Gender-sensitive word embeddings for healthcare.

Abstract

To analyze gender bias in clinical trials, to design an algorithm that mitigates the effects of biases of gender representation on natural-language (NLP) systems trained on text drawn from clinical trials, and to evaluate its performance. We analyze gender bias in clinical trials described by 16772 PubMed abstracts (2008-2018). We present a method to augment word embeddings, the core building block of NLP-centric representations, by weighting abstracts by the number of women participants in the trial. We evaluate the resulting gender-sensitive embeddings performance on several clinical prediction tasks: comorbidity classification, hospital length of stay prediction, and intensive care unit (ICU) readmission prediction. For female patients, the gender-sensitive model area under the receiver-operator characteristic (AUROC) is 0.86 versus the baseline of 0.81 for comorbidity classification, mean absolute error 4.59 versus the baseline of 4.66 for length of stay prediction, and AUROC 0.69 versus 0.67 for ICU readmission. All results are statistically significant. Women have been underrepresented in clinical trials. Thus, using the broad clinical trials literature as training data for statistical language models could result in biased models, with deficits in knowledge about women. The method presented enables gender-sensitive use of publications as training data for word embeddings. In experiments, the gender-sensitive embeddings show better performance than baseline embeddings for the clinical tasks studied. The results highlight opportunities for recognizing and addressing gender and other representational biases in the clinical trials literature. Addressing representational biases in data for training NLP embeddings can lead to better results on downstream tasks for underrepresented populations.