Perception of response governance and collaborative attitude of hospital infection outbreak and their association with perceived stress among healthcare workers: a cross-sectional study in North-Eastern China

Abstract

Objectives

Healthcare-acquired infection outbreaks impose significant psychological stress on healthcare workers, yet limited research has examined stress perception and protective factors during institutional-level outbreak scenarios. Grounded in the Job Demands-Resources model, this study examines the relationships among healthcare workers’ perceptions of response governance, collaborative attitudes and perceived stress, and tests the moderating role of collaborative attitudes.

Design

A cross-sectional study was conducted using an online questionnaire from May to October 2025.

Setting

Healthcare workers from hospitals in North-Eastern China.

Participants

1017 valid healthcare workers, including clinicians, nurses, clinical technicians, administrative personnel, support staff and others with at least 1 year of continuous employment at the hospital. Purposive and snowball sampling were used.

Main outcome measures

Perception of response governance collaborative attitude towards hospital infection outbreaks and perceived stress scores were assessed using validated self-report questionnaires.

Results

The mean perceived stress score among study participants was 2.03 (SD=0.97, range 1–5), indicating low to moderate levels. However, 11.9% of individuals still experienced high stress levels. Perception of response governance showed a significant negative correlation with perceived stress (r=−0.224, p<0.001) and the collaborative attitude towards hospital infection outbreaks also showed a significant negative correlation with perceived stress (r=−0.305, p<0.001). Hierarchical multiple regression, after adjusting for demographic covariates, showed that perception of response governance (β=−0.400, p<0.001) and collaborative attitude (β=−0.210, p=0.008) negatively predicted perceived stress, thereby increasing model explanatory power by 10.2% (ΔR²=0.102, p<0.001). Interaction analysis revealed that the interaction term between perception of response governance and collaborative attitude significantly predicted perceived stress (β=−0.202, p=0.022). Simple slope tests indicated that at higher levels of collaborative attitude, the negative predictive effect of perception of response governance on perceived stress intensified (low level: simple slope=−0.247; high level: simple slope=−0.402). Additionally, years of service between 11 and 15 years (β=0.381, p=0.012) and prior experience in managing healthcare-associated infection outbreaks (β=0.468, p<0.001) were identified as risk factors for perceived stress.

Conclusion

While most healthcare workers reported low-to-moderate stress levels, identifiable high-risk subgroups remained. Given the cross-sectional design, causal inferences cannot be drawn; however, perceived governance and collaborative attitudes showed a synergistic moderating effect on perceived stress. Administrators should implement integrated interventions combining cognitive empowerment with collaborative support to strengthen organisational resilience.

STRENGTHS AND LIMITATIONS OF THIS STUDY.

• The study constructed an interactive model of ‘cognition–attitude–stress’ based on the Job Demands-Resources theory, and empirically tested it using regression analysis and moderation effect analysis.

• The study employed a validated questionnaire with good internal consistency (Cronbach’s α ranging from 0.772 to 0.939) and a large sample size (N=1017), which enabled robust multivariate analysis.

• The use of self-reported measures may introduce reporting bias, and the non-probability sampling method (purposive and snowball sampling) may limit the generalisability of the findings.

• The cross-sectional design of the study limits our ability to make causal inferences.

Introduction

Healthcare-associated infection (HAI) outbreaks pose a persistent threat to global public health security, presenting not only a severe test of healthcare institutions’ technical capabilities but also a challenge to the resilience of their human resources.^{1 2} Healthcare workers are at the frontline of hospital infection outbreak response. Their psychological capital, including stress levels, emotional states and resilience, and their cognitive capital, such as understanding of infection control, professional knowledge and awareness of governance, serve as critical soft factors determining prevention success, patient safety and healthcare system stability.^3–7 High stress levels may lead to fatigue, anxiety or even depression, thereby diminishing their capacity to respond to emergencies.⁸ Meanwhile, cognitive capital influences workers’ judgement and decision-making in complex situations.^{9 10} However, high job demands during outbreaks, such as heavy workload, time pressure and infection risk, can elevate perceived stress and undermine these capitals.^{11 12} In particular, in-depth research on two specific dimensions, namely perception of response governance (a cognitive-capital element) and collaborative attitude (a psychological-capital element), is essential for optimising healthcare resource allocation and enhancing service quality. Therefore, enhancing healthcare workers’ psychological and cognitive capital is not only crucial for improving the effectiveness of hospital infection outbreak control but also a vital means of ensuring patient safety and maintaining the stability of the healthcare system.

Previous studies, including (Hu et al,¹³ Tan et al,¹⁴ Çivilidağ et al¹⁵) in the medical sciences, have extensively investigated healthcare workers’ stress, anxiety and burnout during epidemics and disease outbreaks, while studies, including (Burau et al,⁴ Chamla et al⁶) in the social sciences, have examined the psychosocial and organisational impacts of such crises on healthcare professionals. Another stream of research on knowledge, attitudes and practices has described healthcare workers’ operational understanding of infection control (eg, hand hygiene, personal protective equipment use).^16–18 However, these studies are largely descriptive and are not grounded in a theoretical framework that explains how psychosocial factors jointly influence perceived stress. Moreover, existing systematic reviews have pointed out that most interventions remain at the individual level, with a notable lack of organisational-focused strategies to support healthcare workers during public health emergencies.¹⁹

Moreover, research on institutional response strategies has increasingly highlighted the importance of leadership and organisational support,²⁰ and studies on healthcare workforce resilience have demonstrated that resilience levels vary significantly across hospital departments, suggesting the need for tailored institutional support.²¹

Nevertheless, existing research predominantly treats ‘stress’ as an isolated outcome variable or limits ‘knowledge’ to operational infection control knowledge. There is a lack of in-depth exploration into how the broader perception of response governance and collaborative attitudes towards prevention and control—key psychosocial factors—jointly influence perceived stress. This ‘cognition–attitude–stress’ interaction mechanism, crucial for building resilient emergency teams, remains understudied. Specifically, existing research fails to adequately explain why some healthcare workers maintain high psychological resilience under identical external pressures while others succumb to anxiety and burnout. This disparity likely correlates with their depth of risk perception in emergency management and their attitudes towards teamwork. Therefore, systematically analysing the interactions among these factors is essential to provide theoretical foundations and practical guidance for enhancing healthcare workers’ psychological resilience and emergency response capabilities.

This study employs the Job Demands-Resources (JD-R) model as its theoretical framework. According to the JD-R model, working conditions can be classified into two broad categories: job demands and job resources. Job demands refer to physical, psychological and social aspects of the job that require sustained effort and are associated with physiological and psychological costs, such as exhaustion. In the context of hospital infection outbreaks, healthcare workers face high job demands such as workload, time pressure and risk of infection. Job resources, on the other hand, refer to those aspects of the job that help to achieve work goals, reduce job demands and their associated costs, and stimulate personal growth and development.²²

In this study, we conceptualise two specific job resources: perception of response governance, which refers to clarity of outbreak protocols and institutional support, and collaborative attitude, which refers to the willingness to work with other departments. Although the JD-R model originally focused on personal resources as buffers of stress, subsequent elaborations have emphasised that job resources can be both organisational, such as support and feedback, and interpersonal, such as collaboration. Our choice of these two variables is directly derived from the JD-R model’s proposition that resources, whether personal, social or organisational, can mitigate the negative impact of job demands on employee well-being.

Figure 1 presents the conceptual framework of this study based on the JD-R model. As illustrated in the figure, job demands are expected to increase perceived stress, while the two job resources are hypothesised to buffer this effect. Based on the JD-R model, we hypothesise that these two job resources not only directly reduce perceived stress through their main effects but also interact with each other: a higher level of collaborative attitude strengthens the protective effect of governance perception on stress, representing a moderating effect. Specifically:

Figure 1. Conceptual framework based on the Job Demands-Resources model.

H1: Healthcare workers’ perception of response governance is negatively correlated with their perceived stress.

H2: Healthcare workers’ collaborative attitude is negatively correlated with their perceived stress.

H3: The interaction between perception of response governance and collaborative attitude significantly influences perceived stress.

Drawing on the aforementioned theoretical framework and hypotheses, the specific objectives of this study are threefold: (1) to examine the bivariate correlations among perception of response governance, collaborative attitude and perceived stress in a large sample of healthcare workers from North-Eastern China; (2) to test the main effects of perception of response governance and collaborative attitude on perceived stress after adjusting for demographic and occupational covariates; and (3) to explore the moderating role of collaborative attitude in the relationship between perception of response governance and perceived stress. By achieving these objectives, this study seeks to provide scientific evidence for developing a dual-track intervention strategy centred on cognitive empowerment and collaborative support.

Methods

Study design and settings

The study employed a cross-sectional design using an online questionnaire. Considering factors such as human resources, funding constraints and population representativeness, Heilongjiang Province was selected as the primary survey area. We employed a combination of purposive sampling and snowball sampling.

Purposive sampling was chosen because there was no existing sampling frame for all eligible healthcare workers in Heilongjiang Province. This method allowed us to intentionally select participants from different hospital types, including tertiary, secondary and primary hospitals, and from various professional roles, such as clinicians, nurses, technicians and administrative staff, to ensure a diverse and institutionally representative sample. Probability sampling was not feasible due to the lack of a complete sampling frame and the practical constraints of time and budget.

Snowball sampling was then adopted to further increase the sample size and to reach healthcare workers in smaller or more remote facilities, who might otherwise be difficult to access. This approach also helped expedite data collection within the limited study timeframe from May to October 2025. Participants were asked to forward the online questionnaire link to other eligible colleagues.

A questionnaire survey was conducted via the Questionnaire Star platform from May to October 2025. At the beginning of the survey, participants were informed of the study’s purpose and procedures. They were required to read and confirm their agreement to the consent form before proceeding with the questionnaire, and could exit at any time during completion. All questions were required to be answered completely on submission. Based on IP addresses recorded by the Questionnaire Star platform, each participant could submit only one questionnaire.

Study participants and sample size

The target population of this study was all hospital-based healthcare workers meeting the inclusion criteria in Heilongjiang Province during the study period. According to the Heilongjiang Statistical Yearbook 2024,²³ the total number of hospital-based healthcare workers in the province was approximately 227 022, which served as the estimated size of the target population. As a complete sampling frame covering all eligible hospital healthcare workers across the province was not available, non-probability sampling (purposive and snowball sampling) was adopted. Inclusion criteria were as follows: (1) hospital staff members, including clinicians, nurses, clinical technicians, administrative personnel, support staff and others from tertiary general hospitals, tertiary specialised hospitals and primary/secondary hospitals; (2) individuals with at least 1 year of continuous employment at the hospital. A total of 1278 individuals completed the survey. The research team conducted logical checks on the questionnaires. Invalid questionnaires were excluded based on the following criteria: identical responses across all options, contradictions between age and years of service, completion time under 3 min, and other obvious errors. A total of 261 questionnaires with logical errors were excluded. Ultimately, 1017 valid questionnaires (80.0% of the total responses) were included in the data analysis.

This study employed non-probability sampling (purposive and snowball sampling). According to Peduzzi et al,²⁴ the minimum sample size for regression analysis should be at least 10 times the number of independent variables. In our model, the independent variables included 12 sociodemographic items (recoded into 36 dummy variables) and two continuous variables from the core scales, giving a total of approximately 38 independent variables. Based on this, the minimum required sample size was 380. A total of 1017 valid questionnaires were finally collected, which met the requirement for analysis.

Patient and public involvement

No patients or members of the public were involved in this study.

Study tools

The questionnaire was developed specifically for this study because no existing scale adequately captured the specific dimensions of response governance, collaborative attitude and perceived stress in the context of hospital infection outbreaks. Based on domestic and international literature^{25 26} and the current status of HAI management in China,²⁷ the questionnaire was developed with reference to domestic and international scales^{28 29} and following expert consultation. The questionnaire consisted of four sections: basic information; perceptions of hospital infection outbreak risk management; attitudes towards hospital infection prevention and control collaboration; and perceived stress triggered by hospital infection outbreaks. The full questionnaire has been provided as online supplemental additional file 1.

1. Basic information: 12 items covering sociodemographic characteristics (gender, age, education level, years of service, monthly income, administrative position, professional title, type of healthcare facility, grade of healthcare facility, position, whether employed in an infection control department and history of experiencing HAI outbreaks).

2. Perception of hospital infection outbreak risk management: comprises two items. Example item: ‘I think hospital infection control management is important’. Higher scores indicate a more comprehensive understanding of outbreak management. Scored using a 5-point Likert scale ranging from 1 (‘strongly disagree’) to 5 (‘strongly agree’). A total average score above 3 is considered high perception. This scale demonstrates good consistency and reliability with a Cronbach’s alpha coefficient of 0.772.

3. Attitude towards collaborative hospital infection outbreak prevention and control: comprises six items. Example item: ‘I know about hospital infections and the consequences of hospital infection outbreaks’. Higher scores indicate more positive attitudes towards outbreak management and a clearer understanding of collaborative governance. Scored using a 5-point Likert scale ranging from 1 (‘strongly disagree’) to 5 (‘strongly agree’). A total average score above 3 indicates a positive attitude. This scale demonstrates good consistency and reliability, with a Cronbach’s alpha coefficient of 0.868.

4. Perceived stress from hospital infection outbreaks: comprises six items. Example item: ‘In response to infection outbreaks, I often feel unable to control key aspects such as pathogen tracing’. Scored using a 5-point Likert scale ranging from 1 (‘never’) to 5 (‘always’). A total mean score exceeding 3 indicates high perceived stress. This scale demonstrates good consistency and reliability, with a Cronbach’s alpha coefficient of 0.939.

Statistical analysis

The collected questionnaire data were organised and verified before undergoing statistical analysis. Descriptive statistics (means, SD, frequencies, percentages) were used to summarise participants’ demographic and occupational characteristics, as well as their scores for perception of response governance, collaborative attitude and stress perception. Independent-samples t-tests were used to compare stress perception scores between two groups (eg, gender, infection control experience, experience of HAI outbreaks). One-way analysis of variance was used to compare stress perception scores across three or more groups (eg, age groups, education levels, years of service, monthly income, administrative position, professional title, type of healthcare facility, grade of healthcare facility, position). Pearson correlation analysis was performed to examine the bivariate correlations among perception of response governance, collaborative attitude and stress perception. Hierarchical (block-wise) multiple regression was performed with variables entered in three sequential blocks: step 1 included covariates (demographic and occupational characteristics); step 2 added the main independent variables (perception of response governance and collaborative attitude); step 3 added the interaction term (perception of response governance×collaborative attitude). Multicollinearity was assessed using tolerance and variance inflation factor (VIF); all VIF values were well below 10, indicating no serious multicollinearity. To further probe the significant interaction, simple slope tests were conducted using the SPSS PROCESS macro program (model 1) developed by Hayes (2022). The significance level was set at α=0.05. All statistical analyses were carried out with SPSS V.27. The correlation heatmap was created using the pheatmap package in R V.4.3.2, with a blue–white–red colour scheme.

Hierarchical multiple regression was chosen as the primary analytical strategy because this study employed directly measured manifest variables and the core hypothesis concerned a moderating effect. This approach is straightforward, transparent and widely accepted for testing interaction terms. Structural equation modelling would introduce additional complexity, such as measurement model specification and estimator selection, without providing theoretical benefit for our manifest-variable model. Given that our sample size of 1017 is adequate for regression and the results are readily interpretable for clinical audiences, we considered hierarchical regression to be the more appropriate method for addressing our research questions.

Results

Characteristics of study participants

Table 1 presents the general characteristics of the 1017 healthcare workers who participated in this study; the majority were female (85.3%), and the mean age was 37.87 years (SD=8.09). Most participants possessed high educational attainment, with 85.6% holding a bachelor’s degree or higher. Over half had worked for more than 10 years (57.2%); 62.1% had monthly incomes exceeding ¥5000; most held non-administrative positions (86%); over half worked in general hospitals (54.6%) or tertiary hospitals (74.3%); clinical nursing staff constituted the majority (53.4%); most were not infection control department personnel (76.4%) and had no prior experience with hospital-acquired infection outbreaks (89.3%); senior professional titles were held by a minority (20.6%); the mean score for perception of response governance was 4.14 (SD=0.72), and the mean score for collaborative attitude towards outbreak prevention and control was 4.71 (SD=0.49), both indicating moderately high levels, accounting for 89.2% and 98.9%, respectively. The mean score for perceived stress regarding infection outbreaks was 2.03 (SD=0.97), indicating a low level of perceived stress. However, 11.9% of respondents reported high-stress levels.

Table 1. Univariate analysis of general characteristics of scores on stress perception for healthcare-associated infection outbreaks (N=1017, score, M±SD).

Item	Category	n (%)	Stress perception average score	P value
Gender	Male	150 (14.7)	2.19±0.99	0.026
	Female	867 (85.3)	2.00±0.96
Age (years)	<30	202 (19.8)	1.85±0.92	0.004
	31−50	727 (71.4)	2.05±0.96
	>50	88 (8.7)	2.22±1.01
Education level	Junior high school or lower	11 (1.1)	2.48±1.17	<0.001
	High school/technical secondary school	21 (2.1)	2.46±1.13
	College degree	114 (11.2)	1.88±0.98
	Bachelor degree	639 (62.8)	1.97±0.96
	Master’s degree or higher	232 (22.8)	2.19±0.92
Years of service (years)	1−2	113 (11.1)	1.87±0.84	0.026
	3−5	88 (8.7)	1.86±0.84
	6−10	137 (13.5)	1.96±1.03
	11−15	270 (26.5)	2.09±0.98
	16−20	175 (17.2)	2.00±0.98
	>20	113 (11.1)	2.15±0.97
Monthly income (¥)	<2000	28 (2.8)	1.69±0.85	0.099
	2001−5000	336 (33)	1.99±0.99
	5001-8000	277 (27.2)	2.01±0.97
	8001−10 000	213 (20.9)	2.06±0.96
	10 000−20 000	142 (14)	2.19±0.92
	>20 000	28 (2.8)	1.78±0.73
Infection control (IC) physician/nurse	Yes	100 (9.8)	1.90±0.88	0.296
	No	777 (76.4)	2.01±0.94
Administrative position	Members of the leadership team	7 (0.7)	2.23±1.19	0.226
	Department head/director/head nurse	96 (9.4)	2.21±1.04
	Deputy department head/deputy director/deputy head nurse	39 (3.8)	2.09±0.98
	No administrative position	875 (86)	2.00±0.95
Professional title	Senior	48 (4.7)	2.27±0.93	<0.001
	Associate senior	162 (15.9)	2.26±1.08
	Middle	450 (44.2)	1.97±0.93
	Junior	246 (24.2)	1.89±0.96
	None	111 (10.9)	2.10±1.04
Type of healthcare facility	General hospital	555 (54.6)	2.05±0.96	0.777
	Specialty hospital	399 (39.2)	1.99±0.97
	Traditional Chinese Medicine (TCM) Hospital	21 (2.1)	1.93±0.93
	Integrated Traditional Chinese and Western Medicine Hospital	8 (0.8)	2.29±0.83
	Township (Subdistrict) Health Center	34 (3.3)	2.01±0.93
Grade of healthcare facility	Tertiary hospital	756 (74.3)	2.06±0.96	0.208
	Secondary hospital	224 (22.1)	1.92±0.98
	Primary hospital	37 (3.6)	2.03±0.89
Position	Administrative/logistics department	100 (9.8)	2.18±0.93	0.002
	Doctor	210 (20.6)	2.21±1.12
	Nurse	543 (53.4)	2.26±1.12
	Medical technology department	124 (12.2)	2.01±0.96
	Others	40 (3.9)	1.93±0.93
Experience of healthcare-associated infection (HAI) outbreaks	Yes	109 (10.7)	2.44±1.08	<0.001
	No	908 (89.3)	1.98±0.94

Column percents may not total 100 due to rounding.

^*p<0.05.

Univariate analysis of healthcare workers’ perceived stress levels

The results of the univariate analysis showed that statistically significant differences (p<0.05) in the mean perceived stress scores during outbreaks were observed among personnel categorised by gender, age, educational attainment, years of service, professional title, job position and prior experience with outbreaks. Results are presented in table 1.

Association between healthcare workers’ perception of response governance, collaborative attitude towards hospital infection outbreak and perceived stress

The results of the correlation analysis indicated that perceptions of response governance, attitudes towards collaborative prevention and control, and perceived stress showed significant negative correlations (all p<0.001). Perception of response governance and attitude towards collaborative prevention and control showed a significant positive correlation (p<0.001). Results are presented in table 2 and visualised as a correlation heatmap in figure 2.

Table 2. Correlation analysis between perception of response governance and collaborative attitude for healthcare-associated infection and stress perception (r values).

Category	Perception of response governance	Collaborative attitude	Stress perception
Perception of response governance	1
Collaborative attitude	0.451*	1
Stress perception	−0.224*	−0.305*	1

^*Correlation is significant at the 0.001 level (two-tailed).

Figure 2. Correlation heatmap of perception of response governance, collaborative attitude and stress perception. Numbers represent Pearson correlation coefficients (all p<0.001). Colour intensity indicates the strength of correlation: red for positive, blue for negative.

Hierarchical multiple regression analysis of factors influencing healthcare workers’ perceived stress

Using the mean perceived stress score during infection outbreaks as the dependent variable, the first block included variables that were statistically significant in the univariate analysis (gender, age, education level, years of service, professional-technical title, job position). Variable values are detailed in online supplemental additional file 2.

Results indicated that staff who had experienced an infection outbreak (β=0.468, p<0.001) and those with 11–15 years of service (β=0.372, p=0.012) reported higher stress levels, suggesting occupational maturity may exert a psychological protective effect. The second block incorporated the mean scores of perception of response governance and collaborative attitude into a stratified regression analysis. Results revealed that after controlling for demographic factors, introducing perception of response governance and collaborative attitude significantly increased the model’s explanatory power for perceived stress variation by 10.2% (ΔR²=0.102, p<0.001). Specifically, perception of response governance (β=−0.400, p<0.001) and collaborative attitude (β=−0.210, p=0.008) exerted negative predictive effects on perceived stress. This indicates that sufficient knowledge reserves and positive collaborative attitudes help alleviate psychological stress. At the third level, the interaction term (perception of response governance×collaborative attitude) was included, with independent variables centred. Results showed that this model explained 15.6% of the variance in perceived stress (adjusted R²=0.156, F=9.167, p=0.022). The negative predictive effect of perception of response governance on perceived stress was significant (β=−0.384, p<0.001). The interaction term significantly predicted perceived stress, increasing explanatory power by 0.4% (ΔR²=0.004, p<0.05). This indicates that the collaborative attitude significantly moderated the relationship between perception of response governance and perceived stress (β=−0.202, p=0.022). These results are summarised in table 3.

Table 3. Hierarchical multiple regression analysis of influencing factors of perceived stress.

Step and variables	Control variables	Unstandardised coefficients B	SE	Standardised coefficient beta	P value	95% CI	Adjusted R2	ΔR2	F value
Step 1: control variables									3.688*
(constant)		1.980*	0.412		<0.001	1.171 to 2.788	0.050	0.069
Experience of healthcare-associated infection (HAI) outbreaks	No
Yes		0.372*	0.099	0.119	<0.001	0.177 to 0.566
Years of work experience (years)	1–2
11–15		0.318*	0.16	0.145	0.047	0.004 to 0.632
Step 2: main effects									9.304*
(constant)		4.452*	0.516		<0.001	3.439 to 5.466	0.152	0.102
Experience of healthcare-associated infection (HAI) outbreaks	No
Yes		0.468*	0.094	0.150	<0.001	0.283 to 0.653
Years of work experience (years)	1–2
11–15		0.388†	0.152	0.177	0.011	0.091 to 0.686
Perception of response governance		−0.400*	0.048	−0.284	<0.001	−0.493 to −0.306
Collaborative attitude		−0.210	0.079	−0.088	0.008	−0.366 to −0.055
Step 3: interaction effect									9.167†
(constant)		5.076*	0.582		<0.001	3.933 to 6.219	0.156	0.004
Experience of healthcare-associated infection (HAI) outbreaks	No
Yes		0.468*	0.094	0.150	<0.001	0.284 to 0.653
Years of work experience (years)	1–2
11–15		0.381*	0.151	0.174	0.012	0.084 to 0.678
Perception of response governance		−0.384*	0.048	−0.273	<0.001	−0.478 to −0.289
Collaborative attitude		−0.348*	0.099	−0.145	<0.001	−0.542 to −0.153
Perception of response governance×collaborative attitude		−0.202†	0.088	−0.085	0.022	−0.375 to −0.030

^*p<0.001.

^†p<0.05.

Relationship between healthcare workers’ perceived stress and perception of response governance: the moderating role of collaborative attitudes

A simple slope test was further employed to validate the moderating effect of collaborative attitudes towards infection outbreaks on the relationship between perception of response governance and perceived stress. Participants were categorised into high/low risk governance perception and high/low collaborative attitudes based on thresholds 1 SD above (M+1 SD) and below (M−1SD) the mean, respectively. Results indicate that perception of response governance significantly negatively predicts perceived stress across all levels of collaborative attitude. However, this negative predictive effect progressively strengthens as the level of collaborative attitude increases. Specifically, at low levels of collaborative attitude, the predictive effect of perception of response governance on perceived stress was significant (simple slope=−0.247, t=−3.885, p<0.001). At the mean level, the predictive effect remained significant (simple slope=−0.342, t=−7.240, p<0.001), and at high levels, the predictive effect intensified (simple slope=−0.402, t=−8.135, p<0.001). This indicates that collaborative attitudes amplify the buffering effect of perception of response governance on perceived stress. Results are presented in table 4 and figure 3.

Table 4. The moderating effect of collaborative attitude on the relationship between perception of response governance and stress perception.

Collaborative attitude	Simple slope	SE	T	P value	95% CI
Low level (−1 SD)	−0.247	0.064	−3.885	<0.001	−0.371 to −0.122
Mean level	−0.342	0.047	−7.240	<0.001	−0.435 to −0.249
High level (maximum value)	−0.402	0.049	−8.135	<0.001	−0.499 to −0.305

The moderator values represent the mean minus 1 SD (−0.403), the mean (0.000), and the observed maximum value (0.256). As the ‘mean plus 1 SD’ value fell outside the observed data range, the maximum observed value was used to represent the high level. Independent variables were mean centred.

Figure 3. Moderation model of collaborative attitude on the relationship between perception of response governance and stress perception.

Discussion

This study systematically examined the relationships among perceptions of response governance, collaborative attitudes towards infection prevention and control, and perceived stress among hospital staff in North-Eastern China. Key findings encompass three dimensions: first, participants demonstrated generally moderate-to-high levels of risk governance awareness and positive collaborative attitudes towards infection prevention and control, while overall perceived stress remained low. However, 11.9% of respondents reported high stress levels. Second, both perceptions of response governance and collaborative attitudes showed significant negative correlations with perceived stress. Together, they explained an incremental 10.2% of variance in perceived stress in stratified regression (ΔR²=0.102, p<0.001). Third, collaborative attitudes exerted a positive moderating effect on the relationship between perception of response governance and perceived stress—higher collaborative attitudes amplified the negative predictive effect of perception of response governance on perceived stress, with an incremental explanatory power of 0.4% (ΔR²=0.004, p=0.022). This finding validates the synergistic relationship between cognitive and collaborative resources in emergency management for HAI outbreaks, offering a new perspective on protective factors for healthcare workers’ psychological stress.

This study found a significant negative correlation between perception of response governance levels and perceived stress, aligning closely with multiple studies conducted during the COVID-19 pandemic. Chigwedere et al’s³⁰ systematic review explicitly identified ‘lack of virus-related knowledge’ and ‘insufficient training’ as key risk factors for acute stress disorder and post-traumatic stress symptoms among healthcare workers. This study extends this association from a nationwide infectious disease pandemic to the institutional level of hospital-acquired infection outbreak emergency scenarios, confirming the cross-context stability of the buffering effect of the perception of response governance. However, our findings partially diverge from those of Roshanshad et al³¹ among Iranian healthcare workers: their study suggested that both risk perception and perceived stress reflected fear levels towards COVID-19, with perceived stress showing a weaker association with sociodemographic characteristics. In contrast, our study found significantly higher perceived stress among those with 11–15 years of service and prior exposure to HAI outbreaks. A possible explanation for this discrepancy lies in differing stressor attributes: COVID-19, as a nationwide, persistent stressor with high media exposure, partially obscured differences in occupational characteristics and prior experiences. In contrast, HAI outbreaks—as institutional, episodic events with stronger professional relevance—allowed the influence of occupational maturity (years of service) and prior traumatic experiences to emerge more clearly. This finding suggests that psychological interventions during healthcare outbreak scenarios should prioritise the precise identification of ‘experienced individuals’ and ‘those with prior trauma exposure’.

Psychological sensitisation effects of exposure history: from traumatic memory to persistent vulnerability. This study found that individuals with prior experience of hospital-acquired infection outbreaks exhibited significantly higher levels of perceived stress (β=0.468, p<0.001), a result highly consistent with that of Lee et al³² on COVID-19 in South Korea. Lee et al explicitly stated that healthcare workers who experienced Middle East Respiratory Syndrome isolation had a 1.70 times higher risk of developing post-traumatic stress disorder (OR=1.70; 95% CI 1.22 to 2.37) compared to those without such exposure, proposing that ‘prior epidemic exposure constitutes an effect-modifying factor’. This study replicated this pattern in the context of general hospitals in North-Eastern China, providing the first validation of the psychological sensitisation effect of trauma exposure in a non-pandemic setting involving a hospital-acquired infection outbreak.

Notably, this finding resonates with the theoretical discussions of domestic and international scholars^33–35 on ‘psychological resilience’. Their research indicates that psychological resilience significantly buffers the impact of stress on depression and anxiety. However, this study further reveals that the protective effect of psychological resilience is not uniformly distributed; prior exposure history may constitute a ‘depletion window’ for resilience resources. Hospital-acquired infection outbreaks are not ‘one-time immunity’ events; rather, prior exposure may foster a persistent state of psychological sensitisation. This group should be identified as a key target population for psychological interventions, with psychological resilience tracking mechanisms initiated following infection outbreaks.

The most significant theoretical contribution of this study lies in identifying and validating the positive moderating effect of collaborative attitude towards infection prevention and control on the relationship between perception of response governance and perceived stress. Previous research has predominantly examined cognition and attitude as parallel protective factors in separate analyses. For instance, Lyu et al³⁶ conducted a structural equation modelling analysis among 2795 nurses across six hospitals in Hubei Province. Their findings indicated that infection control knowledge had a limited direct impact on risk perception, whereas workplace safety perception (which can be viewed as an organisational-level manifestation of a collaborative attitude) significantly and negatively predicted risk perception. However, this study did not further examine the interaction effect between these two factors.

This study explicitly identified the interaction term between cognition and collaborative attitude (β=−0.202, p=0.022) in the third block of hierarchical regression. Simple slope analysis revealed that at high levels of collaborative attitude, the negative predictive effect of perception of response governance on perceived stress increased from −0.247 (low collaborative attitude) to −0.402. The deeper implication of this interaction effect is that cognition and collaboration are not independent psychological resources but exhibit a ‘catalyst effect’: a high-level team collaboration atmosphere activates the psychological efficacy of cognitive reserves, enabling individuals to more effectively translate their cognitive understanding of risk governance systems into a sense of control under stressful conditions. Conversely, in a weak collaborative atmosphere, even with ample knowledge reserves, the stress-reducing effect is significantly inhibited.

This finding advances the core proposition of the JD-R model—‘job resources buffer job demands’—by demonstrating synergistic effects among different categories of job resources, rather than a simple additive relationship. It also explains why previous single-track intervention models centred on ‘knowledge training’ yielded limited results: cognitive empowerment must be embedded within collaborative organisational soil to fully unleash its psychological protective potential.

Previous studies have explicitly pointed out^{37 38} that ‘effective psychological interventions and organizational support must transcend the individual level and shift towards addressing systemic issues such as authority allocation, resource allocation, and communication systems’. Another study³⁹ further proposed that ‘structural stress allocation’ is the root cause of excessive stress overload in clinical frontline positions. From an individual-organisation interaction perspective, this study quantitatively validates for the first time the conditional activation effect of ‘collaborative attitudes’ as an organisational social resource on cognitive efficacy.

In other words, while prior research revealed structural flaws where ‘collaborative failure exacerbates stress’ (eg, high-risk positions lacking commensurate support), this study further addresses: ‘Under identical structural conditions, what factors enable some individuals to utilize cognitive resources more effectively?’. The answer lies in individual differences in perceptual attitudes towards collaboration. This finding translates the macro-level calls for ‘governance cognition’ and ‘collaborative communication’ during the COVID-19 pandemic into measurable, actionable organisational psychological variables, providing micro-mechanism evidence for emergency management of HAI outbreaks.

Based on the above findings, this study offers clear practical guidance for public health policy and hospital management system interventions. First, a dual-track intervention framework combining ‘cognitive empowerment’ and ‘collaborative support’ should be established, rather than merely ‘enhancing training’. A systematic review⁴⁰ indicates that in high-income and upper-middle-income countries, multimodal interventions—such as care bundles integrating education, audit feedback and leadership commitment—significantly reduce hospital infection rates, particularly for bloodstream infections and surgical site infections. Conversely, knowledge-only training yields limited effects in low-income countries, with training experiences often showing no significant correlation with knowledge or practice scores.^41–43 This paradox suggests the crux lies not in ‘whether training exists’, but in ‘training format’ and ‘organisational embedding methods’. This study recommends integrating knowledge modules with team collaboration modules into hospital outbreak emergency plans, and employing ‘learning by doing’ approaches such as scenario simulations, interdepartmental joint debriefings and non-punitive error reporting. This synchronises cognitive internalisation with collaborative trust-building within the same operational context.

Second, elevate outbreak stress management from individual psychological support to systematic organisational resilience building. We recommend adding an ‘emergency response psychological support’ indicator to hospital accreditation standards and integrating collaborative organisational culture development into hospital quality management systems. As demonstrated by German public health authorities, the quality of ‘internal communication’ and ‘work atmosphere’ correlates closely with managers’ perceived levels of crisis stress,^{44 45} a finding replicated and deepened in this study’s quantitative data.

Third, establish a dynamic identification and tiered intervention mechanism for high-risk individuals. This study, along with others,^46–48 warns that prior exposure history not only enhances cognitive awareness but also sensitises psychological vulnerability. It is recommended to activate a psychological resilience tracking mechanism following healthcare outbreak incidents, and to incorporate survivors into phased follow-ups. Targeted psychological interventions based on prior trauma exposure should be provided, such as cognitive restructuring and post-traumatic growth counselling.

This study possesses the following strengths. First, the sample covers multiple general and specialised hospitals in North-Eastern China, closely aligning with the regional nursing workforce in terms of gender, age, professional title and job distribution. This ensures strong ecological validity and external generalisability. Second, this study pioneers the introduction of perceptions of response governance and collaborative attitudes towards infection prevention and control as moderating interaction variables in research on infection outbreak stress. The measurement tools demonstrated good reliability and validity, providing an operationalisable pathway for future related studies. This study also has the following limitations. First, as a cross-sectional study, it cannot establish causal temporal relationships between variables. Future longitudinal tracking designs could examine the lagged effects of cognitive and attitudinal dynamics on stress perception. Second, all variables were assessed using self-report scales, which may have introduced common-method bias. However, detecting interaction effects partially mitigates this threat. Third, the sample predominantly comprised clinical healthcare workers, with limited representation from administrative and support staff. Future studies should consciously broaden occupational coverage.

Conclusion

In summary, this study confirms that perception of response governance and collaborative attitude towards hospital infection outbreaks are not only independent protective factors against perceived stress but also exhibit a significant synergistic buffering effect. This finding suggests that public health emergency management systems must equally prioritise fostering a collaborative organisational culture while strengthening technical standards and knowledge training. Only by embedding ‘cognitive empowerment’ and ‘collaborative support’ into daily hospital operations and institutional design can a robust organisational immunity barrier against the psychological impact of infection outbreaks be truly established.

Data availability statement

Data are available on reasonable request.