Study design
This study was designed as a two-armed cluster-controlled trial with an intervention group and a control group, where the clusters were independent hospital ICUs in China, and the participants consisted of registered nurses in clinical practice. In China, all nursing directors in public hospitals are required to organize an occupational health training program, including comprehensive training and routine specialist training[34]. Comprehensive training involves uniform training of basic theories and skills for nurses by the hospital. Routine specialist training is organized by the head nurse and is conducted in the form of lectures once or twice a year, depending on the occupational risks of the unit. Routine specialist training includes special disease care as well as information and training on occupational health risks[35]. After baseline data were collected (as described below), the participating ICUs were randomly divided into the intervention group and the control group. A two-month multidimensional training program was implemented in the intervention group, whereas only routine specialist training was implemented in the control group. Stratified randomization was not possible because the towns where the participants were located were far apart, and the potential propensity to communicate with each other was present, considering that the participants were working in the same hospital.
This study followed the criteria for the development and evaluation of training interventions for healthcare professions recommended by the Equator Network[36]. No other specific risks were associated with participating in this program other than those associated with the adoption of preventive actions. The Research Ethics Committee of Xiangya Nursing School of Central South University approved this study (2017025).
Setting
The programs for the intervention group and the control group were implemented in ICU rooms for the convenience of the nurses. All ICUs participating in this study had similar levels of health care complexity. There were 2.5–3 nurses employed, on average, for each ICU bed. Each nurse had experience in routine specialist training with similar professional training methods and content.
Recruitment of ICUs
The ICUs had to meet the following criteria to ensure the selection of appropriate clusters and the implementation of the intervention: (1) the ICU was located in a tertiary public hospital; (2) the ICU admitted mixed cases (i.e., ICUs with critically ill patients could have been transferred from any department); (3) commitment and explicit interest to implement the programs and evaluation were shown by the hospital nursing directors, head nurses, and nurses; and (4) the ICU was exposed to significant risks of WRMDs, as assessed by the previous survey (prevalence of WRMDs > 90%)[14].
The units were recruited from ICUs in Hunan Province, China, that participated in the previous cross-sectional survey. Under the aforementioned criteria, four mixed ICUs in four tertiary public hospitals were recruited. An independent researcher randomly assigned the four ICUs to the intervention group (two ICUs) or to the control group (two ICUs) by using the random grouping function in Excel.
Recruitment of participants
During recruitment, the purpose and methods of the intervention study were explained orally to the participants in each cluster before they were assigned to their respective group: the intervention group or the control group. Informed consent and the baseline questionnaire were obtained from each participating ICU nurse. After the questionnaires were completed and returned, the four units were randomized, and the participants were informed of whether they were in the intervention group or the control group. Subsequently, the intervention started. Follow-up questionnaires were administered at 3 and 6 months after the start of the intervention.
Eligibility criteria of participants
To be included in this study, the participants needed to be registered nurses, including nurses who were on sick leave and those who were engaged in patient care daily, and wanted to volunteer. Nurses who were pregnant and performed only administrative work were excluded from this study.
Intervention
Our previous cross-sectional survey identified the following risk factors for WRMDs among ICU nurses in Chinese tertiary hospitals: risk perception, physical factors (frequency of handling patients and physical workload), psychosocial factors (job stress), and workplace environmental factors[14]. Accordingly, a personalized and multidimensional intervention program was designed to reduce WRMD symptoms, as reported by Chinese ICU nurses. The multidimensional intervention program consisted of three components: improvement of risk perception, health behavior training, and promotion of a safe working environment.
The multidimensional intervention program was implemented in the intervention group. The control group received routine specialist training on WRMDs, including two lectures on WRMDs and safe working environments. The schedule of the specific interventions is presented in Figure 1. A working group was organized for each hospital ICU. The intervention group included a head nurse, an ergonomics specialist, an orthopedist, a nurse representative, and a researcher responsible for the development and implementation of the intervention. The control group included a head nurse, a nurse representative, and a researcher responsible for routine specialist training on WRMDs.
Improvement of risk perception of WRMDs
Risk perception plays an important role in preventing occupational risks and can be used as an incentive to promote safe work behaviors[37]. The measures used to improve risk perception were as follows: (1) A lecture on WRMDs consisting of a 40-min course that had been determined by the working group. The lecture covered the type, symptoms, epidemiology, risk factors, and consequences of WRMDs and was presented by clinical nursing experts with considerable years of experience in WRMDs. (2) WRMD Awareness Month. The Health Belief Model holds that individual behavior is influenced by cues that motivate people to change their behavior, such as media reports, advice from others, and reminder brochures[28, 29]. Weeks 5 to 8 of the intervention program were assigned as the WRMD Awareness Month. The working group distributed brochures on WRMDs to nurses. The brochures covered the concepts of WRMDs, influencing factors, protective exercises, and application of the principles of ergonomics. A competition on WRMD knowledge and skills was held during the awareness month. It focused on strengthening the knowledge and skills of nurses in order to prevent WRMDs. Nurses with excellent grades were rewarded.
Health behavior training
Some studies have shown that physical interventions, such as the use of ergonomic aids, training in patient handling, and physical activity, positively affect the reduction of musculoskeletal injuries and pain among nurses[24]. The health behavior training in our study included the following: (1) A 40-min lecture on ergonomics. An ergonomic expert introduced the concept of ergonomics, research content, and related principles and applications of nursing practice. (2) Health behavior guidance. A science-based guidance plan was developed by the working group and implemented in the intervention group. Specifically, the behavior guide included suggestions on how to (i) move and carry bedridden patients, (ii) use slides when moving an awake patient from the bed to a wheelchair as well as carrying and moving objects, (iii) lift items, (iv) walk, squat, and turn around at work; (v) adopt the correct sitting posture; (vi) adjust the chair height for fit and comfort; and (vii) perform physical exercises. Each procedure was demonstrated and guided by ergonomics and clinical nursing experts. (3) Health behavior reinforcement and questioning. The Transtheoretical Model of Change claims that changes in human behavior should undergo consolidation and recurrence[38]. Thus, a weekly session starting at week 5 was organized by the working group to reinforce the health behavior of nurses and to solve problems. At the meeting, the nurses would review correct behaviors via scenario simulation and group discussion.
Providing a safe work environment
Poor perception of a safe working environment represents a stressor that may increase the number of WRMDs reported by employees[39]. Based on the inspection results and the recommendations by the ergonomic expert and the clinical nurse expert, the working group proposed and implemented the following: (1) Lectures on a safe working environment, with each session lasting 40 min. A clinical nursing expert introduced management support for working safety, barriers to work safety, safety awareness training, teamwork, and communication. (2) An improvement plan for an ICU layout. Work chairs (5–7) were replaced with height-adjustable chairs for nurses of different heights, a step stool (30 cm high) was provided to help nurses reach for items at a higher location, and slides (89 cm × 50 cm) were purchased to assist nurses when transferring patients from the bed to a wheelchair. At the same time, we ensured that every nurse mastered how to use the slide.
Routine specialist training in WRMDs
The control group received only routine specialist training, including: (1) lectures on WRMDs and a safe working environment, consisting of two 40-min sessions, which were consistent with the content of the intervention group. The training is updated yearly to meet the requirements for unit development.
Finally, all components of the intervention group and the control group required a leader to integrate, coordinate, and lead the research. The tasks involved communication, organizing meetings, and facilitating the implementation of the intervention.
Measurement
Data were collected using self-reported online questionnaires. The online questionnaire was comprised of two parts. The first part consisted of informed consent. If nurses were willing to participate in the study, they completed the succeeding questionnaire, but if not, the nurse closed the application. The second part was the questionnaire itself. Baseline demographic information was collected, including age, gender, height, weight, marital status, job title, education, and number of years working in the ICU. Baseline and follow-up primary and secondary outcomes were collected, with follow-up conducted at 3 and 6 months.
This online questionnaire was developed using the Chinese-based questionnaire software Sojump (Sojump, Hu Xiao, China). The software generated an online link and a two-dimensional image code. By clicking on the link or scanning the two-dimensional image code, the nurses could enter and complete the questionnaire via the WeChat application on a mobile device. To prevent incomplete and duplicate data, the questionnaire contained mandatory fields and limited submission to one online questionnaire for every WeChat account. All respondents who completed the questionnaire received ¥5.0 (about US$0.71) as an incentive via the WeChat mobile payment red envelope function.
All sample data were exported from the Sojump software to SPSS 19.0 and were double-checked to identify inconsistencies and errors. Data for incomplete studies were not used for statistical analysis.
Primary outcome
Report rate of WRMDs in the past 7 days
The Chinese version of the Nordic Musculoskeletal Questionnaire[40] was used to measure self-perceived symptoms of WRMDs in nine regions of the body during the last year and the past 7 days. A diagram of the body was included to allow nurses to identify the affected areas. No checkmarks or multiple checkmarks were allowed. Baseline and follow-up (3 and 6 months) data were collected.
Secondary outcomes
Risk perception
Risk perception was assessed using the Chinese version of the Risk Perception of Musculoskeletal Injury developed by S. J. Lee et al. (2013) [41]. This tool was translated from English to Chinese by the researcher, with the permission of the author. The respondents estimated the risk of WRMDs as perceived by themselves or by other nurses in their respective units. A six-point Likert scale from 1 (extremely unlikely) to 6 (extremely likely) was used. The score was calculated as the mean of the eight items; the higher the score, the greater the WRMD risk perceived.
Application of health behavior
The application of health behavior was measured using the Nursing Physical Factors Evaluation Questionnaire, including the frequency of patient handling (6 items) and physical workload (9 items). This tool was designed by the author for this study A five-point Likert-type scale from 0 (never) to 4 (very often) was used. The total score of the 15 items was considered as the final score; the higher the score, the greater the ergonomic risk.
Perception of a safe working environment
The Chinese version of the Hospital Safety Climate Questionnaire[42] was used to measure the level of awareness regarding workplace and environmental safety, which was slightly modified to fit the context of this study. All items were answered using a four-point Likert-type scale from 1 (strongly disagree) to 4 (strongly agree). The score was calculated as the sum of the items; the lower the score, the safer the environment was perceived to be. Baseline and follow-up (3 and 6 months) data were collected.
Additional details on the questionnaires are provided in the previous article on WRMDs by the author[14].
Data analysis
The data were statistically analyzed using SPSS 19.0 (IBM, NY, USA). Descriptive statistical analysis was used to summarize the demographic characteristics of the participants. The Student’s t-test and the chi-squared test were used to determine whether a statistical difference existed between the intervention group and the control group at baseline. Analyses of the effectiveness of the primary outcome and the secondary outcomes were performed after intervention for 6 months by using a generalized estimation equation (GEE). The subject variable was the number of nurses, and the internal variable was the time point. The model type (linear regression or binary logistic regression) was selected based on the type of outcome indicator. We first analyzed the single-factor GEE, followed by the multifactor GEE, including the demographic factors affecting the outcomes.