Kids who lack motor skills tend to be less physically active than kids who have strong motor skills. This makes a lot of intuitive sense. If you have trouble catching and throwing a ball, it’s going to be difficult to play sports like baseball or football. More importantly, it’s also going to be difficult to throw around a ball with your friends. Ditto for riding a bike. And skating. (For more on the importance of motor skills, view this previous post, or listen to the below audio interview with motor skills researcher Meghann Lloyd).
If this were just a temporary problem it wouldn’t be a big deal. The problem is that physical activity levels track over time. In other words, kids who are inactive become adults who are inactive. This means that a kid who avoids physical activity due to a lack of motor skills is likely on a trajectory towards a physically inactive adulthood.
While the above may make intuitive sense, it is really just a hypothesis. Could childhood motor skills really matter that much?
That’s the question that Meghann Lloyd, Emily Bremer, Mark Tremblay and I recently attempted to answer in a paper published in the journal Adapted Physical Activity Quarterly (available for download here). In the paper, we examined a group of individuals who were first tested in 1991, while they were students in grade 1.
The students went through a range of tests to determine if they had high or low motor skills (students with average motor skills were omitted from the study), as well as testing their height, weight, body composition and physical fitness. The students were also asked a variety of questions about their level of physical activity (this was before more objective measures of physical activity were widely available). As you might expect, kids with high motor skills were more fit and had better body composition than those with low motor skills at baseline (age 6).
This same group of students was assessed again in 1996 (age 11). Those who had high motor skills at age 6 continued to have better motor skills and report being more physically active at age 11. A 10-year follow-up was performed in 2001 (age 16), and once again, those with higher motor skills at baseline continued to report being more physically active than those with lower motor skills.
For this new study, we decided to do a 20-year follow-up of those initial tests in 1991. This presented numerous logistical problems – the original students now live in a variety of far-flung locations, making it impossible to do any direct measures. This meant that we had to rely on questionnaires, rather than objective measures of physical activity and sedentary behaviour. As with any longitudinal study, we also had high levels of attrition (e.g. many people that participated in 1991 no longer wanted to participate in 2011). As a result, our initial group of 100 participants in 1991 was whittled down to just 17 participants who provided all requested information. This is a big limitation, so keep it in mind while you read on.
What did we find?
From the paper:
In females, overall motor skill proficiency in 1991 was positively associated with self-reported leisure time moderate (r = .85, p = .001) and vigorous (r = .63, p = .029) physical activity in 2011 at the age of 26.
In males, locomotor proficiency in 1991 positively correlated with self-reported active transportation in 2011 (r = 90, p = .036), and locomotor proficiency in 1996 (age 11) was positively correlated with vigorous leisure time physical activity in 2011 (r = .98, p = .016).
In other words the better a child’s motor skills at age 6, the higher their self-reported activity levels at age 26. I should point out that physical activity did not just mean “sports” – among males better motor skills at baseline were associated with increased use of active transportation (walking, cycling, etc) at age 26.
Strengths & Limitations
As with any study, this one has a lot of limitations. Actually, this study has even more limitations than usual. As mentioned above, we had huge levels of attrition – we dropped from 100 participants in 1991 to just 17 in 2011. Our primary outcomes were also self-reported, which leaves them open to error and bias (however, this bias would be likely to actually reduce the strength of the observed associations, rather than inflate them). Given these limitations, why did we still publish the paper?
The main reason (and one that was mentioned frequently by peer reviewers) was that there is currently very little data regarding the longitudinal impact of motor skills on health/behavioural outcomes. As I mentioned at the beginning of this post, it seems likely that motor skills could have a longterm impact, but there’s not much hard data that can actually be applied to the issue. Despite its many limitations, this study is currently the the best data we’ve got on the issue. It’s not going to settle things by any means. In that sense it is hypothesis generating more so than hypothesis testing. Hopefully these results will lead others to perform larger, better studies that can more rigorously test the longterm impact of motor skills on health.
[NERD WARNING: the next paragraph quickly gets into esoteric issues related to academic publishing. Feel free to skip to the take-home message below.] I’m sure that some people will argue that weak data is worse than no data at all, and that studies with weak data therefore shouldn’t be published. Not surprisingly, I disagree with that approach. We are rapidly moving towards a situation where individual studies mean little, while systematic reviews carry the day (we’re actually pretty much there in health-related fields like my own). And in that situation, you want as much data as you can possibly get, warts and all (if only some research is deemed publishable, you wind up with biased systematic reviews that don’t tell the full story; this is a big problem when those reviews are being used to determine public health guidelines or what government programs get funding). So long as you’re transparent about the warts, you will eventually get a good sense of the truth, regardless of what any individual study may say.
What is the take-home message?
This study suggests that motor skills not-only track throughout childhood (e.g. kids with poor motor skills at age 6 are likely to still have relatively poor motor skills later on), but that motor skills at a very early age may be associated with important health-related behaviours well into adulthood. This suggests that we should probably take childhood motor skills seriously, especially for kids who are well behind their peers. It also suggests that someone needs to do a much larger study looking at these same issues. The best time to start a 20-year follow study was 20 years ago. The second best time is today. Get a move on!
Lloyd M, Saunders TJ, Bremer E, Tremblay MS. Long-term importance of fundamental motor skills: a 20-year follow-up study. Adapted Physical Activity Quarterly. 2014, 31, 67-78. Available for download here.