...This suggests that the different rules have only a marginal impact on the safety of bicycle traffic in the dark. Only a small number of nighttime accidents can be clearly attributed to the lack of lights: Other major risk factors are driving or riding under the influence of alcohol, higher driving speeds on empty roads at night and impaired night vision especially in older drivers.
Bolding mine.
And just why is night vision impaired? Oh yea, because there is not enough light, but lights on a bike won't help?
Sounds to me like the study "can clearly be attributed to lack of lights" really means can solely be attributed to lack of lights.
It should be the other way round, what should be counted is every case where lack of lights is a contributing factor.