Optimizing pain management in breast cancer care: Utilizing 'All of Us' data and deep learning to identify patients at elevated risk for chronic pain.
Jung In Park, Steven Johnson, Lisiane Pruinelli
PMID 39056443WHAT IT FOUND
A deep learning model predicted chronic pain risk in breast cancer patients with 82% accuracy.
It identified age at diagnosis and survey responses about healthcare respect as key predictors, but low precision means many false alarms, so it is not yet ready for clinical use.
Key findings
01The model achieved an Area Under the Curve (AUROC) of 82.0% and a recall of 68.4% on the test set, but precision was only 38.4%.
02Age at cancer diagnosis was the most significant demographic feature, with an importance score of 0.139.
03Survey questions about whether healthcare providers acted as if the patient was not smart or treated them with less respect were among the top predictors of chronic pain.
STILL TO COME
How it was doneWhat they foundWhat it means for RNs
Read the rest of this summary
You get three full summaries a month, free, and we do not ask for a card. Search, the TL;DRs and your library stay unlimited either way.
What it does not show
The dataset lacked diversity, with 88.2% of patients being White and 98% non-Hispanic, limiting generalizability. The model had low precision (38.4%), meaning it produced many false positives, making it unreliable for screening individual patients. The study only included patients with survey data, which may introduce selection bias. The model did not include medication or genetic data, which are known factors in pain outcomes.
Declared interests
The study used the 'All of Us' Research Program's Controlled Tier Dataset. The text does not explicitly state funding sources or conflicts of interest for the authors, though it notes the data was de-identified and exempt from IRB review.
The easy way to misread this
Do not use this model to screen patients for chronic pain risk in clinical practice. The low precision (38.4%) means it incorrectly flags many patients as high risk who do not develop chronic pain, leading to potential overtreatment or unnecessary anxiety.