Background on Clinical Documentation Practices
Clinical notes in intensive care units serve as critical records of patient encounters, capturing observations, decisions, and care plans. Language choices within these notes can reflect broader cognitive and relational dynamics between clinicians and patients. The recent publication in CHEST Critical Care examines one specific linguistic feature: the use of personal pronouns by note authors.
Personal pronouns such as "I" or "we" represent instances of clinician self-referentiality. This practice may signal a sense of personal agency, team collaboration, or direct involvement in the patient's narrative. Understanding when and why these pronouns appear offers insights into documentation styles across different healthcare roles.
Overview of the Research Team and Publication
The study, titled Clinician self-referentiality: factors associated with author pronoun use in critically ill patient notes, was led by Kristen Vossler along with co-authors Zoë Kopp, Natalie Stumpf, Teva Brender, Samuel K. McGowan, Shannen Kim, Edie Espejo, John Boscardin, Hunter Mills, Allyson C. Chapman, Steven Z. Pantilat, and Julien Cobert. Most contributors are affiliated with the University of California, San Francisco, with additional ties to Stanford University and the San Francisco Veterans Affairs Medical Center.
The work appeared online on June 23, 2026, in CHEST Critical Care and is available through ScienceDirect at https://www.sciencedirect.com/science/article/pii/S2949788426000614. It builds on an earlier abstract presented in 2025 and represents a large-scale analysis of electronic health record data from academic medical centers.
Study Design and Data Sources
Researchers conducted a retrospective cohort study using notes written for adult patients admitted to intensive care units at UCSF between 2012 and 2022. The dataset encompassed 842,953 notes across 5,799 unique patients, providing substantial statistical power for identifying patterns.
The primary outcome measured personal pronoun frequency per note, expressed as a rate per sentence. Covariates spanned patient sociodemographics, author characteristics including gender and professional role, gender concordance between author and patient, ICU type, and selected clinical variables. Statistical modeling employed cluster-robust Poisson regression, with sensitivity analyses to test robustness.
Key Findings on Pronoun Prevalence
Personal pronouns appeared in 53 percent of the examined notes. This prevalence indicates that self-referential language is common yet far from universal in critical care documentation. The distribution varied significantly depending on who authored the note rather than on patient-specific factors.
Patient demographics and clinical characteristics showed minimal association with pronoun frequency. Similarly, gender concordance between the note writer and the patient did not influence usage rates. These results suggest that author identity and professional context drive linguistic choices more than the clinical scenario itself.
Photo by Abdulai Sayni on Unsplash
Author Characteristics as Primary Drivers
Female authors demonstrated lower rates of personal pronoun inclusion compared with male authors. Residents used fewer personal pronouns than attending physicians. Advanced practice providers, registered nurses, respiratory therapists, and physical or occupational therapists all showed reduced pronoun use relative to physicians holding MD or DO degrees.
Pronoun patterns also differed by ICU type and the author's medical specialty. Sensitivity analyses restricted to authors with prior pronoun use largely confirmed the main results, while subset analyses excluding attending notes revealed some variations. Distinctions emerged between first-person singular "I" and plural "we," hinting at nuanced differences in how clinicians position themselves individually or as part of a team.
Implications for Medical Education and Training
These findings carry direct relevance for medical schools, residency programs, and interprofessional training at universities worldwide. Documentation styles may be shaped early in training, with attendings modeling higher rates of self-reference that residents and other team members adopt or modify.
Academic medical centers could incorporate targeted modules on reflective documentation practices. Such training might help standardize approaches while preserving individual voice where clinically appropriate. Faculty development programs at institutions like UCSF could use the study data to illustrate how role and experience influence note composition.
Broader Context in Healthcare Communication
Electronic health records have transformed how clinicians record encounters, yet the human elements of language persist. Self-referential pronouns may foster accountability or convey empathy, but excessive use could shift focus away from the patient. The study underscores the need for future work examining both author intent and how readers, including patients and families, interpret these choices.
Similar explorations in other specialties or care settings could reveal whether patterns observed in ICUs generalize. Comparative studies across institutions would further illuminate cultural or regional influences on documentation norms.
Stakeholder Perspectives and Potential Applications
Clinicians, hospital administrators, and health informatics specialists each bring unique viewpoints. Note writers may appreciate data showing that professional role strongly predicts style, potentially reducing concerns about individual scrutiny. Administrators might consider how documentation guidelines align with observed practices to support quality improvement initiatives.
Patients accessing their records through portals could encounter varying levels of personal voice. Understanding these variations helps set expectations and informs discussions about transparency in clinical communication. Professional societies focused on critical care may reference the work when updating best-practice recommendations for note writing.
Photo by National Cancer Institute on Unsplash
Future Research Directions and Limitations
The authors recommend additional studies probing author intent behind pronoun choices and reader perceptions of those notes. Qualitative interviews with clinicians could complement the quantitative findings. Limitations include the single-institution focus at UCSF and the observational design, which precludes causal inferences.
Expanding analyses to multiple centers and incorporating natural language processing techniques could refine understanding of linguistic patterns. Longitudinal tracking of individual clinicians over career stages might reveal how pronoun use evolves with experience.
Practical Takeaways for Academic and Clinical Communities
Institutions training the next generation of healthcare professionals can use these results to foster awareness of documentation habits. Workshops emphasizing patient-centered language alongside appropriate self-reference may enhance both clarity and relational aspects of notes.
Health systems evaluating electronic health record templates or AI-assisted documentation tools might factor in author-role differences. The open-access availability of the paper supports widespread dissemination and discussion across academic and clinical settings.
