Two Innovation Profiles Emerge Among Community Nurses in Landmark Chinese Study

Community nurses, the frontline workers who deliver the bulk of primary healthcare to ageing populations with chronic disease, are not a uniform bunch when it comes to innovation. That is the central message of a new cross-sectional study published in Nursing Open, which used a person-centred statistical technique to reveal that community nurses in southern China fall into two distinct innovation profiles, each shaped by a different mix of experience, training and organizational support. The findings challenge the long-standing habit of treating nurse innovation as a single, average-level trait and instead point toward a precision approach to workforce development.

The research team, led by investigators affiliated with Shenzhen Nanshan People’s Hospital, surveyed 395 full-time community nurses working across 87 public community health service centres connected to a large tertiary hospital in southern China. Data were collected between August and September 2025 through an anonymous online platform, with safeguards including a minimum completion time, one-response-per-account restrictions and screening for contradictory or straight-lined answers. The effective response rate reached 94.1 percent, and the sample exceeded the recommended minimum of 300 participants for the latent profile analysis the researchers employed.

Latent profile analysis, or LPA, is a statistical method that sorts individuals into hidden subgroups based on their response patterns across several measured variables, rather than forcing everyone onto a single continuum. In this study, the indicator variables were the three dimensions of the Nurse Innovative Behaviour Scale: idea generation, support seeking and idea realization. The researchers fitted models with one through four profiles and evaluated fit using the Akaike Information Criterion, the Bayesian Information Criterion, entropy values and likelihood ratio tests. Although information criteria kept dropping as profiles were added, the Lo–Mendell–Rubin test was non-significant for the three- and four-profile models, so the two-profile solution was retained, backed by a high entropy of 0.934 and average posterior probabilities between 0.980 and 0.987.

The resulting picture was strikingly balanced. Profile 1, labelled the Moderate Innovation Profile, contained 199 nurses, or 50.4 percent of the sample, and showed a consistent but modest level of activity across all three innovation dimensions. Profile 2, the High Innovation Profile, contained 196 nurses, or 49.6 percent, and displayed a proactive, high-level pattern on every dimension. Notably, the two groups differed mainly in overall level rather than in shape, suggesting that innovation in this workforce behaves like a cumulative capacity built from resources and experience, rather than a specialized trait possessed by only a few.

Why do some nurses land in the high-innovation group while others with similar demographics do not? To answer this, the team grounded its analysis in the model of individual innovative behaviour developed by Scott and Bruce in 1994, which frames innovation as the product of individual characteristics interacting with the surrounding work environment. The study measured professional identity with a 30-item scale, telehealth readiness with the Chinese version of the Telehealth Readiness Assessment Tools, and perceived organizational support with a 13-item nursing-adapted scale. All instruments showed excellent internal consistency in this sample, with Cronbach’s alpha values ranging from 0.937 to 0.986.

Univariate comparisons flagged several variables that separated the two profiles, including age, education, professional title, years of experience, prior hospital work, research participation, innovation training and managerial knowledge sharing. A multinomial logistic regression then distilled which of these independently predicted membership in the High Innovation Profile. The strongest categorical predictors were access to departmental innovation training, with an odds ratio of 6.203, and prior experience working in a general hospital, with an odds ratio of 5.144. In practical terms, nurses who had received structured innovation training were more than six times as likely to belong to the high-innovation group, and those with hospital backgrounds more than five times as likely, compared with their peers.

Participation in research projects also carried substantial weight, with an odds ratio of 4.173, while managerial knowledge sharing more than doubled the odds at 2.568. Among the continuous psychological measures, professional identity and telehealth readiness were both statistically significant, with odds ratios of 1.025 and 1.077 respectively, though the authors note that these single-unit effects are modest compared with the powerful categorical catalysts of training and experience. Intriguingly, general perceived organizational support was not a significant predictor at all. The study’s interpretation is that broad, diffuse support fails to translate into proactive behaviour, whereas specific, tangible empowering actions, such as a manager who actively shares research knowledge, do.

The authors are candid about the limitations of their design. The cross-sectional structure prevents any causal claims, and convenience sampling from a single metropolitan hospital network may have attracted unusually motivated respondents, potentially inflating the share of high innovators while limiting generalizability to rural or resource-constrained settings. All measures were self-reported, leaving room for social desirability bias, and the innovation scale captures behavioural frequency rather than the objective clinical quality of the innovations produced. The analysis also relied on aggregated dimension scores rather than item-level data, and the two-profile solution has not yet been validated in an independent cohort, so its structural stability remains unverified.

Even with those caveats, the practical implications are concrete. The researchers recommend abandoning one-size-fits-all incentive schemes in favour of a stratified talent development pipeline. For nurses in the Moderate Innovation Profile, who make up roughly half of the workforce and serve as its stabilization baseline, the strategy should be foundational empowerment: quality improvement training, structured continuing education, small-scale funding for incremental micro-innovations such as refined scheduling procedures or localized health-education toolkits, and mentorship pairings with experienced innovators. For the High Innovation Profile, the strategy should be catalytic: granting clinical autonomy, flexible hours and formal roles as innovation champions, quality improvement coaches and peer mentors who lead the knowledge-sharing activities shown to be so influential.

The study also carries a policy dimension that extends well beyond any single hospital network. As populations age and chronic disease burdens shift care out of hospitals and into communities, the demand for adaptive, digitally capable nursing workforces will only intensify, a priority echoed in China’s National Nursing Development Plan for 2021 through 2025. The authors argue that healthcare authorities should institutionalize telehealth competency training and build centralized knowledge-sharing platforms across primary health networks, since digital readiness emerged as a meaningful channel through which community nurses can transcend local resource limits. Future longitudinal research, the team concludes, should track how stable these profiles are over time and link them to objective clinical outcomes, so that the promise of precision empowerment in nursing management can be tested against real-world results rather than self-report alone.

Subject of Research: Latent profiles of innovative behaviour and their predictors among community nurses in China

Article Title: Profiles of Innovative Behaviour Among Community Nurses and Associated Factors: A Cross‐Sectional Study

Article References: Profiles of Innovative Behaviour Among Community Nurses and Associated Factors: A Cross‐Sectional Study. (n.d.). https://doi.org/10.1002/nop2.70819

Image Credits: AI Generated

DOI: 10.1002/nop2.70819

Keywords: community nursing, innovative behaviour, latent profile analysis, telehealth readiness, professional identity, perceived organizational support, nursing management, primary healthcare, cross-sectional study, knowledge sharing, workforce development, China