Preventive physiotherapy interventions for back care in children and adolescents: a meta-analysis

Abstract

Background: Preventive interventions improve healthy behaviours and they also increase knowledge regarding back care in children and adolescents, but studies exhibit great variability in their contents, duration and number of sessions, and in the assessment methods. The purpose of this study was to review the empirical evidence regarding preventive physiotherapy interventions for back care in children and adolescents, and to ascertain the most efficacious treatments, in what way and under which circumstances.

Methods: Studies were located from computerized databases (Cochrane Library, Medline, PEDro, Web of Science and IME) and other sources. The search period extended to May 2012. To be included in the meta-analysis, studies had to use physical therapy methodologies of preventive treatment on children and adolescents, and to compare a treatment and a control group. Treatment, participant, methodological, and extrinsic characteristics of the studies were coded. Two researchers independently coded all of the studies. As effect size indices, standardized mean differences were calculated for measures of behaviours and knowledge, both in the posttest and in the follow-up. The random and mixed-effects models were used for the statistical analyses and sensitivity analyses were carried out in order to check the robustness of the meta-analytic results.

Results: A total of 19 papers fulfilled the selection criteria, producing 23 independent studies. On average, the treatments reached a statistically significant effectiveness in the behaviours acquired, both in the posttest and in the follow-up (d+ = 1.33 and d+ = 1.80, respectively), as well as in measures of knowledge (posttest; d+ = 1.29; follow-up: d+ = 0.76). Depending on the outcome measure, the effect sizes were affected by different moderator variables, such as the type of treatment, the type of postural hygiene, the teaching method, or the use of paraprofessionals as cotherapists.

Conclusions: The interventions were successful in significantly increasing the behaviours and knowledge acquired both in the posttest and in the follow-up. The combined treatment of postural hygiene with physiotherapy exercise exhibited the best results. The small number of studies limits the generalizability of the results.

Figure 1

Forest plot of effect sizes for measures of behaviours in the posttest.

Figure 2

Forest plot of effect sizes for measures of knowledge in the posttest.

Figure 3

Forest plot of effect sizes for measures of behaviours in the follow-up.

Figure 4

Forest plot of effect sizes for measures of knowledge in the follow-up.

Figure 5

Funnel plot of effect sizes for measures of behaviours in the posttest. Full circles are imputed effect sizes from the Duval and Tweedie’s (2000) trim-and-fill method to achieve symmetry in the funnel plot.

Figure 6

Funnel plot of effect sizes for measures of knowledge in the posttest. Full circles are imputed effect sizes from the Duval and Tweedie’s (2000) trim-and-fill method to achieve symmetry in the funnel plot.

Figure 7

Funnel plot of effect sizes for measures of behaviours in the follow-up. The absence of full circles in the graph indicates an approximately symmetric funnel plot.

Figure 8

Funnel plot of effect sizes for measures of knowledge in the follow-up. The absence of full circles in the graph indicates a symmetric funnel plot.