Predicting anxiety in cancer survivors presenting to primary care – A machine learning approach accounting for physical comorbidity

Physical comorbidity and psychological distress are both major concerns for cancer survivors, and primary care providers can play a crucial role in managing both. This study used machine learning to explore potential predictors for anxiety in cancer survivors that present to primary care with physical comorbidity. Data from 496 cancer survivors was analysed, and results identified that physical symptoms (namely, fatigue/weakness, insomnia, and pain) were the most important predictors of anxiety. The degree of physical comorbidity was negligible. Overall, these findings indicate that the prediction of clinically significant anxiety in cancer survivors is feasible, and also highlight the importance of considering physical function in cancer survivors when assessing their psychological well-being, regardless of the degree of comorbidity. Read full text