-
[show abstract]
[hide abstract]
ABSTRACT: Space-time modeling of small area data is often used in epidemiology for mapping chronic disease rates and by government statistical agencies for producing local estimates of, for example, unemployment or crime rates. Although there is typically a general temporal trend, which affects all areas similarly, abrupt changes may occur in a particular area, e.g. due to emergence of localized predictors/risk factor(s) or impact of a new policy. Detection of areas with "unusual" temporal patterns is therefore important as a screening tool for further investigations. In this paper, we propose BaySTDetect, a novel detection method for short-time series of small area data using Bayesian model choice between two competing space-time models. The first model is a multiplicative decomposition of the area effect and the temporal effect, assuming one common temporal pattern across the whole study region. The second model estimates the time trends independently for each area. For each area, the posterior probability of belonging to the common trend model is calculated, which is then used to classify the local time trend as unusual or not. Crucial to any detection method, we provide a Bayesian estimate of the false discovery rate (FDR). A comprehensive simulation study has demonstrated the consistent good performance of BaySTDetect in detecting various realistic departure patterns in addition to estimating well the FDR. The proposed method is applied retrospectively to mortality data on chronic obstructive pulmonary disease (COPD) in England and Wales between 1990 and 1997 (a) to test a hypothesis that a government policy increased the diagnosis of COPD and (b) to perform surveillance. While results showed no evidence supporting the hypothesis regarding the policy, an identified unusual district (Tower Hamlets in inner London) was later recognized to have higher than national rates of hospital readmission and mortality due to COPD by the National Health Service, which initiated various local enhanced services to tackle the problem. Our method would have led to an early detection of this local health issue.
Biostatistics 03/2012; 13(4):695-710. · 2.14 Impact Factor
-
Data Mining Workshops (ICDMW), 2011 IEEE 11th International Conference on, Vancouver, BC, Canada, December 11, 2011; 01/2011
-
Epidemiology (Cambridge, Mass.) 01/2011; 22(1):36-9. · 5.51 Impact Factor
-
[show abstract]
[hide abstract]
ABSTRACT: This study aims to review epidemiologic evidence of the association between exposure to chlorination disinfection by-products (DBPs) and congenital anomalies. All epidemiologic studies that evaluated a relationship between an index of DBP exposure and risk of congenital anomalies were analyzed. For all congenital anomalies combined, the meta-analysis gave a statistically significant excess risk for high versus low exposure to water chlorination or TTHM (17%; 95% CI, 3-34) based on a small number of studies. The meta-analysis also suggested a statistically significant excess risk for ventricular septal defects (58%; 95% CI, 21-107), but based on only three studies, and there was little evidence of an exposure-response relationship. It was observed no statistically significant relationships in the other meta-analyses and little evidence for publication bias, except for urinary tract defects and cleft lip and palate. Although some individual studies have suggested an association between chlorination disinfection by-products and congenital anomalies, meta-analyses of all currently available studies demonstrate little evidence of such association.
Ciencia & saude coletiva 10/2010; 15 Suppl 2:3109-23.
-
[show abstract]
[hide abstract]
ABSTRACT: Indoor air pollution from solid fuel use is a significant risk factor for acute lower respiratory infections among children in sub-Saharan Africa. Interventions that promote a switch to modern fuels hold a large health promise, but their effective design and implementation require an understanding of the web of upstream and proximal determinants of household fuel use. Using Demographic and Health Survey data for Benin, Kenya and Ethiopia together with Bayesian hierarchical and spatial modelling, this paper quantifies the impact of household-level factors on cooking fuel choice, assesses variation between communities and districts and discusses the likely nature of contextual effects. Household- and area-level characteristics appear to interact as determinants of cooking fuel choice. In all three countries, wealth and the educational attainment of women and men emerge as important; the nature of area-level factors varies between countries. In Benin, a two-level model with spatial community random effects best explains the data, pointing to an environmental explanation. In Ethiopia and Kenya, a three-level model with unstructured community and district random effects is selected, implying relatively autonomous economic and social areas. Area-level heterogeneity, indicated by large median odds ratios, appears to be responsible for a greater share of variation in the data than household-level factors. This may be an indication that fuel choice is to a considerable extent supply-driven rather than demand-driven. Consequently, interventions to promote fuel switching will carefully need to assess supply-side limitations and devise appropriate policy and programmatic approaches to overcome them. To our knowledge, this paper represents the first attempt to model the determinants of solid fuel use, highlighting socio-economic differences between households and, notably, the dramatic influence of contextual effects. It illustrates the potential that multilevel and spatial modelling approaches hold for understanding determinants of major public health problems in the developing world.
Environmental Research 10/2010; 110(7):725-32. · 3.40 Impact Factor
-
[show abstract]
[hide abstract]
ABSTRACT: Maheswaran et al. (2006) analysed the effect of outdoor modelled NO(x) levels, classified into quintiles, on stroke mortality using a Poisson Bayesian hierarchical model with spatial random effects. An association was observed between higher levels of NO(x) and stroke mortality at the small area (enumeration district) level. As this model is framed in an ecological perspective, the relative risk estimates suffer from ecological bias. In this paper we use a different model specification based on Jackson et al. (2008), modelling the number of cases of mortality due to stroke as a binomial random variable where p(i) is the probability of dying from stroke in area i. The within-area variation in outdoor modelled NO(x) levels is used to determine the proportion of the population in area i falling into each of the five exposure categories in order to estimate the probability of an individual dying from stroke given the kth level of NO(x) exposure assuming a homogeneous effect across the study region. The inclusion of within-area variability in an ecological regression model has been demonstrated to help reduce the ecological bias (Jackson et al., 2006, 2008). Revised estimates of relative risk are obtained and compared with previous estimates.
Spatial and spatio-temporal epidemiology. 07/2010; 1(2-3):123-31.
-
04/2010;
-
Statistics in Medicine 11/2009; 28(25):3081-2. · 1.88 Impact Factor
-
[show abstract]
[hide abstract]
ABSTRACT: This paper summarizes the epidemiological evidence for adverse health effects associated with disinfection by-products (DBPs) in drinking water and describes the potential mechanism of action. There appears to be good epidemiological evidence for a relationship between exposure to DBPs, as measured by trihalomethanes (THMs), in drinking water and bladder cancer, but the evidence for other cancers including colorectal cancer is inconclusive and inconsistent. There appears to be some evidence for an association between exposure to DBPs, specifically THMs, and little for gestational age/intrauterine growth retardation and, to a lesser extent, pre-term delivery, but evidence for relationships with other outcomes such as low birth weight, stillbirth, congenital anomalies and semen quality is inconclusive and inconsistent. Major limitations in exposure assessment, small sample sizes and potential biases may account for the inconclusive and inconsistent results in epidemiological studies. Moreover, most studies have focused on total THMs as the exposure metric, whereas other DBPs appear to be more toxic than the THMs, albeit generally occurring at lower levels in the water. The mechanisms through which DBPs may cause adverse health effects including cancer and adverse reproductive effects have not been well investigated. Several mechanisms have been suggested, including genotoxicity, oxidative stress, disruption of folate metabolism, disruption of the synthesis and/or secretion of placental syncytiotrophoblast-derived chorionic gonadotropin and lowering of testosterone levels, but further work is required in this area.
Philosophical Transactions of The Royal Society A Mathematical Physical and Engineering Sciences 11/2009; 367(1904):4043-76. · 2.77 Impact Factor
-
[show abstract]
[hide abstract]
ABSTRACT: BUGS is a software package for Bayesian inference using Gibbs sampling. The software has been instrumental in raising awareness of Bayesian modelling among both academic and commercial communities internationally, and has enjoyed considerable success over its 20-year life span. Despite this, the software has a number of shortcomings and a principal aim of this paper is to provide a balanced critical appraisal, in particular highlighting how various ideas have led to unprecedented flexibility while at the same time producing negative side effects. We also present a historical overview of the BUGS project and some future perspectives.
Statistics in Medicine 07/2009; 28(25):3049-67. · 1.88 Impact Factor
-
Mark J Nieuwenhuijsen,
Rachel Smith,
Spyros Golfinopoulos, Nicky Best,
James Bennett,
Gabriella Aggazzotti,
Elena Righi,
Guglielmina Fantuzzi,
Luca Bucchini,
Sylvaine Cordier,
Cristina M Villanueva,
Victor Moreno,
Carlo La Vecchia,
Cristina Bosetti,
Terttu Vartiainen,
Radu Rautiu,
Mireille Toledano,
Nina Iszatt,
Regina Grazuleviciene,
Manolis Kogevinas
[show abstract]
[hide abstract]
ABSTRACT: There appears to be very good epidemiological evidence for a relationship between chlorination by-products, as measured by trihalomethanes (THMs), in drinking water and bladder cancer, but the evidence for other cancers, including colorectal cancer appears to be inconclusive and inconsistent. There appears to be some evidence for a relationship between chlorination by-products, as measured by THMs, and small for gestational age (SGA)/intrauterine growth retardation (IUGR) and preterm delivery, but evidence for other outcomes such as low birth weight (LBW), stillbirth, congenital anomalies and semen quality appears to be inconclusive and inconsistent.The overall aim of the HIWATE study is to investigate potential human health risks (e.g. bladder and colorectal cancer, premature births, SGA, semen quality, stillbirth, congenital anomalies) associated with long-term exposure to low levels of disinfectants (such as chlorine) and DBPs occurring in water for human consumption and use in the food industry. The study will comprise risk-benefit analyses including quantitative assessments of risk associated with microbial contamination of drinking water versus chemical risk and will compare alternative treatment options. The outcome will be improved risk assessment and better information for risk management. The work is divided into different topics (exposure assessment, epidemiology, risk assessment and management) and studies.
Journal of Water and Health 07/2009; 7(2):185-207. · 1.37 Impact Factor
-
[show abstract]
[hide abstract]
ABSTRACT: Almost half of global child deaths due to acute lower respiratory infections (ALRIs) occur in sub-Saharan Africa, where three-quarters of the population cook with solid fuels. This study aims to quantify the impact of fuel type and cooking practices on childhood ALRI mortality in Africa, and to explore implications for public health interventions.
Early-release World Health Survey data for the year 2003 were pooled for 16 African countries. Among 32,620 children born during the last 10 years, 1455 (4.46%) were reported to have died prior to their fifth birthday. Survival analysis was used to examine the impact of different cooking-related parameters on ALRI mortality, defined as cough accompanied by rapid breathing or chest indrawing based on maternal recall of symptoms prior to death.
Solid fuel use increases the risk of ALRI mortality with an adjusted hazard ratio of 2.35 (95% CI 1.22 to 4.52); this association grows stronger with increasing outcome specificity. Differences between households burning solid fuels on a well-ventilated stove and households relying on cleaner fuels are limited. In contrast, cooking with solid fuels in the absence of a chimney or hood is associated with an adjusted hazard ratio of 2.68 (1.38 to 5.23). Outdoor cooking is less harmful than indoor cooking but, overall, stove ventilation emerges as a more significant determinant of ALRI mortality.
This study shows substantial differences in ALRI mortality risk among African children in relation to cooking practices, and suggests that stove ventilation may be an important means of reducing indoor air pollution.
Journal of epidemiology and community health 06/2009; 63(11):887-92. · 3.04 Impact Factor
-
[show abstract]
[hide abstract]
ABSTRACT: Chronic obstructive pulmonary disease (COPD) is an important cause of mortality with marked geographic variations in Great Britain. Additional factors beyond cigarette smoking are likely to influence these variations, but direct information on smoking by area is not readily available. We compared methods of jointly modeling the spatial distribution of mortality from COPD and lung cancer, using the latter as a proxy for smoking, to identify areas in which risk factors other than smoking may be important.
We obtained district-level mortality and population data for men aged 45 years or older in 1981-1999 in Great Britain. Three models were compared: Bayesian ecological regression using observed (model 1) or spatially smoothed (model 2) lung cancer standardized mortality ratio (SMR) as a smoking proxy, and bivariate regression (model 3) treating smoking as a spatial latent variable common to both diseases.
Model selection criteria favored models 2 and 3 over model 1. Between 9% (model 3) and 25% (model 2) of spatial variation in COPD mortality was estimated to be unrelated to smoking. After adjustment for lung cancer as a proxy for smoking, both models showed similar geographic patterns of higher COPD mortality in conurbation and mining areas, historically associated with heavy industry and higher air pollution levels.
Joint modeling of multiple diseases can be used to investigate geographic variations in risk. These models reveal patterns that are adjusted for the effects of shared area-level risk factors for which no direct data are available.
Epidemiology (Cambridge, Mass.) 04/2009; 20(3):400-10. · 5.51 Impact Factor
-
[show abstract]
[hide abstract]
ABSTRACT: One method of assessing biologic fertility is to measure time to pregnancy (TTP). Accidental pregnancies do not generate a valid TTP value and lead to nonrandom missing data if couples experiencing accidental pregnancies are more fertile than the general population. If factors affecting the rate of accidental pregnancies, such as availability of effective contraception and induced abortion, vary over time, then the result may be protection bias in the estimates of fertility time trends. Six European data sets were analyzed to investigate whether evidence of protection bias exists in TTP studies of fertility trends in Europe over the past 50 years. Couples experiencing accidental pregnancies tended to be more fertile than the general population. However, trends in accidental pregnancy rates were inconsistent across countries and were insufficient to produce substantial bias in fertility trends in simulated data. Where protection bias is suspected, the authors demonstrate use of 2 multiple imputation methods to generate realizations for the missing TTP values for accidental pregnancies. Simulation studies show that both methods successfully reduce or eliminate protection bias. The authors also demonstrate that standard sensitivity analyses for dealing with accidental pregnancies provide an upper bound on the extent of any bias.
American journal of epidemiology 02/2009; 169(3):285-93. · 5.59 Impact Factor
-
[show abstract]
[hide abstract]
ABSTRACT: We introduce a method for preventing unwanted feedback in Bayesian PKPD link models. We illustrate the approach using a simple example on a single individual, and subsequently demonstrate the ease with which it can be applied to more general settings. In particular, we look at the three 'sequential' population PKPD models examined by Zhang et al. (J Pharmacokinet Pharmacodyn 30:387-404, 2003; J Pharmacokinet Pharmacodyn 30:405-416, 2003), and provide graphical representations of these models to elucidate their structure. An important feature of our approach is that it allows uncertainty regarding the PK parameters to propagate through to inferences on the PD parameters. This is in contrast to standard two-stage approaches whereby 'plug-in' point estimates for either the population or the individual-specific PK parameters are required.
Journal of Pharmacokinetics and Biopharmaceutics 02/2009; 36(1):19-38. · 2.06 Impact Factor
-
Statistics and Computing. 01/2009; 19:395-408.
-
[show abstract]
[hide abstract]
ABSTRACT: Data in the social, behavioural and health sciences frequently come from observational studies instead of controlled experiments. In addition to random errors, observational data typically contain additional sources of uncertainty such as missing values, unmeasured confounders and selection biases. Also, the research question is often different from that which a particular source of data was designed to answer, and so not all relevant variables are measured. As a result, multiple sources of data are often necessary to identify the biases and to inform about different aspects of the research question. Bayesian graphical models provide a coherent way to connect a series of local submodels, based on different data sets, into a global unified analysis. We present a unified modelling framework that will account for multiple biases simultaneously and give more accurate parameter estimates than standard approaches. We illustrate our approach by analysing data from a study of water disinfection by-products and adverse birth outcomes in the UK. Copyright (c) 2009 Royal Statistical Society.
Journal of the Royal Statistical Society Series A. 01/2009; 172(3):615-637.
-
[show abstract]
[hide abstract]
ABSTRACT: The use of Bayesian hierarchical spatial models has become widespread in disease mapping and ecologic studies of health-environment associations. In this type of study, the data are typically aggregated over an extensive time period, thus neglecting the time dimension. The output of purely spatial disease mapping studies is therefore the average spatial pattern of risk over the period analyzed, but the results do not inform about, for example, whether a high average risk was sustained over time or changed over time.
We investigated how including the time dimension in disease-mapping models strengthens the epidemiologic interpretation of the overall pattern of risk.
We discuss a class of Bayesian hierarchical models that simultaneously characterize and estimate the stable spatial and temporal patterns as well as departures from these stable components. We show how useful rules for classifying areas as stable can be constructed based on the posterior distribution of the space-time interactions. We carry out a simulation study to investigate the sensitivity and specificity of the decision rules we propose, and we illustrate our approach in a case study of congenital anomalies in England.
Our results confirm that extending hierarchical disease-mapping models to models that simultaneously consider space and time leads to a number of benefits in terms of interpretation and potential for detection of localized excesses.
Environmental Health Perspectives 09/2008; 116(8):1111-9. · 7.04 Impact Factor
-
[show abstract]
[hide abstract]
ABSTRACT: Retrospective case-control studies are more susceptible to selection bias than other epidemiologic studies as by design they require that both cases and controls are representative of the same population. However, as cases and control recruitment processes are often different, it is not always obvious that the necessary exchangeability conditions hold. Selection bias typically arises when the selection criteria are associated with the risk factor under investigation. We develop a method which produces bias-adjusted estimates for the odds ratio. Our method hinges on 2 conditions. The first is that a variable that separates the risk factor from the selection criteria can be identified. This is termed the "bias breaking" variable. The second condition is that data can be found such that a bias-corrected estimate of the distribution of the bias breaking variable can be obtained. We show by means of a set of examples that such bias breaking variables are not uncommon in epidemiologic settings. We demonstrate using simulations that the estimates of the odds ratios produced by our method are consistently closer to the true odds ratio than standard odds ratio estimates using logistic regression. Further, by applying it to a case-control study, we show that our method can help to determine whether selection bias is present and thus confirm the validity of study conclusions when no evidence of selection bias can be found.
Biostatistics 06/2008; 10(1):17-31. · 2.14 Impact Factor
-
[show abstract]
[hide abstract]
ABSTRACT: Analytes in blood and urine may vary over time according to conditions of transport and storage.
UK Biobank pilot study to investigate stability through time of 42 haematological and clinical chemistry analytes in blood and four analytes in urine, kept in storage for up to 36 h, for 40 individuals. Random effects linear regressions were used to model the change through time in repeated assay results on a sample, allowing for heterogeneity between individuals and assay variability.
Assay results for most analytes tended to show a small negative bias (1-3% per 12 h stored) over time on average. Statistically significant (P < 0.05) heterogeneity in time trends between individuals, found for nearly all analytes, was dominated by differences in the baseline (time 0) assay results with the possible exception of Mean Corpuscular Haemoglobin Concentration (MCHC). Four out of 46 analytes (serum calcium, cholesterol, fibrinogen and HDL cholesterol) had a predicted probability of a negative time trend for a future individual >0.9. Results for freeze-thaw samples were not materially different from those for non-freeze-thaw samples, except that stability of the analyte results was only assessed up to 12 h.
The results suggest that any instability in assay results up to 36 h is likely to be small in comparison with between individual differences and assay error, and that a single assay measurement at any time between 0 and 36 h should give a representative value of the analyte concentration at time zero for that individual.
International Journal of Epidemiology 04/2008; 37 Suppl 1:i16-22. · 6.41 Impact Factor
