John Hale, of Dunedin, asks :-

Has controlled experimentation established any connections between a full moon and the human body?

David Bilkey, a psychologist at the University of Otago, responded.

The belief that the moon influences our physiology and our behavior, including our mental health, can be traced back into antiquity and is an idea that is still widely held. But what happens when we approach this question scientifically? What do the data say?

The results are equivocal at best. Unfortunately, many earlier studies that have claimed to reveal moon-phase effects have not been conducted in a particularly robust manner. Often they have relied on individuals retrospectively analyzing their own behavior, which, as a method, is prone to hindsight bias, our tendency to reinterpret past events on the basis of our current knowledge.

Furthermore, many of these studies have been carried out across short time periods, which means that other factors such as day-of-the-week and weather are also likely to have had an influence on the outcomes. As an example of a more carefully controlled study, when data on daily violent incidents in hospital inpatient settings were collected across a six month period and related to moon phase, no evidence of a connection was evident.

More recently, new technology and access to large data sets have allowed further rigorous testing of these ideas. For example, a study of almost 6000 activity-monitor-wearing children across 12 countries found no effect of moon phase on activity levels during the day and only a tiny effect of moon phase on sleep duration (sleep duration was 1% shorter at full moon). Similarly a test of the hypothesis that births are more likely during a full moon, showed no such pattern when data were analyzed from a half million births occurring over a five year period. So, the scientific studies do not provide compelling evidence to support the moon-phase theory. It seems that Werewolves we are not!

Send questions to: Ask-A-Scientist, PO Box 31-035, Christchurch 8444 Or email: