Reliability and validity of the Chinese version of the nurses’ health-related procrastination scale (NHRPS) in Chinese nurses

Abstract

Background

The physical and mental health of nurses is closely related to the quality of nursing and patient safety. Nurses face greater health risks, which are closely related to health procrastination. At present, China lacks scientifically effective tools to assess the level of health-related procrastination among clinical nurses. The purpose of this study was to conduct cross-cultural debugging of the Nurses’ Health-Related Procrastination Scale (NHRPS) and to test the reliability and validity of the Chinese version of the NHRPS among clinical nurses.

Methods

The NHRPS was translated strictly according to cross-cultural debugging guidelines. The content validity of the Chinese version of the Nurses’ Health-related Procrastination Scale was evaluated by 10 experts. From January to February 2025, 395 clinical nurses were surveyed via the Chinese version of the NHRPS. Item analysis, exploratory factor analysis (EFA), confirmatory factor analysis (CFA), Cronbach’s α coefficient and retest reliability were used to test the reliability and validity of the scale.

Results

The Chinese version of the NHRPS contains 25 items in 4 dimensions. Four common factors were extracted via exploratory factor analysis, and the cumulative variance contribution rate was 64.016%. Confirmatory factor analysis revealed that the chi-square/degree of freedom (CMIN/DF) was 1.515, the root mean square error of approximation (RMSEA) was 0.049, the comparative fit index (CFI) was 0.956, the incremental fit index (IFI) was 0.956, and the Tucker Lewis index (TLI) was 0.951. Convergent validity was The Cronbach’s α coefficient of the scale was 0.930. The ranges of composite reliability (CR) and average variance extracted (AVE) of convergence validity were 0.839–0.902 and 0.497–0.616. The fitting index of the second-order factor model is similar to that of the first-order one and has good structural validity. The first-order factor model is a more economical model, and it is more reasonable to select the first-order confirmatory factor model.

Conclusions

The reliability and validity of the Chinese version of the NHRPS are good, and it can be used as a scientific and effective tool to evaluate the current situation of health-related procrastination behaviours for clinical nurses.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12912-025-03275-7.

Background

The National Nursing Career Development Plan (2021–2025) issued by China clearly states that nursing work is closely related to the safety and health interests of people [1]. The physical and mental health of nurses is closely related to the quality of nursing and patient safety, but the role of nurses’ health behaviour in health promotion is worrying.

Procrastination is widely regarded as the act of voluntarily postponing a planned task, although an individual can foresee the adverse consequences of procrastination [2]. This nonadaptive behaviour can produce harmful results for individuals or groups. Studies have shown that, compared with the general population, nurses face greater health risks [3]. Nurses neglect their own health due to high work pressure, such as delaying rest and exercise [4], and adopt emotional survival strategies when facing complex clinical environments to delay emotional expression and mental health support, which increases the risk of chronic diseases and psychological problems [5, 6]. The occurrence of occupational diseases such as varicose veins and lumbar intervertebral disc protrusion among nurses may also be related to the delay in adhering to the principle of human body effort conservation [7].

According to the procrastination-health model [8, 9], increased stress and poor health-promoting behaviours are two important ways in which habitual procrastination increases the risk of poor physical health. Long-term stress can suppress the immune system, reduce the body’s resistance to diseases, and aggravate the occurrence of cardiovascular diseases. The behavioural approach is to exert a damaging effect on health by avoiding and postponing health-promoting behaviours, such as a healthy diet and regular exercise, as well as engaging in health-risky behaviours [10]. Although the impact of behavioural pathways may not manifest immediately, for the nurse population with health vulnerability, the health risks caused by the cumulative effect of poor health behaviours cannot be ignored. Health-related procrastination is a type of procrastination behavior. It refers to the voluntary and unnecessary procrastination behavior that occurs when an individual performs a healthy behavior or task, despite having the initial intention to start or terminate the task [3]. Compared with general procrastination, health-related procrastination is considered to have a stronger and more direct relationship with health outcomes. Delaying healthy behaviours often leads to a greater level of health risk and stress, which includes a combination of concerns about health and the psychological burden caused by delaying behaviours. Moreover, high pressure and high-intensity work in the professional environment of nurses have exacerbated the occurrence of this phenomenon. The incidence of job burnout among medical staff in China is 45.4–76.9% [11], which is a negative psychological state caused by long-term work pressure, emotional consumption and physical and mental exhaustion [12]. Nurse burnout, as an adverse health outcome, has adverse effects on individual nurses, the quality of care and the patient experience.

When reviewing the literature, most of the current research on procrastination has focused on academic procrastination and general procrastination, whereas health-related procrastination has received little attention. Most existing procrastination scales are designed in academic settings and are used by community adults and college students [13–15]. The existing assessment tools for health-related procrastination have a single dimension and are unable to assess multiple dimensions of health [3, 16]. Owing to the special characteristics of nurses, such as hand washing, vaccination, occupational protection and other health behaviours, the application of these tools in the nursing field is still limited. The Nurse Health-related Procrastination Scale was developed by Iranian scholars Mahdi et al. [17], a high-level research team, after in-depth qualitative interviews were conducted with the nurse population for the first time. It has a strict development process and has currently been verified only in the Iranian nurse population, achieving good reliability and validity. This scale has a high degree of novelty. For example, the entries “I delay getting vaccinated and controlling my antibody titers for hepatitis, although these are essential for nurses” and “I delay adhering to safety principles (gowns, masks, gloves, glasses) when providing nursing care” prominently reflect the particularity of the nurse population, which distinguishes the NHRPS from the universal procrastination scale.

In summary, the purpose of this study is to introduce the NHRPS for cross-cultural debugging and to assess the reliability and validity of the Chinese version of the NHRPS.

Methods

Design and participants

The study used a cross-sectional design. The survey was conducted in a tertiary hospital in Shanghai via convenience sampling from January to February 2025. There are 29 NHRPS items. According to the Kendall estimation method, the sample size should be at least 5 ~ 10 times the number of scale items [18]. For confirmatory factor analysis, a minimum sample size of 200 is needed [19]. Considering a potential 10% rate of invalid questionnaires, the ideal sample size should range from 383 ~ 544. The inclusion criteria for the research subjects were as follows: (1) voluntary participation in the study, (2) obtained a nurse practicing certificate, and (3) engaged in clinical nursing work for 12 months. Nursing interns and learning nurses were excluded. A total of 418 eligible participants were included in this study. A total of 23 questionnaires whose completion time was too short (less than 100 s), repetitive content or illogical content were deleted. There were a total of 395 valid questionnaires, accounting for 94.5% of the total.

Instrument tools

General information questionnaire

The general information questionnaire was designed on the basis of a literature review and according to the purpose of the research, including gender, age, department, education level, marital status, child status, job title, working years, average monthly income, and specialist nurse status, with a total of 10 items.

Nurses’ Health-Related procrastination scale

The NHRPS was originally developed by Professor Mahdi et al. in 2023 [17]. The scale contains 29 items, including five dimensions: procrastination in maintaining physical health, procrastination in physical health promotion, procrastination in quitting high-risk behaviours, procrastination in social and mental health, and procrastination in spiritual health. The Likert 5-point scoring method was adopted. The higher the score is, the more frequently procrastination behaviour occurs. This scale demonstrated excellent internal consistency, with an original Cronbach’s α coefficient of 0.947.

Translation procedures

Phase I: translation process

We received the scale developer’s authorization via email in September 2024. The Brislin translation model was used to translate the NHRPS [20], and the translators were not exposed to the original scale. The original scale was independently translated into Chinese by 2 nursing graduate students (both of whom passed CET6) to form T1 and T2. After the two discuss the results and agree, T3 is formed. Two nursing experts with overseas study experience were invited to translate T3 back into English BT1 and BT2 independently. Members of the research team compared and analysed BT1 and BT2 with the original scale, modified the T3 version until it was consistent with the original scale, and finally determined the preliminary Chinese version of the NHRPS, namely, T4.

Phase II: cultural adaptation

Ten nursing specialists, including five PHDs, four master’s degrees, and one undergraduate degree, were invited to the T4 version of the cultural commission process. Three specialists were from the field of mental health research, and seven specialists were from the field of nurse management. The professional experience of these experts ranges from 7 to 24 years. In accordance with their own theoretical knowledge and clinical experience, the experts made suggestions on the language expression and content relevance of each item of the scale. After two rounds of expert consultation, the content of the scale was adapted to the Chinese cultural background and nursing environment, and the Chinese version of the NHRPS T5 was formed. A second round of expert correspondence will also be used to assess content validity.

In this section, 10 experts made adjustments to T3, providing suggestions for the five items in the NHRPS. Item 3 was amended from “washing my hands” to “hand hygiene” in more standardized terms. Item 7 was expanded from “blood pressure” to “health-related indicators (such as blood pressure)”. Item 13, “I delay quitting substance abuse (i.e., opioid, drugs)”, was deleted because it deviates from the Chinese social code of conduct and is not applicable to the Chinese cultural environment. Owing to the low correlation between the item and the measured content, item 22 “I delay managing my finances.” was deleted. Item 27 was supplemented from “spiritual obligations” to “self-reflection and pursuit of life value” to promote understanding.

Phase III: presurvey

In November 2024, convenience sampling was used to select 20 clinical nurses who met the inclusion and exclusion criteria, all with a bachelor’s degree and more than 10 years of work experience. The researchers used the Chinese version of T3 to conduct the investigation, and the nurses had no doubts about the expression and understanding of the items, the relevance of the topic, or whether there were redundant omissions. In the end, only some language expressions of the scale were revised to form the final Chinese version of the NHRPS T6.

Data collection

The researchers are all nursing graduate students with good professional foundations and language communication skills and have a full understanding of the scale items. The questionnaire was created via the online platform Questionnaire Star [21]. After the permission of the research unit was obtained, the questionnaire was distributed through WeChat. Before the survey was conducted, the researchers explained the purpose of the study to the participants and emphasized the principle of confidentiality. The home page of the electronic questionnaire also explains again the significance and purpose of this study and the precautions for completing it, ensuring that the research subjects can choose by themselves whether to complete the questionnaire after being informed of it. The participants could withdraw at any time during the questionnaire-completing process without being affected. The questionnaire does not include the participants’ names, contact information, etc., and is restricted to being filled out only once. By setting screening conditions, questionnaires that take too short a filling time (less than 100 s) and have repetitive or illogical content will be deleted to ensure the authenticity and reliability of the data. Questionnaires that take too little time to complete (less than 100 s), are repetitive or illogical will be removed to ensure data reliability and validity [22].

Data analysis

SPSS 27.0 and AMOS 24.0 were used for data analysis. The data are expressed as the means and standard deviations or frequencies and percentages, with statistical significance set at P < 0.05. Common method bias (CMB) was tested by the Harman single-factor test. The item analysis, validity analysis, and reliability analysis are described below.

Item analysis

Item analysis was used to test the discrimination and homogeneity of the scale items, and the scale was sorted according to the total score, with the top 27% as the high group and the bottom 27% as the low group. Independent sample t tests were used to judge the differences between the two groups for each item (t < 3.00 or P > 0.05, deletion was considered) [22]. Pearson correlation analysis was used to examine the correlation coefficient between the score of each item and the total score of the scale (P > 0.05 or r < 0.40, deletion was considered) [23].

Validity analysis

In this part, content validity and structural validity are used as test indices. Ten experts were selected to evaluate the content validity. According to the item-level content validity index (I-CVI) and scale-level content validity index (S-CVI), I-CVI ≥ 0.78 and scale-level content validity index/average (S-CVI/Ave) ≥ 0.80 are considered acceptable [24, 25]. Structural validity includes EFA and CFA [26]. A total of 167 samples were randomly selected for EFA, and the remaining 228 samples were used for CFA. The KMO value was greater than 0.80, and the value of the Bartlett sphericity test was significant (P < 0.05), which was suitable for factor analysis. Principal component analysis and the maximum variance orthogonal rotation method were used to determine the factor number and load value (factor loading < 0.40, considering deletion) [27, 28]. In CFA, the structural equation model is used to test the degree of fit of the model. The fit indicators and their critical values include CMIN/DF < 3.00; RMSEA < 0.08; and CFI, IFI, TLI ≥ 0.90 [29]. Convergent validity is used to measure whether the measurement results of the same latent trait by different measurement methods in the measurement tool have a high degree of consistency and similarity. Aggregation validity is measured CR values. If the factor loading value is greater than 0.5, the AVE value is greater than 0.5, and if the CR value is greater than 0.7, the convergent validity is relatively high.

Reliability analysis

The Cronbach’s α coefficient was used to measure the internal consistency of the scale. A Cronbach’s α coefficient ≥ 0.80 was considered good reliability of the scale. The samples were divided into odd and even groups, and the correlation between the two groups was tested to assess the split-half reliability coefficient, with a value greater than 0.80 considered good. Test-retest reliability is usually evaluated via the intraclass correlation coefficient (ICC). Thirty participants were selected for measurement, and the ICC values before and after two weeks were calculated. If the ICC value is greater than 0.7, the test‒retest reliability of the scale is considered reasonable [30].

Ethical considerations

This study was approved by the human Trial Ethics Review Committee of Shanghai General Hospital on January 20, 2025 (No. 2025–038) (see Supplementary Material 4). Informed consent to participate was obtained from all of the participants in the study, and the confidentiality principle was strictly implemented.

Results

Common method bias test

Data collection methods such as the questionnaire survey method and self-reporting method may involve CMB. In this study, the Harmam single-factor test was used for possible common method bias. The results show that the characteristic roots of six factors are greater than 1 and that the variance interpretation rate of the first factor is 35.81%, which is lower than the critical criterion of 40%, indicating that there is no serious common method bias phenomenon in the data of this study.

Characteristics of the participants

A total of 418 questionnaires were sent out in this part. After 23 invalid questionnaires were eliminated after two-person screening, 395 (94.50%) subjects were ultimately included in the study. Table 1 shows the detailed demographic information of the study subjects.

Table 1.

Sample characteristics (N = 395)

Characteristics	N	%
Gender
Male	15	3.8
Female	380	96.2
Age
≤ 25	63	15.9
26 ~ 30	92	23.3
31 ~ 40	176	44.6
≥ 41	64	16.2
Type of clinical unit
Internal Medicine	68	17.2
Surgery Department	67	17.0
Obstetrics and Gynecology	40	10.1
Pediatrics Department	23	5.8
Emergency and ICU	156	39.5
Other	41	10.4
Level of education
Associate Degree	64	16.2
Bachelor Degree	319	80.8
Master Degrees or above	12	3.0
Whether you are married
Yes	270	68.4
No	125	31.6
Whether you have children
Yes	241	61.0
No	154	39.0
Position
Nurse	62	15.7
Nurse practitioner	216	54.7
Nurse in charge	114	28.9
Associate professor of nursing or above	3	0.8
Years of working
≤ 5	111	28.1
6 ~ 10	85	21.5
11 ~ 15	71	18.0
16 ~ 20	83	21.0
≥ 21	45	11.4
Average monthly income
≤ 5000	18	4.6
5001 ~ 7000	34	8.6
7001 ~ 9000	140	35.4
9001 ~ 10,000	139	35.2
≥ 10,001	64	16.2
Whether you’re specialist nurse
Yes	143	36.2
No	252	63.8

Item analysis

According to the decision value analysis results, the CR values of the high group and low group ranged from 3.238 to 14.235, and the difference was statistically significant (P < 0.001), indicating the degree of discrimination of each item. Pearson correlation analysis revealed that the correlation coefficient between items 13 and 14 and the total score of the scale was lower than 0.400, indicating that the degree of homogeneity between the items and the total score was low. The research team members discussed these two items together and decided to delete them after specific analysis of the contents of the items (see Supplementary Material 1: Table 1).

Content validity

After expert consultation, 10 experts determined that the items were “content related (3 points)” and “content is very relevant (4 points)”, accounting for 80–100% of the total number of people. The I-CVI of each item of the scale ranged from 0.80 to 1, all of which were > 0.78. The S-CVI/UA is 0.850, and the S-CVI/Ave is 0.981 (see Supplementary Material 1: Table 2).

Construct validity

After items 13, 14, and 15 are deleted, the third dimension of the scale is deleted, leaving four dimensions. The EFA results revealed that the KMO test value of the scale was 0.900, and the value of Bartlett’s spherical test was 2602.785, which reached a significant level, indicating that the scale could be used for factor analysis. Through principal component analysis and the maximum variance orthogonal rotation method, the fixed factor number is 4, and in the factor load matrix of all items, the factor load of all items is greater than 0.40, accounting for 64.016% of the total variance. Therefore, the scale ultimately retained 25 items (see Table 2).

Table 2.

Rotated component matrix in EFA (N = 167)

Dimension	Item	F1	F2	F3	F4
procrastination in maintaining physical health	1		0.595
	2		0.650
	3		0.741
	4		0.867
	5		0.659
	6		0.749
	7		0.485
	8		0.812
procrastination in physical health promotion	9			0.662
	10			0.805
	11			0.796
	12			0.787
procrastination in social and mental health	15	0.747
	16	0.619
	17	0.617
	18	0.818
	19	0.817
	20	0.705
	21	0.726
	22	0.742
	23	0.696
procrastination in social and mental health	24				0.739
	25				0.733
	26				0.803
	27				0.748
Eigenvalue		9.511	3.152	2.046	1.295
contribution rate (%)		38.044	12.608	8.185	5.179
Cumulative contribution rate (%)	64.016

Amos 24.0 software was used to conduct confirmatory factor analysis on 228 samples and further verify the results of the exploratory factors to obtain the initial model. Indicators such as CMIN/DF, RMSEA, IFI and CFI were used to test the degree of fit of the structural model [30], and the results showed that the model had a good fit (see Table 3). Figure 1 shows the CFA path analysis diagram. The correlation coefficients between the 25 items and their underlying variables range from 0.64 ~ 0.79, indicating that the underlying variables have a high degree of interpretation of the observation items. Moreover, the correlation coefficients between each dimension range from 0.76 to 0.93, suggesting that this model may have a higher-level factor structure. Therefore, the path diagram of the second-order four-factor CFA model is drawn, as shown in Fig. 2. The model fit index is within the acceptable range, indicating that the model is relatively ideal (see Table 4).

Table 3.

Fit indices of the first-order CFA model (N = 228)

Fit Indices	CMIN/DF	RMSEA	CFI	IFI	TLI
Reference value	< 3.00	< 0.080	> 0.900	> 0.900	> 0.900
Initial model	1.551	0.049	0.956	0.956	0.951

Fig. 1.

Estimation results of standardized path coefficients of the first-order CFA model. PMPH: Procrastination in maintaining physical health; PPHP: Procrastination in physical health promotion; PSMH: Procrastination in social and mental health; PSH: Procrastination in spiritual health

Fig. 2.

Estimation results of standardized path coefficients of the second-order CFA model. NHRP: Nurses’ Health-Related Procrastination; PMPH: Procrastination in maintaining physical health; PPHP: Procrastination in physical health promotion; PSMH: Procrastination in social and mental health; PSH: Procrastination in spiritual health

Table 4.

Fit indices of the second-order CFA model (N = 228)

Fit Indices	CMIN/DF	RMSEA	CFI	IFI	TLI
Reference value	< 3.00	< 0.080	> 0.900	> 0.900	> 0.900
Initial model	1.628	0.053	0.949	0.950	0.944

Convergent validity

The standardized factor loadings of all the items in the NHRPS ranged from 0.645 to 0.794, which was greater than 0.5. The CR value ranges from 0.839 to 0.902 and is greater than 0.7. The AVE value ranges from 0.497 to 0.616. The AVE value of the first dimension is slightly lower than the critical value (0.497 < 0.5). Overall, the first-order four-factor model has better convergent validity (see Supplementary Material 1: Table 3). The second-order four-factor model also reflects good convergent validity (see Supplementary Material 1: Table 4).

Reliability analysis

The total Cronbach’s α coefficient of the NHRPS was 0.930, and the Cronbach’s α coefficients of each dimension were 0.876, 0.860, 0.885, and 0.881. The split-half reliability coefficient of NHRPS is 0.778. The ICC value of the scale was 0.924, and the ICC values of each dimension were 0.815, 0.782, 0.838, and 0.743, respectively (see Supplementary Material 1: Table 5).

Discussion

Cultural adaptations enhanced construct validity

On the basis of the Brislin translation model and the Cross-Cultural Adaptation Guide [31], we invited 10 experts to perform cultural debugging and modified 5 entries in the NHRPS to adapt to Chinese culture. Specifically, item 3, “washing my hands”, was modified to “hand hygiene”, which meets the general terms of washing my hands, hand antisepsis and other behaviours of the Ministry of Health of China, and is in line with the phrasings of nurses [32] Item 7 was added from “blood pressure” to “health-related indicators (such as blood pressure)”, so that more health indicators can be paid attention to, the comprehensiveness of health monitoring is emphasized, and the applicability of the scale is enhanced. In an announcement of China’s drug administration [33], the abuse of drugs was a violation of laws and regulations, medical use and social norms and was not applicable to the Chinese cultural environment. It is decided to delete item 13: “I delay quitting substance abuse (i.e., opioid, drugs)”. Item 22, “I delay managing my finances”, is deleted because of its low correlation with the corresponding dimension “Procrastination in social and mental health”. According to the definition of health by the World Health Organization (WHO), social and mental health mainly includes the adaptability of individuals in the social environment, the quality of interpersonal relationships and social participation, and maintaining a positive and stable state at the psychological level. Financial management behaviour aims to achieve financial goals and maximize benefits [34]. There are great differences in the definition and emphasis of the two, and the direct connection is weak. The term spiritual obligation is not inherent in China. From the perspective of ethics, spiritual obligations can be regarded as internal moral requirements; that is, an individual realizes self-improvement and the enhancement of life value through self-reflection [28]. Therefore, item 27, “spiritual obligations”, is interpreted as “self-reflection and pursuit of life value” to reduce ambiguity and enhance understanding. After cross-cultural adjustment, the scale shows good content validity, which is detailed below.

The NHRPS has a good degree of discrimination

In the analysis of the items, the critical ratio of 27 items was greater than 3.0, with significant differences (P < 0.001). Except for items 13 and 14, the correlation coefficient between the other 25 items and the total score ranged from 0.452 ~ 0.727 (P < 0.01). Item 13, “I delay quitting substance abuse (i.e., opioid, drugs)”, and item 14, “I delay quitting drinking alcohol”, are related to smoking and drinking. This is not advocated by policies and society and may cause research subjects to engage in behaviours contrary to their original intentions. This is known as social expectation bias [35]. In China’s nursing work, the proportion of the female population is generally greater than that of the male population, and the proportion of male nurses in acute and critical conditions is relatively high. This study collected data across all departments, which led to a lower proportion of male nurses. Studies have shown that smoking and drinking are more common among adult men in China [36], so the response rate to this question is relatively low. This might be the reason why the correlation coefficients of items 13 and 14 are relatively low.

The NHRPS has good validity

In terms of cultural adaptability, the content that does not conform to Chinese culture and language expression habits is modified to ensure that the scale is closer to the Chinese context while retaining the original meaning. The NHRPS showed good content validity, with I-CVI values ranging from 0.80 to 1.00 (> 0.78), S-CVI/UA = 0.850 (> 0.8), and S-CVI/Ave = 0.981 (> 0.9).

In EFA, the number of factors extracted is determined to be 4, and the results show that the cumulative variance contribution rate is 64.016% (> 60%), and the factor loading of each item ranges from 0.485 to 0.867 (> 0.4). These findings indicate that the Chinese version of the NHRPS four-factor model has good structural validity. In CFA, the reliability of the four-factor structure is verified. The results of the structural equation model show that CMID/DF = 1.551, RMSEA = 0.049, CFI = 0.956, IFI = 0.956, and TLI = 0.951, all of which reach the critical value, indicating that the model fits better. Both models also demonstrated good convergent validity, suggesting good structural validity. On the basis of the high correlation among the potential variables, a second-order four-factor model was further constructed. The results show that the model fits ideally well. The four factors extracted from EFA are procrastination in maintaining physical health and promoting physical health, psychosocial health and mental health. These four dimensions reflect nurses’ health-related procrastination behaviours from different perspectives. From the perspective of social psychology, the physical, psychosocial and mental health of nursing staff are interdependent and influence each other. If only one of them is studied as the dependent variable, one-sided conclusions will be drawn. The second-order four-factor model shows that the relationship between the second-order factors and the first-order factors is relatively strong (0.91, 0.90, 0.98, 0.96), suggesting that the second-order factors can fully express the relationship between the first-order factors. The fit indices of the first-order and second-order factor models do not differ much. Considering that the first-order factor model is simpler and clearer and can express the relationships among variables with fewer parameters, it is more reasonable to select the first-order confirmatory factor model.

The results of the EFA and CFA in this study were different from those of the original scale, which might be due to the change in factor structure after cross-cultural adjustment. From the perspective of dimensional content, the deleted entries include “tobacco”, “drinking alcohol” and “substance abuse (i.e., opioid, drugs)”. Drug abuse in Iran is relatively common in society and is generally more common than that in China [37]. Drug abuse behaviour is not supported by Chinese policies and society, and people often avoid mentioning it. The consulting experts also indicated that the content was inapplicable. The reason for the change in the scale structure might lie in cultural differences. Smoking and drinking behaviours exist in both China and Iran, and the proportion of female nurses is relatively high. The reason for the change in the scale structure may lie in the measurement differences, which are related to the selection of samples and/or data analysis methods. On the basis of the results of the expert inquiries and project analysis, the relevance of the deleted items is lower than the critical value, and deletion is considered. In summary, considering comprehensively the dimensions of content, expert inquiries, project analysis and the fitting effect of the model, the four-factor structure is reasonable and more suitable for evaluating the level of health-related procrastination among clinical nurses in China.

The NHRPS has good reliability

The Cronbach’s α of the entire scale was 0.930, and the ICC value was 0.924. The Cronbach’s α of the original scale was 0.947, and the ICC value was 0.944. Compared with the original scale, the reliability of the Chinese version of the NHRPS is lower. This may be due to the following reasons. Judging from the calculation formula of internal consistency reliability : number of items; : the mean of the correlation coefficients of k items), when the correlation between items is high, the increase in the number of items is accompanied by an increase in the internal consistency coefficient. This difference might be related to the deletion of four items in this study. This type of difference also has an undeniable connection with cultural differences, such as the change in factor structure caused by cultural differences. The reliability of the Chinese version of the NHRPS is similar to that of the original scale, with good internal consistency and reliability. It can be used as a scientific assessment tool to measure the health-related procrastination level of nurses reasonably and accurately. (See Supplementary Material 3)

Innovation and limitations

This study localized the NHRPS for the first time in a standardized process, forming a Chinese version of the NHRPS suitable for the Chinese cultural background and medical environment, effectively measuring nurses’ procrastination behaviors in maintaining physical health and promoting physical health, social and mental health, and mental health. For example, nurses may delay seeing a doctor or postpone health check-ups due to psychological stress or emotional problems. The NHRPS helps nursing managers analyse reasons for procrastination by identifying nurses’ procrastination behaviors in the health field, such as high work pressure, insufficient emotional management ability and low coping efficiency. If research on the relationship between health status and work efficiency among emergency nurses is planned, the NHRPS can assess the health status of emergency nurses to analyse the connection between the two and assist nursing managers in formulating targeted health intervention measures, such as psychological counselling and health education courses.

There are several limitations in this study. First, the sample has geographical limitations and lacks representativeness of nurses from rural or nontertiary hospitals. In addition to China and Iran, NHRPS has not yet been verified and implemented in other countries. In the future, further multicenter validation in different countries and regions is still needed to improve the reliability of the scale. Second, the gender imbalance in the sample may have had a potential impact on the research results. Although the proportion of female nurses is generally greater than that of male nurses, we hope that the scale can be verified among nurses, with a greater proportion of males, to expand the scope of application of the scale. Third, no criterion-related validity was conducted in this study. Since the currently recognized effective procrastination scales are mostly used to assess general procrastination in adults and academic procrastination in students and may not be applicable to clinical nurses with complex working environments and health vulnerabilities, it is rather difficult to select appropriate standard scales. Therefore, no criterion-correlation analysis has been conducted.

Conclusions

The Chinese version of the NHRPS in this study includes 25 items in 4 dimensions (procrastination in maintaining physical health, procrastination in physical health promotion, procrastination in social and mental health and procrastination in spiritual health). It has good reliability and validity and has been adapted to the cultural environment and nursing field in China. It can be used to evaluate the health-related procrastination behavior of clinical nurses and provide a scientific and effective evaluation tool for the study of nurses’ health-related procrastination in China in the future.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Abbreviations

NHRPS

Nurses’ Health-Related Procrastination Scale

EFA

Exploratory factor analysis

CFA

Confirmatory factor analysis

CR

Critical ratio

CMIN/DF

Chi-square/degree of freedom

RMSEA

Root mean square error of approximation

CFI

Comparative fit index

IFI

Incremental fit index

TLI

Tucker Lewis index

I-CVI

Item-Level Content Validity Index

S-CVI

Scale-Level Content Validity Index

S-CVI/UA

Scale-Level Content Validity Index/Universal Agreement

S-CVI/Ave

Scale-Level Content Validity Index/Average

ICC

Intraclass Correlation Coefficient

AVE

Average Variance Extracted

CR

Composite Reliability

WHO

World Health Organization

Author contributions

LC and YS are jointly responsible for the design and direction of this study. HLZ and LX completed the data collection. HLZ and LYC completed the data analysis. HLZ wrote the paper and made figures and tables. LC, YS and HLZ worked together in the revision of the paper. All the authors read and approved the final manuscript.

Funding

The authors declare that they have no competing interests in this section.

Data availability

The datasets used and/or analysed during the current study are available from the corresponding author upon reasonable request.

Declarations

Ethics approval and consent to participate

This study was approved by the human Trial Ethics Review Committee of Shanghai General Hospital (Ethics approval number: 2025–038) (see Supplementary Material 4). Informed consent to participate was obtained from all of the participants in the study. The study participants were voluntary and anonymous. During the study, they could decide to withdraw at any time without any influence. All methods were carried out in accordance with the principles of the Declaration of Helsinki. Clinical trial number: not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.