In this single-blinded randomized controlled trial, 36 female nurses with non-specific CLBP who worked in four university hospitals were recruited as volunteers. This sample was divided into two groups of 18 (intervention and control) by computer-generated random lists. The number of participants in each stage and the number of dropouts are provided in study flow diagram (Fig. 1).

The inclusion criteria were as follows: age of 18–55 years; at least 1 year in the nursing profession; nonspecific CLBP for at least 3 months in the past 6 months; pain between the last rib and lower gluteal fold; no history of spinal trauma or spinal or abdominal surgery; no history of systemic disease (e.g., systemic scleroderma or muscular dystrophy), spinal deformity (e.g., scoliosis and kyphosis), or abdominal wall hernia; and no history of participation in core stability exercises in the past 6 months. Exclusion criteria were any signs of serious spinal cord involvement (such as urine or fecal incontinence, numbness, or limb paralysis) and pregnancy. Nurses were assessed by a sports medicine specialist at the Sports Medicine Research Center. Allocations to each group were prepared before the study: the group names were written and concealed in similar-looking packets; at the beginning of the study, each patient was asked to randomly select a packet and was assigned to the specified group. This study was approved by the Ethics Committee of the Tehran University of Medical Sciences (RCT registration no., IRCT138903314231N1), and all participants received written and verbal information about the study and provided written consent.

The main goal in this exercise protocol was to restore and maintain the stability of the spine by retraining the main stabilizing muscles of the trunk through multi-step progressive stages. Nurses in the intervention group visited the Sports Medicine Research Centre weekly to learn and perform their specific exercises for the week ahead at home. In each session, participants performed the specific exercises under supervision of a physiotherapist and a written pamphlet with illustrations was provided. Because the Swiss ball plays a substantial role in the activation of trunk muscles [15,18,19], for the 8 weeks of training, two floor exercises and two exercises with a Swiss ball were included in each week.

During the first session, the intervention group was educated about the muscles involved in core stability, their influence on LBP, and the effect of core stability exercises in reducing LBP symptoms and recurrence. In addition, the intervention group was taught movements that emphasized on the activation of central core stability muscles. Participants learned the abdominal drawing-in maneuver (ADiM) and activation of lumbar paraspinal muscles (especially LM), as well as how to maintain them in static (supine and quadruped) positions. In subsequent sessions, muscle activation was combined with other functional movements, and participants learned how to retain the maneuver in dynamic positions (while sitting and moving the limbs in a controlled and slow motion on the floor or on a Swiss ball) and in normal physical actions during the day. Each exercise was performed for three sets (morning, mid-day, and night) with ten repetitions and a ten-second holding position in each repetition. Notably, the exercise protocol (number of repetitions and contraction time) was quite flexible based on the physical ability of each nurse. In addition, the core stability movements had a progressive pattern over time, and new and more complicated movements based on the retention of balance and stability were added in each session. The intensity of exercises was increased based on participants’ tolerance, and movements were avoided if they caused pain. The movements performed in the eight sessions are demonstrated in Appendix 1. Throughout the study, the physiotherapist contacted the participants in the middle of each week and encouraged them to complete their exercises; the phone number of the trial’s administrator was given to the participants for contacting in case of any problems. Meanwhile, the control group was kept on a waiting list and did not receive any instruction about an exercise program for their LBP.

The demographic characteristics of all participants were collected at the beginning of the study. The Farsi version of Roland–Morris Disability Questionnaire (RDQ) was used to measure disability in all participants at the beginning of the study and 8 weeks later. The questionnaire comprised 24 questions on daily activities, with scores of 0–24 points reported as total score, and was valid and reliable in patients with LBP. Quality of life was estimated before and after the 8 weeks of the trial using the Farsi version of 36-item Short Form Health Survey (SF-36), which contained 36 items classified into eight domains (including physical function, physical role, bodily pain, general health, vitality, social function, emotional role, and mental health). The score of each question was variable from zero (worse state) to one hundred (ideal state). All eight subscales of SF-36 were used for analysis.

The diameters of lateral abdominal muscles, including external oblique (EO), internal oblique (IO), and TrA, were measured on both sides of participants at rest and during ADiM at the beginning of the study and 8 weeks later. While participants laid in a hook-lying position (hips flexed to almost 30°), a point 25 mm antero-medial to the midpoint between the inferior rib and the iliac crest on the mid-axillary line was set for the linear transducer position [20]. Before taking the muscle diameter measurements, participants were trained to perform the ADiM while their contractions were observed using ultrasound biofeedback [21]. The ultrasound assessor had more than three years of experience in the field of musculoskeletal ultrasound assessment and was unaware of study group allocation.

The thickness of lateral abdominal muscles was measured in B-mode format using a Sonosite Miramaxx (Sonosite Inc., Bothell, WA, USA) ultrasound machine. The linear transducer (6–13 MHz) was transversely positioned at the assessment point, which is described above. In addition, to eliminate the confounding effect of food consumption on the thickness of lateral abdominal muscles, all measurements were performed 4 hours after eating [22,23]. Visual Analog Scale (VAS) was used to evaluate the intensity of pain felt by the participants. In this study, VAS was a 100-mm solid horizontal line, with zero at one end considered as no pain and 100 at the other end considered as the most intense pain. RDQ, SF-36, and VAS were completed by all participants before and after the trial.

The sample size of 18 subjects per group was helpful in detecting a 20-point difference in the VAS between groups, assuming a standard deviation of 11.8, an α level of 0.05, a power of 80%, and a dropout rate of 40% [24,25].

Descriptive data was presented as mean±standard deviation and number (percentage). The paired sample t-test was used to compare the results after 8 weeks in each group, and analysis of covariance (ANCOVA) was used to compare the results between the groups. Analysis was adjusted for age, weight, and baseline value of each variable. The significance level of this study was set at p<0.05, and IBM SPSS ver. 21.0 (IBM Corp., Armonk, NY, USA) was used for statistical analysis. Lastly, our statistician was blinded to the study groups.

In this study, we assessed the effectiveness of a multi-step core stability protocol based on three goals: (1) learning how to activate central core stability muscles and coactivating these muscles during other movements, (2) increasing the intensity of exercises based on tolerance and ability of each patient, and (3) integrating exercises into functional movements, dynamic positions, unstable surfaces, and daily physical motions. In the study by Koumantakis et al. [27], patients with recurrent LBP underwent 8 weeks of trunk muscle stabilization training. Similarly, their protocol included the activation of local stabilizing muscles and integration of this activation with dynamic functional movements. However, the study protocol was not as comprehensive as ours in terms of the application to unstable surfaces, verity, and volume of training. In another study by Norris and Matthews [24], patients with CLBP underwent a 6-week integrated back stability program consisting of optimizing posture, back fitness, and functional exercises. Although their protocol was comprehensive, they did not provide any information regarding application to unstable surfaces or illustration of exact movements. Other studies have also not provided details of exercise protocols or the utilization of unstable surfaces and have conducted only supervised sessions without home exercises [11,28-33]. Variations between these exercise protocols could be an explanation for different results.

The average RDQ score showed that the disability index decreased significantly in the intervention group compared with that in the control group. In addition, the change in RDQ scores (6.1) in the intervention group was >5, which is considered as the minimum clinically important difference (MCID) [25]. Koumantakis et al. [27] found improvement of approximately 4.65 units in RDQ scores in patients with recurrent LBP who performed 20 weeks of stabilization-enhanced exercise. In a separate study, Shaughnessy and Caulfield [31] provided a 10-week core stability exercise program to 41 volunteers with CLBP who had been referred to an orthopedic clinic. In the intervention group, the RDQ score decreased by 5 scores, which was significant. In addition, Bayraktar et al. [16] revealed that patients with CLBP who participated in land- and water-based core stability exercises reported improvement of 5 and 4.5 units in RDQ scores, respectively. As discussed above, most studies have demonstrated the effectiveness of core stability exercises in the amelioration of disability. The possible explanations for these findings are improvement of trunk muscles activity and an increase in the lumbar range of motion, which lead to reduced disability and functional recovery. However, it seems that the discrepancy between levels of improvement could be related to different exercises protocols.

The overall quality of life (SF-36 score) in participants in the intervention group improved significantly compared with that in the control group. Based on the consideration of 30% improvement as a threshold for MCID, most of the SF-36 domain scores meet this threshold. In agreement with our results, Bayraktar et al. [16] showed that healthrelated quality of life, which was measured by Nottingham Health Profile, improved in patients with CLBP due to lumbar disc herniation in both water- and land-based core stability exercise groups. Further, Ota et al. [32] reported improvement in quality of life, based on Japanese Orthopedic Association Back Pain Evaluation Questionnaire, after 3 and 6 months of lumbar stabilization exercises in patients with CLBP. The only factor that was not affected by core stability exercises in the present study was the emotional role. Emotional disorders have mainly multifactorial roots, and in nurses, working factors, such as job stress, health behavior, and work environment, may affect emotional status more than physical complications. However, Shaughnessy and Caulfield [31] and Ota et al. [32] illustrated that core stability exercises improve the emotional aspect of quality of life in patients with CLBP; these findings could be due to a longer study duration and more interactions between patients and their exercise instructors.

The amelioration of LBP was indicated by a significant decrease in the VAS score in the intervention group after the trial, whereas no such difference was observed in the control group. In addition, the VAS mean difference in the intervention group (approximately 30 points from 100) meets the criteria for MCID (15 points) [25]. Koumantakis et al. [27] revealed that in their patients, after 8 weeks of core stability training, mean pain intensity over the past week and past month decreased by approximately 18 points (from 100) and 28 points (from 100), respectively. In addition, Norris et al. demonstrated a 39-point (from 100) reduction in the VAS score, based on McGill pain questionnaire, after 6 weeks of integrated back stability program [24]. Zhang et al. [33] used a randomized controlled trial to evaluate the combination of Chinese massage and core stability exercises in patients with nonspecific LBP and obtained similar results. Their results showed that after eight weeks of intervention, the VAS score significantly decreased (approximately 6 points from 10) in the intervention group compared with that in the control group (approximately 4.5 points from 10). Most of these studies met the criteria for MCID in showing that stabilization exercises could diminish pain intensity. The increase in tissue blood flow during lumbar stabilization exercise in patients with chronic non-specific LBP [34] could be suggested as an integrated mechanism for releasing spasm, improving blood flow, and decreasing the inflammation of local tissues in the lumbar spine, which in return reduced pain.

Another focus of our study was the trunk muscles involved in CLBP. The ultrasound data showed that the thickness of the lateral abdominal muscles (EO, IO, and TrA) in ADiM increased significantly in the intervention group compared with that in the control group. Previous studies have shown an alteration in lateral abdominal muscle activity, particularly TrA activity, in patients with CLBP [9,35]. Because the deep abdominal muscles are used in postural control and lumbopelvic coordination, anticipating their role in the clinical outcomes of LBP seems logical. However, the inconsistency among the findings of previous studies may be related to the multifactorial aspects of etiology of and treatments for LBP [36]. Vasseljen and Fladmark [37] showed no significant changes in resting thickness of lateral abdominal muscles and their contraction thickness ratio during ADiM after 8 weeks of ADiM exercises and their implementation in daily living activities in patients with CLBP. However, Kim et al. [38] demonstrated that increasing the intensity during graded stabilization exercises also significantly increases the activity of the related lumbar stabilizing muscles (including EO), which could be considered for future graded core stability protocols. As they suggested, our exercise protocol intensity progressively increased to match the ability of our participants. In addition, Mannion et al. [20] demonstrated that after 9 weeks of spinal segmental stabilization exercises in patients with CLBP, the voluntarily activation of the TrA increased significantly, but its contraction ratio or recruitment showed no significant relationship with disability scores (clinical outcomes). Lastly, Ferreira et al. [39] demonstrated that patients with CLBP who received 8 weeks of exercise to improve the control of lumbopelvic movements and stability showed a greater improvement in the recruitment of the TrA [39]; this finding is consistent with our study results. In our study, the lack of change in muscle thickness at rest was anticipated due to the short length of the intervention; however, improvement in muscle performance during ADiM was observed after eight weeks of exercise.

There are some clinical implications for the current findings. The multi-step core stability exercise program could be applied to all healthcare workers, especially those with nursing responsibilities in hospitals, clinics, schools, and institutions. Furthermore, nursing students could be educated to use these exercises to prevent LBP during their career.

Several limitations for the present study should be addressed, including the small sample size, notable loss to follow-up, and the short duration of the intervention, any of which may have influenced the results. All study participants in our study were females, which limits the generalization of the results to male nurses. The ultrasonography assessment was only conducted in the hooklying position, although it could have shown significant improvement in functional positions. Lastly, the control group in our study received no intervention and was on a waiting list. Some of them left the waiting list and could not be contacted; however, we considered intention-totreat analysis for who had remained on the list, and they received the multi-step core stability exercise program immediately after the completion of the study.