Effectiveness of intervention on physical activity of children: systematic review and meta-analysis of controlled trials with objectively measured outcomes (EarlyBird 54)
BMJ 2012; 345 doi: https://doi.org/10.1136/bmj.e5888 (Published 27 September 2012) Cite this as: BMJ 2012;345:e5888All rapid responses
Rapid responses are electronic comments to the editor. They enable our users to debate issues raised in articles published on bmj.com. A rapid response is first posted online. If you need the URL (web address) of an individual response, simply click on the response headline and copy the URL from the browser window. A proportion of responses will, after editing, be published online and in the print journal as letters, which are indexed in PubMed. Rapid responses are not indexed in PubMed and they are not journal articles. The BMJ reserves the right to remove responses which are being wilfully misrepresented as published articles or when it is brought to our attention that a response spreads misinformation.
From March 2022, the word limit for rapid responses will be 600 words not including references and author details. We will no longer post responses that exceed this limit.
The word limit for letters selected from posted responses remains 300 words.
Dr Symonds seems affronted by evidence that physical activity interventions do little to change the physical activity of children, but that is to miss (or to deny) the value of systematic review and meta-analysis. It is simply not appropriate to cherry pick individual studies that appear to work best, because of the risk of selection error. Rather, trials should be grouped and compared according to type, setting, design, etc. as they were in our sub-group analysis.[1] We showed that there were no significant differences in effectiveness between the different sub-groups of trials (see Figure 3 of the original article). Indeed, trials providing extra activity sessions were no more effective than those that did not. In the absence of sub-group differences, pooling across all trials is appropriate.[2]
Dr Symonds selects out four trials that appeared to produce large intervention effects. Three of these, the three with the largest effects, were also the three smallest studies in the review, involving just 79 children between them. Their outcomes were, for that reason, the least reliable of all the studies. Larger scale versions of these trials would be unlikely to replicate the same ‘success’, given that larger trials (n>200) produced small effects at best (Figure 4), small-to-negligible on average (Figure 3).
Dr Symonds believes that “improving the activity levels of children should be a major public health objective since low levels of activity correlate strongly with numerous adverse health outcomes in later life”. Nobody, I think, denies that more active children are healthier, as we and others have shown.[3,4] The real issue is how to get less active children to be more active. The findings of our review suggest this may be very difficult to achieve.
References
1.Metcalf B, Henley W, Wilkin T. Effectiveness of intervention on physical activity of children: systematic review and meta-analysis of controlled trials with objectively measured outcomes (EarlyBird 54). BMJ 2012;345:e5888. (27 September.)
2.Borenstein M, Hedges LV, Higgins JP, Rothstein HR. Subgroup analyses. In: Introduction to meta-analysis. 1st ed. John Wiley and Sons, 2009:149-186.
3.Metcalf BS, Voss LD, Hosking J, Jeffery AN, Wilkin TJ. Physical activity at the government-recommended level and obesity-related health outcomes: a longitudinal study (Early Bird 37). Arch Dis Child. 2008 Sep;93(9):772-7.
4.Ekelund U, Luan J, Sherar LB, Esliger DW, Griew P, Cooper A. Moderate to vigorous physical activity and sedentary time and cardiometabolic risk factors in children and adolescents. JAMA 2012;307:704-12.
Competing interests: No competing interests
Metcalfe et al.’s meta-analysis of physical activity interventions in children was a thought-provoking read. It is sobering to realise that we probably have a long way to go in this increasingly important area of health promotion.
However, I think it is important to emphasise exactly what this study concludes. That is: if you pool together all the published trials which measured accelerometer data, the net result is that these interventions have, so far, not been effective. This is very different to concluding that “physical activity interventions do not work in children”, and I would echo Richard Weiler’s point here of not getting carried away with the press releases and headlines.
What unites all the studies included in this meta-analysis is that all the studies included used accelerometer data as an outcome measure (not necessarily as the primary one). The interventions themselves varied widely. Several of the interventions demonstrated statistically significant improvements in physical activity levels (Roemmich 2004, Goldfield 2006, Hughes 2008, Weintraub 2008).
Because the interventions varied so widely, combining them in all in single meta-analysis is a dubious academic move, analogous to doing a combined meta-analysis on all antihypertensive medication trials, when the question we really want to answer is “which one works best?” If I had done a trial on whether providing all children with a TV in their bedroom and closing down all swimming pools improved physical activity levels, would this also have been included in this meta-analysis?
Improving physical activity levels in children should be a major public health objective, since low levels of activity in childhood correlate strongly with numerous adverse health outcomes in later life. We should be looking at these studies carefully, deciding which ones seemed to be effective, and why, and then designing larger scale trials to demonstrate the evidence properly. Meanwhile, it remains our moral duty as health care providers to emphasise the huge importance of regular physical activity for health.
Competing interests: No competing interests
In his rapid response, Dr Jaques claims that we look at BMI as the outcome of childhood physical activity interventions, but this is incorrect - the outcome we look at is not BMI but objectively-measured physical activity. Other reviews have examined the response of BMI to such activity interventions and have found that BMI changes little.[1-3] The finding of our review provides an explanation for this, for it shows that activity interventions actually have very little effect on the overall activity of children (~4 minutes more walking/running per day).
Dr Jaques also comments that fitness, rather than fatness, might be a more appropriate outcome measure of activity interventions. While we agree that fitness is an important benefit to children who undertake more activity, our data transcend the argument by showing how difficult it is to make children more active. The gains in fitness resulting from an extra 4 minutes of daily activity are likely to be small. Indeed, an earlier review of after-school activity intervention trials showed only a limited improvement in fitness.[4]
References
1 Kamath CC, Vickers KS, Ehrlich A, McGovern L, Johnson J, Singhal V, et al. Behavioral interventions to prevent childhood obesity: a systematic review and metaanalyses of randomized trials. J Clin Endocrinol Metab 2008;93:4606-15.
2 Harris KC, Kuramoto LK, Schulzer M, Retallack JE. Effect of school-based physical activity interventions on body mass index in children: a meta-analysis. CMAJ 2009;180:719-26.
3 Waters E, de Silva-Sanigorski A, Hall BJ, Brown T, Campbell KJ, Gao Y, et al. Interventions for preventing obesity in children. Cochrane Database Syst Rev 2011;12:CD001871.
4 Beets MW, Beighle A, Erwin HE, Huberty JL. After-school program impact on physical activity and fitness: a meta-analysis. Am J Prev Med. 2009 Jun;36(6):527-37.
Competing interests: No competing interests
The study produced by Metcalf et al. raised some questions for me. As the President Elect of the Faculty of Sport and Exercise Medicine, and a Consultant in Sport and Exercise Medicine, I feel it is important to reinforce that, while the authors look at BMI as an outcome of physical activity interventions, a reduction in BMI is not necessarily a good method of measuring fitness. I would argue that, as one can be overweight but still physically fit, a lack of reduced BMI does not necessarily prove that physical activity interventions in children do not work.
I would be also be interested to see what physical activities these children took part in, as enthusiasm for an activity, particularly in the under 16s, is critical to the success of children’s adherence. Simply put, if the activity is dull, adherence will be very low. If this was the case in this study, it’s not surprising that trials show poor uptake and an apparent reduction in physical activity levels.
Exercise interventions need to tap into what excites children, and for young people, access and environment are fundamental to regular activity. Finally, the health benefits of exercise in both children and adults are not merely limited to a trim waistline. An emphasis on weight reduction, rather than health benefits inherent in increased physical activity, is not helpful.
Competing interests: No competing interests
Metcalf et al should be commended on their detailed review of the effectiveness of interventions on physical activity in children. [1] This review, like all reviews, is limited by the nature of the interventions used in the included studies (themselves no doubt limited by lack of funding in this important area and a lack of focus on physical activity access and environment, which are fundamental to children’s activity levels).
Media headline grabbing, based on research is usually reported as fact, or in this case demonstrating ‘physical activity interventions for children have little impact’ or 'do little to get kids moving'. Unfortunately, few readers of such articles, national newspapers or radio listeners will benefit from reading the full paper, nor understand the limitations of a review and will therefore be unable to judge for themselves. Such blunt headlines, based on research with many limitations, causes lasting damage to public opinion, public health policy in the real world and in turn shapes future research and initiatives.
Sadly, this is not the first time the BMJ has encouraged misguided headlines and debate in the name of ‘evidence’ on complex issues relating to physical activity promotion. The issue is not the meticulous and detailed research, but the fact that the BMJ seems to actively seek public and commercial attention for itself in the trusted educational guise of research. BMJ knows the media will print ridiculous headlines, without consideration for a balanced approach, making a mockery of the same standards that BMJ purports in its editorial and peer review process and also demonstrates a total disregard for the damaging public opinions it shapes for years to come.
No doubt the BMJ will argue that public debate is vital and self-balancing, but with the headlines printed, read and perceived to be fact, the damage is already done.
1. Metcalf B, Henley W, Wilkin T. Effectiveness of intervention on physical activity of children: systematic review and meta-analysis of controlled trials with objectively measured outcomes (EarlyBird 54). BMJ 2012;345:e5888
Competing interests: No competing interests
Re: Effectiveness of intervention on physical activity of children: systematic review and meta-analysis of controlled trials with objectively measured outcomes (EarlyBird 54)
Metcalf, Henley, and Wilkin [1] report on a well conducted and informative meta-analysis related to the effectiveness of interventions on children’s physical activity. They conclude that physical activity interventions have little effect on the overall activity levels of children.
A few considerations should be kept in mind when reading this paper, and if the authors’ conclusions are correct, there are important implications for the future promotion of physical activity among children.
First, the small pooled intervention effect observed across studies for both total activity and moderate to vigorous physical activity (MVPA) may be the result of artificially inflated pre-intervention measures. Children aware that they are being monitored may increase their activity levels above those that are typical (i.e. the Hawthorne effect or reactivity). [2] When the post-intervention measures are taken the novelty of being monitored may have worn off and activity levels become more typical, resulting in any intervention effect being diluted.
Second, the authors do not report any data on participants’ absolute activity levels, such as average minutes per day spent in MVPA. If participants engaged in, or were close to engaging in the daily physical activity recommendations of 60 minutes of MVPA at pre-intervention, then the lack of intervention effect is not problematic.
Third, obesity results from small imbalances in energy intake and expenditure over time. The small intervention effects observed in the study of Metcalf et al., in conjunction with small dietary changes, may therefore be sufficient to redress this energy imbalance and bring about weight reduction.
Finally, the main finding of this meta-analysis is unsurprising given that sustained behaviour change is notoriously difficult. [3] The lack of observable intervention effects supports the need for alternative approaches to changing behaviour, such as changes to the environment or policy-driven strategies. Car exclusion zones around schools to promote active travel, increased minutes of mandatory physical education per week, and incentivized physical activity schemes are just three examples of how this may be achieved.
Rather than abandoning attempts to increase children’s activity levels because efforts to date have largely been unsuccessful, a more positive approach would be to identify and build on those interventions that have been found to be effective.
1. Metcalf B, Henley W, Wilkin T. Effectiveness of intervention on physical activity of children: systematic review and meta-analysis of controlled trials with objectively measured outcomes (EarlyBird 54). BMJ 2012;345:e5888.
2. Foley JT, Beets MW, Cardinal BJ. Monitoring children’s physical activity with pedometers: Reactivity revisited. JESF 2011; 9(2): 82-6.
3. Michie S, Rothman A J, Sheeran, P. Current issues and new direction in Psychology and Health: Advancing the science of behavior change. Psych & Health 2007; 22(3): 249-53.
Competing interests: No competing interests