Enhancing Provider Documentation in Pediatric Obesity Care: A Quality Improvement Project

Amber Ong, University of Illinois Chicago

Karla Lebedoff, University of Illinois Chicago

DOI: 10.58117/rdqq-7s75

Introduction

The prevalence of childhood obesity is steadily on the rise. In the United States, obesity affects approximately 14.7 million children (CDC, 2024a). This is an increase of 1.2 percent since 2015, and 4.3 percent up from 2005 (CDC, 2024b). This trend resulted in a $1.3 billion cost burden on the healthcare system in 2019 (CDC, 2024a). Patients not only share this financial burden, but face a higher likelihood of developing long-term adverse health outcomes such as diabetes, hypertension, heart disease, and depression (Kumar & Kelly, 2017). Many obesity comorbidities, such as type 2 diabetes, dyslipidemia, and non-alcoholic fatty liver disease, were formerly diagnosed as “adult” diseases. Now, such diseases are increasingly common in youth with obesity (Kumar & Kelly, 2017). Obesity requires chronic management and treatment, with life-long remissions and exacerbations (Hampl et al., 2023).

The etiology of obesity is multifaceted and complex. Biologic, environmental, and social factors are at play, including, but not limited to, lifestyle choices, breastfeeding practices during infancy, emotional distress, genetic predisposition, medication use, and social determinants of health (Kumar & Kelly, 2017). Although complex in nature, obesity is treatable. However, increasing prevalence further demands the action and attention of pediatric providers on a global scale. Pediatric primary care providers are well-positioned to identify, educate, and help manage children who are obese or at risk of becoming obese.

The American Academy of Pediatrics (AAP) currently recommends evaluating and treating pediatric obesity by performing a comprehensive assessment of risk factors, addressing nutrition and activity, and providing or referring children to intensive health behavior and lifestyle treatment (IHBLT) (Hampl et al., 2023). Implementing IHBLT in the clinical setting requires significant time, resources, and collaboration with primary care providers who are trained in lifestyle-related fields. The AAP defines the most effective IHBLT programs as 26 or more face-to-face hours over a three- to 12-month period, which aligns with the recommendation statement from the U.S. Preventive Services Task Force in 2010 (Hampl et al., 2023). Barriers to adherence from providers include lack of training or knowledge on lifestyle counseling and difficulty for busy clinical settings to accommodate frequent, longer visits (Hampl et al., 2023).

Provider adherence to AAP recommendations on evaluation and treatment of pediatric obesity remains inconsistent. In a 2018 survey, 88% of providers reported screening for obesity and only 7% met guidelines to refer or offer weight management services. Additionally, interventions that were offered did not meet the recommended contact hours (Staiano et al., 2017). There remains an ongoing gap between evidence-based recommendations and real-world clinical practice regarding childhood obesity prevention (Staiano et al., 2017).

This report discusses the results of a quality improvement intervention that was implemented at a Midwest primary care practice and offers recommendations for future advanced practice nurses managing obesity in pediatric patients. The aim of this project was to educate providers on improving electronic medical record documentation of health habits and follow-up in efforts to prevent obesity among pediatric patients.

Theoretical Framework

The Plan-Do-Study-Act (PDSA) model guided improvements throughout this project (Langley et al., 2009). This model evaluates the impact of implemented changes and allows for the refinement of interventions before re-evaluating them. In this quality improvement project, an educational intervention for providers was developed and implemented for PDSA cycle one. Data from chart reviews and a provider survey were collected and analyzed by the implementation team to determine the impact of the educational intervention and make changes for the following cycle. This process was performed twice for a total of three PDSA cycles.

Materials & Methods

This QI project was reviewed by the Institutional Review Board at University of Illinois Chicago and determined “IRB Exempt,” meaning it does not meet criteria for human subjects research. This QI project took place in a Midwest pediatric private practice comprising 12 providers, including six physicians and six nurse practitioners. Providers were recruited through an in-person introduction to the project by the project leader during scheduled downtime at the practice. All 12 providers agreed to participate and were scheduled for an educational session developed by the project leader. The educational presentation was developed in PowerPoint and provided a review of the AAP’s recommendations for screening, diagnosis, evaluation, and treatment of pediatric overweight and obesity (Appendix Figure 1; Hampl et al., 2023). According to AAP guidelines, providers were educated that body mass index (BMI) remains the most appropriate clinical tool for diagnosing overweight or obesity, with obesity defined as a BMI at or above the 95th percentile for age and sex. The oral presentation was 30 minutes long, offered on two separate days, and attended by all providers. Following the presentation, the project leader met individually with each provider to sign them off on documentation skills using a checklist. Each provider had to demonstrate how to document fruits and vegetables, sugary beverages, physical activity, screen time, plan for follow-up, and how to put in a referral. This concluded PDSA cycle one.

A chart review of PDSA cycle one data revealed only 8.86% of charts included appropriate follow-up or referral. Based on these results, PDSA cycle two introduced the addition of an EMR SmartPhrase to facilitate quicker documentation of overweight/obesity follow-up in patient plans. Typing in the dot phrase, “.kfbmifu” automatically populated the message: “Patient's BMI is above the [85th/95th] percentile for age and sex. Recommended follow-up visit in 1-3 months to monitor growth trends, support behavior changes, and assess progress. Patient/Family [supports plan/declines at this time].” Providers were informed and encouraged to use the smart phrase addition during a monthly provider meeting and through email.

During PDSA cycle 2, providers were given a physical feedback survey to complete. The survey was developed by the project leader and was not piloted. The survey included seven multiple-choice questions on documentation practices, follow-up, and knowledge of referrals/resources. The surveys were completed anonymously and handed into a labeled folder in the office. Of the 12 providers who agreed to participate in the project, one was unavailable for the post-intervention period due to leave and was considered lost to follow-up. A total of ten survey responses were received, including two that did not complete all survey items.

Survey responses were analyzed using bar charts to visually display the distribution of provider responses for each item (Appendix Figure 2). Survey feedback revealed that only two providers identified being familiar with local or system-based nutritional counseling resources and seven providers indicated a referral list would be helpful. Based on these results, the implementation of a referrals list was decided for PDSA cycle three. The project leader consulted with the practice’s referral coordinator and with providers who were most familiar with nutrition and weight management referrals. A definitive referral list was created in Microsoft Word and saved on the private practice’s shared drive. Copies were made and placed at both nursing stations on the floor to be handed out to families. Lastly, the project leader informed providers individually of the handout and where it could be accessed.

Outcomes were measured through a chart review spanning three months before and three months after the intervention. Eligible charts for review included patients aged 3-18 years, seen in the office for an annual well visit, and with a measured BMI at or above the 85th percentile. Patients under three years of age were excluded from participating because the EHR system did not allow providers to document discussions of BMI with family, provide educational handouts, or refer patients under three years old to nutritional counseling for a BMI above the 85th percentile. The post-intervention patient information was collected using SlicerDicer, a self-service reporting tool in Epic EHR that allows users to pull customizable patient data over a specified period and easily access patients’ charts. After entering the eligibility criteria into SlicerDicer and removing charts from outside providers, the report pulled 1168 patient encounters during the post-intervention period. Duplicate encounters from the same patient were avoided, as the report extracted encounters for annual well visits over a three-month period.

The project leader audited all eligible patient charts to check for provider documentation of (1) fruits and vegetables, (2) sugary beverages, (3) physical activity, (4) screen time, and (5) appropriate follow-up/referral. These outcome measures were selected based on their relevance to early obesity identification, prevention counseling, and adherence to AAP clinical practice guidelines. The first four measures align with the AAP’s 5-2-1-0 healthy lifestyle framework which promotes the following daily recommendation for children and adolescents: 5 or more fruits/vegetables, 2 hours or less of screen time, 1 hour of physical activity, and 0 sugary drinks (Bolling & Rogers, 2019). Follow-up/referral documentation was also included because deficiencies in obesity-related follow-up and referral practices are consistently reported in the literature and were identified as an area for improvement within the practice setting (Busch et al., 2018; Satti et al., 2021). All measures were quantifiable through chart review, relevant to clinical practice, and directly targeted by the interventions. Documentation frequency for each measure was calculated by dividing the number of encounters with documented assessment or follow-up by the total number of eligible encounters during the pre- and post-intervention period. Data was transferred to an Excel spreadsheet to calculate descriptive statistics and create run charts. At the end of the project, pre- and post-intervention percentages in provider documentation were compared using chi-square analysis in Excel. Given the categorial nature of the independent variables and the binary nature of the outcomes (documented vs. not documented), the chi-square test provided the most appropriate method for evaluation and was used to calculate p-values, assessing whether observed changes in documentation were likely due to the intervention rather than chance.

Criteria for appropriate follow-up/referral documentation included: in-office follow-up sooner than one year, continued follow-up with a specialist, or referral to IHBLT. All documented offerings for follow-up or IHBLT were included, even if the patient/family declined. Exclusion criteria for follow-up/referral documentation included documentation of an elevated BMI attributed to a highly muscular build, significant improvement in BMI over time, or a borderline overweight BMI with documentation of healthy eating and being active.

Results

The initial steps of the intervention included developing an educational PowerPoint presentation on AAP guidelines for pediatric overweight and obesity, scheduling time for provider education, and delivering the educational session to providers (Appendix Figure 4). Following the educational intervention, a chart review revealed that follow-up/referral was the least documented measure. As a result, the next intervention focused on developing an EMR SmartPhrase to standardize and streamline follow-up documentation. Subsequently, the project leader developed and administered a provider feedback survey to assess documentation practices, follow-up processes, and knowledge of available referrals and resources. Survey findings revealed limited familiarity with referral resources, with provider interest in a standardized referral list. Therefore, the final intervention involved developing a comprehensive referral list for nutrition and weight management programs to educate providers and provide handouts for families.

When comparing pre- and post-intervention documentation rates, outcomes revealed fruits and vegetables increased by 0.2% (p>0.05), sugary beverages increased by 10.2% (p<0.05), physical activity increased by 1.4% (p<0.05), and follow-up/referral increased by 9.6% (p<0.05) (Appendix Table 1). The increase in documentation of fruits and vegetables was not statistically significant, as its documentation rate prior to intervention was nearly 100%. Screen time documentation declined by 1.2% (p>0.05). However, screen time documentation rates were maintained above 94% throughout the project post-intervention (Appendix Figure 3). Both fruits and vegetables, and physical activity were maintained at 100% throughout all PDSA cycles.

Discussion

This QI project showed that educational interventions can be effective in improving documentation and can support individual practice change to prevent childhood obesity. Improvement in provider documentation was seen in four out of five measures. The use of SlicerDicer in Epic demonstrated that EMR-based monitoring supports ongoing quality improvement.

Limitations include single-site implementation and a short project timeline of three months. Because of the short timeline of this project, the long-term sustainability of these results cannot be determined, indicating the need for further investigation with longer PDSA cycles. Additionally, the project was conducted during an EHR transition. The project site transitioned from using PCC to Epic one week after delivery of the education session to providers. Post-intervention data collection was delayed for two weeks to avoid skewed data due to provider unfamiliarity with the new system. In addition to being a time of heavy learning and adjustment, providers adapted to “well visit” templates through Epic, which included assessment of balanced diet, consumption of juice/soda, physical activity, and screen time. However, this project demonstrated that the presence of these well visit templates did not result in a 100% documentation rate across all measures. Furthermore, the documentation for follow-up and process for referrals was not affected by this EHR change and remained similar across both systems.

Challenges of this project included provider practice variation, disagreements with guideline recommendations, and the deletion of certain assessments within the well-visit templates. After delivery of the provider education session, disagreements were expressed with the AAP’s clinical practice guideline recommendation of follow-up or referral to IHBLT. The suggestion of two contact hours per month and 26 hours over 3-12 months for optimal effect was seen as unrealistic by some providers. Furthermore, some providers stated that they do not routinely refer patients with a BMI over the 85th percentile but will for patients with a BMI over the 95th percentile, despite current recommendations. This provider variation was echoed in the provider feedback survey along with the majority of providers answering that they do not use any EMR SmartPhrases when documenting obesity-related care. Despite the creation of the SmartPhrase to help document follow-up quickly and easily, only two providers used the phrase post-intervention. Potential reasons for low usage include provider preference, perceived usefulness, and competing clinical demands, which may have limited retention of SmartPhrase information introduced through meetings and email reminders. Strategies for future cycles include involving providers in the development and wording of the SmartPhrase, having providers demonstrate use to increase familiarity with how to populate the phrase, and adding auto-populating fields to allow providers to further tailor their documentation. Finally, provider documentation of screen time was the only measure to decline post-intervention despite PDSA cycle interventions, including automatic population of screen time documentation within the well visit template; in fact, three providers deleted this section from ten well visit notes. One provider stated screen time documentation was deleted because the subject was not discussed during the visit. No other patterns were recognized across the included charts.

The results of this QI project are consistent with findings reported in the literature. Similar to the QI project by Busch et al. (2018), provider education was associated with increased rates of referral recommendations for children overweight and with obesity. Additionally, Satti et al. (2021) and Tucker et al. (2024) demonstrated that provider education combined with EMR-based interventions improved adherence to AAP guidelines for the screening and management of pediatric overweight and obesity. Similarities in findings may be attributed to the use of comparable quality improvement methodologies, including PDSA cycles, implementation of provider education and EMR-based tools, and engagement of key stakeholders throughout the project. These interventions likely improved provider awareness, standardized documentation practices, and increased accessibility of referral resources, contributing to improved guideline adherence.

The results of this project have important implications for nursing practice. Implementation of provider-focused interventions can lead to meaningful improvements in documentation of diet, activity, and follow-up. Assessment of diet and activity can identify obesity-related behaviors and lead to early identification of at-risk children. Establishing appropriate follow-up at well visits is equally important as failure to do so limits intervention opportunities for children who are overweight and/or living with obesity. Clinical practices can ensure providers are conducting these assessments and scheduling timely follow-ups objectively through documentation. Therefore, improved documentation may support practice change by prompting providers to consistently assess obesity-related behaviors, deliver targeted counseling, place referrals, and schedule appropriate follow-up visits in alignment with AAP guidelines. Furthermore, consistent documentation supports continuity of care by allowing providers to monitor patient progress over time, reinforce lifestyle recommendations at following visits, and identify patients who may require additional intervention or referral. The components of the intervention helped address identified barriers, including provider knowledge, workflow efficiency, and accessibility of resources. Therefore, these results may be transferable to other practices with similar barriers.

Conclusion

Provider-focused interventions, including education sessions, EMR enhancements, and increased accessibility to resources, can significantly improve provider documentation of nutritional and activity assessments, follow-up, and referral. For advanced practice nurses managing obesity in pediatric patients, increasing adherence to these evidence-based practices supports obesity prevention through early identification and proactive management of children at risk for obesity. Next steps include investigating the sustainability of change over time and the use of provider feedback to tailor interventions and strengthen stakeholder engagement.

Acknowledgments

Amber Ong https://orcid.org/0009-0007-0603-6281

Karla Lebedoff https://orcid.org/0000-0002-3599-2412

Generative AI Statement

I, Amber Ong, hereby confirm that this manuscript is the result of my own (and my coauthors’) independent scholarly work, and that all material from the work of others has been fully acknowledged. No content generated by AI technologies has been used in this assessment.

Funding

No external financial support or grants were received from any public, commercial, or not-for-profit entities for the research, authorship, or publication of this article.

Conflict of Interest

The authors declare no conflicts of interest regarding this manuscript.

References

Appendix

Tables

Table 1. Pre- and Post-Intervention Documentation Outcomes

Note. Percentiles represent the proportion of charts meeting documentation criteria. P-values were calculated using chi-square analysis.

Figures

Figure 1. Pediatric Overweight and Obesity Algorithm

Note. Adapted from Hampl et al. (2023) clinical practice guideline; modified to reflect project implemented components.

Figure 2. Provider Feedback Survey Results

Note. The survey was distributed to 11 providers. One provider did not complete the survey and some providers did not complete all items, resulting in variable response counts across survey questions.

Figure 3. Run Charts of Documentation Performance Before and After Intervention

Figure 4. Evolution of the Intervention