Birthweight and the risk of childhood-onset type 1 diabetes: a meta-analysis of observational studies using individual patient data.

AIMS/HYPOTHESIS: We investigated whether children who are heavier at birth have an increased risk of type 1 diabetes. METHODS: Relevant studies published before February 2009 were identified from literature searches using MEDLINE, Web of Science and EMBASE. Authors of all studies containing relevant data were contacted and asked to provide individual patient data or conduct pre-specified analyses. Risk estimates of type 1 diabetes by category of birthweight were calculated for each study, before and after adjustment for potential confounders.Meta-analysis techniques were then used to derive combined ORs and investigate heterogeneity between studies. RESULTS: Data were available for 29 predominantly European studies (five cohort, 24 case-control studies), including 12,807 cases of type 1 diabetes. Overall, studies consistently demonstrated that children with birthweight from 3.5 to 4 kg had an increased risk of diabetes of 6% (OR 1.06 [95% CI 1.01-1.11]; p=0.02) and children with birthweight over 4 kg had an increased risk of 10% (OR 1.10 [95% CI 1.04-1.19]; p=0.003), compared with children weighing 3.0 to 3.5 kg at birth. This corresponded to a linear increase in diabetes risk of 3% per 500 g increase in birthweight (OR 1.03 [95% CI 1.00-1.06]; p=0.03). Adjustments for potential confounders such as gestational age, maternal age, birth order, Caesarean section, breastfeeding and maternal diabetes had little effect on these findings. CONCLUSIONS/INTERPRETATION: Children who are heavier at birth have a significant and consistent, but relatively small increase in risk of type 1 diabetes.

Investigation of spatial clustering from individually matched case-control studies.

The paper demonstrates how existing theory to assess spatial clustering based on second-moment properties of a labelled point process can be adapted to matched case-control studies. The null hypothesis that cases are a random sample from the superposition of cases and controls is replaced by the hypothesis that each case is a random sample from the set consisting of itself and its k matched controls. We compare the proposed test with other tests of spatial clustering, and describe an application to data on childhood diabetes in Yorkshire, England.

Individual inconsistency across measures of inhibition: an investigation of the construct validity of inhibition in older adults.

Inhibition is a central construct to the frontal lobe theory of ageing, yet its construct validity remains unproven. Furthermore, age effects on measures of inhibition are often reported without adequate control for the effects of global slowing on performance. We investigated inhibitory function in older adults in two experiments. In Experiment 1, 49 people with ages between 59 and 86 (mean=70 years 9 months S.D.=7.54) completed four analogues of the Stroop interference paradigm. To control for global slowing and to enable comparisons across all measures, we used a random effects model based on log-transformed response times. Age did not contribute significantly to the model and the estimated correlation between tasks was not significant. In Experiment 2, 33 people with ages between 62 and 86 (mean=73 years 4 months, S.D.=6.57) were compared on two measures of Stroop-like interference which were very similar in surface task demands. Age did not contribute significantly to the model but the estimated correlation between tasks was robust (r=0.714). We conclude that age may make little contribution to inhibitory function independently of other factors such as speed and intelligence. Second, that the level of individual consistency in the performance of measures of inhibition will depend on the similarity of the tasks used.

Second-order analysis of space-time clustering.

We consider the problem of detecting and describing space-time interaction in point process data. We extend existing second-order methods for purely spatial point process data to the spatial-temporal setting. This extension allows us to estimate space-time interaction as a function of spatial and temporal separation, and provides a useful reinterpretation of a popular test, due to Knox, for space-time interaction. Applications to simulated and real data indicate the method's potential.

Second-order analysis of inhomogeneous spatial point processes using case-control data.

Methods for the statistical analysis of stationary spatial point process data are now well established, methods for nonstationary processes less so. One of many sources of nonstationary point process data is a case-control study in environmental epidemiology. In that context, the data consist of a realization of each of two spatial point processes representing the locations, within a specified geographical region, of individual cases of a disease and of controls drawn at random from the population at risk. In this article, we extend work by Baddeley, Møller, and Waagepetersen (2000, Statistica Neerlandica54, 329-350) concerning estimation of the second-order properties of a nonstationary spatial point process. First, we show how case-control data can be used to overcome the problems encountered when using the same data to estimate both a spatially varying intensity and second-order properties. Second, we propose a semiparametric method for adjusting the estimate of intensity so as to take account of explanatory variables attached to the cases and controls. Our primary focus is estimation, but we also propose a new test for spatial clustering that we show to be competitive with existing tests. We describe an application to an ecological study in which juvenile and surviving adult trees assume the roles of controls and cases.

Effect and duration of lung volume reduction surgery: mid-term results of the Brompton trial.

UNLABELLED: Although many studies have reported improvement in lung function following LVRS, the magnitude of improvement and subsequent decline has not been evaluated against medical therapy after the second year. METHODS: Existing pulmonary function records were collapsed for ech participant since randomisation from Brompton LVRS trial cohort. Longitudinal data analysis was used to profile th history of medically treated patients and the effect of LVRS. RESULTS: Pulmonary function results were collated from survivors over a median of 25 (17 to 39) months. The estimated immediate increase in mean FEV1, following surgery was +0.2591 (0.179, 0.339), with a rate of change of -0.0051 (-0.009, -0.001) per month compared to medical therapy (p < 0.001). The changes in the secondary outcome measures (LVRS compared to medical therapy) were an increase in FVC (p = 0.004), decrease in RV (p < 0.001) and TLC (p < 0.001), with differences that were maintained over time. The initial reduction in RV/TLC ration was sustained (p < 0.001), but the estimated initial increase in peak flow was accompanied by a gradual decline that was not statistically significant (p = 0.062). KCOc showed no immediate change, but there was a gradual sustained increase with time (p = 0.009). Mean oxygen saturations improved and continued to do so compared to patients on medical therapy (p = 0.001). CONCLUSIONS: The immediate increase in FEV1 is not sustained, although the mechanical improvements of LVRS on increasing FVC, reducing both the RV and RV/TLC ratio, appear to be maintained. The important benefits of LVRS may be the gradual and sustained increase in transfer factor accompanied by improved oxygen saturations.

Second-order analysis of spatial clustering for inhomogeneous populations.

Motivated by recent interest in the possible spatial clustering of rare diseases, the paper develops an approach to the assessment of spatial clustering based on the second-moment properties of a labelled point process. The concept of no spatial clustering is identified with the hypothesis that in a realisation of a stationary spatial point process consisting of events of two qualitatively different types, the type 1 events are a random sample from the superposition of type 1 and type 2 events. A diagnostic plot for estimating the nature and physical scale of clustering effects is proposed. The availability of Monte Carlo tests of significance is noted. An application to published data on the spatial distribution of childhood leukaemia and lymphoma in North Humberside is described.

A mixture model for sucking patterns of breast-fed infants.

Previous studies of sucking patterns have mainly been on bottle-fed babies and have assumed that the babies' sucks occur within bursts separated by gaps of predetermined minimum length which is fixed over the feed. This study considers babies that are breast-fed, a more complex and natural process than bottle-feeding, and develops a more sophisticated model for the pattern of bursts and gaps which allows the parameters of the process to vary over the feed. We consider data from four breast feeds of each of 32 babies. We develop a two-component mixture model based on an underlying Markov chain model for the switching between bursts and gaps. We use the model to provide summary statistics for each feed and give estimates of the normal range of each of the model's parameters.

Asymptomatic primary biliary cirrhosis: clinical features, prognosis, and symptom progression in a large population based cohort.

BACKGROUND: Many patients with primary biliary cirrhosis (PBC) are asymptomatic at the time of diagnosis. However, because most studies of asymptomatic PBC have been small and from tertiary centres, asymptomatic PBC remains poorly characterised. AIMS: To describe the features and progression of initially asymptomatic PBC patients. METHODS: Follow up by interview and note review of a large geographically and temporally defined cohort of patients with PBC, collected by multiple methods. RESULTS: Of a total of 770 patients, 469 (61%) were asymptomatic at diagnosis. These patients had biochemically and histologically less advanced disease than initially symptomatic patients. Median survival was similar in both groups (9.6 v 8.0 years, respectively) possibly due to excess of non-liver related deaths in asymptomatic patients (31% v 57% of deaths related to liver disease). Survival in initially asymptomatic patients was not affected by subsequent symptom development. By the end of follow up, 20% of initially asymptomatic patients had died of liver disease or required liver transplantation. The majority of initially asymptomatic patients developed symptoms of liver disease if they were followed up for long enough (Kaplan-Meier estimate of proportion developing symptoms: 50% after five years, 95% after 20 years). However, 45% of patients remained asymptomatic at the time of death. CONCLUSIONS: Although asymptomatic PBC is less severe at diagnosis than symptomatic disease, it is not associated with a better prognosis, possibly due to an increase in non-hepatic deaths. The reasons for this are unclear but may reflect confounding by other risk factors or surveillance bias. These findings have important implications for future treatment strategies.

Type 1 diabetes mellitus in childhood: a matched case control study in Lancashire and Cumbria, UK.

AIMS: The aim of the study was to identify environmental risk factors for insulin-dependent diabetes mellitus (Type 1 DM) in childhood. METHODS: A matched case-control study of Type 1 DM conducted in Lancashire and Cumbria, UK, using a structured interview. Cases (n=196, participation rate 83%) were children under 16 years of age diagnosed prior to October 1998 and attending diabetic clinics. Controls (n=381) were healthy children from the community matched by gender and by age (within a few days of birth). The data were analysed by logistic regression using the technique of Breslow and Day for matched case control studies. RESULTS: The multivariate regression model showed that the following factors were significantly associated with the risk of developing Type 1 DM (odds ratio, 95% confidence intervals): sharing a room with a sibling (0.458, 0.290-0.721), social contact with other children when aged 6-11 months (0.439, 0.256-0.752), consumption of sugary food (0.080, 0.024-0.261), parental insulin dependent diabetes mellitus (10.651, 3.086-36.761), maternal thyroid disease (4.861, 1.681-14.058), consuming more than one pint of milk per day prior to school entry (0.498, 0.310-0.802), maternal smoking during pregnancy (0.373, 0.218-0.636), a father with no academic qualifications (0.504, 0.278-0.913), maternal age at time of birth (0.900, 0.854-0.948), maternal infections in pregnancy (2.453, 1.011-5.948), other maternal illnesses or conditions in pregnancy (2.007, 1.139-3.535), belonging to an Asian family (0.104, 0.028-0.394), and regular contact with pets and other animals (0.552, 0.309-0.987). CONCLUSION: Many of the results are consistent with the hygiene hypothesis which links improved living standards with decreased exposure to microorganisms and increased risk of immune mediated disease in childhood. These findings challenge the idea that improved hygiene acts exclusively through a Th2 mechanism leading to atopic disease as Type 1 DM is mediated by a Th1 reaction. The association with maternal smoking could be due to recall bias but a causal link cannot be excluded with confidence.

The geographical distribution of primary biliary cirrhosis in a well-defined cohort.

The incidence of primary biliary cirrhosis (PBC) varies widely between regions. However, little is known about variation within regions and the degree to which this may reflect environmental risk factors. The aim of this study was to describe the spatial distribution of cases of PBC in a defined region of Northeast England over a defined period, and to assess the magnitude of any departure from random spatial distribution. Seven hundred seventy patients with established PBC were identified in a previous comprehensive case finding study. A total of 3,044 control locations were randomly selected from postcode (zip code) data weighted for number of drop off points per postcode. Geographical analysis was performed by testing both for spatial variation in risk and local clustering by using previously described point process methods. Both tests used the same null hypothesis that risk of disease does not vary spatially and cases occur independently of each other. Statistically significant spatial variations in risk were found in the whole study region (P <.001) and in the major urban area within the region (P <.004). Risk was higher in the urban area of Tyneside than in the surrounding rural area. Within the rural area, spatial variation in risk was equivocal (P =.012), but there was significant (P =.001) clustering of cases (estimated average cluster effect approximately 10 excess cases within a 7-km radius). PBC occurred to a density of 10.7 cases/km(2) in the highest risk areas. In conclusion, PBC is unevenly distributed in Northeast England. This may reflect one or more environmental risk factors in its etiology.