Creating computable algorithms for symptom management in an outpatient thoracic oncology setting

Area of Science:

• Oncology
• Clinical Informatics
• Evidence-Based Medicine

Background:

• Effective symptom management is crucial for quality cancer care but often lacks an evidence base.
• Automating and adapting national guidelines for point-of-care use can improve clinician adherence.

Purpose of the Study:

• To describe the adaptation of research evidence into a clinical decision support system for individualized symptom management.
• To provide clinicians with point-of-care recommendations for common lung cancer symptoms.

Main Methods:

• A modified ADAPTE process and nominal group technique were used by expert panels.
• National guidelines were adapted and integrated with research evidence to create computable algorithms.
• Algorithms were developed for pain, fatigue, dyspnea, depression, and anxiety in lung cancer patients.

Main Results:

• Multidisciplinary groups approved computable algorithms for key symptoms, including various pain levels.
• Algorithms incorporated patient-specific factors like age, comorbidities, and medications for tailored interventions.
• Algorithms were reconciled for managing multiple concurrent symptoms.

Conclusions:

• A modified ADAPTE process and nominal group technique facilitated the creation of approved, locally adapted algorithms.
• The development process was resource-intensive but yielded expert-validated computable algorithms for individualized symptom management.
• These algorithms support evidence-based, tailored symptom management for lung cancer patients.