Career Compromise, Perceived Organizational Support, and Professional Identity in Nurses: A Latent Profile Analysis...

To characterize unobserved patterns of nurses' professional identity (PI), identify factors related to profile membership, and determine whether perceived organizational support (POS) statistically accounts for part of the association between career compromise (CC) and PI.

A cross-sectional electronic questionnaire was completed by 2288 nurses recruited from 10 hospitals in Henan Province, China, between August and November 2025. The assessment tools utilized comprised a general information questionnaire, the Career Compromise Scale, the Perceived Organizational Support Scale, and the Nurses' Professional Identity Scale. Latent profile analysis (LPA) was employed to derive the PI subgroups. Group differences and profile correlates were evaluated with univariate analyses and multinomial logistic regression, respectively. The assessment of the indirect connection between CC and PI via POS was carried out utilizing Pearson correlation and SEM.

The classification of nurses' PI resulted in three latent profiles: low-PI (30.4%), moderate-PI (43.4%), and high-PI (26.2%). Holding a position in nursing management and earning a higher income were associated with an increased likelihood of belonging to the moderate- and high-patient impact groups. Conversely, the likelihood of nurses in acute/critical care units or emergency departments being part of the moderate- and high-PI groups was lower. After accounting for various factors, having more than 10 night shifts monthly was not significantly related to high-PI group membership. According to the SEM, CC was positively linked to POS (β = 0.901, p < 0.01) and PI (β = 0.463, p < 0.01), and POS had a positive link with PI (β = 0.382, p < 0.01); the standardized indirect effect amounted to 0.344.

Nurses' PI was represented by three profiles. Membership varied by income, management role, department, gender, hospital level, and hospital type, whereas monthly night-shift frequency did not independently differentiate the high-PI profile after covariate adjustment. POS also accounted for a significant portion of the positive CC-PI association.