Essential Yet Invisible: Professional Identity Formation Among Academic Internal Medicine Hospitalists

Study Design

This qualitative study employed thematic analysis and was grounded in a constructivist paradigm, which frames identity as continuously produced through individuals’ interpretations of and interactions with their social environment [16]. This stance was well suited to our aim of understanding how AHs construct professional identity within the academic enterprise, complementing Burke’s emphasis on identity as an iterative meaning-making process [17]. This approach enabled us to examine not only how AHs construct their identities but also the institutional and structural factors that enable or constrain this process. To enhance geographical and institutional diversity, we purposefully recruited participants across multiple settings in the U.S., and utilized subsequent snowball sampling to include AH from different institutions, and roles within academic medicine.

Identity formation is a topic that is complex and has many theoretical traditions [18]. Social identity theory suggests that professional identity is constructed through the roles individuals occupy and the interactions through which those roles are negotiated and validated [19]. We selected Burke’s Identity Theory because it conceptualizes identity as an ongoing feedback process between individuals and social structures, offering an analytic lens on a group like AHs whose professional identity is still being negotiated and understood within academic medicine [15]. The theory distinguishes roles – the externally defined expectations such as clinical service, teaching, and scholarly productivity from identity, the internalized meanings and standards individuals hold about who they are and what their work means. For AHs, who occupy multiple roles that are often incompletely recognized by traditional academic reward structures, this distinction allows us to move beyond describing role strain to examining how identity itself is formed, contested, and stabilized through iterative interactions with institutional feedback. In the context of AHs, this means that identity formation is dynamic, relational, and shaped by institutional structures making it both a site of vulnerability and an opportunity for meaningful institutional intervention.

Educational Setting and Participant Sampling

The study was conducted across nine academic medical centers in the United States in 2024–2025. AHs were recruited through the author group’s specialty and professional networks using snowball sampling to ensure geographic and institutional diversity. Sample size was determined using information power, which refers to the degree to which a sample provides information pertinent to the specific study aims [20]. Given our focus on identity formation in a relatively homogenous professional group, we estimated that 20 interviews would provide sufficient information power to identify patterns while maintaining data quality and depth. This estimation was informed by preliminary literature review and refined through iterative team discussion.

Institutional Review Board approval for human subjects research was obtained from The George Washington University prior to study commencement and received an exempt status (NCR245647).

Participant Characteristics

Twenty AHs participated; demographic characteristics are summarized in Table 1. The cohort was predominantly aged 31–40 years (70%), with 0–10 years of experience as attending physicians (85%). Most were employed in urban institutions (85%), with geographic representation from the Northeast (55%), South (25%), West (15%), and Midwest (10%). The majority held assistant professor rank (70%); 65% had engaged with institutional faculty development, 25% had completed longitudinal teaching courses, and 10% held a master’s in medical education. The cohort was 65% White, 25% Asian, 5% Black, and 5% Hispanic/Latino; 55% identified as female. Nearly all identified as straight/heterosexual (95%), and 5% were international medical graduates. Twenty-five percent had prior non-academic medicine experience; and 35% worked night shifts.

Data Collection

A semi-structured interview guide (Supplementary Material 1) was developed by the research team and grounded in Burke’s identity theory. Drawing on the theory’s distinction between externally defined roles and internalized identity standards, questions explored the roles participants occupied as AHs, the meanings they ascribed to each, and how they experienced congruence or dissonance between institutional expectations and their own identity standards.

The interview guide was pilot tested on two AHs and modified based on participant feedback. For example, pilot feedback included that one probe was leading for participants, and this particular probe was removed. All interviews were conducted virtually on Zoom by a trained research assistant (BG) and lasted approximately 45–60 minutes. Demographic information was collected prior to interviews. After obtaining informed consent, interviews were audio-recorded and transcribed using a secure transcription service. Transcripts were de-identified and reviewed for accuracy (SS and BG) before analysis. Data collection and analysis occurred iteratively, with early interviews informing subsequent interviewing and ongoing refinement of analytic focus.

Data Analysis

De-identified transcripts were uploaded into Dedoose® a secure online qualitative analysis platform. Initial coding was informed by identity theory as a sensitizing framework, guiding analytic attention to roles, recognition, and identity negotiation, while remaining open to inductively generated codes. Braun and Clarke’s six-phase thematic analysis method was used to identify patterns in the data [21]. We performed open coding of initial transcripts, followed by consolidation of codes into a codebook. Each transcript was independently coded by two research team members (SS and BG) and coding discrepancies were resolved through discussion during team meetings. Codes were then iteratively compared across transcripts, merged or refined based on conceptual overlap, and consolidated through collaborative team discussion into themes representing coherent and distinct patterns of shared meaning.

We monitored the adequacy of our data throughout concurrent data collection and analysis using the concept of information power [20]. Several features of our study contributed to the richness of our dataset and informed our judgment that the sample held sufficient information power: a narrowly defined aim, participants with characteristics highly specific to the research question, and an analysis grounded in Identity Theory. As interviews progressed, the research team met regularly to review transcripts and discuss emerging themes. We assessed adequacy by monitoring the extent to which successive interviews generated new conceptual insights relevant to the study aims, as opposed to elaborating or confirming themes already identified. After 20 interviews, the team reached consensus that additional interviews were yielding primarily confirmatory rather than novel information, and that the dataset enabled a sufficiently rich and nuanced analysis to address the research questions. At that point, recruitment was concluded.

Critical Reflexivity

The research team engaged in ongoing reflexive practice throughout the study, attending to how our differing positionalities shaped question design, data collection, and interpretation. SS, a woman of color and practicing AH, brought insider knowledge of hospital medicine and the academic expectations placed on AHs, which informed interview guide development and facilitated rapport during recruitment. This proximity required deliberate reflexivity to guard against assumptions of shared meaning and over-identification with participants’ experiences. ZZ, a first-generation immigrant woman of color and academic internist with inpatient and outpatient experience and expertise in qualitative research, contributed a contextually informed yet analytically distanced perspective; she frequently challenged early interpretations and pressed the team to interrogate how identity, power, and institutional structure shaped participants’ narratives. BG, a white medical student with qualitative research experience focused on health disparities and the experiences of sexual and gender minority patients, brought an outsider perspective to academic hospital medicine and conducted all interviews to reduce power differentials and encourage candid discussion. Reflexivity was operationalized through regular team meetings and written exchanges in which we examined our backgrounds, assumptions, and disciplinary commitments, and considered how these might shape our engagement with the data.

Overview of Themes

This study explored how AHs construct and stabilize professional identity within the evolving landscape of academic hospital medicine. Guided by Burke’s Identity Theory, the findings are organized to reflect how participants navigated the iterative feedback between their internalized identity standards and the institutional environments within which those standards were affirmed, challenged, or left unverified. The themes are therefore presented not as discrete topics but as interconnected identity processes, each illustrating how role expectations, institutional feedback, and self-perception interact to shape professional identity under conditions of ambiguity, competing demands, and inconsistent recognition. Four interrelated themes were identified (Figure 1): (1) the absence of consensus: defining what it means to be an Academic Hospitalist; (2) identity non-verification: the mismatch between self-perception and external recognition; (3) competing responsibilities, tensions, and burnout; and (4) mentorship and professional growth.

Figure 1

We interpret these themes as connected through a process rather than as discrete findings. The absence of a shared definition of the AH role (Theme 1) is not merely a conceptual gap; it removes the stable anchor against which AHs and their institutions might align expectations. Without that anchor, the discrepancy AHs described between how they understood their own contributions and how those contributions were recognized by others (Theme 2) becomes difficult to resolve, because there is no agreed standard to appeal to. This unresolved discrepancy was, in our reading, a precondition for the role strain participants articulated (Theme 3): competing clinical, educational, and scholarly demands are more depleting when the work itself is not legibly valued, and when AHs must continually justify a role whose boundaries are contested. Mentorship (Theme 4) functioned in participants’ accounts as the primary site where these tensions were named, interpreted, and partially absorbed — less a solution than a relational mechanism through which AHs made the ambiguity workable. Read together, the themes describe how definitional ambiguity propagates into recognition, strain, and adaptive response, which is why we present them as interconnected rather than parallel.

Theme 1: The Absence of Consensus: Defining What it Means to be an Academic Hospitalist

AHs described pervasive definitional ambiguity about what constitutes their role, which emerged as a foundational source of identity uncertainty shaping how they understood what it means to “be” an AH and what forms of work confer academic legitimacy. While some role ambiguity exists across academic medicine, participants’ accounts pointed to a distinctive form of uncertainty rooted in the structural features of hospital medicine itself. Participants pointed out that unlike established specialties, which developed over decades alongside defined fellowship pathways, recognized scholarly traditions, and stable promotion norms, hospital medicine is a comparatively young, generalist field whose academic roles emerged in response to workforce shortages and clinical service demands rather than a coherent scholarly mandate. Without a founding disciplinary identity organized around a specific organ system, procedural domain, or research tradition, AHs have inherited academic roles whose expectations for scholarship, teaching, and productivity remain inconsistently defined across institutions and, often, within the same division. This structural ambiguity surfaced along three interrelated dimensions: whether teaching alone qualifies as academic work, whether scholarship is required, and whether academic identity is conferred by institutional location or by the nature of the work itself.

Theme 2: Identity Non-verification: The Mismatch Between Self-Perception and External Recognition

AHs described a persistent mismatch between how they understood their contributions and how those contributions were recognized by institutions, which participants experienced as failed identity verification that undermined their sense of professional legitimacy.

Theme 3: Competing Responsibilities, Tensions, and Burnout

AHs described competing clinical, educational, and scholarly demands as a source not only of logistical strain but of identity strain, particularly when institutional structures limited their ability to enact roles aligned with their academic values. These pressures manifested through the expectation to volunteer for academic work before earning protected time, insufficient protected time for non-clinical engagement, pressure to produce scholarship without support, and structural isolation through night shifts.

Theme 4: Mentorship and Professional Growth

AHs described mentorship as central to professional growth and identity formation, yet access to meaningful mentorship remained limited and unevenly distributed. Consistent with identity theory, mentorship functioned as a core mechanism through which identity was negotiated, validated, and made visible, providing role modeling and belonging in the absence of clear institutional pathways. Participants described mentorship as both a personal need and a professional responsibility central to their academic identity.

Interpretation and Meaning

Burke’s identity theory posits that a coherent sense of self develops through the roles individuals occupy and through social validation of behaviors aligned with those roles [19]. Our data suggest that in academic hospital medicine, institutional structures frequently prevent this alignment, producing “identity dissonance” the internal conflict that arises when individuals cannot enact roles consistent with their self-concept or when others fail to recognize them [19].

Many participants described “niche development”, anchoring themselves in secondary domains such as education, quality improvement, or leadership to create clarity and legitimacy. This parallels Varpio et al.’s framing of subjectification, where professional identities are made visible through unique, situated actions [23]. Notably, AHs’ identity construction, often through the educator role, occurred in spite of rather than because of institutional structures. Consistent with Burke’s identity theory, participants engaged in active identity work seeking verification, adjusting strategies, and renegotiating meanings in response to misalignment between institutional expectations and internal identity standards. While these efforts supported identity coherence, they were time-intensive and, participants emphasized, insufficient substitutes for consistent institutional recognition.

Protected time emerged as central to identity formation and sustainability. Participants described it not as a scheduling accommodation but as institutional validation of academic contribution. Without it, academic work becomes unpaid labor that contributes to burnout, a dynamic consistent with broader literature on clinician well-being [24]. Notably, burnout appeared even in early-career participants, challenging the assumption that it is primarily a later-career phenomenon.

Contextualization with Existing Literature

Lang et al. identified mentorship of others as “currency for promotion” and emphasized its role in advancement [14]. Our findings extend this: mentorship is indeed essential, but remains scarce and unevenly distributed, particularly for underrepresented faculty. The mentorship gap reflects hospital medicine’s demographic youth as few senior mentors have spent entire careers as hospitalists. Mentorship is most effective when mentors are attuned to mentees’ evolving goals and identities, and when relational alignment, shared identity, lived experience, or values is present [25, 26]. Mentorship tailored to underrepresented groups better equips faculty for advancement [27]. Our data also demonstrate that AHs’ mentorship identity extends beyond peers and junior faculty to learners, which participants described as central to their sense of purpose.

The expectation that AHs «prove themselves first» before receiving protected time has been described from the perspective of hospitalist group leaders, who frame early uncompensated academic labor as a prerequisite for later investment [28]. Our findings extend this by centering AH perspectives and showing the expectation operates across academic domains including teaching and is experienced as devaluing, contributing to identity dissonance and burnout.

Our findings on institutional non-recognition parallel Burke’s concept of identity verification the process through which behaviors aligned with an identity are socially confirmed [29]. This echoes prior work on medical educators, whose labor is often unseen and accorded lesser capital than clinical work [22]. We extend this literature to inpatient academic hospital medicine, where educational contributions are embedded in clinical service and rendered invisible by RVU-based valuation. When institutions fail to recognize teaching, mentorship, and non-billable work, they prevent verification of salient aspects of AH identity, producing frustration and identity fragmentation. Although learners consistently rate hospitalists as highly effective inpatient educators, this educator identity remains institutionally underappreciated [30].

Limitations

This study was not designed to systematically examine how participant demographics influence PIF, and therefore participants did not consistently frame experiences through these lenses, except in the context of night work and mentorship with shared backgrounds. The sample was predominantly early-to mid-career, reflecting the bottom-heavy AH workforce, and it is plausible that PIF may manifest differently among senior faculty. Purposive snowball sampling through professional networks may have favored individuals more engaged with academic discourse. Finally, the study included only internal medicine AHs and does not capture the experiences of pediatric or family medicine hospitalists.

Clinical and Practical Implications

First, institutions should develop clearer, field-wide definitions of AH work and communicate them explicitly at hiring and throughout tenure, while preserving flexibility for niche development. Definitions alone, however, are insufficient: participants described persistent structural tension in which institutions prioritize clinical service without embracing an academic vision for hospitalists. Without accompanying commitments to protected time, recognition, and advancement aligned with academic values, even well-articulated definitions fail to support PIF. Attention to flexible scheduling and promotion processes that account for those working nights is also warranted [31].

Second, protected time should shift from discretionary to non-negotiable, automatically provided at hiring rather than contingent on “proving yourself first.” The built-in week off in some AH schedules should be recognized as recovery time, not an expectation for additional academic work. Our findings suggest 20% protected time represents a meaningful investment, though most institutions in this study did not offer it.

Third, institutions should establish mentorship infrastructure that supports organic relationship formation while actively addressing gaps for underrepresented faculty, including dedicated mentorship roles with protected time, peer mentorship structures, and visible mentors from diverse backgrounds.

Finally, leadership should formally recognize AH contributions in teaching, mentorship, and education through tangible mechanisms: educational metrics in performance evaluation alongside RVU-based measures, institutional resources for educational initiatives, and compensation structures that account for the full scope of AH work.

Taken together, our findings suggest AHs should be understood as core academic clinicians whose educational, clinical, and relational work is central to institutional missions. When “academic” work is defined by RVU-driven metrics privileging scholarship and procedural productivity, teaching, mentorship, and clinical leadership are marginalized. Without recalibration, AHs are left to individually negotiate identity and legitimacy within systems that structurally constrain their contributions.

Future research should evaluate the effectiveness of institutional interventions standardized role definitions, guaranteed protected time, and relationally aligned mentorship programs on engagement, satisfaction, and retention. The paradoxical relationship between AHs’ younger age and higher burnout also warrants further exploration.