Our study assessed parental understanding of BMI, investigated demographic factors associated with BMI chart understanding, and tested differentials in parental understanding using standard versus color-coded BMI charts. Previous studies have shown that pediatricians hesitate to show BMI charts to parents, believing that parents will not be able to understand their complexity.15
With standard BMI charts, parents answered less than two-thirds of the questions correctly, even when they reported their child(ren)’s doctor had previously discussed BMI. However, parents universally demonstrated greater understanding using the color-coded BMI charts. Parents demonstrated more than four times greater odds of answering questions correctly using color-coded charts than standard charts. Furthermore, parents with the lowest numeracy demonstrated the largest increases with color-coded charts, suggesting color-coding could help reduce numeracy disparities in understanding BMI charting.
It makes sense that parental numeracy skills would be highly correlated with understanding of BMI and BMI charts since BMI and graphs in general are numerical entities. Previous studies have also shown a link between numeracy and adult understanding of health information that is reliant on numerical skills, such as understanding of food labels,24
interpretation of glucose meter readings and insulin adjustment in patients with diabetes,26
and asthma management.27
Similar to our study, several studies found a stronger relationship between comprehension and numeracy than comprehension and literacy.
There are several important limitations of this study. First, our study population had relatively high health literacy, with almost all participants scoring in the “adequate” range on the S-TOFHLA as in other recent studies.28
Despite this “ceiling effect,” we did find an association between literacy level and the ability to both provide a correct definition of BMI in open-ended format and interpret BMI charts in general. However, we were unable to assess the impact of color-coded BMI charting on parents with inadequate health literacy, which had been one of our initial goals. Second, although our “Understanding BMI” questionnaire was designed to assess parental understanding of BMI charting, there is, as yet, no validated tool to test this. Therefore, it is unclear whether correctly interpreting BMI charting on this questionnaire translates to understanding of a child’s weight status in a real office setting, or, more importantly, whether the ability to interpret a BMI chart would contribute to parental motivation to make recommended lifestyle changes. Third, our study was completed using a convenience sample derived from two large academic centers’ pediatrics clinics. While those participating in this study reflected a diverse sample similar demographics to the clinics from which they were recruited, those demographics may or may not be generalizable to other settings. Similarly, due to the non-random nature of recruitment and lack of prior data regarding our population’s children’s BMIs, it is not possible to confirm that our participants’ responses regarding BMI experience and knowledge are typical for the clinic populations in general. However, given that the demographics of our sample are consistent with our clinic populations, there is no particular reason to expect selection bias. Fourth, in one clinic, patients visiting specialty physicians as well as general pediatricians were included, which may account for differences in demographics between sites. To reduce the impact of these between-site differences, such differences were adjusted for in all analyses. Fifth, surprisingly, our study population included no participants with colorblindness on our screen, so it was not possible to assess the interpretability of color-coded BMI charts for those individuals. If color-coded BMI charting (or other color-coded interventions) were to be implemented in practice, categories should be easily distinguishable, even to those who are so-called “red-green” colorblind. Also, our data only allowed the exploration of potential factors associated with understanding BMI; they did not allow for predictive modeling. Therefore, we were unable to ascertain, for example, whether higher education was a proxy for higher numeracy ability, or if higher education itself was an independent predictor of a parents’ ability to understand BMI charting. This leaves remaining questions about which parents are most likely to find which type of BMI charting most helpful in discussions about BMI. Finally, the terminology of BMI categories changed during our study, and it is hard to know how much of a role new terminology of BMI categories would help or hinder BMI charting understanding.
Despite these limitations, our findings have important implications. If color-coding helps improve parents’ understanding of BMI, doctors may be able to use the color-coded BMI chart as a tool in their practice to communicate weight status to parents more effectively. In one study, when asked what resources would be most helpful, 90% of pediatricians endorsed better tools to communicate weight problems to patients.14
We also know that parental readiness to make changes is affected by parents’ perceptions of their children’s weight. One study noted that if parents perceived their children were overweight, they were twice as likely to fall into the preparation/action stage of change to help their children lose weight.29
In another study, most (68%) parents of obese children who thought their children’s weight was unhealthy were told this by a doctor, and more parents whose doctors explained their children’s weight as unhealthy were preparing to make lifestyle changes compared with those whose doctors never stated this (75% vs. 25%, p<0.05).30
In a recent British study of the effects of a structured weight communication program, half the parents of young obese children who were told their children’s weight was unhealthy reported positive health behavior change, and most wanted such weight-based communication regularly.31
Finally, when doctors fail to take obesity seriously at younger ages, parents perceive this as a barrier to change health habits.32
Color-coded BMI charts could be one way to communicate early problems in the universally-understood stop light motif. Whether better communication of BMI to parents leads to increased recognition of their children’s weight status and whether such understanding then leads to improved lifestyle or weight trajectories are projects deserving further research, and we have begun to study these important questions.
To our knowledge, our study is the first to evaluate whether use of color-coded charts aids parental understanding of BMI charting. Our results suggest that color-coded BMI charts increase the number of parents who will understand BMI charting, particularly those with lower numeracy. Color-coded BMI charting may be one element in the growing repertoire of tools to assist pediatricians in effectively communicating with parents and may help start a conversation of therapeutic lifestyle change in an era of a childhood obesity epidemic.