Abstract
Background: Studies indicate that a fraction of nursing professionals believe in a "lunar effect"-a purported correlation between the phases of the Earth's moon and human affairs, such as birth rates, blood loss, or fertility.
Purpose: This article addresses some of the methodological errors and cognitive biases that can explain the human tendency of perceiving a lunar effect where there is none.
Approach: This article reviews basic standards of evidence and, using an example from the published literature, illustrates how disregarding these standards can lead to erroneous conclusions.
Findings: Roman, Soriano, Fuentes, Galvez, and Fernandez (2004) suggested that the number of hospital admissions related to gastrointestinal bleeding was somehow influenced by the phases of the Earth's moon. Specifically, the authors claimed that the rate of hospital admissions to their bleeding unit is higher during the full moon than at other times. Their report contains a number of methodological and statistical flaws that invalidate their conclusions. Reanalysis of their data with proper procedures shows no evidence that the full moon influences the rate of hospital admissions, a result that is consistent with numerous peer-reviewed studies and meta-analyses. A review of the literature shows that birth rates are also uncorrelated to lunar phases.
Conclusions: Data collection and analysis shortcomings, as well as powerful cognitive biases, can lead to erroneous conclusions about the purported lunar effect on human affairs. Adherence to basic standards of evidence can help assess the validity of questionable beliefs.