Inequity Measures for Evaluations of Environmental Justice: A Case Study of Close Proximity to Highways in NYC

Assessments of environmental and territorial justice are similar in that both assess whether empirical relations between the spatial arrangement of undesirable hazards (or desirable public goods and services) and socio-demographic groups are consistent with notions of social justice, evaluating the spatial distribution of benefits and burdens (outcome equity) and the process that produces observed differences (process equity. Using proximity to major highways in NYC as a case study, we review methodological issues pertinent to both fields and discuss choice and computation of exposure measures, but focus primarily on measures of inequity. We present inequity measures computed from the empirically estimated joint distribution of exposure and demographics and compare them to traditional measures such as linear regression, logistic regression and Theil’s entropy index. We find that measures computed from the full joint distribution provide more unified, transparent and intuitive operational definitions of inequity and show how the approach can be used to structure siting and decommissioning decisions.

Limitations of the Odds Ratio in Gauging the Performance of a Diagnostic or Prognostic Marker

A marker that is strongly associated with outcome (or disease) is often assumed to be effective for classifying individuals according to their current or future outcome. However, for this to be true, the associated odds ratio must be of a magnitude rarely seen in epidemiological studies. An illustration of the relationship between odds ratios and receiver operating characteristic (ROC) curves shows, for example, that a marker with an odds ratio as high as 3 is in fact a very poor classification tool. If a marker identifies 10 percent of controls as positive (false positives) and has an odds ratio of 3, then it will only correctly identify 25 percent of cases as positive (true positives). Moreover, the authors illustrate that a single measure of association such as an odds ratio does not meaningfully describe a marker’s ability to classify subjects. Appropriate statistical methods for assessing and reporting the classification power of a marker are described. The serious pitfalls of using more traditional methods based on parameters in logistic regression models are illustrated.

Causes of death and determinants of outcome in critically ill patients

INTRODUCTION: Whereas most studies focus on laboratory and clinical research, little is known about the causes of death and risk factors for death in critically ill patients. METHODS: Three thousand seven hundred patients admitted to an adult intensive care unit (ICU) were prospectively evaluated. Study endpoints were to evaluate causes of death and risk factors for death in the ICU, in the hospital after discharge from ICU, and within one year after ICU admission. Causes of death in the ICU were defined according to standard ICU practice, whereas deaths in the hospital and at one year were defined and grouped according to the ICD-10 (International Statistical Classification of Diseases and Related Health Problems) score. Stepwise logistic regression analyses were separately calculated to identify independent risk factors for death during the given time periods. RESULTS: Acute, refractory multiple organ dysfunction syndrome was the most frequent cause of death in the ICU (47%), and central nervous system failure (relative risk [RR] 16.07, 95% confidence interval [CI] 8.3 to 31.4, p < 0.001) and cardiovascular failure (RR 11.83, 95% CI 5.2 to 27.1, p < 0.001) were the two most important risk factors for death in the ICU. Malignant tumour disease and exacerbation of chronic cardiovascular disease were the most frequent causes of death in the hospital (31.3% and 19.4%, respectively) and at one year (33.2% and 16.1%, respectively). CONCLUSION: In this primarily surgical critically ill patient population, acute or chronic multiple organ dysfunction syndrome prevailed over single-organ failure or unexpected cardiac arrest as a cause of death in the ICU. Malignant tumour disease and chronic cardiovascular disease were the most important causes of death after ICU discharge.

Semiparametric Estimation in General Repeated Measures Problems

This paper considers a wide class of semiparametric problems with a parametric part for some covariate effects and repeated evaluations of a nonparametric function. Special cases in our approach include marginal models for longitudinal/clustered data, conditional logistic regression for matched case-control studies, multivariate measurement error models, generalized linear mixed models with a semiparametric component, and many others. We propose profile-kernel and backfitting estimation methods for these problems, derive their asymptotic distributions, and show that in likelihood problems the methods are semiparametric efficient. While generally not true, with our methods profiling and backfitting are asymptotically equivalent. We also consider pseudolikelihood methods where some nuisance parameters are estimated from a different algorithm. The proposed methods are evaluated using simulation studies and applied to the Kenya hemoglobin data.

Gauss-Seidel Estimation of Generalized Linear Mixed Models with Application to Poisson Modeling of Spatially Varying Disease Rates

Generalized linear mixed models (GLMMs) provide an elegant framework for the analysis of correlated data. Due to the non-closed form of the likelihood, GLMMs are often fit by computational procedures like penalized quasi-likelihood (PQL). Special cases of these models are generalized linear models (GLMs), which are often fit using algorithms like iterative weighted least squares (IWLS). High computational costs and memory space constraints often make it difficult to apply these iterative procedures to data sets with very large number of cases. This paper proposes a computationally efficient strategy based on the Gauss-Seidel algorithm that iteratively fits sub-models of the GLMM to subsetted versions of the data. Additional gains in efficiency are achieved for Poisson models, commonly used in disease mapping problems, because of their special collapsibility property which allows data reduction through summaries. Convergence of the proposed iterative procedure is guaranteed for canonical link functions. The strategy is applied to investigate the relationship between ischemic heart disease, socioeconomic status and age/gender category in New South Wales, Australia, based on outcome data consisting of approximately 33 million records. A simulation study demonstrates the algorithm's reliability in analyzing a data set with 12 million records for a (non-collapsible) logistic regression model.

A Diagnostic Test for the Mixing Distribution in a Generalised Linear Mixed Model

We introduce a diagnostic test for the mixing distribution in a generalised linear mixed model. The test is based on the difference between the marginal maximum likelihood and conditional maximum likelihood estimates of a subset of the fixed effects in the model. We derive the asymptotic variance of this difference, and propose a test statistic that has a limiting chi-square distribution under the null hypothesis that the mixing distribution is correctly specified. For the important special case of the logistic regression model with random intercepts, we evaluate via simulation the power of the test in finite samples under several alternative distributional forms for the mixing distribution. We illustrate the method by applying it to data from a clinical trial investigating the effects of hormonal contraceptives in women.

Source of funding and results of studies of health effects of mobile phone use: systematic review of experimental studies

OBJECTIVES: There is concern regarding the possible health effects of cellular telephone use. We examined whether the source of funding of studies of the effects of low-level radiofrequency radiation is associated with the results of studies. We conducted a systematic review of studies of controlled exposure to radiofrequency radiation with health-related outcomes (electroencephalogram, cognitive or cardiovascular function, hormone levels, symptoms, and subjective well-being). DATA SOURCES: We searched EMBASE, Medline, and a specialist database in February 2005 and scrutinized reference lists from relevant publications. DATA EXTRACTION: Data on the source of funding, study design, methodologic quality, and other study characteristics were extracted. The primary outcome was the reporting of at least one statistically significant association between the exposure and a health-related outcome. Data were analyzed using logistic regression models. DATA SYNTHESIS: Of 59 studies, 12 (20%) were funded exclusively by the telecommunications industry, 11 (19%) were funded by public agencies or charities, 14 (24%) had mixed funding (including industry), and in 22 (37%) the source of funding was not reported. Studies funded exclusively by industry reported the largest number of outcomes, but were least likely to report a statistically significant result: The odds ratio was 0.11 (95% confidence interval, 0.02-0.78), compared with studies funded by public agencies or charities. This finding was not materially altered in analyses adjusted for the number of outcomes reported, study quality, and other factors. CONCLUSIONS: The interpretation of results from studies of health effects of radiofrequency radiation should take sponsorship into account.

Endocrine symptom assessment in women with breast cancer: what a simple "yes" means

GOALS OF WORK: To investigate the self-reported symptoms related to endocrine therapy in women with early or advanced breast cancer and the impact of these symptoms on quality of life (QL) indicators. MATERIALS AND METHODS: Symptom occurrence was assessed by the Checklist for Patients on Endocrine Therapy (C-PET) and symptom intensity was assessed by linear analogue self-assessment (LASA) indicators. Patients also responded to global LASA indicators for physical well-being, mood, coping effort and treatment burden. Associations between symptoms and these indicators were analysed by linear regression models. MAIN RESULTS: Among 373 women, the distribution of symptom intensity showed considerable variation in patients reporting a symptom as present. Even though patients recorded a symptom as absent, some patients reported having experienced that symptom when responding to symptom intensity, as seen for decreased sex drive, tiredness and vaginal dryness. Six of 13 symptoms and lower age had a detrimental impact on the global indicators, particularly tiredness and irritability. CONCLUSIONS: Patients' experience of endocrine symptoms needs to be considered both in patient care and research, when interpreting the association between symptoms and QL.

Testing the structural and cross-cultural validity of the KIDSCREEN-27 quality of life questionnaire

OBJECTIVES: The aim of this study is to assess the structural and cross-cultural validity of the KIDSCREEN-27 questionnaire. METHODS: The 27-item version of the KIDSCREEN instrument was derived from a longer 52-item version and was administered to young people aged 8-18 years in 13 European countries in a cross-sectional survey. Structural and cross-cultural validity were tested using multitrait multi-item analysis, exploratory and confirmatory factor analysis, and Rasch analyses. Zumbo's logistic regression method was applied to assess differential item functioning (DIF) across countries. Reliability was assessed using Cronbach's alpha. RESULTS: Responses were obtained from n = 22,827 respondents (response rate 68.9%). For the combined sample from all countries, exploratory factor analysis with procrustean rotations revealed a five-factor structure which explained 56.9% of the variance. Confirmatory factor analysis indicated an acceptable model fit (RMSEA = 0.068, CFI = 0.960). The unidimensionality of all dimensions was confirmed (INFIT: 0.81-1.15). Differential item functioning (DIF) results across the 13 countries showed that 5 items presented uniform DIF whereas 10 displayed non-uniform DIF. Reliability was acceptable (Cronbach's alpha = 0.78-0.84 for individual dimensions). CONCLUSIONS: There was substantial evidence for the cross-cultural equivalence of the KIDSCREEN-27 across the countries studied and the factor structure was highly replicable in individual countries. Further research is needed to correct scores based on DIF results. The KIDSCREEN-27 is a new short and promising tool for use in clinical and epidemiological studies.

Validation of candidate smear microscopy quality indicators, extracted from tuberculosis laboratory registers

SETTING: Kinshasa Province, Democratic Republic of Congo. OBJECTIVE: To identify and validate register-based indicators of acid-fast bacilli (AFB) microscopy quality. DESIGN: Selection of laboratories based on reliability and variation in routine smear rechecking results. Calculation of relative sensitivity (RS) compared to recheckers and its correlation coefficient (R) with candidate indicators based on a fully probabilistic analysis incorporating vague prior information using WinBUGS. RESULTS: The proportion of positive follow-up smears correlated well (median R 0.81, 95% credibility interval [CI] 0.58-0.93), and the proportion of first smear-positive cases fairly (median R 0.70, 95% CI 0.38-0.89) with RS. The proportions of both positive suspect and low positive case smears showed poor correlations (median R 0.27 and -0.22, respectively, with ranges including zero). CONCLUSIONS: The proportion of positives in follow-up smears is the most promising indicator of AFB smear sensitivity, while the proportion of positive suspects may be more indicative of accessibility and suspect selection. Both can be obtained from simple reports, and should be used for internal and external monitoring and as guidance for supervision. As proportion of low positive suspect smears and consistency within case series are more difficult to interpret, they should be used only on-site by laboratory professionals. All indicators require more research to define their optimal range in various settings.

Admission glycaemia and outcome in patients with acute coronary syndrome

Some studies of patients with acute myocardial infarction have reported that hyperglycaemia at admission may be associated with a worse outcome. This study sought to evaluate the association of blood glucose at admission with the outcome of unselected patients with acute coronary syndrome (ACS). Using the Acute Myocardial Infarction and unstable angina in Switzerland (AMIS Plus) registry, ACS patients were stratified according to their blood glucose on admission: group 1: 2.80-6.99 mmol/L, group 2: 7.00-11.09 mmol/L and group 3: > 11.10 mmol/L. Odds ratios for in-hospital mortality were calculated using logistic regression models. Of 2,786 patients, 73% were male and 21% were known to have diabetes. In-hospital mortality increased from 3% in group 1 to 7% in group 2 and to 15% in group 3. Higher glucose levels were associated with larger enzymatic infarct sizes (p<0.001) and had a weak negative correlation with angiographic or echographic left ventricular ejection fraction. High admission glycaemia in ACS patients remains a significant independent predictor of in-hospital mortality (adjusted OR 1.08; 95% confidence intervals [CI] 1.05-1.14, p<0.001) per mmol/L. The OR for in-hospital mortality was 1.04 (95% CI 0.99-1.1; p=0.140) per mmol/L for patients with diabetes but 1.21 (95% CI 112-1.30; p<0.001) per mmol/L for non-diabetic patients. In conclusion, elevated glucose level in ACS patients on admission is a significant independent predictor of in-hospital mortality and is even more important for patients who do not have known diabetes.

Higher cord blood levels of mannose-binding lectin-associated serine protease-2 in infants with necrotising enterocolitis

Necrotising enterocolitis (NEC) causes significant morbidity and mortality in premature infants. The role of innate immunity in the pathogenesis of NEC remains unclear. Mannose-binding lectin (MBL) recognizes microorganisms and activates the complement system via MBL-associated serine protease-2 (MASP-2). The aim of this study was to investigate whether MBL and MASP-2 are associated with NEC. This observational case-control study included 32 infants with radiologically confirmed NEC and 64 controls. MBL and MASP-2 were measured in cord blood using ELISA. Multivariate logistic regression was performed. Of the 32 NEC cases (median gestational age, 30.5 wk), 13 (41%) were operated and 5 (16%) died. MASP-2 cord blood concentration ranged from undetectable (<10 ng/mL) to 277 ng/mL. Eighteen of 32 (56%) NEC cases had higher MASP-2 levels (> or =30 ng/mL) compared with 22 of 64 (34%) controls (univariate OR 2.46; 95% CI 1.03-5.85; p = 0.043). Higher cord blood MASP-2 levels were significantly associated with an increased risk of NEC in multivariate analysis (OR 3.00; 95% CI 1.17-7.93; p = 0.027). MBL levels were not associated with NEC (p = 0.64). In conclusion, infants later developing NEC had significantly higher MASP-2 cord blood levels compared with controls. Higher MASP-2 may favor complement-mediated inflammation and could thereby predispose to NEC.

RISK OF SHUNT-DEPENDENT HYDROCEPHALUS AFTER OCCLUSION OF RUPTURED INTRACRANIAL ANEURYSMS BY SURGICAL CLIPPING OR ENDOVASCULAR COILING: A SINGLE-INSTITUTION SERIES AND META-ANALYSIS

OBJECTIVE: To compare the risk of shunt-dependent hydrocephalus after treatment of ruptured intracranial aneurysms by clipping versus coiling. METHODS: We analyzed 596 patients prospectively added to our database from July of 1999 to November of 2005 concerning the risk of shunt dependency after clipping versus coiling. Factors analyzed included age; sex; Hunt and Hess grade; Fisher grade; acute hydrocephalus; intraventricular hemorrhage; angiographic vasospasm; and number, size, and location of aneurysms. In addition, a meta-analysis of available data from the literature was performed identifying four studies with quantitative data on the frequency of clip, coil, and shunt dependency. RESULTS: The institutional series revealed Hunt and Hess grade, Fisher grade, acute hydrocephalus, intraventricular hemorrhage, and angiographic vasospasm as significant (P < 0.05) risk factors for shunt dependency after a univariate analysis. In a multivariate logistic regression analysis, we isolated intraventricular hemorrhage, acute hydrocephalus, and angiographic vasospasm as independent, significant risk factors for shunt dependency. The meta-analysis, including the current data, revealed a significantly higher risk for shunt dependency after coiling than after clipping (P = 0.01). CONCLUSION: Clipping of a ruptured aneurysm may be associated with a lower risk for developing shunt dependency, possibly by clot removal. This might influence long-term outcome and surgical decision making.

Towards greater accuracy in individual-tree mortality regression

Background mortality is an essential component of any forest growth and yield model. Forecasts of mortality contribute largely to the variability and accuracy of model predictions at the tree, stand and forest level. In the present study, I implement and evaluate state-of-the-art techniques to increase the accuracy of individual tree mortality models, similar to those used in many of the current variants of the Forest Vegetation Simulator, using data from North Idaho and Montana. The first technique addresses methods to correct for bias induced by measurement error typically present in competition variables. The second implements survival regression and evaluates its performance against the traditional logistic regression approach. I selected the regression calibration (RC) algorithm as a good candidate for addressing the measurement error problem. Two logistic regression models for each species were fitted, one ignoring the measurement error, which is the “naïve” approach, and the other applying RC. The models fitted with RC outperformed the naïve models in terms of discrimination when the competition variable was found to be statistically significant. The effect of RC was more obvious where measurement error variance was large and for more shade-intolerant species. The process of model fitting and variable selection revealed that past emphasis on DBH as a predictor variable for mortality, while producing models with strong metrics of fit, may make models less generalizable. The evaluation of the error variance estimator developed by Stage and Wykoff (1998), and core to the implementation of RC, in different spatial patterns and diameter distributions, revealed that the Stage and Wykoff estimate notably overestimated the true variance in all simulated stands, but those that are clustered. Results show a systematic bias even when all the assumptions made by the authors are guaranteed. I argue that this is the result of the Poisson-based estimate ignoring the overlapping area of potential plots around a tree. Effects, especially in the application phase, of the variance estimate justify suggested future efforts of improving the accuracy of the variance estimate. The second technique implemented and evaluated is a survival regression model that accounts for the time dependent nature of variables, such as diameter and competition variables, and the interval-censored nature of data collected from remeasured plots. The performance of the model is compared with the traditional logistic regression model as a tool to predict individual tree mortality. Validation of both approaches shows that the survival regression approach discriminates better between dead and alive trees for all species. In conclusion, I showed that the proposed techniques do increase the accuracy of individual tree mortality models, and are a promising first step towards the next generation of background mortality models. I have also identified the next steps to undertake in order to advance mortality models further.

Opinions on registering trial details: a survey of academic researchers

BACKGROUND: The World Health Organization (WHO) has established a set of items related to study design and administrative information that should build the minimum set of data in a study register. A more comprehensive data set for registration is currently developed by the Ottawa Group. Since nothing is known about the attitudes of academic researchers towards prospective study registration, we surveyed academic researchers about their opinion regarding the registration of study details proposed by the WHO and the Ottawa Group. METHODS: This was a web-based survey of academic researchers currently running an investigator-initiated clinical study which is registered with clinicaltrials.gov. In July 2006 we contacted 1299 principal investigators of clinical studies by e-mail explaining the purpose of the survey and a link to access a 52-item questionnaire based on the proposed minimum data set by the Ottawa Group. Two reminder e-mails were sent each two weeks apart. Association between willingness to disclose study details and study phase was assessed using the chi-squared test for trend. To explore the potential influence of non-response bias we used logistic regression to assess associations between factors associated with non-response and the willingness to register study details. RESULTS: Overall response was low as only 282/1299 (22%) principal investigators participated in the survey. Disclosing study documents, in particular the study protocol and financial agreements, was found to be most problematic with only 31% of respondents willing to disclose these publicly. Consequently, only 34/282 (12%) agreed to disclose all details proposed by the Ottawa Group. Logistic regression indicated no association between characteristics of non-responders and willingness to disclose details. CONCLUSION: Principal investigators of non-industry sponsored studies are reluctant to disclose all data items proposed by the Ottawa Group. Disclosing the study protocol and financial agreements was found to be most problematic. Future discussions on trial registration should not only focus on industry but also on academic researchers.