unfortunately, the study included information about intensity of night time light but did not included information about the sources of light
So I guess it's user-interpreted intensity
For heart attack, for example, compared with those in the darkest 50th percentile, those who experienced some nighttime light — in the 51st to 70th percentile — had a 20 percent increased risk of heart attack.
Those who experienced more light, in the 71st to 90th percentile, had a 27 percent increased risk.
And those who experienced the most nighttime light, in the 91st to 100th percentile, had a 47 percent higher risk.