The lunar effect is a link between different stages of the 29.5-day lunar cycle and the behavior and physiology of various species, purportedly including humans. The changing phase and position of the Moon in its orbit impacts night lighting and ocean tides on Earth. Various organisms have adapted to this repeating cycle.

A considerable number of studies have examined the effect on humans. By the late 1980s, there were at least 40 published studies on the purported lunar-lunacy connection, and at least 20 published studies on the purported lunar-birthrate connection. Literature reviews and metanalyses have found no correlation between the lunar cycle and human biology or behavior. In cases such as the approximately monthly cycle of menstruation in humans (but not other mammals), the coincidence in timing reflects no known lunar influence. The widespread and persistent beliefs about the influence of the Moon may depend on illusory correlation – the perception of an association that does not in fact exist.

In a number of marine animals, there is strong evidence for the effects of lunar cycles. Observed effects relating to reproductive synchrony may depend on external cues relating to the presence or amount of moonlight. Corals contain light-sensitive cryptochromes, proteins that are sensitive to different levels of light. Coral species such as Dipsastraea speciosa tend to synchronize spawning in the evening or night, around the last quarter moon of the lunar cycle. In Dipsastraea speciosa, a period of darkness between sunset and moonrise appears to be a trigger for synchronized spawning. Another marine animal, the bristle worm Platynereis dumerilii, spawns a few days after a full moon. It contains a protein with light-absorbing flavin structures that differentially detect moonlight and sunlight. It is used as a model for studying the biological mechanisms of marine lunar cycles.

Claims of a lunar connection have appeared in the following contexts:

It is widely believed that the Moon has a relationship with fertility due to the corresponding human menstrual cycle, which averages 28 days. However, no connection between lunar rhythms and menstrual onset has been conclusively shown to exist, and the similarity in length between the two cycles is most likely coincidental.

Multiple studies have found no connection between birth rate and lunar phases. A 1957 analysis of 9,551 births in Danville, Pennsylvania, found no correlation between birth rate and the phase of the Moon. Records of 11,961 live births and 8,142 natural births (not induced by drugs or cesarean section) over a 4-year period (1974–1978) at the UCLA hospital did not correlate in any way with the cycle of lunar phases. Analysis of 3,706 spontaneous births (excluding births resulting from induced labor) in 1994 showed no correlation with lunar phase. The distribution of 167,956 spontaneous vaginal deliveries, at 37 to 40 weeks gestation, in Phoenix, Arizona, between 1995 and 2000, showed no relationship with lunar phase. Analysis of 564,039 births (1997 to 2001) in North Carolina showed no predictable influence of the lunar cycle on deliveries or complications. Analysis of 6,725 deliveries (2000 to 2006) in Hannover revealed no significant correlation of birth rate to lunar phases. A 2001 analysis of 70,000,000 birth records from the National Center for Health Statistics revealed no correlation between birth rate and lunar phase. An extensive review of 21 studies from seven different countries showed that the majority of studies reported no relationship to lunar phase, and that the positive studies were inconsistent with each other. A review of six additional studies from five different countries similarly showed no evidence of relationship between birth rate and lunar phase. In 2021, an analysis of 38.7 million births in France over 50 years, with a detailed correction for birth variations linked to holidays, and robust statistical methods to avoid false detections linked to multiple tests, found a very small (+0.4%) but statistically significant surplus of births on the full moon day, and to a lesser extent the following day. The probability of this excess being due to chance is very low, of the order of one chance in 100,000 (p-value = 1.5 x 10-5). The belief that there is a large surplus of births on full moon days is incorrect, and it is completely impossible for an observer to detect the small increase of +0.4% in a maternity hospital, even on a long time scale.

It is sometimes claimed that surgeons used to refuse to operate during the full Moon because of the increased risk of death of the patient through blood loss. One team, in Barcelona, Spain, reported a weak correlation between lunar phase and hospital admissions due to gastrointestinal bleeding, but only when comparing full Moon days to all non-full Moon days lumped together. This methodology has been criticized, and the statistical significance of the results disappears if one compares day 29 of the lunar cycle (full Moon) to days 9, 12, 13, or 27 of the lunar cycle, which all have an almost equal number of hospital admissions. The Spanish team acknowledged that the wide variation in the number of admissions throughout the lunar cycle limited the interpretation of the results.

In October 2009, British politician David Tredinnick asserted that during a full Moon "[s]urgeons will not operate because blood clotting is not effective and the police have to put more people on the street.". A spokesman for the Royal College of Surgeons said they would "laugh their heads off" at the suggestion they could not operate on the full Moon.