The study was conducted across every municipal-level maternal and child health hospital in Hunan Province, a region of central China with tens of millions of residents. Between late March and early July 2024, the research team invited all 2,839 staff nurses in these hospitals to participate. Ultimately, 1,593 nurses — just over 55 percent — completed a comprehensive questionnaire that measured three things simultaneously: missed nursing care behaviors using a validated survey instrument, abusive supervision using the Abusive Supervision Questionnaire, and the quality of the leader-member relationship using the widely used Leader-Member Exchange-7 Scale. The scale of the dataset matters, because the statistical method the team employed — latent profile analysis, or LPA — needs large samples to work reliably. Rather than treating every nurse as an average point on a continuum, LPA searches for hidden subgroups: clusters of individuals who share a common underlying pattern that isn’t directly observable but reveals itself through the combination of their measured responses.

What emerged was a three-profile structure that the researchers labeled the out-group, the in-group, and the middle-group. The out-group, comprising 142 nurses or about 11.8 percent of the sample, was characterized by low-quality leader-member exchange, occasional experiences of abusive supervision, and comparatively fewer missed care behaviors. The in-group — the overwhelming majority at 866 nurses, or 72.1 percent — enjoyed high-quality relationships with their leaders, experienced little abusive supervision, and reported only occasional care omissions. But the middle-group, 193 nurses or 16.1 percent, was the alarming one: these nurses reported moderate leader-member exchange, frequent abusive supervision, and frequent missed nursing care. In other words, the nurses most likely to skip essential patient care were not those with the worst relationships overall, but those trapped in a specific and toxic middle zone — connected enough to their leaders to be embedded in the hierarchy, yet subjected to frequent hostility from the very people who controlled their daily working lives.

Missed nursing care is one of the most significant yet underreported patient safety threats in modern healthcare. The concept refers to any element of required care that is omitted or significantly delayed — a missed dose of medication, an overlooked comfort check, a delayed response to a call bell, inadequate documentation, or skipped patient education. Globally, studies have found that the vast majority of nurses admit to leaving at least some care undone on their last shift, and the consequences are well documented: higher rates of medication errors, falls, hospital-acquired infections, and readmissions. What has been less clear is why care gets missed beyond the obvious resource constraints. The new Hunan study adds a crucial psychological dimension, suggesting that the interpersonal climate between nurses and their immediate supervisors is not background noise but an active driver of whether patients receive the care they are owed.

The concept of abusive supervision, drawn from organizational psychology, describes a manager’s sustained display of hostile verbal and nonverbal behavior toward subordinates — ridiculing them in public, taking credit for their work, using demeaning language, or breaking promises. It is distinct from physical violence or formal disciplinary action; it is the slow corrosive drip of everyday contempt. Leader-member exchange, by contrast, measures the quality of the working relationship between a leader and each individual subordinate, ranging from high-quality partnerships built on trust, loyalty, and mutual obligation to low-quality transactions limited to the formal employment contract. Decades of research have shown that nurses in high-quality exchanges receive more support, more resources, and more latitude, while those in low-quality exchanges are often relegated to the margins. The Hunan study’s innovation was to examine how these two leadership variables interact with care omissions within the same individuals, rather than studying each in isolation.

The statistical machinery behind the findings deserves attention because it represents a shift in how health services research is conducted. Traditional analyses test whether variable A predicts variable B on average across a population. Latent profile analysis instead asks whether the population contains distinct types — and the answer here was emphatically yes. The researchers evaluated competing models with different numbers of profiles using a battery of fit indices, including the Akaike Information Criterion, the Bayesian Information Criterion, the sample-size adjusted BIC, and the bootstrap likelihood ratio test, alongside the adjusted Lo-Mendell-Rubin likelihood ratio test. The three-profile solution won out, and the team then used the Three-Step Approach to Auxiliary Variable Selection — a method designed to test which external factors predict membership in each profile without distorting the underlying classification — to probe the sociodemographic drivers of profile membership.

Two factors stood out with statistical significance: education level and average daily overtime hours. Nurses with higher formal education were more likely to occupy certain profiles than others, a finding the researchers interpret as a call for targeted support — mentoring programs, professional development pathways, and recognition structures aimed specifically at highly educated nurses, who may face unique pressures or expectations within hospital hierarchies. Meanwhile, overtime emerged as a potent predictor in its own right. Nurses working extended daily shifts were disproportionately represented in the higher-risk profiles, reinforcing a growing body of evidence that fatigue and excessive workload do not merely tire nurses out but reshape their entire relationship with the workplace, including how they experience and respond to their supervisors. The researchers suggest that flexible scheduling for nurses on extended shifts could meaningfully reduce care omissions.

The practical implications of the study are unusually concrete for research of this type. Because the middle-group — frequent abusive supervision combined with frequent missed care — was identified as a distinct and sizable cluster, the authors argue that future interventions should target these nurses and their leaders directly rather than applying blanket policies across the workforce. That could mean leadership training aimed at curbing hostile supervisory behaviors, structured feedback channels for nurses caught in the middle zone, and organizational audits that flag units where the combination of moderate relationship quality and frequent hostility is concentrated. The logic is one of precision public health applied to management: identify the highest-yield population, understand its specific configuration of risk, and intervene there first, where a fixed investment of resources is likely to produce the largest reduction in patient harm.

There are, of course, limits to what a single cross-sectional study can establish. The data were collected at one point in time, in one province, in one specific type of hospital — maternal and child health facilities — so it remains an open question whether the same three-profile structure would appear in tertiary hospitals, rural clinics, or healthcare systems outside China. Cross-sectional designs also cannot prove causation: it is possible that abusive supervision drives care omissions, but equally possible that units with high rates of missed care breed frustration that supervisors express as hostility, or that some third factor — chronic understaffing, perhaps — generates both. The authors themselves call for longitudinal and multi-setting research to test whether the profile structure holds. Still, the study’s framing of missed care as a relational and organizational phenomenon, rather than purely an individual failing, aligns with a broader international movement to treat healthcare worker wellbeing as patient safety infrastructure.

For hospital leaders everywhere, the message is uncomfortable but actionable. The nurses most at risk of omitting care are not the disengaged outsiders at the bottom of the relationship spectrum, but the embedded professionals in the middle — competent enough to be relied upon, mistreated enough to be eroded. One in six nurses in this study fell into that category. Every one of them represents a patient who may wait longer for pain medication, a family member who may not receive an update, a chart that may go unfinished. The finding that education and overtime shape these patterns gives administrators two concrete levers to pull. But the deeper lesson may be simpler and harder: how nurse managers speak to their staff on an ordinary Tuesday is not a soft issue of workplace culture. It is a measurable determinant of whether patients get their care — and the statistics now say so.

Subject of Research: The co-occurrence of missed nursing care, abusive supervision, and leader-member exchange among hospital nurses

Article Title: Co-occurrence of missed nursing care, abusive supervision, and leader-member exchange: a latent profile analysis

Article References: Hu, S., Zhu, X., Liu, S., Hu, J., Hu, H., Batool, S., Jiang, Z., Cheng, Y., Xu, M., & Li, X. (2026). Co-occurrence of missed nursing care, abusive supervision, and leader-member exchange: a latent profile analysis. BMC Nursing. https://doi.org/10.1186/s12912-026-05529-4

Image Credits: AI Generated

DOI: 10.1186/s12912-026-05529-4

Keywords: missed nursing care, abusive supervision, leader-member exchange, latent profile analysis, nursing research, patient safety, nurse leaders, overtime, healthcare management, BMC Nursing, China, workplace psychology