over a 3- to 4-month period to plan and secure funding for the program; the team did this work as a part of their ongoing job responsibilities. Additional staff time was needed to design and program the embedded asthma support tools and the alert system, and to train the practices on how to use them. As noted, training involved two individuals who collectively spent 8 hours at each of the 12 practices. Once the alert system became operational, existing staff incorporated its use into their regular workflow.
- Costs: Planning, development, and implementation costs for the program, including research-related expenses, totaled $487,303, including $293,556 in direct costs and $193,747 in indirect costs.
Funding SourcesAgency for Healthcare Research and Quality
An AHRQ Translating Research into Practice grant funded program development and the cluster-randomized trial.
Getting Started with This Innovation
- Design system in collaboration with clinicians: Physicians, nurses, and other clinicians will likely not accept decision support tools unless they fit in well within existing workflows and/or make their lives easier. Even moderate disruptions that take just 1 or 2 minutes will not be tolerated by most physicians. To avoid this problem, include physicians and other clinicians who will be using the system in its design, and provide opportunities for them to provide input on proposed screen design, use of color, timing of the alerts, and other important issues. Often such input leads to system improvements. For example, the hospital did not initially plan to use colors to distinguish between alerts that require action and those that do not, but then switched to the red/green system in response to physician feedback.
- Provide ample training and support: Clinicians will not routinely use decision support unless they feel comfortable doing so. To help them get used to the system, invest the necessary time and effort in upfront training, and create a way for users to access support as needed if they run into problems after implementation.
Sustaining This Innovation
- Elicit ongoing feedback: Periodically survey or interview users to ensure their continued satisfaction with the system, and to elicit and address any concerns or issues they may be facing.
- Share information on updates, program impact: Keep clinicians informed of any changes or new developments with respect to the decision support tools and the guidelines on which they are based. Periodically share data demonstrating the impact of the system on adherence with guidelines and, if available, on clinical outcomes. Evidence documenting improvements will serve to keep clinicians engaged in and enthusiastic about the program.
- Consider revising or abandoning tools if ineffective: If the evidence suggests that decision support is not consistently being used and/or not producing positive results, consider revamping the tools or abandoning them. If such steps are not taken, clinicians may become overburdened with multiple decision support tools for various diseases and patient populations, leading them to stop using all of them (even those that work).
Additional ConsiderationsThis type of system is likely most effective when implemented in organizations that already own and regularly use an EHR or electronic medical record as part of their clinical workflow. Organizations without such systems may be able to use "freestanding" decision support tools, but this approach often proves ineffective, as such systems tend to disrupt physician workflow and may be viewed negatively by physicians not accustomed to using electronic tools.
Contact the InnovatorLouis M. Bell, MD
Chief, Division of General Pediatrics, Children's Hospital of Philadelphia
The Patrick S. Pasquariello, Jr, MD Endowed Chair in Pediatrics
Associate Chair for Clinical Activities, Department of Pediatrics, University of Pennsylvania
34th and Civic Center Boulevard
Philadelphia, PA 19104
Innovator DisclosuresDr. Bell has not indicated whether he has financial interests or business/professional affiliations relevant to the work described in this profile; however, information on funders is available in the Funding Sources section.
References/Related ArticlesBell LM, Grundmeier R, Localio R, et al. Electronic health record based decision support to improve asthma care: a cluster-randomized trial. Pediatrics. 2010;125(4):e770-7. [PubMed]
Bell LM, Grundmeier R, Localio R, et al. Electronic health record based decision support to improve asthma care: a cluster-randomized trial. Pediatrics. 2010;125(4):e770-7. [PubMed]
|Disclaimer: The inclusion of an innovation in the Innovations Exchange does not constitute or imply an endorsement by the U.S. Department of Health and Human Services, the Agency for Healthcare Research and Quality, or Westat of the innovation or of the submitter or developer of the innovation. Read more.|
Original publication: September 29, 2010.
Original publication indicates the date the profile was first posted to the Innovations Exchange.
Last updated: June 19, 2013.
Last updated indicates the date the most recent changes to the profile were posted to the Innovations Exchange.
Date verified by innovator: May 06, 2013.
Date verified by innovator indicates the most recent date the innovator provided feedback during the annual review process. The innovator is invited to review, update, and verify the profile annually.