Some statistical methods for investigating the date of birth as a disease indicator

A novel Bayesian functional spatial partitioning method with application to prostate cancer lesion detection using MRI

Spatial partitioning methods correct for nonstationarity in spatially related data by partitioning the space into regions of local stationarity. Existing spatial partitioning methods can only estimate linear partitioning boundaries. This is inadequate for detecting an arbitrarily shaped anomalous spatial region within a larger area. We propose a novel Bayesian functional spatial partitioning (BFSP) algorithm, which estimates closed curves that act as partitioning boundaries around anomalous regions of data with a distinct distribution or spatial process. Our method utilizes transitions between a fixed Cartesian and moving polar coordinate system to model the smooth boundary curves using functional estimation tools. Using adaptive Metropolis-Hastings, the BFSP algorithm simultaneously estimates the partitioning boundary and the parameters of the spatial distributions within each region. Through simulation we show that our method is robust to shape of the target zone and region-specific spatial processes. We illustrate our method through the detection of prostate cancer lesions using magnetic resonance imaging.

Tests and diagnostic plots for detecting lack-of-fit for circular-linear regression models

Regression diagnostics and lack-of-fit tests mainly focus on linear-linear regression models. When the design points are distributed on the circumference of a circle, difficulties arise as there is no natural starting point or origin. Most classical lack-of-fit tests require an arbitrarily chosen origin, but different choices may result in different conclusions. We propose a graphical diagnostic tool and a closely related lack-of-fit test, which does not require a natural starting point. The method is based on regional residuals which are defined on arcs of the circle. The graphical method formally locates and visualizes subsets of poorly fitting observations on the circle. A data example from the food technology is used to point out the before-mentioned problems with conventional lack-of-fit tests and to illustrate the strength of the methodology based on regional residuals in detecting and localizing departures from the no-effect hypothesis. A small simulation study shows a good performance of the regional residual test in case of both global and local deviations from the null model. Finally, the ideas are extended to the case of more than one predictor variable.

Seasonal birth patterns in myositis subgroups suggest an etiologic role of early environmental exposures

OBJECTIVE

To evaluate whether seasonal early environmental exposures might influence later development of autoimmune disease, by assessing distributions of birth dates in groups of patients with idiopathic inflammatory myopathies (IIMs).

METHODS

We assessed birth patterns in groups of patients with juvenile-onset IIM (n = 307) and controls (n = 3,942) who were born between 1970 and 1999, and in groups of patients with adult-onset IIM (n = 668) and controls (n = 6,991) who were born between 1903 and 1982. Birth dates were analyzed as circular data. Seasonal clustering was assessed by the Rayleigh test, and differences between groups by a rank-based uniform scores test.

RESULTS

The overall birth distributions among patients with juvenile IIM and among patients with adult IIM did not differ significantly from those among juvenile and adult controls, respectively. Some subgroups of patients with juvenile IIM had seasonal birth distributions. Hispanic patients with juvenile-onset IIM had a seasonal birth pattern (mean birth date October 16) significantly different from that of Hispanic controls (P = 0.002), who had a uniform birth distribution, and from that of non-Hispanic patients with juvenile-onset IIM (P < 0.001), who had a mean birth date of May 2. Juvenile dermatomyositis patients with p155 autoantibody had a birth distribution that differed significantly from that of p155 antibody-negative juvenile dermatomyositis patients (P = 0.003). Juvenile IIM patients with the HLA risk factor allele DRB1*0301 had a birth distribution significantly different from those without the allele (P = 0.021). Similar results were observed for juvenile and adult IIM patients with the linked allele DQA1*0501, versus juvenile and adult IIM patients without DQA1*0501, respectively. No significant patterns in birth season were found in other subgroups.

CONCLUSION

Birth distributions appear to have stronger seasonality in juvenile than in adult IIM subgroups, suggesting greater influence of perinatal exposures on childhood-onset illness. Seasonal early-life exposures may influence the onset of some autoimmune diseases later in life.

Season of death and birth predict patterns of mortality in Burkina Faso

BACKGROUND

Mortality in developing countries has multiple causes. Some of these causes are linked to climatic conditions that differ over the year. Data on season-specific mortality are sparse.

METHODS

We analysed longitudinal data from a population of approximately 35,000 individuals in Burkina Faso. During the observation period 1993-2001, a total number of 4,098 deaths were recorded. The effect of season on mortality was investigated separately by age group as (i) date of death and (ii) date of birth. For (i), age-specific death rates by month of death were calculated. The relative effect of each month was assessed using the floating relative risk method and modelled continuously. For (ii), age-specific death rates by month of birth were calculated and the mean date of birth among deaths and survivors was compared.

RESULTS

Overall mortality was found to be consistently higher during the dry season (November to May). The pattern was seen in all age groups except in infants where a peak was seen around the end of the rainy season. In infants we found a strong association between high mortality and being born during the time period September to February. No effect was seen for the other age groups.

CONCLUSIONS

The observed excess mortality in young children at or around the end of the rainy season can be explained by the effects of infectious diseases and, in particular, malaria during this time period. In contrast, the excess mortality seen in older children and adults during the early dry season remains largely unexplained although specific infectious diseases such as meningitis and pneumonia are possible main causes. The association between high infant mortality and being born at around the end of the rainy season is probably explained by most of the malaria deaths in areas of high transmission intensity occurring in the second half of infancy.