Discussion
Through inter-professional collaboration with hematology/oncology, endocrinology, hospital medicine, pharmacy, and IT, we successfully improved the testing, diagnosis, supplementation and monitoring of Vitamin D deficiency and insufficiency for children with newly-diagnosed cancer and demonstrated sustainability over 24 months. We achieved and sustained our aim to improve Vitamin D testing, adequate Vitamin D supplementation and accurate and timely follow-up testing post-supplementation. We exceeded our SMART Aim of ≥80% compliance with the newly developed institutional guideline and clinical decision-tree after implementation and sustained our aim of ≥80% over time.
Children with cancer are at high risk for Vitamin D deficiency and insufficiency due to inadequate sun exposure and treatment-related complications, such as poor diet, hepatotoxicity and/or nephrotoxicity of chemotherapeutic agents, impaired absorption due to mucositis and/or colitis, and interference of Vitamin D metabolism by glucocorticoids. Current Endocrine Society Clinical Practice Guidelines on the evaluation, treatment, and prevention of Vitamin D deficiency recommend testing patients at risk for deficiency with the measurement of serum 25(OH)D concentrations, followed by treatment with either oral Vitamin D2 or Vitamin D3 supplementation in patients found to be deficient or insufficient.13 Research from our group showed that Vitamin D supplementation significantly increases serum 25(OH)D concentrations.12
Low bone mineral density has been reported in pediatric cancer survivors even several years after completion of cancer therapy.16, 17 Current national survivorship guidelines18 recommend a baseline bone mineral density evaluation followed by laboratory assessments, in order to evaluate bone health beginning two years after therapy completion, particularly for patients with history of acute lymphoblastic leukemia.17, 18
Our QI methodology led to successful establishment of the QI team, with specific roles for team members; continued communication of the QI initiative aim and progress to staff; two-way feedback on processes implementation; and accurate and efficient patient identification for data analysis. The EMR-based interventions implemented by the IT team for automated BPA alerts and triggers for Vitamin D laboratory testing and monitoring may have contributed to the observed increase in compliance with the newly developed institutional guideline for Vitamin D testing and supplementation. After conducting educational interventions regarding the standardized criteria for Vitamin D deficiency and insufficiency diagnoses, and the clinical decision-making tree for adequate supplementation, we observed a significant increase in testing and supplementation rates. This finding is consistent with previous studies showing that passive EMR features such as BPA alerts need to be accompanied by complementary strategies such as educational interventions to improve compliance with guidelines and achieve sustainability over time.19
Regular feedback about group performance, compared to prior performance were positively received, and helped motivate oncology providers to further adhere to the newly implemented practices. Although additional research is needed to establish the benefit of performance feedback (individual or group) in improving guideline compliance and sustained culture change, our findings support previous studies showing the positive impact of providing timely feedback on improving clinical practice and EMR documentation.20 Additionally, hospital leadership support and awareness of the clinical need, and frequent communication with key stakeholders contributed to achieving and sustaining our aim.
There are several limitations in our project to be discussed. First, we focused on one aspect of clinical care in a specific setting. Second, we reviewed data for three months prior to our interventions on a small group of patients. Studying pre-intervention data in a larger group and over a longer period may help to determine temporal variation. Third, our EMR interventions were geared toward BPA alerts and automated testing options in our admission and chemotherapy order sets, which are specific components of our EMR and may not be reproducible in other practices that do not use a similar EMR. However, as most centers managing pediatric cancer have had EMRs in place for many years, this limitation should be resolvable. While motivations for performing testing or prescribing supplementation were not explored, providers’ perceived utility of the decision-making tree and automated triggers, and positive feedback of frequent reports and reminders, contributed to steady improvement over the first 4 months of the QI initiative and sustainability for 24 months. Additionally, our QI initiative could have been further strengthened by assessing balancing measures to determine whether there were unintended consequences in other parts of the clinical workflow as a result of our QI initiative, such as documentation of time spent by the provider reviewing the clinical decision-making tree, ordering tests and prescribing supplementation.