Dear Mike,

Thanks for your message - sorry for the delay.  I hope these references are
helpful.
Let me know if you need any further info.

John H

Acta Psychiatr Scand 1999 Jun;99(6):412-8 

Attempted suicide and major public holidays in Europe: findings from the 
WHO/EURO Multicentre Study on Parasuicide.
Jessen G, Jensen BF, Arensman E, Bille-Brahe U, Crepet P, De Leo D,
Hawton K, Haring C, Hjelmeland H, Michel K, Ostamo A, Salander-Renberg E,
Schmidtke A, Temesvary B, Wasserman D
Centre for Suicidological Research, Odense, Denmark. 
[Medline record in process]

OBJECTIVE: The aim of the study was to examine the relationship between
suicide 
attempts and major public holidays in Europe. METHOD: The analysis was
based on 
data on 24 388 suicide attempts by persons aged 15 years or older in the
period 
1989-1996. Data from 13 centres (representing 11 countries)
participating in the 
WHO/EURO Multicentre Study on Parasuicide were analysed. The analysis of
the 
fluctuation of suicide attempts around public holidays was based on the
daily 
number of suicide attempts for each centre. For each day in the period
under 
examination a mean number of suicide attempts (mu) was calculated. The
analysis 
was based on the assumption that the data followed a Poisson
distribution. The 
observed number of daily suicide attempts was compared with the expected
number 
of attempts. A multiplicative model for the expected number in each
centre was 
developed. RESULTS: Before Christmas there were fewer suicide attempts
than 
expected, and after Christmas there were approximately 40% more attempts
than 
expected. In addition, more attempts than expected were registered on
New Year's 
Day. In countries where people have the day off work on Whit Monday
there were 
significantly fewer attempts during the 3 days before, but where Whit
Monday is 
a normal working day significantly fewer attempts occurred on the Monday
to 
Wednesday after Whit Sunday. CONCLUSION: There appears to be a
transposition of 
a significant number of suicide attempts from before (and during) a
major public 
holiday until after it. The division of holidays into non-working and
working 
days showed that a 'holiday effect' could only be found around major
public 
holidays, particularly Christmas, Easter and Whitsun. These findings
support the 
theory of the 'broken-promise effect' for major public holidays. 
PMID: 10408262, UI: 99334747 



Forensic Sci Int 1997 Oct 6;89(3):167-74 

Prison suicide in Finland, 1969-1992.
Joukamaa M
University of Oulu, Oulu University Hospital, Department of Psychiatry,
Finland. 

BACKGROUND: All prisoners' suicides in Finland during 1969-1992 (n =
184) were 
studied. METHOD: The data were collected from official documents.
RESULTS: Of 
all prisoners' deaths, 47% were suicides. The rate of suicide among male 
prisoners was three-fold compared to the normal adult Finnish male
population. 
More than half of those committing suicide had a psychiatric disturbance
and one 
half of them had visited the prison health services because of a
psychiatric 
problem not more than one week before the suicide. Almost one third of
the 
suicides were committed in isolation rooms. Contrary to the findings of
many 
previous studies, there was no concentration of suicides at weekends, on 
religious holidays, in different seasons nor at the beginning of the 
confinement. CONCLUSIONS: The most important finding of this study was
the 
common use of health care facilities in the prison just before the
suicide. It 
is important to try to develop the means to recognize suicidal ideation
among 
all prisoners seeking psychiatric care. The surveillance of prisoners in 
isolation cells needs to tighten up and should probably be continuous. 
PMID: 9363625, UI: 98030076 




Acta Psychiatr Scand 1995 Sep;92(3):214-9 

Timing of completed suicides among residents of Olmsted County,
Minnesota, 
1951-1985.
Panser LA, McAlpine DE, Wallrichs SL, Swanson DW, O'Fallon WM, Melton LJ
3rdSection of Health Services Evaluation, Mayo Clinic, Rochester, MN
55905, USA. 
Previous studies of suicide timing may have been biased by incomplete 
ascertainment of suicides and by delays between the suicidal act and
subsequent death. Those potential biases were assessed and minimized in
this population-based sudy by using the unique resources of the
Rochester Epidemiology Project in Olmsted County, Minnesota. Using these
more accurate data, we confirmed previous reports of no excess suicides
on birthdays (+/- 3 days), or during 3 United States national holidays.
While most prior reports found excess post-holiday suicides and suicide
peaks on Mondays, those findings were not observed in Olmsted County.
Because 93% of deaths occurred on the date of the suicidal act, using
date of death instead of the actual date of suicide is sufficient for
most research purposes. 
PMID: 7484201, UI: 96051022 




Med J Aust 1994 Dec 5-19;161(11-12):728 

There are days ... and moons ... and public holidays. Self-poisoning is
not lunacy.
Buckley NA, Dawson AH, Whyte IM
Publication Types:

Letter 
    PMID: 7830660, UI: 95131845 



Acta Psychiatr Scand 1992 Sep;86(3):253-4 

Parasuicide during Ramadan in Jordan.
Daradkeh TK
Faculty of Medicine & Health Sciences, United Arab Emirates University,
Al-Ain. 
To investigate the impact of national and religious events on the rate
of parasuicide, a comparison was made between the number of reported
parasuicides during the month of Ramadan and the month before and after
Ramadan in Jordan, for the years from 1986 to 1991. Significantly fewer
parasuicides were reported during Ramadan than the month preceding it
and the month that follows Ramadan. The findings confirm previous
observations that national events reduce the rate of parasuicide, but
the protective effect does not persist into the month that follows
Ramadan. 
PMID: 1414423, UI: 93034362 





Br J Psychiatry 1991 Feb;158:155-7 

Monthly and seasonal variation in parasuicide. A sex difference.
Masterton G
Department of Psychological Medicine, Royal Infirmary of Edinburgh. 
In a total of 22,169 parasuicides over 19 years, there was evidence for
a seasonal cycle in parasuicide admissions among women, with an increase
during the summer and a fall over the winter, including a pronounced
December nadir. Further analysis of the winter trough suggests a
sociological explanation, the focal point being Christmas. There was no
evidence for monthly or seasonal variation among men. PMID: 2012904, UI:
91191241 



Percept Mot Skills 1990 Dec;71(3 Pt 1):994 

Suicide and homicide on St. Valentine's Day.
Lester D
Richard Stockton State College, Pomona, NJ 08240. 
PMID: 2293198, UI: 91156459 



BMJ 1990 Mar 24;300(6727):783-4 

Association between parasuicide and Saint Valentine's Day.
Davenport SM, Birtle J
University Department of Psychiatry, Queen Elizabeth Hospital,
Birmingham. 
PMID: 2322739, UI: 90213016 





Epidemiol Prev 1990 Mar;12(42):23-9 

[Analysis of temporal distribution of suicides. Methodological notes on 
comparison of dates of death and dates of suicidal act].
[Article in Italian] 

Stefanini P, Biggeri A, Geddes M, Comodo N
U.O. Epidemiologia, Firenze. 
The short-term temporal cycles in the frequency of suicide for the
Province of Florence 1985-1987 has been analyzed. The authors initially
present a comparison between the Date of Suicidal Act and the recorded
Death Certificate Date. In their data the difference, still debated in
the literature, appears of no practical importance. The results show
that suicides peak on Spring and Fall,with a temporal variation with
regard to the first decade of the month, high at the beginning and low
at the end. No evident pattern has been explained by day of the week,
unlikely any reports from the literature. On the contrary, the effect of
the National Holidays are striking, showing a decrease on the holidays 
and a sharp rise on the subsequent day. 
PMID: 2147146, UI: 91059890 



Nursing (Lond) 1989 Dec 7-20;47:28-9 

Christmas in casualty.
Crouch S
MeSH Terms:

Alcoholic Intoxication/nursing 
    Emergency Service, Hospital* 
    Holidays* 
    Human 
    Nursing Staff, Hospital/psychology* 
    Suicide, Attempted 
    PMID: 2616082, UI: 90137596 



 

JAMA 1988 Dec 23-30;260(24):3654 

A piece of my mind. Thanks for the memories.
Collins TM
Publication Types:

Historical article 
    MeSH Terms:

Adult 
    Emergency Medicine 
    History of Medicine, 20th Cent. 
    Holidays* 
    Human 
    Male 
    Suicide* 
    Vietnam 
    War/history* 
    PMID: 3057256, UI: 89056010 



     

Am J Public Health 1988 May;78(5):541-3 

A comparison of injury date and death date in 42,698 suicides.
Phillips DP, Sanzone AG
Department of Sociology, University of California at San Diego, La Jolla
92093. 
The date of death was compared with the date of injury (the onset of the 
suicidal act) for 42,698 California suicides, 1973-85. Findings revealed
92.6 per cent of all California suicides occurred within one day of the
injury date. Graphs of daily suicides by injury date and by death date
displayed essentially identical patterns around the holidays, by date of
the week, and after publicized suicide stories. These findings suggest
that the difference between date of death and date of injury need not be
taken into consideration in the analysis of temporal fluctuations, as
was previously supposed. 
PMID: 3354737, UI: 88181301 



     

Psychol Rep 1987 Aug;61(1):224 

Suicide and homicide at Easter.
Lester D
PMID: 3671609, UI: 88041662 




Psychol Rep 1987 Apr;60(2):414 

Suicide and homicide rates on national holidays.
Lester D
PMID: 3495812, UI: 87232448 



 

Psychol Rep 1987 Feb;60(1):245-6 

Traditional holidays and suicide.
Sparhawk TG
PMID: 3562724, UI: 87176198 



Suicide Life Threat Behav 1987 Spring;17(1):1-12 

A drop in suicides around major national holidays.
Phillips DP, Wills JS
To study the fluctuation of suicides around the holidays, we examined
188,047 U.S. suicides from 1973 to 1979. Correcting for the effects of
extraneous variables, such as seasons and days of the week, we found
that suicides dropped by an annual average of -102.5 in the period
surrounding the holidays. One set of holidays (Memorial Day,
Thanksgiving, and Christmas) was associated with an unusually low risk
of suicide before, during, and after the holiday. Another set of
holidays (New Year's Day, July 4th, and Labor Day) was associated with a
low risk of suicide before the holiday and a high risk just afterwards.
Almost all demographic groups experienced a low risk of suicide around
the holidays: whites, blacks, males, females, retired persons, and
persons of working age. Only white teenagers deviated from this pattern.
Thus, the evidence does not support the widespread popular assumption
that holidays are a risk factor in suicide. 
PMID: 3495908, UI: 87235574 


Am J Psychiatry 1985 Jun;142(6):782 

Suicide at Christmas.
Lester D
Publication Types:

Letter 
    MeSH Terms:

Holidays* 
    Human 
    Seasons 
    Suicide/epidemiology* 
    United States 
    PMID: 4003607, UI: 85222774 



Demography 1983 Feb;20(1):45-59 

Temporal variations in mortality: a comparison of U.S. suicides and
motor 
vehicle fatalities, 1972-1976.
Bollen KA
Among the various types of mortality, motor vehicle fatalities (MVF) and 
suicides are two that are very responsive to social, economic, and
psychological factors. The analysis of the temporal patterns of suicides
and MVF points to important but neglected forces affecting them. This
paper examines the daily patterns of MVF and suicides for the United
States from 1972 to 1976. The effects of the day of the week, month,
year, and holidays on these types of mortality are estimated. Total MVF
and suicides, as well as daily differentials by sex, age, and race, are
analyzed with regression techniques. The results show that temporal
variations are stronger in MVF than in suicides but are present in 
both. MVF tend to peak on Saturday, in the summer months, in 1972 and
1973, and on holidays. Suicides are generally highest on Monday, in the
spring months, in 1975 and 1976, and on nonholidays. Differentials in
the temporal patterns are evident. For instance, whites generally have a
sharper decline in suicides on holidays than do nonwhites. The holiday
peak in MVF for the 60 and over population is Christmas, while it is New
Year's for the younger age groups. Males exhibit a greater increase in
suicides from March to May than do females. Finally, the results of a
comparison of the daily patterns of MVF and suicides revealed a negative
correlation. This contrasted with the strong positive relationship
between MVF and suicides generally found in cross-sectional studies. The
implications of these findings are briefly discussed. 
PMID: 6832433, UI: 83158323 



     

Community Ment Health J 1983 Fall;19(3):177-86 

Myths of mid-winter depression.
Christensen R, Dowrick PW
An investigation was conducted into the effect of a major Alaskan annual
winter festival upon the rates of crisis data. Analysis of rates of
suicide, attempted suicide, family disturbance calls, crisis calls, and
mental health admissions indicated no significant effect of the
festival. Statewide statistics over several years indicate that demands
for depression-related services appear to peak in either the summer or
the fall. These results do not support the widely held belief that
depression is more common during the winter in the North or that mid-
winter festivals help to promote psychological well-being. It is 
concluded that the pervasiveness of such myths may lead to misdiagnosis
or mistreatment, and that other folklore should be examined for its
validity. 
PMID: 6681268, UI: 85202645 




Suicide Life Threat Behav 1980 Spring;10(1):41-50 

The frequency of suicides around major public holidays: some surprising 
findings.
Phillips DP, Liu J
This paper analyzes the recorded number of U.S. daily suicides, 1972-
1976. The paper shows that: there is a dip in suicides on and just
before major public holidays; there is a peak in suicides just after
these holidays; and, the dip in suicides is somewhat larger than the
peak--hence, there is a net decrease in suicides around the holidays. On
balance, the holidays appear to reduce the number of suicides, not
increase them, as previously believed. In addition, previous research
has focussed on the negative psychological impact of holidays. 
Our findings indicate that holidays have a positive component, which
should be taken into account in future research. 
PMID: 7361344, UI: 80148664 



     

Am J Epidemiol 1979 May;109(5):517-20 

Temporal variation in suicide and homicide.
Lester D
In this study of all of the violent deaths in the US over two years,
clear monthly and daily variations were found. Suicide peaked in the
Spring and Fall; homicide in July and December. Suicide was more common
on Mondays; homicide on Saturdays and Sundays. Homicide was more common
on national holidays, while suicide tended to be less common. No lunar
variation was found. 
PMID: 453175, UI: 79206598 



 

Psychol Rep 1977 Jun;40(3 Pt 2):807-10 

Seasonal suicide attempts and forms of loneliness.
Wenz FV
PMID: 866519, UI: 77194431 



 
Northwest Med 1969 Mar;68(3):232-8 

Market analysis for suicide prevention. Relationship of age to suicide
on 
holidays, day of the week and month.
Blachly PH, Fairley N
PMID: 5784438, UI: 69204443 
At 11:34 09/07/99 +0100, you wrote:
>It is commonly stated that psycho-social crises and acts of deliberate self
>harm increase around emotionally charged festivals like Christmas and New
>Year ( and perhaps will even more over the coming Millennium celebrations)
>Does anyone know of any evidence to support this assertion. I found one
>reference "It has previously been reported [Davenport & Birtle, 1990] that
>there is a positive association between nonfatal deliberate self-harm in
>adolescents and emotionally charged festivals, such as St Valentine's Day
>and Christmas Day. ", do you know of any others? 
>
>Thanks
>
>Mike Lambert.
>A&E Department, 
>Norfolk and Norwich Hospital
>
>
Professor John Henry
Academic Department of Accident and Emergency Medicine
Imperial College School of Medicine
St Mary's Hospital
London W2 1NY

Tel 44-171 886 6187
Fax 44-171 886 6315


%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%