No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Death has a preference for birthdays-an analysis of death time series
Abstract
To examine the relation between the day of death and the day of birth. To determine whether the "death postponement" hypothesis or the "anniversary reaction" hypothesis is more appropriate.
We analyzed data from the Swiss mortality statistics 1969-2008. Deaths below the age of 1 were excluded from the analysis. Time series of frequencies of deaths were based on differences between the day of death and the day of birth. We applied autoregressive integrated moving average modeling with intervention effects both in straight and reverse time series.
The overall death excess on the day of birth was 13.8%, mainly because of cardiovascular and cerebrovascular diseases (more in women than in men) as well as suicides and accidents (in particular, falls in men). Unexpectedly, we also found an excess of deaths in cancers. An (negative) aftereffect was found in cancers, and (positive) anticipatory effects were found in falls in men.
In general, birthdays do not evoke a postponement mechanism but appear to end up in a lethal way more frequently than expected ("anniversary reaction").
... This was partly due to insufficient sample sizes (Barraclough and Shepherd, 1976;Lester, 1986). More recent studies, however, found evidence supporting the "birthday blues" hypothesis (Williams et al., 2011;Ajdacic-Gross et al., 2012;Pena, 2015). These studies used large general population data collected over an extended period of time and estimated daily patterns of deaths in relation to one's birthday in England and Wales (Williams et al., 2011), Switzerland (Ajdacic-Gross et al., 2012), and the United States (Pena, 2015). ...
... More recent studies, however, found evidence supporting the "birthday blues" hypothesis (Williams et al., 2011;Ajdacic-Gross et al., 2012;Pena, 2015). These studies used large general population data collected over an extended period of time and estimated daily patterns of deaths in relation to one's birthday in England and Wales (Williams et al., 2011), Switzerland (Ajdacic-Gross et al., 2012), and the United States (Pena, 2015). Their findings indicated that the risk of death increased on the day of one's birthday from 6 to 40%, depending on the cause of death. ...
... Williams et al. (2011) focused only on suicide deaths, while Pena (2015) did not differentiate causes of death due to data limitations. Ajdacic-Gross et al. (2012) examined daily patterns of deaths by causes of deaths, but not by age-at-death groups or other demographic groups except for sex. Thus, they did not provide comprehensive evidence on the possibility that the effect of birthdays on mortality is heterogeneous. ...
This study examined whether people have a higher risk of death on and around their birthday using a large national mortality data. We examined 2,073,656 death records of individuals who died in Japan from major external causes between 1974 and 2014. Poisson regression analysis showed that people were more likely to die on their birthday than on any other calendar day by means of suicide, traffic accidents, accidental falls, drowning, and choking. For suicide cases in particular, people are 50% more likely to die on their birthday compared to any other dates. Excess deaths on birthdays were observed regardless of gender, marital status, and age-at-death subgroups. For suicide deaths, our results provided strong support for the “birthday blues” hypothesis that predicts excess deaths on birthdays. With regards to traffic accidents and other unintentional accidents, however, our results suggest that excess deaths on birthdays may be related to birthday celebrations. For the elderly population, our analysis indicates that special activities associated with birthday celebrations, which often involves going out to consume festive food and drinks, may be contributors to a sudden increase in the number of accidental deaths on their birthday. In contrast, a notably sharp increase in the number of motor vehicle accidents was observed for individuals in the 20s on their birthday, which may be attributable to birthday celebrations that involved both driving and drinking.
... The use of the quadratic in the exponent isn't arbitrary, it is motivated by the central limit theorem. We will find it beneficial to reparametrize this as (2) where B will play the role of a concentration parameter and λ is the average of (p 1 − p 2 ) 2 over the unit square. ...
... The question we pose is Can older people hold on to die just past their birthdays? This question was popularized by Phillips [21] and is still of current research interest [2]. The effect is claimed to be potent on important people. ...
... Posterior distribution of the difference p 1 − p 2 for different values of the concentration parameter B as defined in(2). ...
Recent advances in Monte Carlo methods allow us to revisit work by de Finetti who suggested the use of approximate exchangeability in the analyses of contingency tables. This paper gives examples of computational implementations using Metropolis Hastings, Langevin, and Hamiltonian Monte Carlo to compute posterior distributions for test statistics relevant for testing independence, reversible or three-way models for discrete exponential families using polynomial priors and Gröbner bases.
... Several empirical studies have searched for evidence of deviations from expected mortalitydnamely spikes or dipsdon or around holidays and birthdays, and they have found mixed evidence (Phillips and Smith, 1990;Young and Hade, 2004;Williams et al., 2011;Skala and Freedland, 2004;Reulbach et al., 2007;Abel and Kruger, 2009;Phillips et al., 2010;Ajdacic-Gross et al., 2012). Some studies have found mortality dips before holidays and birthdays and spikes on and after those days, while others have found either the opposite or no dips or spikes at all. ...
... The variation in results could be attributed to differences in the subjects studied and the causes of death considered. Except for one study in Switzerland (Ajdacic-Gross et al., 2012), those studies focus on a specific subgroup of the population (e.g., baseball players, psychiatric patients, elderly Chinese-American women, Roman Catholic priests) or on a subset of the causes of death (e.g., natural causes, suicide, cancer). In addition, some studies estimate daily-level patterns, while others estimate patterns at a weekly-or monthlylevel, making direct comparisons difficult to interpret. ...
... The contributions of the study can be summarized as follows. First, it provides estimates of excess death rates on birthdays using the largest sample to datedroughly ten times the size of the largest samples among previous studies (Ajdacic-Gross et al., 2012;Phillips et al., 1992). Second, it presents estimates of daily-level dips and spikes around birthdays for the entire populationdas opposed to estimates of weekly-or monthly-level patterns or estimates for a specific subgroup of the population found in other studies (Skala and Freedland, 2004). ...
This study estimates average excess death rates on and around birthdays, and explores differences between birthdays falling on weekends and birthdays falling on weekdays. Using records from the U.S. Social Security Administration for 25 million people who died during the period from 1998 to 2011, average excess death rates are estimated controlling for seasonality of births and deaths. The average excess death rate on birthdays is 6.7% (p < 0.0001). No evidence is found of dips in average excess death rates in a ±10 day neighborhood around birthdays that could offset the spikes on birthdays. Significant differences are found between age groups and between weekend and weekday birthdays. Younger people have greater average excess death rates on birthdays, reaching up to 25.4% (p < 0.0001) for ages 20–29. Younger people also show the largest differences between average excess death rates on weekend birthdays and weekday birthdays, reaching up to 64.5 percentage points (p = 0.0063) for ages 1–9. Over the 13-year period analyzed, the estimated excess deaths on birthdays are 4590.KeywordsTiming of deathDeath on birthdaysDeath on weekends
... The work of other authors mainly confirm this 'birthday deadline phenomenon' that the mortality actually increases before the birthday. The authors of [1] analyse Swiss mortality databases from years 1969-2008 containing over 2 million death records with the help of the ARIMA model. They show a 13.8% increase in death rate around birthdays with variations of between 11 to 18 per cent in men and women older than 60. ...
It is a common belief that people close to death from natural causes can postpone their imminent death if they see a strong reason to survive a bit longer. This is known as the Postponement hypothesis: that a meaningful occasion can act as a motivator to prolong life for a short amount of time. A few studies have already addressed this hypothesis but their conclusions are contradictory. To check the postponement hypothesis, we analysed almost 249 thousand cases in the dataset for South African people who died in the year 2015. We took a person's birthday as the meaningful occasion and analyse the death rate around this date using statistical models offered by survival analysis. If the hypothesis is true, it can be expected that the mortality rate should be lower a period just before the birthday and, perhaps, higher shortly afterwards. The results of our analysis show that no postponement of death can be seen for the examined dataset. In fact, to the contrary, the data suggest that the mortality rate is higher both before and after the birthday. Speculations as to why this is the case might be a higher risk associated with the stress of expectations for the birthday as well as an earlier start of celebrations with associated departure from the recommended regime.
... Is the risk of suicide higher on one's birthday? This question is known as the "birthday blues" hypothesis and recently, large population-based data studies strongly support it [1][2][3][4] . Daily suicide rates were estimated in Switzerland 1 , Japan 2,3 , England, and Wales 4 and these studies found that the suicide risk was 6-40% higher on birthdays than other days. ...
Recent studies suggest that the risk of suicide is higher during and around birthdays. The so-called “birthday blues” might be stronger on birthdays at milestone ages (e.g., 20, 30, 40), as these symbolic ages might represent occasions for existential stock-taking that may highlight failures and underachievement in life. Moreover, in some countries (including Japan), certain symbolic birthdays come with the expectation of celebration with family and friends, and thus such special birthdays may elevate the birthday blues if there is nobody to celebrate the occasion with. This study examines the possibility that there are more suicides on milestone birthdays than on other birthdays or days other than birthdays, using approximately one million individual death records between 1974 and 2014 in Japan. Graphical analysis and Poisson regression analysis showed that suicides occurred more frequently on milestone birthdays when people turn 20, 30, 40, and 60. This pattern was predominately observed in men. Our findings suggest that it is crucial for health professionals and family members to pay close attention to vulnerable individuals as their birthdays approach. In particular, individuals are at a higher risk when birthdays coincide with occasions of social significance, including the ages of adulthood (age 20) and retirement (age 60).
... Ajdacic-Gross et al. [25] used the SARIMA model with an interventional effect to evaluate the relationship between birth and death days on Swedish mortality data between 1969 and 2008. Grundy et al. [26] utilized an interrupted time series analysis method to investigate the effect of creating 20 mph traffic speed zones on RTIs and RTM in London between 1986 and 2006. ...
Objectives
Every day, an average of 3,400 deaths and tens of millions of injuries occur as a result of traffic accidents. This study aims to model and validate road traffic injury (RTI) times series, specifically considering gender.
Study design
Time trend studies of monthly road traffic injuries (RTI) in Iran from March 2005 to February 2016, as well as those of males and females from March 2009 to February 2016 were performed.
Methods
The seasonal auto-regressive integrated moving average method (SARIMA) was employed to predict RTI time series. The final model was selected from various SARIMA models based on the Akaike information criterion (AIC) and the Bayesian information criterion (BIC). To examine whether the residuals were white noise, the Ljung-Box (LB) test and residuals plots were used for un-correlation, and the zero mean and stationarity, respectively. Additionally, smoothing methods were utilized to validate the SARIMA models for fitting and out-of-range prediction of the time series models under study. The sample auto-correlation function (ACF) and the partial autocorrelation function (PACF) with 20 lags were employed to determine the order of models and to ascertain if the residuals of the model were uncorrelated.
Results
Based on the obtained results, SARIMA (2,1,0)(0,1,1)12, SARIMA (0,1,1)(0,1,1)12, SARIMA (1,1,1)(0,0,1)12, and SARIMA (2,0,0)(1,0,0)12 were chosen for the time series including incidence rates of total road traffic injuries (IRTI), IRTI of males, females, and males-to-females, respectively. The AIC values were -87.57, 413.38, -732.91, and -85.32, respectively. The LB test for the residuals of the time series models of (0.539) IRTI, (0.3) IRTI of males, (0.23) females, and (0.237) males-to-females indicated that residuals were uncorrelated. Furthermore, prediction values for the next 24 months (2016 to 2018) showed no decline in the incidence rate of male and female traffic injuries. Results of the predictions using exponential smoothing methods indicated out-of-range prediction validity of the SARIMA models.
Conclusion
This study exemplified the high efficiency of SARIMA models in predicting road traffic injuries (RTIs). Based on observations, the IRTI mean in Iran was 35.57 in 2016. The predicted values of the IRTI for 2016–2018 by the SARIMA model showed no decreasing trend. During the studied period, the observed values of IRTI for males were two to three times the female values. Thus, prediction of RTI can provide a useful tool for traffic safety policymaking by simulating interrupted time series when applying new traffic enforcement interventions and regulations in the future. Additionally, IRTI analysis of males and females showed that men had a non-increasing trend but higher incidence of traffic injuries, whereas the IRTI for women revealed an increasing trend from 2009 to 2012 with a lower incidence of injuries. This growth could be attributed to the impact of increased outdoor activities of women and the increased number of issued driving licenses in the period of 2009–2012.
... 3,23 ED visits increase during the above-mentioned vulnerable times, and individuals often describe feelings of increased stress, especially with alcohol intake. 43,44 In addition, patients who are considered high risk should have their records fl agged to reassess for suicide risk during each visit or possibly during home visits. ...
Suicide in older adults is continuing to rise and, as the older population increases, so will the rate of suicide. By learning more about the risk factors, assessment areas to explore, and ways to improve treatment, primary care providers can help decrease the incidence of suicidal behaviors in this population.
... Hillard et al. 35 revealed that there was no data based increase in acute psychological disorders on and before Christmas, which weakens a psychological explanation. However, there are tendencies for a depressed mental status during the Christmas period, mainly due to concerns such as loneliness and absence of family 36 , which supports our findings and leads to a possible interaction of social support and distress. As we focused on the specific collective of cancer patients other mechanism such as particularities in social relations or deprivation 37 , and pathological affections of the body, for example, chronic inflammation is associated with a depressed mood, 38 might be in action. ...
Постановка проблемы. Ранее было показано, что в России и в мире праздники сопровождаются ростом смертности, главным образом в Рождество и Новый год. А как обстоит дело с диагнозами избыточной смертности при праздновании дней рождений?
Материал и методы. С этой целью исследовали данные 16,77 млн умерших с 01.01.2011 по 31.12.2019 на предмет соотношения дат рождения и смерти. Во всех случаях даты рождения и смерти получили номер дня года. При совпадении этих номеров случаю смерти присваивали значение 0, при несовпадении – число дней различия с днем рождения до или после этого дня от ±1 до ±182. Эту процедуру проделывали как для смертности в целом, так и для 6 больших классов смертей и 8 видов внешних причин. Избыточную смертность в день рождения оценивали с помощью скользящей средней LOWESS (локально взвешенная полиномиальная регрессия), которая описывала смертность в непраздничные дни, а в праздничные дни ее перерывы заполняли линейной регрессией, которая служила показателем «естественной», непраздничной смертности в праздники.
Результаты. Выявлен значительный рост смертности в день рождения (4,0 тыс. в день за 9 лет), а всего избыточная смертность, связанная с днем рождения, составила 19,7 тыс. или 3,5% от «естественной» смертности. Главный вклад в избыточную смертность внесли болезни системы кровообращения (9,1 тыс.) и внешние причины (5,2 тыс.), а среди внешних причин самоубийства (1,3 тыс.), отравления алкоголем (1,0 тыс.) и повреждения с неопределенными намерениями (0,8 тыс.). Избыточная смертность в день рождения была пропорциональна непраздничной смертности. Исключение составили отравления алкоголем, которые в 2 раза превысили ожидаемое по будням число смертей. Особенность самоубийств состоит в том, что увеличение их количества начинается за 2 месяца до дня рождения с максимумом за 35 дней, последующим снижением и резким ростом в день рождения. Дальнейшая смертность от самоубийств имеет цикличность с максимумами на 7-й и 15-й дни после дня рождения. Сравнение с другими праздниками показало, что дни рождения по размерам избыточной смертности занимают промежуточное положение между Новым годом и другими государственными праздниками. Главным фактором избыточной смертности в праздники является злоупотребление крепкими спиртные напитками.
Заключение. День рождения, как и государственные праздники, сопровождается ростом смертности. Легко предположить, что потери этим не исчерпываются: заболевших в день рождения много больше, чем умерших в связи с этим праздником.
Formulation of the problem. It is known that holidays are accompanied by an increase in morbidity and mortality. The aim of this study was to summarize the experience of previous studies examining therelationship between national holidays and the time of death, as well as to analyze the Russian characteristics of mortality during holidays. The article poses the following questions: 1) Is the risk of dying on holidays increasing in Russia? 2) If so, on which ones? 3) Is this related to alcohol abuse? 4) To what extent are men and women involved? 5) What is theestimate of excess deaths on holidays?Methods. The study used daily data on the number of deaths in Russia from all causes and from alcohol poisoning, disaggregated by sex for 2000-2017. In total, 35.4 million people died during the period under review. We studied 9 public holidays in Russia. The LOWESS moving average, calculated for non-holidays and extrapolated to holidays, was taken as zero. We took into account holidays that exceeded the LOWESS confidence interval. To estimate losses on a birthday, linear regression and its confidence interval were taken as zero. We used data from Rosstat and Google Trends to indirectly estimate legal and illegal alcohol consumption as well.Results. Five out of nine public holidays in Russia are accompanied by an increase in the number of deaths. The greatest increase is observed in connection with the New Year on January 1-15. At this time, the excess number of deaths amounted to 113.6 thousand people over 18 years, or 6.3 thousand people per year, with a maximum on January 1 (2.0 thousand per day). This is 14.0% more deaths than on weekdays. The excess number of deaths on the holidays on February 23, March 8 and May 9 amounted to 1.7 thousand per year. 80.9% of excess deaths in January are of men. The all-cause mortality and mortality from alcohol poisoning coincides with the maximum on January 1. After 2005, when the duration of the New Year holidays increased, the maximum on January 1 and the subsequent dynamics of mortality did not change. The annual maximum sales of alcoholic beverages are in December. Birthdays are also accompanied by an increase in total mortality by 9.1 thousand per year, and this is also associated with alcoholism.Conclusion. In Russia, on holidays, mainly on New Year's and birthdays, there is a significant increase in mortality, which is mainly due to alcohol abuse and does not depend on the duration of the holidays in January. This damage can be reduced by a decrease in the availability of strong alcohol, the maximum sales of which occur in December.
Постановка проблемы. Известно, что праздники сопровождаются ростом заболеваемости и смертности. Целью данного исследования являлось обобщение опыта предыдущих работ, изучающих связь между национальными праздниками и временем смерти, а также анализ российских особенностей смертности в праздничные дни. Мы ставили следующие вопросы: 1) «увеличивается ли в России риск умереть в праздники?»; 2) «если да, то в какие?»; 3) «связано ли это со злоупотреблением алкоголем?»; 4) «в какой степени в это вовлечены мужчины и женщины?»; 5) «какова оценка избыточного числа смертей в праздничные дни?».
Методы. В исследовании использованы ежедневные данные о числе умерших в России от всех причин и от алкогольных отравлений с разделением по полу за 2000-2017 гг. Всего в рассматриваемый период умерло 35,4 млн человек. Исследовали 9 государственных праздников в России. За ноль отсчета принимали скользящую среднюю LOWESS, рассчитанную для непраздничных дней и экстраполированную на праздничные дни. Учитывали праздничные дни, которые превышали доверительный интервал LOWESS. Для оценки потерь в день рождения за ноль отсчета принимали линейную регрессию и ее доверительный интервал. Для косвенной оценки легального и нелегального потребления алкоголя использовали данные Росстата о помесячных продажах алкогольной продукции и данные Google Trends о популярности поисковых запросах на алкогольную продукцию.
Результаты. Пять из девяти государственных праздников в России сопровождаются ростом числа смертей. Наибольший рост наблюдается 1-15 января в связи с Новым годом. В это время избыточное число смертей составило 113,6 тыс. человек за 18 лет или 6,3 тыс. человек в год с максимумом 1 января (2,0 тысячи в день). Это на 14,0% больше, чем в будни. В праздники 23 февраля, 8 марта и 9 мая избыточное число смертей составило в сумме 1,7 тыс. в год. 80,9% избыточных смертей в январе приходится на мужчин. Общая смертность и смертность при отравлении алкоголем совпадает по максимуму 1 января и в последующей динамике. После 2005 г., когда увеличилась продолжительность новогодних праздников, максимум 1 января и последующая динамика смертности не изменилась. Годовой максимум продаж алкогольных напитков приходится на декабрь. Дни рождения также сопровождаются приращением общей смертности на 11,1 тыс. в год, и это также связано с алкоголизацией.
Заключение. В России в праздники, главным образом новогодние и дни рождения, происходит значительный прирост смертности, который обусловлен преимущественно злоупотреблением алкоголем и не зависит от продолжительности праздников в январе. Снизить этот урон может уменьшение доступности крепкого алкоголя, максимум продаж которого приходится на декабрь.
Background
Previous studies show contradictory findings on the relationship between birthday and deathday, in particular whether people postpone death until after their birthday. We examine the phenomenon in eight groups of famous people.
Methods
Birthday and deathday for the following groups were recorded: British prime ministers, US presidents, Academy Award best actor, best female actor, best director, Nobel Prize winners, Wimbledon men’s and ladies' singles winners, all from when records began. For each group, the difference in days between the deathday and birthday was calculated. Under the hypothesis of no association, one can expect the difference to have a uniform distribution. This is assessed using goodness-of-fit tests on a circle.
Results
All groups showed some departure from the uniform and it occurred around the birthday in all groups. British prime ministers, US presidents, Academy Award actors and directors, Nobel Prize winners and Wimbledon men show a ’dip' in deaths around the birthday. The length of the ’dip' varied between the groups and so they gave different p-values on different test statistics. For Academy Award female actors and Wimbledon ladies, there was rise in deaths before and after birthday. When Nobel Prize winners were subdivided into their categories, Science and Literature had a ’dip' around the birthday, but not other categories.
Conclusions
We conclude ’something' happens to deathday around the birthday. Some groups of famous people show a ’dip' in death rate around the birthday while for others, particularly women, the association is in the opposite direction.
Background. Seasonal as well as weekly cycles in suicide have been described, replicated and poorly understood for a long time. In Western countries, suicides are typically least frequent on weekends and most frequent on Mondays and Tuesdays. To improve understanding of this phenomenon a strategy is required which focuses on anomalous findings beyond the regular patterns. Here, we focused on instances where the weekly suicide patterns disappear or are interrupted. Methods. We used data from Swiss and Austrian mortality statistics for the periods 1969–2010 and 1970–2010, respectively. First, the data were cross-tabulated by days of the week and the available socio-demographic information (sex, age, religious affiliation and region). Second, time series of cumulated daily frequencies of suicide were analysed by seasonal Autoregressive Integrated Moving Average (ARIMA) models which included intervention effects accounting for Easter and Pentecost (Whit) holidays. Results. First, the cross tabulations showed that weekly cycles may be smoothed above all in young persons and smoothed in drowning, jumping and car gas exhaustion suicides. Second, the ARIMA analyses displayed occasional preventive effects for holidays Saturdays and Sundays, and more systematic effects for holiday Mondays. There were no after effects on Tuesdays following holiday Mondays. Conclusions. In general, the weekend dip and the Monday backlog effect in suicide show striking similarities to the Advent season effect and are interpretable within the same template. The turning points between low and high frequencies possibly provide promising frames for the timing of prevention activities.
An Geburtstagen und biografisch bedeutsamen Jahrestagen kommt es signifikant häufiger zu bestimmten Todesfällen und psychischen Dekompensationen. Analysen von Sterbestatistiken haben diese klinische Evidenz eindrucksvoll bestätigt. Das Stressmodell und das Konzept der „Anniversary reactions“ sind die beiden überzeugendsten Erklärungsversuche für diese Zusammenhänge. Das Konzept der Anniversary reactions hat den Vorteil, dass es bewusste und unbewusste Gedächtnisinhalte integriert. Die biografische Signifikanz der jeweiligen Jahrestage ist im autobiografischen Gedächtnis verankert. Die neueren Ergebnisse der Hirnforschung zu Gedächtnisfunktionen und Hirnrepräsentanzen bestätigen die Grundannahmen der Theorie der Anniversary reactions. Abstract Certain fatal casualties and psychic decompensations occur significantly more often on birthdays and other biographically important anniversaries. Analyses of mortality statistics have impressively confirmed this clinical evidence. The stress model and the concept of anniversary reactions are the two most convincing explanations for these associations. The advantage of the concept of anniversary reactions lies in the integration of conscious and unconscious memory contents. The biographic significance of the respective anniversaries is stored in the autobiographic memory. Latest results of brain research on memory functions and brain representation confirm the assumption of anniversary reactions.
Seasonal as well as weekly cycles in suicide have been described, replicated and poorly understood for a long time. In Western countries, suicides are typically least frequent on weekends and most frequent on Mondays and Tuesdays. To improve understanding of this phenomenon a strategy is required which focuses on anomalous findings beyond the regular patterns. Here, we focused on instances where the weekly suicide patterns disappear or are interrupted.
Methods.
We used data from Swiss and Austrian mortality statistics for the periods 1969–2010 and 1970–2010, respectively. First, the data were cross-tabulated by days of the week and the available socio-demographic information (sex, age, religious affiliation and region). Second, time series of cumulated daily frequencies of suicide were analysed by seasonal Autoregressive Integrated Moving Average (ARIMA) models which included intervention effects accounting for Easter and Pentecost (Whit) holidays.
Results.
First, the cross tabulations showed that weekly cycles may be smoothed above all in young persons and smoothed in drowning, jumping and car gas exhaustion suicides. Second, the ARIMA analyses displayed occasional preventive effects for holidays Saturdays and Sundays, and more systematic effects for holiday Mondays. There were no after effects on Tuesdays following holiday Mondays.
Conclusions.
In general, the weekend dip and the Monday backlog effect in suicide show striking similarities to the Advent season effect and are interpretable within the same template. The turning points between low and high frequencies possibly provide promising frames for the timing of prevention activities.
To determine if people dying from cancer are able to prolong their own life in order to experience a certain biographical event, or whether the appearance of such an occasion leads to increased deaths before the event.
We compared numbers of cancer deaths during a period of 1 week before and after biographically important occasions, which were birthday, Christmas and Easter. As a psychogenic postponement or hastening of death is most likely in chronic diseases (as opposed to accidents or cardiovascular events), we included cancer deaths only. We estimated relative risks (RRs) with their corresponding Bonferroni corrected CIs to assess effects of biographical events. All registered cancer deaths in Germany from 1995 to 2009 were included (3 257 520 individual deaths). Numbers of deaths were corrected for seasonality.
Considering all cases, there were noticeably more deaths than expected in the week preceding Christmas, leading to an RR of dying after the event of 0.987 (CI 0.978 to 0.997). Estimates indicating a hastening of death were consistent over several subgroups. Other occasions showed inconsistent results, especially there was no convincing postponement effect in our data.
While there is no evidence of different death numbers before and after Easter and birthdays, the appearance of Christmas seems to increase deaths.
To examine the frequency of suicides on holidays and special days of the year, specifically in Queensland, Australia.
We analysed data from the Queensland Suicide Register between 1990 and 2009. The days examined were: Easter, Christmas, New Year, Valentine's Day and Anzac Day. We compared suicide cases on these days with the average counts for periods before and after the days.
There was a statistically significant increase in suicides on Christmas Eve and on New Year's Day. Our results are discussed in light of trends reported in the literature.
The beginning and end of the festive season are times when special attention is warranted for those with a heightened risk of suicide.
The present study, which contained a sample of 3,948 suicides and controls for seasonal effects, did not replicate the results of Barraclough and Hughes's (1987) study. People are not significantly more likely to die near the time of their birthdays than they are at other times of the year.
The Fédération Internationale de Football Association (FIFA) World Cup, held in Germany from June 9 to July 9, 2006, provided an opportunity to examine the relation between emotional stress and the incidence of cardiovascular events.
Cardiovascular events occurring in patients in the greater Munich area were prospectively assessed by emergency physicians during the World Cup. We compared those events with events that occurred during the control period: May 1 to June 8 and July 10 to July 31, 2006, and May 1 to July 31 in 2003 and 2005.
Acute cardiovascular events were assessed in 4279 patients. On days of matches involving the German team, the incidence of cardiac emergencies was 2.66 times that during the control period (95% confidence interval [CI], 2.33 to 3.04; P<0.001); for men, the incidence was 3.26 times that during the control period (95% CI, 2.78 to 3.84; P<0.001), and for women, it was 1.82 times that during the control period (95% CI, 1.44 to 2.31; P<0.001). Among patients with coronary events on days when the German team played, the proportion with known coronary heart disease was 47.0%, as compared with 29.1% of patients with events during the control period. On those days, the highest average incidence of events was observed during the first 2 hours after the beginning of each match. A subanalysis of serious events during that period, as compared with the control period, showed an increase in the incidence of myocardial infarction with ST-segment elevation by a factor of 2.49 (95% CI, 1.47 to 4.23), of myocardial infarction without ST-segment elevation or unstable angina by a factor of 2.61 (95% CI, 2.22 to 3.08), and of cardiac arrhythmia causing major symptoms by a factor of 3.07 (95% CI, 2.32 to 4.06) (P<0.001 for all comparisons).
Viewing a stressful soccer match more than doubles the risk of an acute cardiovascular event. In view of this excess risk, particularly in men with known coronary heart disease, preventive measures are urgently needed.
This study shows that mortality dips before a symbolically meaningful occasion and peaks just afterward. Mortality among Chinese (n = 1288) dips by 35.1% in the week before the Harvest Moon Festival and peaks by the same amount (34.6%) in the week after. We chose to study mortality among Chinese and a Chinese holiday for two reasons. First, the holiday moves around the calendar, thus allowing separation of the effects of the holiday from fixed, monthly mortality effects. Second, the holiday appeals strongly to one (experimental) group and not to others (which can be used as control groups). In terms of percentage, cerebrovascular diseases displayed the largest dip/peak pattern, followed by diseases of the heart, and then malignant neoplasms. The dip/peak mortality pattern does not appear in various non-Chinese control groups. The statistical significance of the findings was demonstrated with linear and curvilinear regression analysis and with two nonparametric tests. After testing alternative explanations for the findings, we concluded that the dip/peak pattern occurs because death can be briefly postponed until after the occurrence of a significant occasion.
(JAMA. 1990;263:1947-1951)
The relationship between suicide and birthdays, and suicide and public holidays has been studied from data on 32,291 Danish suicides by persons ages 15 years and older in the 25-year period 1970–1994. Evidence was found to support the theory of the “broken-promise effect” for major public holidays in that there appears to be a postponement of a significant number of suicides from before a holiday until after. The division of holidays into nonworking and (half-time) working days showed that a “holiday effect” could only be found around major public holidays, particularly Christmas, Easter, and Whitsun. The postponing or transpositioning effect is relevant to prevention, especially because of the availability and accessibility of help at the end of and after major public (nonworking) holidays.
The authors study the date of death in relationship, in the annual cycle, to date of birthday of every person who died of natural causes in Switzerland between 1 st January, 1969 and 31 st December, 1992 (N=12275033). They highlight a maximal mortality (17% over expected) on the actual anniversary of birth with the entire population, a slightly lower figure being found under 80 years of age, higher over this age. They tie this phenomenon to the classical group of “anniversary reactions”, in which problems of identity probably play a role. An uncertainty hangs nevertheless over the value of these findings, which are not confirmed by any published material, and even denied by some.
Socioculturally meaningful events have been shown to influence the timing of suicide, but the influence of psychiatric disorder on these associations has seldom been studied.
To investigate the association between birthday and increased risk of suicide in the general population and in a national sample of psychiatric patients.
Data on general population suicides and suicide by individuals in recent care of mental health services were examined for day of death in relation to one's birthday using Poisson regression analysis.
An increased risk of suicide was observed on day of one's birthday itself for males in both the general population (IRR = 1.39, 95% CI = 1.18-1.64, p < .01) and the clinical population (IRR = 1.48, 95% CI = 1.07-2.07, p = .03), especially for those aged 35 years and older. In the clinical population, risk was restricted to male patients aged 35-54 and risk extended to the 3 days prior to one's birthday.
Birthdays are periods of increased risk for men aged 35 and older in the general population and in those receiving mental health care. Raising health-care professionals' awareness of patient groups at greater risk at this personally significant time may benefit care planning and could facilitate suicide prevention in these individuals.
An excess of deaths was observed in the 30 days before and after a birthday for a sample of self-inflicted deaths aged 75 and over.
In epidemiology, data often arise in the form of time series e.g. notifications of diseases, entries to a hospital, mortality rates etc. are frequently collected at weekly or monthly intervals, Usual statistical methods assume that the observed data are realizations of independent random variables. However, if data which arise in a time sequence have to be analysed, it is possible that consecutive observations are dependent. In environmental epidemiology, where series such as daily concentrations of pollutants were collected and analysed, it became clear that stochastic dependence of consecutive measurements may be important. A high concentration of a pollutants today e.g. has a certain inertia i.e. a tendency to be high tomorrow as well.
Since the early 1970s, time series methods, in particular ARIMA models (autoregressive integrated moving average models) which have the ability to cope with stochastic dependence of consecutive data, have become well established in such fields as industry and economics. Recently, time series methods are of increasing interest in epidemiology.
Since these methods are not generally familiar to epidemiologists this article presents their basic concepts in a condensed form. This may encourage readers to consider the methods described and enable them to avoid pitfalls inherent in time series data. In particular, the following topics are discussed: Assessment of relations between time series (transfer function models). Assessment of changes of time series (intervention analysis), forecasting and some related time series methods.
This study shows that mortality dips before a symbolically meaningful occasion and peaks just afterward. Mortality among Chinese (n = 1288) dips by 35.1% in the week before the Harvest Moon Festival and peaks by the same amount (34.6%) in the week after. We chose to study mortality among Chinese and a Chinese holiday for two reasons. First, the holiday moves around the calendar, thus allowing separation of the effects of the holiday from fixed, monthly mortality effects. Second, the holiday appeals strongly to one (experimental) group and not to others (which can be used as control groups). In terms of percentage, cerebrovascular diseases displayed the largest dip/peak pattern, followed by diseases of the heart, and then malignant neoplasms. The dip/peak mortality pattern does not appear in various non-Chinese control groups. The statistical significance of the findings was demonstrated with linear and curvilinear regression analysis and with two nonparametric tests. After testing alternative explanations for the findings, we concluded that the dip/peak pattern occurs because death can be briefly postponed until after the occurrence of a significant occasion.
This paper presents evidence that there are fewer deaths than expected before three ceremonial occasions: the birthday, Presidential elections, and the Jewish Day of Atonement. The investigation of mortality before the birthday was based on biographical information on 1,333 famous persons; official vital statistics tables were used to study mortality before the other two occasions. Alternative explanations of our findings are examined; the evidence suggests that the dip in deaths before ceremonies results from some persons' postponement of death. These results are interpreted in terms of Durkheim's discussion of social integration and ceremonies.
Notifications of diseases, entries in a hospital, injuries due to accidents, etc., are frequently collected in fixed equally spaced intervals. Such observations are likely to be dependent. In environmental medicine, where series such as daily concentrations of pollutants are collected and analysed, it is evident that dependence of consecutive measurements may be important. A high concentration of a pollutant today has a certain 'inertia', i.e. a tendency to be high tomorrow as well. Dependence of consecutive observations may be equally important when data such as blood glucose are recorded within a single patient. ARIMA models (autoregressive integrated moving average models, Box-Jenkins models), which allow the stochastic dependence of consecutive data to be modelled, have become well established in such fields as economics. This article reviews basic concepts of Box-Jenkins modelling. The methods are illustrated by applications. In particular, the following topics are presented: the ARIMA model, transfer function models (assessment of relations between time series) and intervention analysis (assessment of changes of time series).
Both acute and chronic use of alcohol are associated with suicidal behavior. However, the differing relationship of each component of alcohol use and possible causal mechanisms remain unclear.
This article reviews and summarizes associations between acute alcohol consumption (with and without intoxication) and suicidal behavior (both completed suicide and suicide attempts) among adults 19 years and older, as presented in literature published between 1991 and 2001. Possible mechanisms and methodologic challenges for evaluating the association are also discussed. An application of a research design (the case-crossover study) that has the potential for addressing the effects of acute alcohol use over and above usual or chronic use is presented.
The majority of articles reviewed were restricted to descriptive studies that documented the prevalence of suicide completers or attempters who tested positive for alcohol use. A wide range of alcohol-positive cases were found for both completed suicide (10-69%) and suicide attempts (10-73%). Common methodologic limitations included the lack of control groups (for evaluating risk conferred by alcohol use), selection and ascertainment bias, and small sample sizes. The results of the case-crossover pilot study indicated substantially higher risk of suicide during or shortly after use of alcohol compared with alcohol-free periods.
Although there is a substantial literature of published studies on acute alcohol use and suicidal behavior, the majority of studies focus on completed suicide and report prevalence estimates. Findings from such studies are subject to several possible sources of bias and have not advanced our knowledge of mechanisms in the association between acute alcohol use and suicidal behavior. The case-crossover design may help to overcome some limitations of these studies and facilitate evaluation of associations and possible causal mechanisms by which acute alcohol use is linked to suicidal behavior.
Articles in the medical literature and lay press have supported a belief that individuals, including those dying of cancer, can temporarily postpone their death to survive a major holiday or other significant event, but results and effects have been variable.
To determine whether, for the patient dying of cancer, a "death takes a holiday" effect showing a reduction in deaths in the week before a significant event was associated with Christmas, the US holiday of Thanksgiving, or the date of the individual's birthday.
Analysis of death certificate data for all 1,269,474 persons dying in Ohio from 1989-2000, including 309,221 persons dying with cancer noted as the leading cause of death.
We measured the total number of cancer deaths in the 2 weeks centered on the event of interest and the proportion of these deaths that occurred in the week before the event to determine whether this proportion was significantly different from 0.5 by using an exact binomial test.
The proportion of persons dying of cancer in the week before Christmas, Thanksgiving, and the individual's birthday was not significantly different from the proportion dying in the week after the event (P = .52, .26, and .06, respectively). However, among black individuals there was an increase in cancer deaths in the week before Thanksgiving (P = .01), whereas women showed an increase in cancer deaths in the week before their birthday (P = .05). There was no statistically significant excess of deaths in the postevent week in any subgroup.
We found no evidence, in contrast to previous studies, that cancer patients are able to postpone their deaths to survive significant religious, social, or personal events.
Several researchers have claimed that a connection exists between deathdays and birthdays or other special occasions. Some people, especially women, appear to postpone death until after their birthdays or other special occasions to celebrate these events one more time. The purpose of this retrospective study was to determine if a deathday-birthday or deathday-Christmas connection was evident in cancer-related deaths. Data were obtained from the Rochester (New York) Regional Tumor Registry (N = 2566), the New York State Cancer Registry (N = 50,562), and the Ohio State Department of Health (N = 73,907) for three samples of individuals who died from cancer. Only a few deathday-birthday connections were evident. However, these connections were not replicated across states and years.
To examine the influence of birthdays on the onset and course of vascular events such as stroke, TIA, and acute myocardial infarction (AMI).
This population-based study included all emergency department (ED) admissions due to ischemic stroke, TIA, or AMI from April 2002 to March 2004 in Ontario, Canada. All cases were identified through the National Ambulatory Care Reporting System. Calculations of daily and weekly numbers of events were centered on the patient's birthday and the week of the birthday. Statistical analyses include binomial tests and logistic regression.
During the study period, there were 24,315 ED admissions with acute stroke, 16,088 with TIAs, and 29,090 with AMI. The observed number of vascular events during the birthday was higher than the expected daily number of visits for stroke (87 vs 67; p = 0.009), TIA (58 vs 44; p = 0.02), and AMI (97 vs 80; p = 0.027) but not for selected control conditions (asthma, appendicitis, head trauma). Vascular events were more likely to occur on birthday (242 vs 191; odds ratio [OR] = 1.27). No significant differences were observed during the birthday week for any of the conditions. Multivariate logistic regression showed that birthday vascular events were more likely to occur in patients with a history of hypertension (OR = 1.88; 95% CI 1.09 to 3.24). Sensitivity analyses with alternative definitions of birthday week did not alter the results.
Stress associated with birthdays may trigger vascular events in patients with predisposing conditions.
Several studies have reported evidence that death is often associated with significant ceremonial events like holidays and birthdays, a phenomenon variously called the "anniversary reaction," "holiday effect," "birthday blues," etc. This article, using data from the social security death index (SSDI), shows how evidence for this phenomenon can arise from an artifact in death records called "death heaping" in which certain dates, usually the 1st or the 15th, are systematically entered for missing data. In this study, each calendar birthday for the year 2003 in the SSDI was paired with either the 14th or 15th calendar day of death. For each pairing, there were more deaths for the 15th than the 14th. A second analysis compared the correspondence between dying on the same day of the month as one's birthday for the first 28 days of the month. A significantly greater number of birth-death combinations occurred for the 1st and 15th day of the month. These results suggest that studies of "anniversary reactions" should conduct preliminary evaluations to consider the possibility that statistically significant relationships do not arise from "death heaping."
Die Sterblichkeitsstatistik in der Schweiz Datenqualität der Todesursachen und der Berufsbezeichnungen Amtliche Statistik der Schweiz, Nr. 155 [Mortality statistics of Switzerland. Data quality of the causes of death and the professions
- Minder Ce
- Zingg
Minder CE, Zingg W. Die Sterblichkeitsstatistik in der Schweiz. Datenqualität der Todesursachen und der Berufsbezeichnungen. Amtliche Statistik der Schweiz, Nr. 155 [Mortality statistics of Switzerland. Data quality of the causes of death and the professions. Official Statistics of Switzerland, No 155]. Bern (Switzerland): Bundesamt für Statistik; 1989.
Time-series analysis. Cambridge (UK)
- Jm Gottman
Gottman JM. Time-series analysis. Cambridge (UK): Cambridge University Press; 1981.
Einführung in die Zeitreihenanalyse
- B Schmitz
Schmitz B. Einführung in die Zeitreihenanalyse [Introduction in time series
analysis].
- V Ajdacic-Gross
V. Ajdacic-Gross et al. / Annals of Epidemiology 22 (2012) 603e606
Amtliche Statistik der Schweiz, Nr. 155 [Mortality statistics of Switzerland. Data quality of the causes of death and the professions
- C E Minder
- W Zingg
Minder CE, Zingg W. Die Sterblichkeitsstatistik in der Schweiz. Datenqualität
der Todesursachen und der Berufsbezeichnungen. Amtliche Statistik der
Schweiz, Nr. 155 [Mortality statistics of Switzerland. Data quality of the causes
of death and the professions. Official Statistics of Switzerland, No 155]. Bern
(Switzerland): Bundesamt für Statistik; 1989.
Time-series analysis
- J M Gottman
Gottman JM. Time-series analysis. Cambridge (UK): Cambridge University
Press; 1981.