The social media intervention for lower back pain education study (SMILE): a protocol for a randomized trial to reduce occupational low back pain in nursing professionals


 Background Low back pain in nursing profession remains an important health issue. This study aims to assess whether a theory based interactive social media intervention could change job-related risk behaviors among nursing personnel. Methods This is a community-randomized controlled trial. Three settings of hospitals will randomly be assigned to the study arms: Intervention setting 1 receiving an in-person educational package; intervention setting 2 receiving the same material a web-based; and a control setting receiving nothing. The intervention would contain materials related to occupational risk behaviors and will try to promote the knowledge, attitude, skill, and self-efficacy of nurses on low back pain preventive behaviors. The primary outcome will be reduction in low back pain. The study also will assess whether the intervention could reduce pain-related disability and improve the quality of life. Discussion Occupational low back pain among nurses could adversely affect both patients and nurses’ health. This study is an attempt to see if social media could play a role to guide the development and implementation of future interventions.

Musculoskeletal disorders account for 7% of all diseases in the community; of which low back pain is one of the most prevalent disorders (4). Low back pain (LBP) is a common, recurrent and costly health problem worldwide (5). Low back pain affects between 51 and 90% of people at some point during their lifetime (6,7). It has been evidenced that low back pain is one of the main reasons for seeking medical care (8,9). It causes a massive medical and economic load on individuals, families, communities, industry, and governments (10)(11)(12).
The prevalence of low back pain is higher in some professions than in others (13).
For instance, nursing professions are particularly experiencing a higher risk of low back pain and it accounts for 60% of the reported occupational disorders this population (14)(15)(16). Nursing has been recognized between the top professions at risk of low back pain (17), with low back pain rates exceeding those employed in heavy industry (18,19). The year prevalence of low back pain in nurses has a mean of 70% (20)(21)(22) and the lifetime prevalence ranges from 35 to 80% (17,(23)(24)(25).
Recurrence rates of low back pain in nurses exceed 70% (26).
Low back pain in the nursing profession occur for several reasons including the nature of nursing care, patient mobility, long-standing, rotational movements, and repeated bending (27,28). Even those who are working in emergency departments and intensive-care units are likely to suffer more from low back pain due to their specific working conditions (27). The impact of low back pain for nurses is large and includes work absenteeism, increased risk of chronicity, associated personal and economic costs, reduced nursing workforce efficiency, disability and decreased quality of life (17).
There are myriad factors have been identified for low back pain. Biomechanical risk factors such as physical load, force, frequency, vibration, bending, twisting, lifting, pushing/pulling, carrying, heavy physical work, posture. Psychosocial risk factors may affect a workers' psychological response to their work and influence the risk of low back disorders. For example, the mental workload, job stress, job satisfaction, social relations, job security, job demands, organizational level. Individual or personal risk factors such as age, gender, Body Mass Index (BMI), family history, genetics, smoking, physical activity, work experience (29). Although several factors play a role for low back pain in nurses, one of the most important reasons for low back pain due to occupation among nurses are behavioral factors (30,31). In other words, it is argued that if nursing professions could take care of their behaviors during working hours then it would be possible to reduce or lessen their pain and suffering.
It is argued that the main barriers to nurse education are time constraint, shortness of classrooms in hospitals, several job commitments, and the costs. Indeed, to overcome these limitations we decided to use an interactive social media intervention. The use of social media interventions is increasingly becoming popular in public health and a number of studies showed that they were promising platform for promoting healthy behaviors especially when they were theory driven. (32)(33)(34). Several studies indicate interventions that were strongly based in theory had a greater impact than those that were not (35). Thus, we decided to indicate behavioral factors that cause low back pain among nurses, design and develop an appropriate intervention based on the PRECEDE model (Predisposing, Reinforcing and Enabling Constructs in Educational/Environmental Diagnosis and Evaluation), and finally implement the intervention.
The PRECEDE is part of the PRECEDE-PROCEED (Policy, Regulatory and Organizational Constructs in Educational and Environmental Development) model and it is the most popular and commonly used health education planning model (36). As shown in Figure 1, a behavior can be influenced by predisposing, reinforcing, and enabling factors and thus an educational intervention or program based on the PRECEDE model seeks to identify these three factors and then if necessary make changes to predisposing factors (including knowledge, attitudes, beliefs and values), reinforcing factors (including attitudes and behaviors among those who are involved), and enabling factors (including access to resources, availability of health services, policies and legislation, and existing regulations, and behavioral skills that affect the adoption of a health behavior). Therefore, the reason for using this model to develop an interactive social media intervention for the prevention of low back pain is mostly comes from the multidimensional nature of job-related low back pain. In fact, this study tries to find out predisposing, reinforcing and enabling factors that can be applied to a program in the workplace in order to reduce low back pain among nursing professions. Another important reason for using the PRECEDE model is the unique ability to use it in designing and implementing educational interventions in the work environment. This is why the model has been the foundation of many health education and health interventions in the past three decades (36).
Several studies exist that examine the effectiveness of different educational interventions in this population (37)(38)(39)(40)(41). For instance, a study reported that stretching exercises in nurses with low back pain resulted in significant lower pain scores at follow-up compared to the control group (41). However, a recent systematic review of the literature on the efficacy of interventions for low back pain in nurses concluded that 'at present there is no strong evidence of efficacy for any intervention in preventing or treating low back pain in nurses. The authors recommended that it might be worth exploring high quality individualized interventions (17). Yet, many health behavior change websites are not theory driven and fail to incorporate proven, evidence-based approaches. A study by Evers et al. (2003) found that of 37 public health behavior change sites, few were theory driven or used evidence-based approaches (42).
The overall aim of this study is to develop and evaluate a theory based interactive social media intervention in order to reduce occupational low back pain in nurses working in teaching hospitals.

Study overview
This study consists of three phases. An overview of the different phases containing aims, methods and participants are depicted in Table 1.

Aim
The overall aim of this study is to develop and evaluate a theory based interactive social media intervention in order to reduce occupational low back pain in nurses working in teaching hospitals.

Trial design
This is a community-randomized controlled trial. The study will conduct in three hospitals. First, we will provide a list of all hospitals, and three hospitals will select randomly. Then selected hospitals allocated randomly by Roll of a dice to two intervention settings and one control setting. The intervention settings will receive educational programs while the control setting will receive nothing. Participants will 7 be assessed at three points in time: at the baseline, three and six months follow up.
The study procedure from enrollment through follow up data collection and analysis are shown in Figure 2.

Study setting
The study setting will be hospitals affiliated to Mazandaran University of Medical

Intervention
The main aim of this study was low back pain reduction. Thus, we will identify the individual and environmental factors affecting the back pain. For recognizing risk factors, and effective factors in promoting the health of the low back and design intervention, we will use the PRECEDE-PROCEED model. The findings of this phase will be obtained through semi-structured interviews.
On based the Educational and ecological assessment phase, we can determine factors that, if modified, would be most likely to result in behavior change and to sustain this change process. These factors are generally classified as predisposing, enabling, and reinforcing factors (36). According to administrative and policy assessment phase, we will identify resources, organizational barriers and facilitators, and policies for intervention implementation and sustainability (43). This educational material will be evidence-based and elaborated using understandable language and different formats, including pictures, video and 2 -3D animation. The nurses will be able to contact a researcher by email, call and website. Furthermore, to reinforce nurses' motivation and participation, encouraging certificate (for annual evaluation) will implement.
The intervention setting 1 will receive the intervention via in-person education.
They will receive education content in two sessions and 60 minutes per session and through group discussions, role-playing, questions-answers, lectures, educational films and, animations.
The intervention setting 2 will receive the intervention via a web site. They will receive training on how to use the site and they will be monitored by the main investigator. The content of the education will upload to the site in two days and at a specified time, such as the In-person intervention hospital. The content of intervention will be similar to setting one. Both of the intervention settings will receive a weekly reminder during the study period.
The control setting will receive nothing. However, after completion of the study the control group will receive one of the interventions based on their interest.

Primary outcomes
The primary outcome will be, reduced lower back pain. Lower back pain reduction will be assessed by using a VAS questionnaire.

Secondary outcomes
Secondary outcomes will include reduced pain-related disability and increased quality of life.

Data collection
Data collection instrument will the standard questionnaire contained Visual Analog Scale, The Quebec Back Pain Disability Scale, Health Survey SF-36 and self-design questionnaire. The questionnaires will complete by nurses in 3 points time; before intervention, 3 months and 6 months after intervention. The questionnaires will nameless to participation's confidence towards the intervention.
The low back pain visual analog scale (VAS) has been widely used for measuring pain (44). Operationally a VAS is usually a horizontal line, 100 mm in length, anchored by word descriptors at each end, from none to an extreme amount of pain (none, mild, moderate and severe) (45). The validity and reliability of this scale have been repeatedly confirmed (46)(47)(48).
Reduced pain-related disability will be assessed by using the Quebec Back Pain Disability Scale (QBPDS) and increased quality of life will be assessed by using the

Other variables
Other variables also will be measured, including knowledge, attitude, self-efficacy, reinforcing factor, enabling factor and behavior (by self-design questionnaire). The self-designed questionnaire will design of 30 items and six components.
Components will include knowledge, attitude, self-efficacy, reinforcing factor, enabling factor and behavior. Items with the Likert spectrum will be five parts. Its validity and reliability will be calculated.

Sample size
One hundred and eighty nurses working in hospitals affiliated to Mazandaran University of Medical Sciences who are eligible to enter into the study will be recruited. The sample size including 10% drop, 60 people are estimated for each study group in order to detect at least 20 percent differences in the primary outcome between intervention and control groups. As such the study would have a power of 80% at 5% significant level.

Randomization
First, we will provide a list of all hospitals from Mazandaran University of Medical Sciences. The name of hospitals will write on the card and placed in the box. Then the cards will blend and select 3 cards one after another. Then selected hospitals will allocate to intervention and control setting based on Roll of a dice (two hospitals as intervention settings and one hospital as the control setting). That way the numbers 1 and 2 will for the intervention setting 1, numbers 3 and 4 for the intervention setting 2 and the numbers 5 and 6 for the control setting.
In each hospital, participants will select based on the random number table and staff ID. We will label the staff ID. Then, we will select the starting point on the table and will be continued until the number of samples is completed.

Implementation
The researcher will generate the random allocation sequence and will enroll participants, then will assign participants to interventions.

Blinding
This study will be Single blind and participants will not be informed about setting allocation.

Statistical analysis
The quantitative data will be analyzed using SPSS V.23 (and if necessary, AMOS software). Descriptive statistics will include frequencies, means, and standard deviations. The Kolmogorov-Smirnov test will be used to check the normal distribution of data.
Between three groups the primary and secondary outcomes will assess using One Way ANOVA, Post-hoc (LSD) for determining the mean differences between which group. Also, for each group, we can use from ANOVA with a repeated measure for assessing the specific variable in three-time points. Continuous outcomes measured at the baseline and the third month will assess using baseline-adjusted ANCONA in the third month. For each group, continuous outcomes measured in the third month only will assess using paired-sample t-tests. Then the six-month outcomes will analyze and compare with the baseline and the third-month outcomes.
Sensitivity analysis for the primary outcome analysis used linear regression models adjusting for baseline prognostic variables, BMI, working hours, smoking, longstanding, heavy lifting, patient displacement, frequent bending and stress at the workplace.
For analysis assessing the correlation between demographic variables, using Correlation Tests (Pearson for parametric data and Spearman or Kendall for nonparametric data) and Chi-square test.

Ethics
The ethics committee of Tarbiat Modares University approved the study. All participants will ask for permission and completing the informed consent prior to the study commence.

Discussion
One of the features of this study is to pay attention to the promotion of health in the workplace. This study will investigate the efficacy of interventions for the reduction of low back pain in nurses. The main framework of this study will be the theory-driven and evidence-based approach to develop an interactive social media intervention for nurses. Moreover, we will execute the In-person intervention. Then we will compare the result of two education methods. Finally, this study will achieve the result of intervention based on interactive social media according to the workload and multiple shifts in the hospitals.

Strengths and limitations:
Strengths of this study include the community randomized controlled study design and that the study will be designed and implemented according to specific planning.
The interactive social media will provide flexibility and convenience for users, supporting adherence to the program. Also, this study will compare two different educational methods but the same content.
One of the concerns will be that participants will not be assessed by a clinician and LBP will commonly be diagnosed through self-report. A second concern, during the course some participants might use pain relief medications which will affect the outcome of the study.
One of the strengths and weaknesses of the study is the 12-month follow-up that brings two concepts: 1. Will education interventions be maintained long-time?
2. Or in the long-time, its effect will be paled?

Availability of data and materials
The data (when ready) will be available from the corresponding author on request.    SPIRIT_checklist.doc