EFFECTS OF POSTURAL EDUCATION IN ELEMENTARY SCHOOL CHILDREN: A SYSTEMATIC REVIEW

ABSTRACT Objective: To determine the effect of postural education on the learning and postural habits of elementary school children without physical intervention. Methods: We searched PubMed, Lilacs, SciELO, Cochrane, and Science Direct data bases and reference lists of studies in February 2020. The eligibility criteria were randomized clinical trials related to the effect of postural education in children aged between 6 and 12 years old. Two authors independently assessed trials for inclusion and risk of bias: randomization process, deviations from intended interventions, missing outcome data, measurement of the outcome, and selection of the reported result. Data were extracted in standardized tables including information on author, publication year, country, sample size, age, sex, intervention characteristics, outcome measurements and results. Results: We found seven clinical trials (involving 2,568 children) for the review. The studies were conducted between 2000 and 2018: four in Belgium, two in Spain, and one in Germany. All seven included trials underwent evaluation: only one had a clear process of randomization and allocation concealment. All included studies were judged as having high risk of bias in at least one domain or have concerns for multiple domains. Conclusions: The positive effects of acquired knowledge and postural habits found in the studies cannot be used to reliably support postural education in elementary school children due to a high risk of bias in the evaluated studies.


INTRODUCTION
During development, children acquire postural habits that they tend to adhere to for the rest of their lives. 1 According to Noll et al., 2 most students use an inadequate posture to carry out activities such as writing, using the computer, and picking up objects from the ground. Furthermore, children and adolescents frequently suffer from musculoskeletal pain, particularly in the back and neck, 3 but can learn healthy habits which could prevent future pain. 4 Zapater et al. 5 highlighted the higher efficiency of preventive approaches to musculoskeletal problems when the child is in a growth phase and propose further research on educational programs on seated posture in the classroom so that this issue may be effectively addressed. Grors et al. 6 raised important reflections on the incidence of spinal column pain in the population as well as necessary strategies to achieve genuine social change. They also argued that educational initiatives should be directed toward individuals in their formative age, a phase in which attitudes and beliefs are being shaped. They also discussed how strategies, such as public education, social marketing, and intervention policies should be aimed at the child population.
The classroom is among the diverse contributing factors to the manifestation of musculoskeletal symptoms in school children. 7 For instance, the use of school bags, a common practice in elementary school children is a risk factor for musculoskeletal discomfort. 8 Marques et al. 9 discussed the prolonged time children spend sitting, which is a risk factor for lumbar pain. Furthermore, different studies have revealed the existence of shortcomings in the anthropometric measurements of service users and the furniture used in schools. 9,10 As a result, studies have advanced toward detecting the impact of poor posture, both in relation to pain and to postural deficiencies, and as a barrier to concentration and learning. [9][10][11] Health and education professionals play an important role in schools, and given the known risk of children developing inappropriate behaviors and postures as time passes, these can entail a functional compromise. 12 Whereas this is a relatively current issue, 13 the short-, medium-, and long-term effects of postural education strategies for elementary school children are not clear. Thus, the present systematic review aims to evaluate postural education effects relating to acquired knowledge and postural habits in children from 6 to 12 years old.

METHOD
The study was based on the guidelines of the Preferred Reporting Items for Systematic Reviews-PRISMA, but there is no protocol registration. 14 This study included original articles on clinical trials relevant to the effect of postural education in children between 6 and 12 years old (Chart 1). Exclusion criteria were recommendation studies, incomplete texts, duplicated articles, study protocols, pilot studies, and studies classified as quasi-experimental.
In February 2020, we systematically searched five databases: PubMed, Latin American and Caribbean Health Sciences Literature (Lilacs), Scientific Electronic Library Online (SciELO), Cochrane Central Register of Controlled Trials (CENTRAL), and Science Direct. Search terms included (child* OR students OR pediatr*) AND (postu* OR spine OR spinal curvatures) AND (health promotion OR school health services OR educ* OR quality of life) AND (trial). We set no limitations as to language or publication date.
Two authors (PJV and FUC) independently reviewed the titles and abstracts of the identified articles. Subsequently, the complete texts of potentially relevant studies were analyzed, and any disagreements were resolved by a third examiner (JCJN). In addition, we attempted to identify other potentially eligible trials by searching the reference lists of the retrieved included trials (other source). No contact was made with study authors to identify additional studies. The number of articles in each screening stage is shown in Figure 1. EndNote X8.2 was used to manage bibliographic references and visualize duplicated references.
Two reviewers (PJV and FUC) independently evaluated the risk of bias and, if necessary, consulted a third review author (JCJN) for all included studies, in accordance with the recommendations by the Cochrane Collaboration, which recommends using version 2 of the Cochrane risk-of-bias tool for randomized trials (RoB 2). 15 The following items were evaluated: randomization, allocation concealment, blinding of the participant and researchers, blinding of the evaluation, incomplete data, selective publication and other biases ( Figure 2). Data extraction and summary were undertaken according to author, publication year, country, sample size, age, sex, intervention characteristics, outcome measurements and results. More detailed data on intervention were collected, such as the professionals involved, the composition of the postural education program, and additional interventions. For all continuous variables, we extracted sample sizes, means and standard deviations for each intervention and control group. The data were inserted into a spreadsheet in Excel program. However, since the meta-analysis was not possible due to methodological heterogeneity, the data were organized in a form, and we reported findings descriptively.

RESULTS
Of the total of 5,378 studies, 55 were excluded because of duplication, 4,676 were excluded based on the analysis of their titles and 614 were excluded based on abstracts. Of the 33 selected for complete text analysis, 26 were excluded due to the eligibility criteria. As a result, seven studies were chosen for systematic review ( Figure 1).
Regarding the risk of bias, each trial was rated as high risk, unclear risk or low risk on the following domains: 1-randomization process, 2-deviations from intended interventions, 3-missing outcome data, 4-measurement of the outcome, and 5-selection of the reported result. The RoB 2 tool by Cochrane includes overall risk-of-bias judgement. Whereas all the studies had low risk of bias for missing outcome data (domain 3), [16][17][18][19][20][21][22] and four studies had low risk of bias for measurement of the outcome (domain 4), [16][17][18][19] only one 16 had a clear process of randomization and allocation concealment clear (domain 1) and, in addition, all included studies were judged as high risk of bias in at least one domain for their results or were presented some concerns for multiple domains in a way that substantially lowers confidence as to their the result. Figure 2 shows the scores for these studies.
The characterization of the included studies is shown in Table 1. A total of 2,568 elementary school children participated in the clinical trials. The studies were undertaken between 2000 and 2018: four in Belgium, 17,18,21,22 two in Spain, 16,20 and one in Germany. 19 Acquired knowledge was evaluated in five studies, 16,17,19,21,22 all using questionnaires to measure changes related to this outcome, although no questionnaires were the same ( Table 2). Dullien et al. 19 showed a significant short-term improvement in acquired knowledge with postural education. Two studies 16,17 found a significant increase in acquired knowledge by the experimental group both in the short-and medium-term. Cardon et al. 21 showed a significant short-, medium-, and longterm improvement in acquired knowledge by the experimental group, along with an increase in the percentage of correct answers regarding general and specific knowledge in the same group one year post intervention. Cardon et al. 22 found that after postural education, both the postural education group and the group associated with a physical activity program significantly increased their knowledge regarding the care for the spinal column in the short term, with no significant differences between them (Table 3). Postural habits were investigated in six studies: four in practical tests/filmed-movement sessions, [17][18][19]22 and two with questionnaires (Table 2). 20,21 In the study by Cardon et al., 17 the filming was individual, and significant short-and medium-term improvements were found as to the intervention. Cardon et al. 18 also evaluated postural habits using a hidden camera and tasks performed in pairs and showed a significant effect of the intervention with an improvement in scores in the medium and long term. Meanwhile, Dullien et al., 19 using task observation, found that only the experimental group improved their behavior in the water crate-carrying task. Cardon et al. 22 used filming, based on the study by Cardon et al., 18 and found that the group that received postural education and postural education associated with a physical  activity program presented significant behavioral improvements related to spinal care than the control groups. They also found that the group that underwent postural education had a significantly higher score than the group that received postural education associated with a physical activity program in the short term. Cardon et al. 21 evaluation of a questionnaire with items on self-reported behavior showed that the postural education group presented a significant increase in the frequency of checking school bag weight in the medium and long term, in their posture when taking off shoes in the medium term, and in picking things up and carrying them in the short and medium term. Meanwhile, Vidal et al. 20 used a questionnaire on daily postural habits and showed a significant improvement in scores for healthy habits in the short and medium term (Table 3).
Regarding postural education, the interventions varied from one to six sessions. Five studies used six sessions for postural education, of which four reported a one-week interval between sessions and only three reported 60 minutes for each session. There was a lack of information on number and duration of sessions in only one study. 19 In the Belgium studies, 17,18,21,22 the interventions were conducted by physiotherapists. Cardon et al. 22 presented an additional intervention for postural education that included two intervention groups: postural education and postural education associated with promoting physical activity. Meanwhile, Dullien et al. 19 included static and dynamic exercises associated with postural education. The detailed characteristics of the interventions related to postural education are shown in Table 4. As the theoretical basis for structuring educational programs, the authors relied on the literature, 22-26 studies on biomechanics, 27 the German Back School, 28,29 previous studies by the same authors, 17,18,21,30 and referent literature to the "Back Book", 22,33 as well as the cooperation of Orthopedic residents, psychologists, sports scientists, and teachers (Table 1). 19 Table 3 Results of the included studies.

Study Results
Cardon et al. 17 • ↑Knowledge acquired in the immediate post-test and follow-up (p<0.001) • Postural habits: -Better scores in the EG in the immediate post-test period and follow-up (p<0.001) Cardon et al. 18 • Postural habits: -EG presented a higher score after the intervention, after three months and one year for all the items and total score -The increase in the total score for the practical test pre intervention evaluation and evaluation after a one-year follow-up was +1.14 for CG and +26.5 for EG -In the evaluation with the hidden camera, the score was significantly greater in EG (p<0.001) one year after the intervention Cardon et al. 21 • Knowledge acquired: -↑Knowledge acquired in the immediate post-test period and follow-up of three months and one year (p<0.001) -The improvement in general knowledge in the immediate post-test at one year was 33% in EG and 12% in CG; for specific knowledge, 21% in EG and 6% in CG • Postural habits: -Self-reporting of checking schoolbag weight: EG scored higher in the pre-and all post-tests (p<0.001).
-Posture when taking the shoes off: EG scored significantly higher in the post-test at three months, whereas posture when picking things up and carrying them was significantly higher in the immediate post-test and at three months • Knowledge acquired: -EG significantly improved their knowledge; there was a significant interaction between "group" and "test time" (F (1.123)=11.87, p=0.001) • Postural habits: -EG improved their behavior in the water crate-carrying task; there was a significant interaction between the factors "group" and "test time"

DISCUSSION
Health and education services that align, integrate, and collaborate in partnership can improve efficiency, reduce resource consumption, and produce better results. 34 According to the Centers for Disease Control and Prevention (CDC), 35 establishing healthy behaviors in children is more advantageous and easier than trying to change already-established unhealthy habits in adulthood. In this regard, the schools perform a fundamental role. The present review identified interventions with various components adapted to children's age range, and which were tested in randomized trials as options for providing postural education to elementary school children. After analyzing the postural education sessions, all the proposals were adapted to the child population, including active methodology, games, comic books, and characters, among others, and worked on the concepts of biomechanics, the spinal column, and posture. In this regard, Jachyra and Fusco 36 discussed the potential benefits of schools' implementation of play-based learning to children's health and well-being, given that playing is a fundamental right of children and an opportunity for them to be active. In addition, three studies included an information session for parents and teachers. Cardon, de Bourdeaudhuij, and de Clercq 21 emphasized the premise that parents have a fundamental role in shaping their children's health choices.
Lewallen et al., 37 discussing the role of health education for students, provided by qualified and trained teachers, emphasized that health education helps students acquire knowledge, attitudes, and skills necessary for adopting health-enhancing behaviors and for becoming agents of health promotion in their communities. The authors also highlighted that the initiatives and collaborative actions of health professionals, such as nurses, dentists, and physicians are important in addressing school children's actual and potential health problems. In the present review, most professionals involved were physiotherapists who, in the body of the profession's knowledge, undertake in-depth studies into Anatomy, Pathology, and Biomechanics, including a deep and broad understanding of normal movement and impaired function. As a result, these professionals are critical agents in the promotion of health and well-being, who educate individuals and their family members on managing their health conditions to maximize their quality of life. 38 Future concerns remain, such as those related to children's increasing use of computers, but the ergonomic guidelines remain below adult standards, 39 as shown in the study by Howie et al., 40 which found that to minimize potential musculoskeletal and sedentary lifestyle risks, playing with "non-screen" toys should be encouraged, along with education and advice provided to parents and caregivers. Balkó et al. 41 discussed the increase in studies showing a rising trend toward a sedentary lifestyle in elementary schoolchildren and proposed, as a preventive measure, an increase in physical education classes in schools or interaction between state institutions, schools, families, and sports clubs to improve the amount of children's daily activity.
For the main results, the positive effects as to acquiring knowledge and postural habits found in the studies cannot be used to reliably support postural education in elementary school children. The findings were limited by the high risk of bias in the evaluated studies, and the heterogeneity in the research methodologies did not allow meta-analysis of the results. Checking reference lists of included trials was undertaken to minimize the potential source of bias of the search strategy, which may not have retrieved all relevant papers. Another limitation is that there is no protocol registration in the PROSPERO registration record.
Evidence available at the time of writing cannot be used to reliably support postural education in elementary school children, thus reinforcing the importance of researching postural education for school children's health and the role played by professionals in its promotion.

Funding
This study was partly financed by the Brazilian Coordination for the Improvement of Higher Education Personnel (CAPES) -Finance Code 001.