Sir,

The study by Dharani et al.1 represents a very informative comparison between objective light meter measurements and the more subjective diary monitoring of time spent outdoors, both of which have implications for myopia. They found a poor to fair correlation. They reasonably used a cut-off value of >1000 Lux to compare light meter measurements with time spent outdoors, as this was similar to cut-offs in previous studies. If they had used a higher cut-off (which might be justifiable as their pilot test revealed that being indoors with a stream of bright sunlight could yield means of 1573 and 4445 Lux), then this could lower the values for the light meter measurements in their Table 1, which would bring them closer to the diary measurements of time outdoors, and could potentially affect the strength of the correlation. Thus, it would be useful to calculate means and test correlation with different cut-off values, or even to test the strength of the correlation between the raw total light meter measurements and time outdoors.

Also, they excluded days on which all intensity readings were consistently <100 Lux, as this implied the light meter was not worn. It would be useful to know how many days were excluded as a result. If it were the case, for example, that the light meter was more likely to be left at home on days when children were spending more time outdoors, which is potentially plausible, then this would suggest that the overall measurements underestimate light exposure and outdoor activity. The authors could compare the outdoor diary for these days to investigate this hypothesis. Such an effect would of course be small if very few days were excluded.