Impact of cervical pain, neck mobility, and body mass index on teachers’ postural control

ABSTRACT Purpose: to analyze the impact of neck pain, neck mobility, and body mass index on teachers’ postural control. Methods: a cross-sectional study with 54 state public school teachers, 68.5% (n = 37) being females, with a mean age of 46.5 ± 9.3 years. Data were collected with the following instruments: Craniocervical Dysfunction Index (Brazilian version), force platform in bipedal and semi-tandem stance, visual analog scale, cervical mobility index, and body mass index. Data were analyzed with nonparametric statistics and multiple linear regression; the significance level was set at p<0.05, with 95% confidence intervals. Results: teachers with neck pain and severely impaired neck mobility had greater postural control changes in the semi-tandem stance. In the bipedal stance, those with mild mobility changes and neck pain had a smaller total displacement. Obese teachers had a smaller movement amplitude in the anteroposterior and mediolateral directions. Conclusion: teachers presented with neck pain and severely impaired neck mobility had a worse postural control. Obese teachers had a smaller total amplitude in both movement directions.


INTRODUCTION
Cervical pain affects approximately 50% of the adult population in some phase of life and is one of the most common musculoskeletal dysfunctions 1,2 .Neck (or cervical) pain helps develop biomechanical changes that impair the cervical spine regarding joint positioning, which in turn affects postural control 3 .Like the visual and vestibular sensory systems, cervical somatosensory systems receive information on posture and send it to cortex regions, triggering efferences necessary to zygapophyseal joints and neck muscles to protect joint movements and maintain postural control 3 .However, when it is not maintained, these alterations of postural control, sometimes manifest as cervical kinesthesia, leading to compensatory body posture because of difficulties positioning the head and eyes to maintain balance 3 .
Posture can be influenced by physiological changes (such as the aging process, chronic diseases, and pharmacological interactions), and by the work setting (as people take wrong postures often due to changes in the work process).Such changes result from social transformations and educational reforms that influence the teaching activity -e.g., cuts in financial resources and an excessive commitment to quality education and productive results, even when there is mass production and scarcity of material and human resources, devaluing teachers [4][5][6][7] .
Changes imposed on the teachers' work make them have an excessive workload; thus, they remain long hours in the same position, decrease their physical activity, increase sedentarism, and possibly increase obesity 8,9 .An increased body mass index (BMI) changes the mechanics of step and plantar pressure when standing, thus overloading the plantar region, which can cause structural changes in the feet and lower limbs 10,11 .
Such structural changes can change postural control 10,11 , which is mostly assessed with the force platform.Data acquired in this instrument can be visualized in a statokinesigram (with a postural control map in the anteroposterior [AP] direction vs the mediolateral [ML] direction) or a stabilogram (temporal postural control series in each of the said directions).The stabilometric measure commonly used in postural assessment is the center of pressure (COP) 12 .
Few resources are available to assess craniocervical disorders.Hence, the validation of the Craniocervical Dysfunction Index (CDI -Brazilian version) provides health professionals with a detailed and early investigation of such disorder, assessing cervical pain, muscle tension, and joint sounds during cervical spine movements, the amplitude of movement, pain during movement, and craniocervical posture 13 .They can be related to metabolic changes, interfering with BMI or functional structures, causing postural instability.
Teachers' quality of life can be directly affected by the impact of musculoskeletal dysfunction of the cervical spine, increasing obesity, and postural control changes.Thus, it is important to assess cervical pain and mobility and the interference of obesity with the postural control of elementary and middle school teachers.Hence, this study aimed to analyze the impact of cervical pain, neck mobility, and BMI on teachers' postural control.

METHODS
This cross-sectional study is part of a research project named PRÓ-MESTRE -Health, Lifestyle, and Work of Teachers in the Public Health Network of Londrina.Its objective is to assess the health, lifestyle, and work of state public school teachers 14,15 .The project was approved by the Human Research Ethics Committee of the Universidade Estadual de Londrina, Brazil (protocol no.45285015.1.0000.0108)(CAAE: 33857114.4.0000.5231).All participants were initially informed of the research objectives and procedures and signed an informed consent form.
The broad research project was conducted in three stages -the present study belongs to the third one.Data were collected between September 2015 and June 2017.The teachers were invited to participate in auditory, vestibular, vocal, postural control, and cervical mobility assessments, as well as related factors, such as physical activity and metabolic and circulatory changes.
The inclusion criteria were as follows: teachers of both sexes; aged 18 to 60 years; working in public elementary and high schools, responsible for at least one school subject; teaching for more than 12 months; who were not on a leave of absence of more than 30 days in the last 12 months.The exclusion criteria were the following: having physical or sensory limitations that prevented them from verifying postural control and taking the balance tests (e.g., inability to understand and respond to simple verbal commands and/or incapacity to use required postures); severely impaired visual and/ or hearing acuity and incapacity to perform activities of daily living; orthopedic disorders, limited movements, or lower limbs prostheses; self-reported central vestibular dysfunction; self-reported consumption of alcoholic beverages within 24 hours of the assessment or drugs that affected the central nervous system (such as tranquilizers and antidepressants) or the vestibular system within 48 hours of the assessment; patients who underwent vestibular rehabilitation after medical discharge.
Initially, 88 teachers were invited and agreed to be assessed.However, 10 participants were excluded for being above 60 years old; 11 for being on a leave of absence for more than 30 days in the previous year; three for reporting alcohol ingestion in the previous 24 hours; one for having severe visual impairment; one for having a severely diminished eyesight (who was reportedly waiting for a cornea transplant); and six for not attending assessments after having scheduled them three times to different days and hours.Thus, 54 teachers were included in the cervical mobility, cervical pain, BMI, and postural control assessments.
The statistical power of the sample was calculated based on a post hoc test, using GPower software 3.1.9.2 for Windows 16 .The means and standard deviations of total displacement in the semi-tandem stance were used regarding the groups without cervical pain (9.44 ± 10.8; n = 35) and with cervical pain (19.7 ± 16.5; n = 19) in a two-tailed test, with α = 0.05.The resulting statistical power was 70%.

Clinical information
The participants' clinical information necessary to the research included age, sex, dizziness, tinnitus, and so forth, according to Miller's medical history protocol 17 .Routine audiological assessments were conducted at the Department of Audiology at the Speech-Language-Hearing Clinic (UNOPAR).Moreover, cervical pain was assessed with a visual analog scale (VAS), CDI (Brazilian version), cervical mobility, BMI, and postural control.

Postural control assessment
Postural control was assessed with the ground vertical force of reaction on a force platform (BIOMEC400, EMG System of Brazil, SP, Brazil), sample at 100 Hz.All force signals were filtered with a second-order 35-Hz low-pass Butterworth filter and converted into COP data, using MATLAB routines (The MathWorks, Natick, MA, USA) 16 .The following data on postural control parameters were extracted: COP ellipse area (95% confidence) (A-COP, in cm 2 ), COP velocity (VEL) in both movement directions (AP and ML, in cm/s) 17 , total displacement (D-TOTAL, in cm), and amplitude of displacement (AMP, in cm) in both directions [18][19][20] .
Participants first got familiarized with the equipment and protocol until they felt comfortable with the test.Balance was assessed with a standardized protocol: barefoot, arms by their sides or parallel to the trunk (PA), in bipedal stance 21 ; and in semi-tandem (ST) on the platform, the right ankle 2.5 cm away from the posterior hallux 22 .The two stances (PA and ST) were assessed on a rigid surface and then on a foam surface.
The test used the experimental protocol with eyes open, asking participants to fix their eyes on a target (black cross, 14.5 cm high x 14.5 cm across x 4 cm deep) on the wall, placed 2 m away from them at eye level 21 .Three 30-second attempts were made, with 30-second intervals between them 19 .The mean of the three measures was used for subsequent analysis 21 , which was based on the difference between the means obtained on foam and rigid surfaces, in both stances (PA and ST).

Cervical Mobility Index and CDI (Brazilian version)
The cervical mobility index corresponds to item "A" in CDI (Brazilian version), which is based on the Cervical Mobility Index described by Wallace and Klineberg 6 .The amplitude of cervical spine movement was assessed with a flexometer manufactured by Sanny.During the assessment, it was verified whether they had any joint sounds, which may help the evaluator to identify functional impairments or joint degeneration.Each movement (flexion, extension, rotation, and lateral flexion) scored 0, 1, or 5 (respectively normal, mild, or severe), according to the patient's active amplitude of movement 6 .
The pain was assessed during neck flexion, extension, rotation, and lateral inclination.Muscle pain was analyzed by palpating the trapezius, sternocleidomastoid, occipital, masseter, and mandible muscles.

Statistical analysis
Data were statistically analyzed with the Statistical Package for the Social Sciences, version 20.0 (SPSS, UK), with a 95% confidence index and 5% significance level (p < 0.05) in all tests.The parametric distribution of data was verified with the Shapiro-Wilk test; without the assumption of normality, the Mann-Whitney and Kruskal-Wallis tests with Dunn post hoc were used for the continuous variables.The effect size of nonparametric tests was also calculated.Mann-Whitney used the following equation: r = Z / √ n, in which "r" is the correlation coefficient, "Z" is the standardized U value, and "n" is the number of observations 25 .Kruskal-Wallis used the estimated square epsilon (E r 2 ) with the equation: " is the coefficient, whose value ranges from 0 (no relationship) to 1 (perfect relationship); "H" is the valued obtained in Kruskal-Wallis, and "n" is the number of observations [26][27] .The effect sizes followed the classification by Cohen (1988) 7 .The Spearman correlation test analyzed the correlation between COP variables and VAS, Mobility Index, and BMI.The Spearman correlation was classified as follows: values below 0.4 were considered weak, and between 0.4 and 0.5 were considered moderate 28 .The chi-square test analyzed the association between categorical variables.A multiple linear regression analysis was performed to verify possible predictors for each COP variable.It included VAS cervical score, cervical pain, and BMI; BMI classification and mobility were transformed into dummy variables to enter the model.Index difference in D-TOTAL in the semi-tandem stance (p = 0.037; E r 2 = 0.12) (Table 2); the Dunn post hoc showed that severe mobility difficulties in the semitandem stance were worse in the assessment process (severe vs normal: adjusted p = 0.024; severe vs mild: adjusted p = 0.017).The analysis of the cervical mobility index subgroup found a difference in the mild group regarding A-COP in the bipedal stance (p = 0.037; r = 0.36); the cervical pain group had a smaller A-COP in this stance (Table 3).CDI data indicated a significant difference and a moderate effect size in the mild dysfunction group regarding VEL ML (p = 0.028; r = 0.4) and D-TOTAL (p = 0.002; r = 0.5) in the semi-tandem stance.Those who reported cervical pain had higher values in these COP variables (Table 4).
The comparison between groups with and without cervical pain indicated a D-TOTAL difference in the semi-tandem stance (p = 0.008; r = 0.3).Those who reported cervical pain had greater postural sway (Table 1).The Kruskal-Wallis test found a Mobility   group had a smaller amplitude in both directions.The chi-square test found no association between cervical pain and BMI (p > 0.05).
Concerning BMI, a difference with a small effect size was found in the obesity group in bipedal stance regarding AMP-AP (p = 0.003; E r 2 = 0.02) and AMP-ML (p = 0.030; E r 2 = 0.01), as shown in Table 5.The obesity The data also indicate a weak correlation in semitandem stance between VAS and D-TOTAL (r s: 0.367; p = 0.007) and between BMI and D-TOTAL (r s: 0.304; p = 0.027).In other words, as pain intensity or movement difficulties increase, total displacement also increases.There was also a weak correlation in the bipedal stance regarding AMP-AP (r s: -0.299; p = 0.033) and AMP-ML (r s: -0.340; p = 0.015).That is, in this case, as BMI increases, the amplitude in both directions decreases.There was no correlation between BMI and VAS for cervical pain.
Concerning multiple linear regression results after the model application was regressed, the bipedal stance variables were not considered predictors of A-COP, VEL AP, VEL ML, or D-TOTAL.On the other hand, BMI, obesity, and overweight were predictors of AMP-AP, while only obesity was a predictor of AMP-ML.In the semi-tandem stance, no predictors of VEL AP, VEL ML, or AMP-ML were found.On the other hand, severe mobility and obesity were predictors of A-COP, while cervical pain, severe mobility, and obesity were predictors of D-TOTAL.These data are shown in Table 6.

DISCUSSION
Studies demonstrate that cervical pain increasingly impacts the socioeconomic life of the general population, particularly the professionals addressed in this study.It can have a disabling effect on them, due to changes in cervical spine mobility and postural control, thus decreasing physical activity, and possibly increasing BMI 1,2,4,6,[30][31][32] .Most teachers assessed in this study were females (68.5%), 35.2% of the sample reported cervical pain, 20.4% had mobility difficulties, and 30.8% were obese.As in this study, there is a consensus in the literature that head and neck pain predominates in women.Even though such pain did not predominate in some studies analyzed, it is among those that most impair performance, possibly increasing the risk of occupational diseases 22,23,[32][33][34][35] .
Cervical dysfunctions can change the proprioceptive and somatosensory systems, impairing neuromuscular and postural control, and thus causing imbalance 36 .This was demonstrated in the present study, as teachers who had cervical pain also had greater D-TOTAL than those without such pain, corroborating the study by Soares et al.The postural control assessment indicated a greater displacement in the cervical pain group 37 .
Given the complexity of postural control, cervical pain affects the whole body.It can impair the function of deep muscles, such as joint and tendon mechanoreceptors, interfering with cervical spine mobility and stability through changes in the information sent to the central nervous system and the vestibular and visual systems 38 .These information changes interfere with postural control, causing the person with pain to increasingly lose postural control -which also agrees with the data in this study.
Postural control depends on both internal and external factors, and the capacity to maintain postural control depends on genetic factors, age, positioning of the center of mass, flexibility, and visual control 39 .The foam used in the assessments is an external factor that affects postural control, disturbing the posture and requiring neutralizing reactions to the stimuli.Not only foam but also other internal and external disturbances affect balance, forcing the body and its systems to master postural control 40 .
Besides the surfaces, the bipedal and semi-tandem stances challenge individuals in postural control assessment.The absence of D-total differences in the bipedal stance may be due to the larger base of support, as the less demanding task ensures functional adaptations that diminish postural sway.In its turn, the semi-tandem stance poses a greater challenge: since the base of support is smaller, body instability increases, and so does the postural imbalance.Hence, standing on foam in a semi-tandem stance requires attention.Continuous muscle action may interfere with this task, and the metabolic level may increase, likewise increasing body sway and specific modulations of the neuromotor system to regulate postural control 40,41 .
Thus, this study showed that cervical pain and limited cervical spine mobility, associated with a smaller base of support in the semi-tandem stance (which increases task demand), increased body sway in the group with severe mobility difficulties, in contrast with the bipedal stance, which is less difficult and more stable.
Furthermore, the analysis of A-COP in the group with mild cervical mobility difficulties showed they had a smaller area in the bipedal stance.This agrees with position analysis and demonstrates better performance in more stable stances, with a larger base of support, and milder mobility difficulties -unlike the subgroups with medium and severe cervical spine mobility difficulties in the semi-tandem stance 20,42 .
The analysis of the subgroup with mild mobility difficulties indicates better performance in the position with a lower degree of lesion of the neck proprioceptor stimuli and less interference of disturbances with cervical sensorimotor control.This leads to the hypothesis that the person had a mild dysfunction, without intense constant pain or the perception of pain.Participants seldom reported this finding; instead, it was verified in the detailed CDI assessment.Structural and functional cervical impairment and mechanoreceptor activation may be absent or diminished -thus, the reflex of these changes on postural control may likewise be reduced, decreasing postural control.
Few studies have addressed postural control in obese teachers.Nonetheless, this is a greatly important topic because teachers have been working harder, often seated, which limits periodical physical activity.After the COVID-19 pandemic, teachers are more sedentary, as the working hours are often longer than the established limits either because they had to be in social isolation or had to readapt to new technologies and telework.This limited their physical activity, social interaction, and nutrition and increased postural impairment due to the lack of an adequate work setting.These circumstances increase obesity, also affecting posture and postural control [43][44][45][46] .
Most studies demonstrate a greater postural control imbalance in the obese population, as increased body mass changes the weight load on the forefoot and hindfoot, changing the person's balance.Almost all participants were assessed with open and closed eyes, which demonstrates the importance of sight as a source of proprioception in balance control 47,48 .However, the study by Sasaki et al. reported no change in postural dynamics with eyes open or closed, suggesting that the absence of sight did not have any effect on postural control 49 .
High body mass increases plantar shear strength in standing posture.Despite the predominance of females in the assessments in the present study, the results agree with another one conducted on males, whose results did not indicate changes in postural control.This may have happened because such participants adapted more easily and flexibly to acute external stimuli on static posture 50 .
Data in the present study differ from those found by other authors.One of the main problems of postural control regulation is maintaining the center of mass on the foot triangle, which is done by the COP, as pressure on the heel changes.However, when the ankle and foot joint sway, neural responses are adjusted in the calf muscles, calcaneal tendon, toe flexors, and opposed hip and trunk muscles to maintain balance 10,11 .Hence, smaller AP and ML amplitudes found in this study concerning BMI in the bipedal stance lead to the analysis of balance in relation to the base of support.
Increased body mass alone does not fully explain postural sway.On the other hand, a high BMI increases not only the plantar region in both AP and ML directions but also deformities in medial sectional and longitudinal plantar arches.Tactile perception also increases due to high plantar pressure, which suggests that mechanoreceptor sensitivity decreases as sensory feedback increases.Another factor is that young individuals have greater activation sensitivity in the neuromuscular system and can easily detect changes in stimuli and thus adapt the functions with compensations.Moreover, obese individuals may concentrate more to keep postural control during the assessment, using the proprioceptive, visual, and vestibular systems.Hence, they compensate for the decreased sensitivity of plantar mechanoreceptors, preventing the decrease in standing postural control 50 .
When the base of support increases, so does the area of mass in which the body can move without losing balance.In the case of overweight, the mass in joint regions (including the ankle) increases, expanding the base of support.Thus, the muscles involved, whether weak or strong, are recruited to neutralize external forces.As the increased base of support, muscle and joint structures, and systems involved in balance (including sight) readapt, obese individuals have body adaptations that help maintain postural control 47 .
Another hypothesis that agrees with the results of this study is that postural control changes may be different in static assessment.Therefore, assessments must include dynamic tasks, such as gait, to analyze whether the adaptation to external stimuli interferes with postural control in the obese population 50 .This topic is little discussed, though greatly important to teachers' health, especially because they are the basis of all education.All kinds of professionals who are in the learning process need special attention not to impair the quality of their education.Regarding this study, some limitations must be pointed out.Postural control assessments were performed with eyes open, but other sensory conditions might have resulted in different findings -although the participants' occupational activities are carried out with eyes open.Another limitation was the scarcity of studies reporting teachers' postural control variables.

CONCLUSION
In the semi-tandem stance, teachers with cervical pain and severe neck mobility impairment had a greater postural imbalance.BMI was not directly related to cervical pain.CDI (Brazilian version) was compared with cervical pain and COP variables, indicating a greater displacement velocity in the semi-tandem stance.Further studies addressing obese teachers and cervical pain are needed to confirm these findings.

Figure 1 .
Figure 1.Anatomical reference points in the SAPO protocol

Table 3 .
Analysis of subgroups of the cervical mobility index and their relationship with cervical pain and variables of the center of pressure Captions: COP (center of pressure); n (number of participants); † (p-value with the Mann-Whitney test); a (median and interquartile range [Q 1 -Q 3 ]); A-COP (area of the center of pressure); VEL AP (anteroposterior velocity); VEL ML (mediolateral velocity); D-TOTAL (total displacement); AMP-AP (anteroposterior amplitude); AMP-ML (mediolateral amplitude); *(statistically significant difference).

Table 5 .
Comparison between categories of body mass index and variables of the center of pressure Captions: COP (center of pressure); n (number of participants); a (median and interquartile range [Q 1 -Q 3 ]); A-COP (area of the center of pressure); VEL AP (anteroposterior velocity); VEL ML (mediolateral velocity); D-TOTAL (total displacement); AMP-AP (anteroposterior amplitude); AMP-ML (mediolateral amplitude); *(statistically significant difference)

Table 6 .
Analysis of the multiple linear regression for variables of the center of pressure