A scoping review of ethics considerations in clinical natural language processing

Objectives To review through an ethics lens the state of research in clinical natural language processing (NLP) for the study of bias and fairness, and to identify gaps in research. Methods We queried PubMed and Google Scholar for articles published between 2015 and 2021 concerning clinical NLP, bias, and fairness. We analyzed articles using a framework that combines the machine learning (ML) development process (ie, design, data, algorithm, and critique) and bioethical concepts of beneficence, nonmaleficence, autonomy, justice, as well as explicability. Our approach further differentiated between biases of clinical text (eg, systemic or personal biases in clinical documentation towards patients) and biases in NLP applications. Results Out of 1162 articles screened, 22 met criteria for full text review. We categorized articles based on the design (N = 2), data (N = 12), algorithm (N = 14), and critique (N = 17) phases of the ML development process. Discussion Clinical NLP can be used to study bias in applications reliant on clinical text data as well as explore biases in the healthcare setting. We identify 3 areas of active research that require unique ethical considerations about the potential for clinical NLP to address and/or perpetuate bias: (1) selecting metrics that interrogate bias in models; (2) opportunities and risks of identifying sensitive patient attributes; and (3) best practices in reconciling individual autonomy, leveraging patient data, and inferring and manipulating sensitive information of subgroups. Finally, we address the limitations of current ethical frameworks to fully address concerns of justice. Clinical NLP is a rapidly advancing field, and assessing current approaches against ethical considerations can help the discipline use clinical NLP to explore both healthcare biases and equitable NLP applications. Lay Summary The objective of this work is to explore the ethical considerations of clinical natural language processing (NLP) in the context of bias. Bias here refers to systematic differences in terms of representation or application of NLP models between group identities like race, gender, and sexuality. We searched PubMed and Google Scholar for articles concerning NLP, ethics, and bias between 2015 and 2021. We analyzed articles against a framework that combines different stages of the machine learning (ML) development process (design, data, algorithm, critique) and important ethical principles in the medical domain. We included 22 out of 1162 prescreened articles in this review. Articles were categorized into: design (N = 2), data (N = 12), algorithm (N = 14), and critique (N = 17). Clinical NLP can be used to study bias in research that relies on clinical text as well as explore biases in the healthcare setting. We identify 3 areas of active research at the intersection of clinical NLP and ethics: (1) selecting performance metrics that interrogate bias in ML; (2) opportunities and risks of identifying sensitive patient information like gender, and sexuality; and (3) best practices in balancing individual autonomy, leveraging patient data, and inferring and manipulating sensitive information of subgroups.