Study design
This single-blind randomized clinical trial with parallel group’s design was performed in Imam Reza (AS) hospital in Kermanshah, IRAN from 04 February 2019 (first assessment) to 20 March 2019 (last assessment). After obtaining permission from the ethics committee of Kermanshah University of Medical Sciences (Reference number: IR.KUMS.REC.1397.815) and registration in the Iranian Registry of Clinical Trials (IRCT) database (Reference number: IRCT20180408039227N1 on 03/02/2019), all methods were performed in accordance with the Declaration of Helsinki. Furthermore, the trial was reported based on the Guidelines for Consolidated Standards of Reporting Trials (CONSORT).
Participants
Among the nurse staff and nurse aides working in the operating room of Imam Reza Hospital, forty-four participants (30 women, 14 men, mean age: 37.86 ± 6.74) with chronic non-specific LBP were selected based on their history and examination by an experienced physiotherapist. The inclusion criteria were having at least 3 months of LBP, no history of lumbar surgery or spinal fractures, pregnancy, rheumatic diseases, history of cancer, pacemaker, severe deformity, no congenital abnormalities of the spine and anti-inflammatory and analgesic drug used during the study, In addition, unwillingness to continue to participate in the study was considered as an exclusion criterion. All participants gave written informed consent before entering the trial. Using permuted block randomization method with a block size of four, they were randomly assigned to two groups: back exercises (n = 22), or TENS (n = 22) groups (Fig. 1). In order to control the potential selection bias, the random allocation sequence was generated by a person who was not involved in the enrollment or screening of participants. The statistician who analyzed the data was blind to the group allocation of participants. Because of the nature of interventions (using an instrument in one group), blinding the participants was not possible. In addition, the researchers that provided the treatments for participants were not blind, because they were involved in the intervention process and data collection.
Based on the reduction in the pain scores according to a previous study (Mean difference (95% CI) = − 1.2 (− 2 to − 0.4) and − 0.1 (− 0.1 to 0.7), respectively for TENS group and control group), the required sample size was estimated. A 95% confidence level and a power of 80% to be 22 for each group were considered [16].
Interventions
After selecting the samples, participants were informed about this study, how to perform the intervention, the duration and the place of the study. The training of the samples was done in the first session. The length of times of applying the two types of intervention, including exercises and TENS were equal. It was 15 min. The interventions were performed three sessions every week for 6 weeks in the workplace of the participants (in the restroom of the hospital operating room ward), so there was no need to leave the work.
Also, the back exercises group performed exercises for strengthening and stretching of the back and pelvis muscles which included 1) Pelvic tilt 2) Single knee to chest 3) double knee to chest 4) hip flexor stretches 5) squat (Fig. 2) [17]. In each set, these exercises were repeated 10–12 times. They did as many sets of exercise as up to 15 min.
In the TENS group, the Trio 300 dual-channel TENS device from ITO. Co was used. The electrodes of the device were placed in the lower back region. A frequency of 100 Hz, duration of 0.2 ms and an intensity that increased as the participants felt a comfortable prickling sensation (about 15 mA) was applied [18].
Primary outcome measures
Before the intervention and at the end of the sixth week (Immediately after the intervention), the McGill pain questionnaire and the Oswestry questionnaire were completed to assess the pain intensity and performance of the participants, respectively.
Pain
The pain was assessed by the short-form of McGill Pain Questionnaire (MPQ). This questionnaire has 20 items, and its purpose is to assess the subject’s perception of pain in different aspects (four aspects of sensory perception of pain, affective emotional perception, evaluative perception and perception of diverse pain (Miscellaneous)) [19]. Internal consistency was found by Cronbach’s alpha to be 0.76 for total scores. Also, the reliability of this questionnaire has been assessed and confirmed in previous studies (intraclass correlation coefficient (ICC) for total SF-MPQ ranged from 0.90 to 0.94 and Cronbach’s alpha ranged from 0.84 to 0.91) [20,21,22].
Disability
To assess the disability, the Persian version of the Oswestry Disability Index (ODI) was administered. It is a scale with a range from 0 to 100 describing the level of disability about pain intensity, sexual function, sleep quality, and level of performance in daily activities such as sit, walk, lift, stand, work, and travel. The reliability of the Persian version of this questionnaire has been tested in this current study (Cronbach’s alpha = 0.84) and approved in a study by Mousavi, et al. (Cronbach’s alpha = 0.83; ICC = 0.91) [23].
Statistical analysis
Statistical analyses were performed using the IBM SPSS statistics software (version 23) and for all statistical tests, p < .05 was set as the significant level. Mean ± SD and frequency (percentage) were used to express the continuous and categorical variables, respectively. According to the results of the Shapiro-Wilk test, all variables of the study had a normal distribution. Differences in general characteristics of the participants between the two groups were analyzed using the independent t-test and chi-square test.
A mixed ANOVA test was conducted as follows: (1) the main effect of groups (back exercises vs. TENS) was determined by any conspicuous difference in pain and disability observed between two groups; (2) the main effect of time (pre vs. post) was determined by any conspicuous difference in pain and disability observed amidst the two different time points; (3) and the time × group (interaction) effect.
In this regard, to assess the time effect within each group separate repeated measure ANOVA tests were performed. Thereafter, a 2 × 2 (time × Group) ANOVA with repeated measures on time was conducted to assess potential interaction. Also, Cohen’s d effect sizes were determined by dividing the difference between the means by the standard deviation. The values of 0.2, 0.5, and 0.8 were regarded as small, moderate, and large effect sizes, respectively [24]. P-values less than 0.05 were considered statistically significant for all of these tests.