The research team, led by Li Zhang and Yaxuan Zhao of the Department of Adult Psychiatry at the Brain Hospital Affiliated to Nanjing Medical University, together with Yanhong Zhang of the hospital’s nursing department, set out to answer a deceptively simple question: do all psychiatric nurses care in the same way? Conventional studies typically average caring behavior scores across an entire workforce, an approach that treats the group as homogeneous. But averages can hide critical differences. A hospital whose nurses score moderately well on average might contain a cluster of exceptional caregivers alongside a cluster of struggling ones, and each cluster may need entirely different interventions. To capture that hidden variation, the team turned to a statistical technique called latent profile analysis.

Latent profile analysis, or LPA, is a person-centered method that sorts individuals into unobserved subgroups based on their patterns of responses across multiple measured variables. Rather than asking how a single factor, such as years of experience, shifts an average score up or down, LPA asks whether distinct types of responders exist at all. The researchers used Mplus version 8.3 to fit a series of competing models, comparing them with standard fit indices including the Akaike Information Criterion, the Bayesian Information Criterion, an adjusted version of the BIC, the Lo-Mendell-Rubin likelihood ratio test, and a bootstrap likelihood ratio test. These tools collectively determine how many latent profiles best describe the data without overfitting, a balance that is essential when the goal is to identify genuinely distinct groups rather than statistical noise.

The evidence base for the analysis came from 473 valid questionnaires collected between March and June 2025 from nurses working at 20 tertiary psychiatric hospitals spread across six provinces of China. Of the 485 questionnaires distributed, 473 were returned in usable form, a strikingly high response rate of 97.53 percent that lends considerable weight to the findings. Participants completed a demographic questionnaire alongside three validated instruments: the Caring Behaviors Inventory, which measures concrete caring actions across dimensions such as respect, supportive protection, and connectedness; the Chinese version of the Spiritual Climate Scale, which assesses whether the workplace nurtures meaning, values, and inner purpose; and the Professional Identity Scale for Nurses, which captures how strongly nurses see themselves as belonging to and valuing their profession.

When the models were compared, three profiles emerged. The first, and largest in terms of overall level, was labeled the medium level high support and guarantee group, nurses whose caring behaviors were moderate overall and anchored most strongly in providing support and safeguarding patients. The second was the high level lower relative fluctuation group, nurses who scored high overall but whose caring showed more unevenness across dimensions. The third was the high level low respect and contact group, a profile in which overall caring was high but the dimensions of respect and human connection lagged behind. That such a gap could exist, with nurses delivering technically strong care while scoring lower on respect and interpersonal contact, is precisely the kind of heterogeneity that average-based analyses would have smoothed away entirely.

The existence of that third profile carries particular weight in psychiatric settings. Patients with mental disorders often require prolonged care in closed hospital environments, where the nurse-patient relationship is not merely a comfort but a core component of treatment itself. Respect and genuine contact are the currency of psychiatric nursing; without them, even technically competent care can feel custodial to a vulnerable patient. The finding suggests that some nurses who excel at the protective and supportive elements of the job may still need targeted support in the relational dimensions, something no hospital-wide average would ever reveal.

Having identified the profiles, the researchers used logistic regression analysis in IBM SPSS Statistics version 27.0 to ask what predicted membership in each. Three factors reached statistical significance: the number of night shifts worked, the spiritual climate of the workplace, and professional identity. Each of these is notable, but two stand out for a practical reason: unlike fixed demographic characteristics such as age or tenure, spiritual climate and professional identity are modifiable. Hospitals can, in principle, act on them. The finding that night shift burden influences caring behavior profiles also echoes a long line of occupational health research linking shift work to fatigue, emotional exhaustion, and degraded patient interactions, and it gives hospital administrators a concrete staffing lever to examine.

The concept of spiritual climate deserves particular attention. In nursing research, spiritual climate refers to the extent to which a workplace supports employees’ sense of meaning, values, and connectedness to something larger than routine tasks. In psychiatric hospitals, where staff confront chronic illness, stigma, and occasionally aggression, a workplace that nourishes inner purpose may buffer the emotional toll that otherwise erodes caring behavior. The study’s results suggest that when that climate is strong, nurses are more likely to belong to the higher-level caring profiles. Similarly, professional identity, the internalized sense of being a competent and valued nurse, appears to act as a psychological resource that sustains caring even under difficult conditions.

The authors emphasize that, to their knowledge, few studies have applied latent profile analysis to caring behaviors among nurses, making this one of the first efforts to map the internal structure of caring in psychiatric settings rather than merely measuring its average level. They argue that the identified profiles and their associated factors can inform the development of targeted interventions. Instead of delivering the same communication training or resilience workshop to every nurse, hospital leaders could direct resources toward the specific subgroups that need them: bolstering respect and contact skills for the high level low respect and contact group, stabilizing consistency for the fluctuating group, and lifting the overall level of the medium group through climate and identity-building initiatives.

The study was approved by the ethics committee of the Brain Hospital Affiliated to Nanjing Medical University and conducted in accordance with the Declaration of Helsinki, with all participants providing informed consent and completing anonymous questionnaires. Funded by the Nanjing Medical University Science and Technology Development Fund, the work arrives at a moment when mental disorders represent a significant and growing global disease burden and when health systems worldwide are grappling with nursing shortages and burnout. Its central message is both methodological and practical: caring is not a single dial that turns up or down uniformly across a workforce, but a mosaic of distinct patterns shaped by working conditions and inner resources. Hospitals that learn to see those patterns, and to act on the modifiable factors behind them, may find that the most effective way to improve patient care is not to train harder, but to understand better who their caregivers are and what they need.

Subject of Research: Caring behaviors and their influencing factors among nurses in Chinese tertiary psychiatric hospitals

Article Title: Current status and influencing factors of caring behaviors among nurses in tertiary psychiatric hospitals in China: a cross-sectional study based on latent profile analysis

Article References: Zhang, L., Zhao, Y., & Zhang, Y. (2026). Current status and influencing factors of caring behaviors among nurses in tertiary psychiatric hospitals in China: a cross-sectional study based on latent profile analysis. BMC Nursing. https://doi.org/10.1186/s12912-026-05393-2

Image Credits: AI Generated

DOI: 10.1186/s12912-026-05393-2

Keywords: psychiatric nursing, caring behavior, latent profile analysis, spiritual climate, professional identity, night shifts, cross-sectional study, nurse-patient relationship, tertiary psychiatric hospitals, China, nursing research, mental health