Seasonality of type 1 diabetes mellitus in children and its modification by weekends and holidays: retrospective observational study.

BACKGROUND: Diagnoses of type 1 insulin dependent diabetes mellitus are generally more common in winter, although this seasonal pattern has not been observed in children of preschool age (0-4 years) or in all countries. AIMS: To confirm the persistence of seasonality and the influence of age, holidays, and weekends. METHODS: We extracted data on date of birth, date of presentation, age, and sex of children diagnosed with diabetes and registered with the Scottish Study Group for the Care of Diabetes in the Young. Cosinor analysis was applied to monthly and mid-monthly data. Two sample Z tests were used to compare the epochs 1984-1992 and 1993-2001. RESULTS: Some 4517 children between 0 and 14 years of age (2407 male and 2110 female) presented with IDDM between 1 January 1984 and 31 December 2001. Seasonality was evident in children above 4 years of age with amplitudes of 19.5-25.7% and peaks between mid December and mid January. Presentation was strongly influenced by weekends and holiday periods, with reduced presentations in December compared with November and January, and with the lowest presentations in July (the main Scottish holiday month). Using mid-month to mid-month data did not change the overall seasonality but did improve the fits for cosinor analysis. Mondays and Fridays were the most common days for presentation. CONCLUSION: Initial presentation of IDDM in Scotland follows a stable seasonal pattern in all but the youngest children with lower rates of presentation in holiday periods and at weekends for all age groups.

Higher incidence of SIDS at weekends, especially in younger infants.

BACKGROUND: An increased incidence of sudden infant death syndrome (SIDS) at weekends has been previously noted, although it has not been found in all studies. Where a weekend effect has been found it is most prominent on Sundays. AIM: To identify the magnitude of this effect in British data before and after the fall in prevalence associated with the "back to sleep" campaign. METHODS: Aggregated data from England Wales, Scotland, and Northern Ireland for the years 1986-98 during which there were 12,762 deaths from SIDS were examined. RESULTS: Comparing the periods before (1986-90) and after (1993-98) the fall in incidence associated with the "back to sleep" campaign, the percentage of infants dying at the weekend in the later period (32.24%) was slightly greater than for the earlier period (31.30%), although this difference was not statistically significant. In both periods the observed weekend incidences were significantly higher than the expected weekend percentage of 28.6%. Although the proportion of infants aged < or =4 months and > or =5 months at death did not change between the two periods, the weekend effect was more marked in the younger infants, particularly in the 1993-98 period. There was no consistent Saturday effect. In 1993-98 deaths appeared to be more evenly distributed throughout the week for infants > or =5 months. CONCLUSION: The excess of SIDS at weekends still appears to be present despite the overall fall in prevalence associated with the "back to sleep" campaign. Increased risks at weekends might have implications for support to high risk families.

Multivariate outlier detection applied to multiply imputed laboratory data.

In clinical laboratory safety data, multivariate outlier detection methods may highlight a patient whose laboratory measurements do not follow the same pattern of relationships as the majority of patients, although their individual measurements are not found to be outlying when considered one at a time. Missing data problems are often dealt with by imputing a single value as an estimate of the missing value. The completed data set may then be analysed using traditional methods. A disadvantage of using single imputation is the underestimation of variability, with a corresponding distortion of power in hypothesis testing. Multiple imputation methods attempt to overcome this problem, and in this paper a study is described which considers the application of multivariate outlier detection methods to multiply imputed clinical laboratory safety data sets. Three different proportions of missing data are generated in laboratory data sets of dimensions 4, 7, 12 and 30, and a comparison of eight multiple imputation methods is carried out. Two outlier detection techniques, Mahalanobis distance and generalized principal component analysis, are applied to the multiply imputed data sets, and their performances are discussed. Measures are introduced for assessing the accuracy of the missing data results, depending on which method of analysis is used.

Seasonality, but not prevalence of sudden infant death syndrome varies by region in mainland Britain.

This study aimed to investigate whether seasonal variation in day length contributed to winter/summer variation in sudden infant death syndrome (SIDS) at different latitudes in mainland Britain. Over 11 yrs 13,973 deaths were studied. Using appropriate analytic techniques a sine curve was fitted to monthly rates with the amplitude indicating magnitude of seasonal change. The rate of SIDS per 1,000 live births was the same (1.73) in the north as in the south. The amplitude was a quarter less in the north (41.3%) than in the south (54.2%) (p<0.001). While annual rates did not differ, the within year distribution did. The findings for seasonality of SIDS births were similar (amplitudes: north 213%, south 32.3%). Correlations were made between SIDS amplitude and individual environmental factors, particularly temperature and day length. These complex issues, while reported briefly, do not allow firm conclusions. In the north the winter day length is shorter, sunshine hours are less and temperature is lower, but the winter increment in SIDS is less. The extent of seasonal variation of sudden infant death syndrome is greater in the south as compared with the colder, darker north but this has no effect on sudden infant death syndrome rates. Changing photoperiod by latitude, amongst other environmental influences, may hold clues to the aetiology of sudden infant death syndrome.

Seasonality of sudden infant death syndrome (SIDS) by age at death.

Seasonality of sudden infant death syndrome (SIDS) is a well-established epidemiological finding. The purpose of the study was to determine whether this feature varied significantly with age at death. In total, 13990 cases of SIDS in Scotland, England and Wales during 1982-1992 were studied by age group at death. Seasonality was established by fitting a sinusoidal curve and for each set of monthly data the peak position in the year and its magnitude were determined. Weighted regression revealed significant differences in peak position and amplitude of seasonal variation between those dying at < or = 4 months and those aged > or = 5 months at death. Those infants in the younger age group were more likely to die earlier in the winter months and had a smaller variation in seasonality. The peak (acrophase) months were January for < or = 4 months and February for > or = 5 months at death. Weighted regressions of peak position and amplitude on age at death had p-values of <0.001 and <0.01, respectively. A log linear model relating SIDS incidence to month of birth, month of death and age was able to explain some of these findings. The findings support the hypothesis that in SIDS there may be more than one infant cohort, each of which passes through a vulnerable developmental window at different ages.

Seasonality of sudden infant death syndrome in mainland Britain and Ireland 1985-95.

By the end of 1995 four years had passed since the dramatic fall in the incidence of sudden infant death syndrome (SIDS), following the "back to sleep" campaign. This time lag permitted a more definitive epidemiological reassessment than had been possible before. The extent of seasonal variation from 1992-5 fell by about half, occurring mainly as a single step down in 1992. The fall was relatively greater in winter than in summer. Before 1992 the extent of seasonal variation (amplitude) was greater in those age > or = 5 months compared with those aged < or = 4 months. Since 1992 the falls in incidence and amplitude have been greater in the younger group, suggesting that SIDS deaths in younger and older babies may have different causes.

Modelling memory in coal tits: an illustration of the EM algorithm.

Some species of birds are known to use memory to retrieve previously stored food. A series of experiments on one of those species, coal tits, investigated various aspects of such memory. For one particular experiment, a number of statistical models describing the memory were fitted. However, some of the data were unavoidably incomplete. The expectation maximization (EM) algorithm provides a means of incorporating the incomplete data into the fitting procedure.

Month of birth as an independent variable in the sudden infant death syndrome.

Well-known epidemiological features of sudden infant death syndrome (SIDS) are age at death and the increased numbers in winter. There are more SIDS deaths in late autumn/early winter and there is a seasonal rhythm of births with a peak in late summer and early autumn. The data set was 14033 SIDS deaths from Scotland, England and Wales over the 11 years 1982-92. Using log-linear models, which accounted for age at death and month of death, birth month was found to be a statistically significant risk factor for SIDS independent of age at death and winter environment (P < 0.001). Although winter season had the largest effect (relative risk 2.7 in January compared with August), the independent effect of birth month was of clinical as well as statistical significance with a relative risk for August births of 1.37 compared with those born in April. The analysis of each birth month cohort revealed a change in age distribution with infants born in early winter (December) dying at a younger age (mean 108 days) than those born in midsummer (June) (mean 146 days). Although winter season and age are the most influential factors, the substantial effect of month of birth requires explanation and points to as yet unidentified environmental influences during pregnancy.

Seasonal variation in sudden infant death syndrome and bronchiolitis--a common mechanism?

The purpose of this study was to compare epidemiologic features of sudden infant death syndrome (SIDS) and bronchiolitis and to apply statistical examination in order to examine the hypothesis that similar mechanisms could be at work in both conditions. The setting was Scotland from 1982 through 1990. We compared 1,211 deaths from SIDS with 10,058 hospital admissions for bronchiolitis in infancy. The comparisons included age, sex ratio, and seasonality. The sex ratios were similar (SIDS [M:F]:1.61:1; bronchiolitis: 1.63:1), but age distribution was different (chi 2 = 104.6, p < 0.001). When monthly rates throughout the year were compared using correlation of residuals from average season variation and by autocorrelation of residual series, no significant relationships were found between the two conditions (r2 = 0.0004). Once the seasonal pattern common to both conditions was accounted for, SIDS was not autocorrelated between months whereas bronchiolitis exhibited a high level of autocorrelation indicating an epidemic pattern for the latter but not for the former. While a common seasonal variation may indicate some shared etiologic factors associated with winter season, the two conditions do not appear to be closely related. The hypothesis that a common host susceptibility is at work is not supported. Further investigations of seasonal influences are warranted.