2.1 Study design
We conducted an assessor-blinded, parallel-group (1:1 allocation ratio) randomized controlled trial. Data were collected at the Sport Rehabilitation Laboratory. The study’s protocols were approved by the research ethics committee of the Medical Sciences (Registration no.: IR.UMSHA.REC.1395.470). The protocols were also registered (Registration no.: IRCT20170615034554N2). The subjects provided informed consent forms.
2.2 Participants
Community participants were recruited between Jun 2018 and October 2018. They were screened using the posture assessment grid. RSP and FHP were defined as the anterior displacement of the middle point between the anterior borders of the acromion and tragus in relation to a vertical reference line, and HK was defined as the posterior displacement of the middle point between the thoracic curve apex in relation to a vertical reference line [23]. The inclusion criteria were an FHP >46 degrees; an RSP >52 degrees [24] and kyphosis >42 degrees [25]; pain in the head and neck, shoulder, and spine during laptop or cellphone use; and an age range of 18 to 22 years old [26]. The exclusion criteria were a history of elite exercise; membership in professional sports clubs; a history of fractures, surgery, or joint diseases, especially in the spine, shoulder girdle, and pelvis; skeletal malalignment in the ankle and knee; visual impairment irreparable through glasses; vestibular system impairment; obesity, water allergy, respiratory problems, or cardiovascular disease; and a history of migraines and analgesic usage. The screening identified 34 eligible subjects with concurrent deformities (Fig. 1). G*Power 3.1 software [27], a testing power of 0.85, an effect size of 0.8, and a significance level of 0.05 were used to estimate the minimum sample size, which, for this study, was 12 [27,28]. Allowing for the possibility of dropouts, we selected all the 34 individuals and divided them randomly into experimental (17 males) and control (17 males) groups (Table 1). Out of this sample, four withdrew from participation for various reasons, leaving us with a final sample of 30 (14 in the experimental group and 16 in the control group).
Fig. 1. CONSORT flow diagram (insert here)
2.3 Randomization and Masking
The subjects were randomized using the Random Number Generator, after which they were assigned to experimental and control groups on the basis of allocation concealment via the sequentially numbered, opaque sealed envelopes method by an uninvolved person and stored in a locked location. Another independent person opened the next sequential envelope and informed the therapist of treatment allocation by phone. The assessor was unaware of the exercises and interventions prescribed for the groups, but blinding could not be imposed on the subjects and the statistician with respect to the correctional training and the grouping and their assigned exercises, respectively.
2.4 Interventions
Water-Based Corrective Exercises
Previous research recommended treatment to be performed in three stages: the normalization of peripheral structures (inhibition), the restoration of muscle balance (elongation and activation), and the facilitation of afferent system and sensory motor training (integration) [3]. In the present study, the exercises involved over the course of these steps were carried out in a water environment. In the first stage, trigger points were released through massage in water, and the myofascial release of stiff muscles was induced using a foam roller. In the second stage, static stretching exercises were implemented to relieve muscle tightness. Given the tightness in respiratory accessory muscles owing to concurrent deformities, muscle stretching was combined with corrective exercises related to breathing patterns. Then, strengthening exercises (activation) were performed. Finally, proprioceptive exercises were conducted for neck and shoulder joints. Because the last stage of the interventions was functional (integrative) in nature, the subjects were asked to participate in a game in water while assuming correct posture. The eight-week training program involved three sessions per week, which comprised warm-up exercises (10–15 minutes), water-based corrective exercises (35–45 minutes), and cool down activities (5–10 minutes) (Appendix 1).
2.5 Outcome measures
2.5.1. FHP and RSP Postural Assessment
Imaging, which has been used in numerous studies, has been found to exhibit good reliability [16, 24 and 29]. In the current work, lateral view images were used to evaluate FHP and RSP angles. Three anatomical landmarks (the tragus, C7, and the acromion) were marked (Fig. 2), and a digital camera (Sony Cyber-shot DSC-RX100VI, Japan) mounted on a tripod was placed at a distance of 265 cm from the wall, aligned with the shoulder of a subject. He was then asked to stand in a position with which he was completely comfortable and directed to look at the point (a cross) on the wall (eyes along the horizon). After five seconds, the examiner took three images from the lateral view [16] and inputted them into a computer. AutoCAD 2013 was used to measure the angle of the perpendicular line connecting the tragus to C7 (FHP angle) and the angle between the perpendicular line running from the acromion process to C7 (RSP angle). The means of three angles obtained for each deformity were recorded as the desired angles for FHP and RSP (Fig. 2) [16, 24].
Fig. 2. Calculation of FHP and RSP angles using AutoCAD software (insert here)
2.5.2. Measurement of Thoracic Kyphosis Angle
Different studies showed that a flexible ruler has better sensitivity and validity (r = 0.72) than that exhibited by a radiograph [30, 31]. Accordingly, the present research employed a flexible ruler (24-inch Staedtler Mars, Nurnberg, Germany) in measuring thoracic kyphosis angles. Before measurement was initiated, the T2 and T12 vertebrae were marked. First, the T2 vertebra was identified by asking a subject to flex his head forward, at which position the spinous processes of the C6 and C7 vertebrae could be visibly recognized. The examiner placed two index and middle fingers on these segments, and as the subject returned to a normal posture. The remaining spinous process was thus indicated as relating to C7. Counting was subsequently initiated from the spinous process of C7 downward to determine the location of the T2 vertebra. Second, to identify the spinous process of T12, the examiner touched the lower edge of the 12th rib on two sides using his thumb and then simultaneously moved two fingers on both sides of the body upwards and inwards until the rib disappeared. The distance between the two fingers was connected, and the middle point was marked as the spinous process of T12 [32]. The ruler was placed on T2 and T12 and placed entirely on the spine by hand. The curvature was drawn on A3 paper, and the kyphosis angle was measured using equation (1) (Fig. 3) [33]. The ICC of the flexible ruler was 0.92 [34].
[Due to technical limitations, the formula could not be displayed here. Please see the supplementary files section to access the formula.]
Fig 3. Measurement of kyphosis angle using flexible ruler (insert here)
2.5.3. Assessment of Neck and Shoulder Pain
The intensity of neck and shoulder pain was assessed using a 10 cm visual analog scale (VAS), which presents good reliability and validity (intraclass correlation = 0.92) [35]. Along this scale, 0 indicates the absence of pain, 1 denotes minimal pain, and 10 is equivalent to intolerable pain. The participants were asked about the severity of pain that they experience during daily activities, when they use a mobile phone or laptop for a long time, and when they study for long periods. The intensity of subjective pain was measured before, during, and after the interventions.
2.6 Statistical Analysis
The Shapiro–Wilks test and Levene’s test were carried out to assess the normal distribution of data and the homogeneity of variances. An independent sample t-test was performed to evaluate intergroup differences before the interventions. Repeated measures ANOVA with a 2 × 2 (group × time) mixed design was used to analyze intergroup changes over time. Cohen’s d was adopted in calculating effect size in the pre- and post-test comparison. Values of 0.01, 0.06, and 0.14 were regarded as corresponding to small, moderate, and considerable effects, respectively [36]. The data were analyzed using the SPSS software (version 20), and a P value of 0.05 was considered statistically significant.