Epilepsy & Behavior 13 (2008) 549–550

Contents lists available at ScienceDirect

Epilepsy & Behavior

journal homepage: www.elsevier.com/locate/yebeh

Brief Communication

Moonstruck? The effect of the lunar cycle on seizures

Sallie Baxendale *, Jennifer Fisher
Department of Clinical and Experimental Epilepsy, Institute of Neurology UCL, Queen Square, London, UK
National Society for Epilepsy, Chalfont St Peter, Buckinghamshire, UK

a r t i c l e i n f o a b s t r a c t

Article history: Recent reports on the effects of the lunar cycle on seizure occurrence have yielded mixed results. If the
Received 3 June 2008 moon phase is influential, we hypothesized that this would be due to the moon’s contribution to noctur-
Revised 15 June 2008 nal illumination, rather than its waxing or waning state, and that significant correlations would not be
Accepted 17 June 2008
apparent if local cloud cover were controlled for. We found a significant negative correlation between
Available online 22 July 2008
the mean number of seizures and the fraction of the moon illuminated by the sun (q = –0.09, P < 0.05)
in 1571 seizures recorded in a dedicated epilepsy inpatient unit over 341 days. This correlation disap-
Keywords:
peared when we controlled for the local clarity of the night sky, suggesting that it is the brightness of
Epilepsy
Seizures
the night and the contribution the moon phase makes to nocturnal luminance, rather than the moon
Lunar cycle phase per se, that may influence the occurrence of epileptic seizures.
Nonepileptic attack disorder Ó 2008 Elsevier Inc. All rights reserved.
Meteorology
Moon
Light

1. Introduction and may simply reflect statistical probability combined with a

Superstitions regarding the influence of the moon on many con-
ditions, including epilepsy, date back to antiquity. Many of the
remedies for epilepsy recommended by Pliny the Elder more than
2000 years ago, ranging from the ingestion of camel brains to a
helping of lion fat garnished with fennel, were to be taken at spe-
cific times during the lunar cycle to maximize their efficacy. The
idea that the moon can influence human behavior remains popular
to this day, and these beliefs are particularly prevalent among
nurses and mental health professionals [1].
Recent studies of the effect of the moon on seizure occurrence
have yielded mixed results. Polychronopoulos et al. [2] found an
increase in emergency room admissions for seizures during the full
moon quarter in a series of 859 patients. Conversely, Ruegg et al.
[3] reported a peak in episodes of status epilepticus in the opposite
phase of the lunar cycle, 3 days after a new moon. Benbadis et al.
[4] failed to find an increase in epileptic seizures in people with
an established diagnosis of epilepsy at the time of the full moon,
but reported that such increases were more likely to occur in the
final quarter of the lunar phase.
Discrepancies in the reports of an increase in seizures related to
the moon phases suggest that it is an inconsistent phenomenon
* Corresponding author. Address: Department of Neuropsychology (Box 37),
National Hospital for Neurology and Neurosurgery, Queen Square, London WC1N
3BG, UK. Fax: +44 207 813 2516.
E-mail address: sallieb@ion.ucl.ac.uk (S. Baxendale).
publication bias toward positive findings. Nevertheless, a number
of theories have been developed to explain these apparent lunar ef-
fects. These range from rather unlikely putative explanations
involving the gravitational effect of the moon on the water mass
of the body and brain and cerebral reactions to changes in the
earth’s electromagnetic field [5] to the nocebo effect of patients’
expectations of seizures creating self-fulfilling prophesies.
Before the advent of modern lighting, the brightness of the
moon at night undoubtedly played an important role in shaping
human behaviors by allowing after-dark activities at certain times
and not others. The effects of nocturnal illumination on the sleep–
wake cycle have been postulated as the genesis of many of the an-
cient associations of the moon with madness and epilepsy that
continue to this day [6]. In modern urban living, the effects of
the moon on nocturnal illumination are greatly reduced. Neverthe-
less, a recent sleep study reported that normal healthy individuals
slept, on average, 20 minutes less and that their morning fatigue
was greater during the full moon phase than during the new moon
[7].
If the moon phase does have an effect on the frequency of epi-
leptic seizures, we hypothesized that this would be due to the
moon’s contribution to nocturnal illumination rather than its wax-
ing or waning state. The traditional phases of the moon were deter-
mined millennia ago using essentially arbitrary cutoffs, based on
the angle of the sun and moon to the earth. In this study, we there-
fore used a precise measure of the fraction of the moon illuminated
by the sun to determine the brightness of the moon on each day of

1525-5050/$ - see front matter Ó 2008 Elsevier Inc. All rights reserved.
doi:10.1016/j.yebeh.2008.06.009
550 S. Baxendale, J. Fisher / Epilepsy & Behavior 13 (2008) 549–550
our study and controlled for the effects of contemporaneous, local
cloud cover on this illumination.
2. Methods
2.1. Lunar measures
The percentage of the moon’s surface illuminated by the sun is a
more sophisticated, quantitative measure of the moon’s appear-
ance than the phase. Like the phase, the percentage of the moon
illuminated is the same regardless of the observer’s position on
earth. Data on the fraction of the moon visible from the earth from
January to December 2006 were provided by the Astronomical
Applications Department of the U.S. Naval Observatory.
2.2. Clear skies
The amount of light received on the earth’s surface from the sun
in the daytime and reflected from the moon’s surface during the
night depends on the amount, type, and height of the cloud cover
and any associated precipitation, together with other factors such
as fog, mist, or dust in the atmosphere. Combined measures of total
sunshine hours and precipitation provide a measure of the clarity
of the skies in any 24-hour period. The meteorological data for
the ward in 2006 for each 24-hour period were provided by the
UK Met office.
2.3. Seizure data
This retrospective study was conducted in a dedicated epilepsy
inpatient unit. Patients are admitted for diagnostic clarification,
supervised changes to their antiepileptic medications, specialist
medical review, and presurgical assessments. Patients on the unit
are monitored 24 hours a day by specialist staff, and all observed
and self-reported seizure events are logged, with many videotaped.
Seizure events are classified in the seizure log when they occur.
We examined the seizure log for 2006 over 341 days; 1571 of
the events logged were recorded as clear epileptic seizures in the
contemporaneous records, confirmed by EEG studies or expert
medical review. An additional 2658 of the events logged were clas-
sified as not clearly epileptic in nature. Although many of these
events were nonepileptic attacks, atypical seizures may also have
been classified as ‘‘other” during the initial stages of assessment.
3. Results
3.1. Bright nights
There was a significant negative correlation between the mean
number of epileptic seizures per 24 hours and the fraction of the
moon illuminated during that period (Spearman’s q = 0.09,
P < 0.05). Conversely, the frequency of unclassified seizures was
positively correlated with the fraction of the moon illuminated
(q = 0.17, P < 0.01).
These findings suggest that although epileptic seizures are less
likely to occur on brighter nights, the frequency of unclassified
events increases.
Partial correlations were conducted to control for cloud cover
(total sunshine hours, amount of precipitation) on the brightness
of the moon during its lunar cycle. These correlations also con-
trolled for the total number of patients experiencing seizures on
each day to ensure that individuals in the midst of a cluster of at-
tacks did not exert an undue influence on the seizure frequencies.
Controlling for these factors, the fraction of the moon illuminated
was not significantly correlated with the number of seizure events
(corr = 0.03, P > 0.05). However, the correlation with unclassified
seizures remained positive and significant (corr = 0.14, P < 0.01).
4. Discussion
There is no relationship between the fraction of the moon illu-
minated and the frequency of epileptic seizures when the overall
brightness of the night sky is controlled for. This suggests that it
is the brightness of the night and the contribution the moon phase
makes to that, rather than the phase of the moon per se, that may
have an influence on the frequency of epileptic attacks. Epileptic
seizures were more likely to occur on dark nights than brighter
nights. This would be consistent with the recent reports of an in-
crease in seizures in the darker phases of the lunar cycle in popu-
lations with established epilepsy [3,4].
The mechanisms underlying this pattern are unclear, although
the effects of increased nocturnal luminance on sleep patterns,
even in urban areas [7], and the production of melatonin [8,9]
may both have an impact on seizure thresholds [10–12]. However,
it remains to be seen whether the observed effects of nocturnal
illumination in this study will pan out more consistently than
the lunar phase literature.
Consistent with the findings of Benbadis et al. [4], we also found
a relationship between the moon phase and the frequency of sei-
zures that were not clearly epileptic in nature, with an increase
in these events around the time of the full moon, regardless of
the brightness of the night sky and the local visibility of the moon,
suggesting a possible nocebo effect associated with beliefs sur-
rounding the full moon in this group.
Acknowledgment
The authors are grateful to Jane Yeomans and the nursing team
on the Gowers Epilepsy Assessment Unit at the National Society for
Epilepsy for their diligence in keeping the seizure logs used in this
study.
References
[1] Vance DE. Belief in lunar effects on human behavior. Psychol Rep
1995;76:32–4.
[2] Polychronopoulos P, Argyriou AA, Sirrou V, et al. Lunar phases and seizure
occurrence: just an ancient legend? Neurology 2006;66:1442–3.
[3] Ruegg S, Hunziker P, Marsch S, Schindler C. Association of environmental
factors with the onset of status epilepticus. Epilepsy Behav 2007;12:66–73.
[4] Benbadis SR, Chang S, Hunter J, Wang W. The influence of the full moon on
seizure frequency: myth or reality? Epilepsy Behav 2004;5:596–7.
[5] Persinger MA, Psych C. Sudden unexpected death in epileptics following
sudden, intense, increases in geomagnetic activity: prevalence of effect and
potential mechanisms. Int J Biometeorol 1995;38:180–7.
[6] Raison CL, Klein HM, Steckler M. The moon and madness reconsidered. J Affect
Disord 1999;53:99–106.
[7] Roosli M, Juni P, Braun-Fahrlander C, Brinkhof MW, Low N, Egger M. Sleepless
night, the moon is bright: longitudinal study of lunar phase and sleep. J Sleep
Res 2006;15:149–53.
[8] Bojkowski CJ, Aldhous ME, English J, et al. Suppression of nocturnal plasma
melatonin and 6-sulphatoxymelatonin by bright and dim light in man. Horm
Metab Res 1987;19:437–40.
[9] Hebert M, Martin SK, Lee C, Eastman CI. The effects of prior light history on the
suppression of melatonin by light in humans. J Pineal Res 2002;33:198–203.
[10] Sandyk R, Tsagas N, Anninos PA. Melatonin as a proconvulsive hormone in
humans. Int J Neurosci 1992;63:125–35.
[11] Schapel GJ, Beran RG, Kennaway DL, McLoughney J, Matthews CD. Melatonin
response in active epilepsy. Epilepsia 1995;36:75–8.
[12] Sizonenko PC, Moore DC, Paunier L, Beaumanoir A, Nahory A. Melatonin
secretion in relation to sleep in epileptics. Prog Brain Res 1979;52:549–51.